Patient-centered Care and Treatment in HIV Infection

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

In this issue of EclinicalMedicine, Diaco and colleagues describe a retrospective, secondary analysis of routinely collected data of 103 patients followed by two infectious diseases specialists in the HIV out-patient clinic of the University Department of Medicine and Infectious Diseases Service, Kantonsspital Baselland, Basel, Switzerland. They found that 70 patients were de-escalated to either dolutegravir/emtricitabine or dolutegravir/lamivudine by stopping tenofovir disoproxil fumarate or abacavir [1]. Viremia remained undetectable in all de-escalated patients at least 48 weeks later. These results confirm those of eight previous, heterogeneous studies in experienced patients on dolutegravir plus lamivudine [reviewed in [2]], which showed a low risk of virologic failure (6 of 727 patients, without emergence of specific resistance mutations) and a high tolerability (only 18 patients interrupted the treatment due to side effects). Although Diaco and colleagues included emtricitabine much more often than lamivudine in their regimen, they acknowledge that the two drugs are interchangeable; lamivudine is available as a low cost generic and a fixed-dose lamivudine-dolutegravir combination is more likely to be available in the future. Diaco's and colleagues' study is yet another example that the once quite neglected issue of patient-centered medicine in HIV care is finally becoming important and that tailoring of therapy and individual patients' preferences must be taken into strong account. Indeed following patients' preferences is central to the important issue of adherence to treatment, so vital in HIV therapy. The aging HIV-infected population (in 2020, around 50% of infected individuals will be 50 years of age or above in high income countries) [3] suffers from a number of non-communicable diseases, and HIV-infected people are not immune from the epidemic of non-communicable diseases that affects sub-Saharan Africa [4]. Comorbidities imply the use of several drugs with potential interactions and therefore deintensification, i.e. reduction of number of anti-HIV drugs, is an option that will be increasingly adopted in the future and is particularly attractive in those patients who have achieved good control of HIV replication following the initial use of a three-drug regimen. The dual dolutegravir plus lamivudine or emtricitabine regimen combines a drug with high antiviral potency, long half-life, high genetic barrier to resistance and tolerability (dolutegravir) with one with a good tolerability (lamivudine or emtricitabine). In addition, interactions with other drugs are limited, and dual therapy with dolutegravir and lamivudine does not appear to increase genital shedding of HIV-1 [5], hence should not increase the risk of sexual transmission of the virus. The concern regarding the reduced “forgiveness” (i.e., the possibility that viral replication increases) should patients be not fully adherent was highly mitigated in Diaco's patients by the fact that they were followed for many years by the same physician, who had established a very good relationship with them and whom they trusted. This is yet again another extremely important aspect of patient-centered care in HIV infection; patients need to be familiar with their physicians to be able, among other aspects, to openly let them know whether they adhere to treatment, and how much so. We have demonstrated the value of this aspect of care in previous publications [6], [7]. Dolutegravir monotherapy was not considered due to a high risk of virological failure but Diaco and colleagues think that it “may still emerge as a valid de-escalation option in selected patients in the future” [1]. Indeed, in a recent report only three cases (4.8%) of virologic failure (all prior to 24 weeks of treatment) were observed in a median follow-up time of eighteen months in 63 patients on dolutegravir monotherapy, [8] and we have shown that this option could be pursued in highly adherent patients with a zenith HIV-RNA below 100,000 copies/ml [7]. In conclusion, the study from Diaco and colleagues confirms once again the virologic safety and feasibility of a dual regimen including dolutegravir. The results of the ongoing TANGO study (Switch study to evaluate Dolutegravir plus Lamivudine in virologically suppressed human immunodeficiency virus type 1 positive adults), a phase III, randomized, open-label, active-controlled, multicenter, parallel-group study that started enrolling patients in 2018 [9] will obviously be more relevant and should finally show whether indeed this regimen can become the preferred option for deintensification therapy in non-hepatitis B virus coinfected patients. Additional investigations are also needed to identify the factors predicting success of other deintensification strategies and further improve patient-centered anti-HIV therapy. Author Contributions Massimiliano Lanzafame wrote the Commentary and Sandro Vento modified/amended it. Declaration of Interests No conflicts for either author.

Related collections

Most cited references 7

Record: found
Abstract: found
Article: found

Is Open Access

Tackling Africa's chronic disease burden: from the local to the global

Ama de-Graft Aikins, Nigel Unwin, Charles Agyemang … (2010)

Africa faces a double burden of infectious and chronic diseases. While infectious diseases still account for at least 69% of deaths on the continent, age specific mortality rates from chronic diseases as a whole are actually higher in sub Saharan Africa than in virtually all other regions of the world, in both men and women. Over the next ten years the continent is projected to experience the largest increase in death rates from cardiovascular disease, cancer, respiratory disease and diabetes. African health systems are weak and national investments in healthcare training and service delivery continue to prioritise infectious and parasitic diseases. There is a strong consensus that Africa faces significant challenges in chronic disease research, practice and policy. This editorial reviews eight original papers submitted to a Globalization and Health special issue themed: "Africa's chronic disease burden: local and global perspectives". The papers offer new empirical evidence and comprehensive reviews on diabetes in Tanzania, sickle cell disease in Nigeria, chronic mental illness in rural Ghana, HIV/AIDS care-giving among children in Kenya and chronic disease interventions in Ghana and Cameroon. Regional and international reviews are offered on cardiovascular risk in Africa, comorbidity between infectious and chronic diseases and cardiovascular disease, diabetes and established risk factors among populations of sub-Saharan African descent in Europe. We discuss insights from these papers within the contexts of medical, psychological, community and policy dimensions of chronic disease. There is an urgent need for primary and secondary interventions and for African health policymakers and governments to prioritise the development and implementation of chronic disease policies. Two gaps need critical attention. The first gap concerns the need for multidisciplinary models of research to properly inform the design of interventions. The second gap concerns understanding the processes and political economies of policy making in sub Saharan Africa. The economic impact of chronic diseases for families, health systems and governments and the relationships between national policy making and international economic and political pressures have a huge impact on the risk of chronic diseases and the ability of countries to respond to them.

0 comments Cited 117 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: found

Is Open Access

HIV Glasgow 2018, 28–31 October 2018, Glasgow, UK

(2018)

KL1 HIV/AIDS, global health and the Sustainable Development Goals K De Cock CDC Country Office, US Centers for Disease Control and Prevention, Nairobi, Kenya Sustainable Goal (SDG) 3 calls for an end to the epidemics of AIDS, tuberculosis, malaria and neglected tropical diseases by 2030, and the concomitant UNAIDS Fast‐Track Strategy aims to reduce new HIV infections to no more than 500,000 annually by 2020 and 200,000 by 2030. Central to the global effort is the UNAIDS 90‐90‐90 initiative which requires 90% of persons with HIV to be diagnosed, 90% of those to receive ART and 90% of the treated to be virally suppressed. There is controversy around how “the end of AIDS” is defined, about whether this ambitious goal is achievable and whether AIDS exceptionalism is still appropriate. UNAIDS has targeted 30 million people to be on ART by 2020, when fiscal requirements are expected to be 26 billion US dollars annually; current expenditure is about 7 billion US dollars less. This presentation will review progress in the AIDS response in the overall context of current global health. It honours Jacqueline Van Tongeren and Joep Lange and their work, and is dedicated to their memory. KL2 Strategies to reduce HIV incidence in Europe A Pharris European Centre for Disease Prevention and Control (ECDC), Stockholm, Sweden HIV incidence is increasing in the European region as a whole, although there are large epidemiological differences between Western, Central and Eastern Europe. Whilst overall 80% of people in the European region have been diagnosed with HIV, this varies greatly across sub‐regions with 86%, 83% and 76% of people diagnosed in Western, Central and Eastern Europe respectively. Among those diagnosed, 64% are estimated to be on treatment and this, too, differs across the region with 90%, 73% and 46% of those diagnosed on treatment in Western, Central and Eastern sub‐regions, respectively. Among those on treatment in the European region, 85% are virally suppressed with variations across sub‐regions in Europe (92%, 78% and 74% in Western, Central and Eastern). Within sub‐regions and among key populations within countries there is considerable diversity in diagnosis, proportion on treatment and viral suppression rates. While some countries within the region have been successful in meeting and surpassing the 90‐90‐90 targets, others are facing enormous challenges and are lagging behind. While the tools to prevent HIV – including diversified testing strategies, treatment as prevention, PrEP and harm reduction – have multiplied in recent years, their application across Europe is uneven and, in most settings, far lower than needed to impact incidence. Differences in epidemiology of HIV and health systems across Europe necessitate context‐specific strategies to strengthen and control HIV prevention and care efforts. KL3 PrEP: what's happening in Europe and the world in general S McCormack MRC Clinical Trials Unit, University College London, London, UK Within and beyond Europe, PrEP is undoubtedly contributing to the decline in new diagnoses reported in gay and other MSM, but the public health benefit is difficult to assess precisely and the impressive decline seen in some city clinics is not universal. San Francisco, central London and New South Wales have seen the largest gains. In all these settings testing and treatment were already at scale when PrEP was introduced. The contribution of PrEP to the toolkit is most accurately captured in New South Wales where they observed a 35% reduction in state‐wide new HIV diagnoses in MSM following rapid scale‐up of PrEP in the EPIC trial, two seroconversions amongst 3927 years of follow‐up amongst trial participants [1]. TDF/FTC PrEP is extremely effective biologically, but it is costly and needs to be delivered as part of a comprehensive package of interventions to reduce the risk of sexually transmitted infections including HIV – a package that is not available to everyone in Europe or globally in spite of the current burden of sexually transmitted infections. Introducing PrEP is therefore an opportunity to strengthen prevention services, and one of the most cost‐efficient methods is to employ key populations to deliver services when and where convenient to eligible peers (AIDS 2018). Adherence remains the Achilles heel for PrEP, and the products in the pipeline may go some way to addressing this: vaginal rings, long‐acting injectables and implants. However, first and foremost is the need to empower key populations with the information they need to understand their risk of HIV/STIs and how to reduce this during the various phases of their sexual lifetime. Abstract KL3 – Figure 1. Status of formal PrEP implement in Europe. Reference [1] Grulich et al. Rapid reduction in HIV diagnoses after targeted PrEP implementation in NSW, Australia. CROI 2018; Abs 88. Oral Abstracts O11 – Living Well with HIV: Ongoing Challenges O111 Retention and re‐engagement in care: a combination approach again required F Burns Centre for Sexual Health and HIV Research, and Royal Free Hospital, London, UK Effective ART remains the cornerstone of successful HIV management, with life expectancy in those successfully treated similar to that of the general population. ART is also an effective means of reducing population HIV transmission with the goal of zero new infections. However, suboptimal engagement in HIV care threatens to derail this success and is associated with serious consequences for both individual and public health. Engagement in care for any individual is dynamic and disengagement may happen at any time. Indeed, in the UK as many as one in four HIV clinic appointments are missed. While ‘living well with HIV’ is the current mantra, it is still denied to many. The population groups most at risk of disengagement are invariably those most marginalised and with the least advocacy. They include people who struggle with HIV‐related stigma, those with insecure residency and/or employment and people living with mental health, alcohol and drug dependency issues – problems that may increase in the current political and economic environment. Sustainable engagement will require a combination of biomedical, behavioural and structural strategies that recognise and address individual level factors and create a more enabling environment for health over the life course. Community participation and partnership in this process will be vital, with peer and social support services playing a key role. To tailor and target interventions appropriately, mechanisms are needed to predict those at risk of subsequent disengagement as well to respond once this occurs. Effective mechanisms for predicting and monitoring engagement are either limited or lacking, and the pool of evidence‐based interventions to improve engagement small. Future investment in research and services to tackle engagement is required to ensure the health inequalities we see across our cohorts reduced. O112 HIV and aging: challenges and goals J Falutz Department of Medicine, McGill University, Montreal, Canada Currently, overall long‐term survival of treated PLWHIV world‐wide approaches that of the general population. An increasing minority will live as long as their seronegative peers. As a result, the average age of PLWHIV, currently in the mid‐50s in resource‐rich countries, has increased. The proportion of older PLWHIV who are long‐term survivors compared to those who seroconvert at an older age varies according to local factors. The salutary impact on survival has nevertheless been challenged by several developments. The increasing proportion of PLWHIV approaching a typical geriatric age range will significantly impact health care delivery; their clinical features are similar to that of the general population about 5 to 10 years older. In addition to the earlier occurrence of common age‐related conditions, with increased multimorbidity compared to controls, several common geriatric syndromes have also impacted this younger population. These often difficult‐to‐evaluate and ‐manage conditions may include: sarcopenia, impaired mobility and falls, sensory complaints (neuropathy, visual and auditory deficits), cognitive decline and, significantly, frailty. This latter condition, a state of increased vulnerability to biologic and environmental stressors, with reduced ability to maintain homeostasis, remains challenging to evaluate and operationalize. In the general population, a simple and reliable metric to diagnose frailty in the usual clinical setting remains elusive. This is compounded by the poorly understood biologic basis for frailty, distinct from its increased risk of concurrent disabilities and comorbidities. Research into common determinants of frailty between the geriatric population and PLWHIV related to immune‐senescence, chronic inflammation, epigenetics and mitochondriopathy provide clues to potential avenues for prevention and management. Frailty may be key to understanding the discordance between chronologic and biologic age. Concurrently, investigation of predictors of successful aging in PLWHIV is progressing. Insights into the concepts of both psychological and physical resilience in seronegatives may be an important bridge contributing not only to increased lifespan but also to improved health‐span for PLWHIV. O113 48‐week changes in biomarkers in subjects with high cardiovascular risk switching from ritonavir‐boosted protease inhibitors to dolutegravir: the NEAT022 study E Martinez 1, L Assoumou2, G Moyle3, L Waters4, M Johnson5, P Domingo6, J Fox7, H Stellbrink8, G Guaraldi9, M Masia10, M Gompels11, S de Wit12, E Florence13, S Esser14, F Raffi15, A Pozniak3, J Gatell16 1Infectious Diseases, Hospital Clínic & University of Barcelona, Barcelona, Spain. 2Institut Pierre Louis d’Épidémiologie et de Santé, INSERM, Sorbonne Universités, Paris, France. 3St Stephen's AIDS Trust, Chelsea and Westminster Hospital, London, UK. 4Mortimer Market Centre, Central and North West London NHS Foundation Trust, London, UK. 5Infectious Diseases, Royal Free Hospital, London, UK. 6Infectious Diseases Unit, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. 7Infectious Diseases, Guy's and St Thomas’ Hospital, London, UK. 8Infectious Diseases, Infektionsmedizinisches Centrum, Hamburg, Germany. 9Infectious Diseases, University of Modena and Reggio Emilia, Modena, Italy. 10Infectious Diseases Unit, Hospital de Elche, Elche, Spain. 11Infectious Diseases, Southmead Hospital, Bristol, UK. 12Infectious Diseases, Saint Pierre Hospital, Universite Libre de Bruxelles, Brussels, Belgium. 13Infectious Diseases, Institute of Tropical Medicine, Antwerp, Belgium. 14Infectious Diseases, Universitatsklinikum, Essen, Germany. 15Infectious Diseases, University Hospital of Nantes, Nantes, France. 16Medicine, University of Barcelona, Barcelona, Spain Background: Switching from ritonavir‐boosted protease inhibitors (PI/r) to dolutegravir (DTG) in subjects with a high cardiovascular risk resulted in a better lipid profile at 48 weeks than continuing PI/r. Whether this strategy may have an impact on biomarkers involved in the pathogenesis of cardiovascular disease in HIV‐infected subjects is unknown. Materials and methods: Within a pre‐planned sub‐study, we assessed 48‐week changes in several biomarkers including serum high‐sensitivity C reactive protein (hsCRP), interleukin‐6 (IL‐6), intercellular adhesion molecule‐1 (ICAM‐1), vascular adhesion molecule‐1 (VCAM‐1), selectin E and P, adiponectin, insulin, oxidised LDL, malondialdehyde, soluble CD14 (sCD14) and CD163 (sCD163), and cystatin C, and urine beta‐2 microglobulin. The median percent changes from baseline were compared with Mann‐Whitney test, and the association between the percent changes in biomarkers and lipid fractions or other variables of interest with Spearman correlation test. All p values were two‐sided with a significance level of 0.01 to account for the multiplicity of tests. Results: Of 415 randomised patients, 313 (147 DTG, 166 PI/r) remained on their allocated therapy for 48 weeks and had samples available. We observed significant decreases in sCD14 (−11%, p 1 year contributed 773 hospitalisations from 553 individuals. Seven percent, 18% and 10% of hospitalisations in MSM, MSW and women were AIDS‐related. Non‐Black MSW and women remained at higher risk of hospitalisation, but the association was weaker than that seen in the first year after diagnosis (Table 1). Lack of viral suppression, lower CD4, older age and earlier diagnosis date were also independently associated with hospitalisations. Abstract O114 – Table 1. Association between gender / sexual orientation & ethnicity and all‐cause hospitalisation rate in HIV‐positive individuals in the first year after diagnosis (analysis A) and >1 year after diagnosis (analysis B) A: Hospitalisation in first year after diagnosis B: Hospitalisations from one year after diagnosis N (PY) Ratea Unadjusted HR (95% CI) Adjusted HR (95% CI)b N (PY) Ratea Unadjusted RR (95% CI) Adjusted RR (95% CI)c MSM 655 (592) 6.1 1.0 1.0 2310 (15,013) 2.2 1.0 1.0 Black MSW 138 (103) 27.2 4.2 (2.5 to 6.8) 2.4 (1.5 to 4.0) 391 (2514) 3.7 1.7 (1.4 to 2.2) 1.1 (0.9 to 1.4) Other ethnicity MSW 169 (127) 30.6 4.7 (3.0 to 7.5) 3.2 (2.0 to 5.1) 431 (2314) 4.8 2.2 (1.8 to 2.8) 1.8 (1.4 to 2.2) Black women 245 (185) 25.4 3.8 (2.5 to 5.9) 2.4 (1.6 to 3.8) 752 (4621) 3.7 1.7 (1.4 to 2.0) 1.4 (1.1 to 1.7) Other ethnicity women 100 (81) 19.8 3.1 (1.7 to 5.6) 2.3 (1.3 to 4.2) 327 (2090) 3.3 1.5 (1.2 to 2.0) 1.4 (1.1 to 1.9) All p values 30). Poisson regression compared mortality between genders adjusting for demographics, traditional risk factors and HIV‐related parameters. Results: Data from 24,383 PLWHIV (19.2% females) were included (see Table 1). At cART initiation the mean number of non‐AIDS comorbidities in males (0.26) and females (0.25) were similar (p = 0.34). At last available follow‐up in 2016 the mean number of comorbidities had increased in both males (0.59) and females (0.59), p = 0.18. Mortality risk increased with number of comorbidities, from 6.83 deaths per 1000 person‐years in PLWHIV with zero comorbidities, to 13.8, 28.2, 65.6 and 139 per 1000 person‐years with 1, 2, 3, ≥4 comorbidities, respectively. Poisson regression confirmed the relationship between multimorbidity and mortality: risk ratio (RR) 2.66 (2.54 to 2.79) per additional comorbidity. Overall mortality risk, adjusted for the number of comorbidities, was significantly lower in women than men (RR 0.78 [0.67 to 0.91], p = 0.002). However, there was a significant interaction between gender, number of comorbidities and mortality (p 50 years). Average number of acute and chronic comorbidities was 0.3 (range 0 to 3) and 1 (range 0 to 6), respectively. Mean annual total healthcare costs including ART were 22,817€, excluding ART 7609€; mean inpatient costs were 1467€, mean outpatient costs were 1589€, mean medication costs excluding ART were 4196€, and mean ART costs were 14,232€. Estimated incremental annual costs were 4791€ for acute cardiovascular disease, 14,525€ for acute hepatitis C, 7366€ for acute renal disease, 3511€ for bone fractures due to osteoporosis, 16,023€ for chronic hepatitis C, 2581€ for diabetes mellitus type 2, ‐867€ for dyslipidaemia and ‐1678€ for being female (Table 1, Figure 2). Abstract O116 – Figure 1. Evaluation periods. CCI = Charlson comorbidity index. Abstract O116 – Figure 2. Incremental cost estimates of acute and chronic non‐HIV related comorbidities. Abstract O116 – Table 1. Prevalence and incremental cost estimates for acute and chronic non‐HIV related comorbidities included in the final model Comorbidities Prevalence Incremental cost (€) 95% CI of cost (€) (lower) 95% CI of cost (€) (upper) p value Intercept 4766 4460 5093 Gender, being female 17.4% ‐1,678 ‐1929 ‐1327 1 year of randomised treatment. Conclusions: In this meta‐analysis of 13 randomised clinical trials of PrEP in 15,678 participants, there was no significant difference in risk of Grade 3/4 clinical adverse events or SAEs between TDF/FTC (or TDF) and control (Table 1). Furthermore, there was no significant difference in risk of specific renal or bone adverse outcomes. The safety profile of TDF/FTC would support more widespread use of PrEP in populations with a lower risk of HIV infection. Abstract O143 – Table 1. Results of the meta‐analyses of overall risk difference between treatment and control study arms for each outcome of interest Outcome Events (PrEP) Total participants (PrEP) Events (control) Total participants (control) Risk difference (95% CI) Significance Grade 3/4 adverse events 1305 7504 1259 7502 0% (‐1% to 2%) p = 0.56 Serious adverse events 738 7843 795 7835 0% (‐1% to 1%) p = 0.74 Renal (creatinine elevations) 11 7620 5 7622 0% (0%‐0%) p = 0.38 Bone (fractures) 202 5588 184 5596 0% (0%‐0%) p = 0.69 Reference: [1] Hill A, Hughes SL, Gotham D, Pozniak AL. Tenofovir alafenamide versus tenofovir disoproxil fumarate: is there a true difference in efficacy and safety? J Virus Erad. 2018;4:72‐9. O144 Dual therapy with PI/r+3TC or PI/r+TDF shows non‐inferior HIV RNA suppression and lower rates of discontinuation for adverse events, versus triple therapy. Meta‐analysis of seven randomised trials in 1624 patients Z Liew 1, A Hill2, B Simmons1 1Faculty of Medicine, Imperial College London, London, UK. 2Department of Translational Medicine, University of Liverpool, Liverpool, UK Background: Using fewer nucleos(t)ide analogues could improve safety, increase adherence and lower treatment costs. Generic versions of lamivudine (3TC), tenofovir (TDF), atazanavir/r (ATVr), darunavir/r (DRV/r) and lopinavir/r (LPV/r) are becoming available worldwide. Several randomised trials have evaluated two drug combinations of a ritonavir‐boosted protease inhibitor (PI/r) in combination with 3TC or TDF in naive patients or those with HIV RNA suppression at baseline. Methods: A systematic search of PubMed, Embase, conference proceedings and trial registries was conducted to identify all randomised controlled trials comparing PI/r+3TC or PI/r+TDF dual therapy to triple therapy in treatment‐naïve and treatment‐experienced, suppressed patients. Using inverse‐variance weighting, pooled risk differences (RD) were calculated for virological suppression (FDA Snapshot), protocol‐defined virological failure, treatment‐emergent resistance and discontinuation due to adverse events. Virological suppression was assessed for non‐inferiority (FDA non‐inferiority margin delta=‐4%). Results: Seven studies were identified of three different ritonavir‐boosted PIs in 1624 patients, three treatment naïve (ANDES n = 145, GARDEL n = 306, Kalead n = 152) and four treatment experienced (ATLAS n = 236, DUAL n = 249, OLE n = 239, SALT n = 267). The pooled risk difference for viral suppression at 48 weeks of dual therapy compared to triple therapy was +2% (95% CI ‐2% to +6%) which met the FDA criteria for non‐inferiority. Results were consistent in treatment‐naïve and switching studies (p = 0.94). There were 5/822 patients on dual therapy with treatment‐emergent primary IAS NRTI drug resistance mutations, versus 5/802 on triple therapy (p = 0.98) (Table 1). Treatment discontinuation for adverse events was significantly lower for dual therapy at Week 48 (RD ‐2.6%, 95% CI ‐4.2 to ‐0.9%, p = 0.002). Conclusions: In this meta‐analysis of seven randomised trials in 1624 patients, rates of HIV RNA suppression b D/C+F/TDF (baseline — switch) N = 363 D/C/F/TAF (switch— Week 96)c N = 295 p‐valuea,b Patient‐years exposure 323 303 626 512 109 AEs, any grade 312 (86) 246 (73) 334 (92) ND 326 (90) 125 (42) ND Grade 3–4 AEs 20 (6) 29 (9) 45 (12) ND 33 (9) 15 (5) ND Serious AEs 17 (5) 24 (7) 39 (11) ND 36 (10) 8 (3) ND AE‐related discontinuations 8 (2) 2 (1) 10 (3) ND 17 (5) 1 (0.3) ND Median change in eGFR eGFRcyst, mL/min/1.73 m2 +4.0 ND +4.4 0.007 +1.6 0.0 0.130 eGFRcr, mL/min/1.73 m2 ‐5.5 ND ‐5.6 0.001 ‐8.0 +2.3 0.001 Median changes in renal biomarkers UPCR (mg/g) ‐15.7 ND ‐15.5 0.001 ‐10.5 ‐1.4 0.112 UACR (mg/g) ‐0.6 ND ‐0.7 0.001 ‐0.2 ‐0.5 0.001 RBP:Cr (µg/g) +6.9 ND +13.7 0.001 +35.1 ‐35.5 0.001 B2M:Cr (µg/g) ‐30.4 ND ‐27.0 0.001 +18.4 ‐40.5 0.001 Median change in fasting lipids TC (mg/dL) +28.6 ND +34.0 0.001 +10.4 +21.8 0.001 HDL‐C (mg/dL) +4.4 ND +5.0 0.001 +1.5 +1.9 0.001 LDL‐C (mg/dL) +17.4 ND +21.7 0.001 +5.0 +15.1 0.001 Triglycerides (mg/dL +24.4 ND +29.2 0.001 +14.2 +14.2 0.001 TC/HDL‐C ratio +0.20 ND +0.25 0.001 +0.08 +0.24 0.001 Change in BMD N = 113 N = 113 N = 99 N = 83 Lumbar spine Mean % change ‐0.7 ND ‐0.9 0.039 ‐2.6 +0.5 0.223 Increase by ≥3% 12% ND 16% ND 5% 18% ND Decrease by ≥3% 27% ND 34% ND 43% 10% ND Total hip Mean % change +0.1 ND ‐0.3 0.473 ‐2.8 +0.5 0.160 Increase by ≥3% 12% ND 17% ND 2% 18% ND Decrease by ≥3% 13% ND 23% ND 48% 11% ND Femoral neck Mean % change ‐0.3 ND ‐1.3 0.005 ‐3.1 +0.2 0.660 Increase by ≥3% 14% ND 11% ND 5% 21% ND Decrease by ≥3% 23% ND 30% ND 58% 17% ND awithin treatment arm comparisons for change at Week 96 from reference assessed by: Wilcoxon signed‐rank test (eGFR, renal biomarkers and fasting lipids) and paired t‐test (BMD). breference for the D/C/F/TAF arm is study baseline and for the control arm is the last value before the switch. crespectively 2.5%, 41.3%, 36.4% of patients randomised to the control arm switched to D/C/F/TAF at Week 60, Week 72 and Week 84. eGFRcyst = eGFR based on serum cystatin C (CKD‐EPI formula); eGFRcr = eGFR based on serum creatinine (CKD‐EPI formula); UPCR = urine protein: creatinine ratio; UACR = urine albumin: creatinine ratio; RBP:Cr = urine retinol binding protein: creatinine ratio; B2M:Cr = urine beta‐2‐microglobulin: creatinine ratio; TC = total cholesterol; HDL‐C = high density lipoprotein‐cholesterol; LDL‐C = low density lipoprotein‐cholesterol; BMD = bone mineral density; ND = not determined. Conclusions: High virological response and low failure rates were seen at Week 96 in both arms, with no development of resistance to darunavir or TAF. FoR D/C/F/TAF, safety findings at Week 96 were consistent with those at Week 48. Bone, renal and lipid safety were consistent with known TAF and cobicistat profiles. In the control arm, safety findings were consistent with those in the D/C/F/TAF arm. D/C/F/TAF combines the efficacy and high genetic barrier to resistance of darunavir with the safety benefits of TAF for ART‐naïve, HIV‐1‐infected patients. O213 Comparable viral decay with dolutegravir plus lamivudine versus dolutegravir‐based triple therapy J Gillman 1, P Janulis2, R Gulick3, C Wallis4, B Berzins2, R Bedimo5, K Smith6, M Aboud6, B Taiwo2 1Prism Health North Texas, Dallas, TX, USA. 2Division of Infectious Diseases, Northwestern University, Chicago, IL, USA. 3Division of Infectious Diseases, Weill Cornell Medicine, New York, NY, USA. 4BARC South Africa and Lancet Laboratories, Johannesburg, South Africa. 5Infectious Diseases Section, VA North Texas Health Care System, Dallas, TX, USA. 6ViiV Healthcare, Research Triangle Park, NC, USA Background: The GEMINI studies demonstrated non‐inferiority of dolutegravir (DTG) plus lamivudine (3TC) compared to DTG plus tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) in treatment‐naïve HIV‐1 infected individuals with pre‐treatment plasma HIV‐1 RNA (viral load, VL) 1000 to 500,000 copies/mL. However, rapidity of viral suppression with DTG plus 3TC, which may influence the risk of viral transmission or selection of resistant variants, has not been studied adequately, particularly at higher pre‐treatment VL. A substudy of the PADDLE trial showed comparable viral decay between DTG plus 3TC and DTG‐based three‐drug therapy in participants with pre‐treatment VL 100,000 copies/mL) and an interaction term of the drug therapy and baseline VL stratum. Results: The VL change from baseline with two‐drug therapy was non‐inferior to three‐drug therapy (p 100,000 copies/mL = 0.725; three‐drug – baseline VL 100,000 copies/mL = 0.596. Conclusions: Viral decay with the two‐drug regimen of DTG plus 3TC is comparable to the viral decay with DTG‐based triple therapy, even in individuals with higher pre‐treatment VL up to 500,000 copies/mL. O214 The impact of M184V/I mutation on the efficacy of abacavir/lamivudine/dolutegravir regimens prescribed in treatment‐experienced patients F Olearo 1, H Nguyen2, F Bonnet3, G Wandeler4, M Stoeckle5, V Bättig5, M Cavassini6, A Scherrer7, P Schmid8, H Bucher9, H Günthard7, J Böni10, S Yerly11, A D'Armino Monforte12, M Zazzi13, P Bellerive14, B Rijnders15, P Reiss16, F Wit16, R Kouyos17, A Calmy18 1Infectious Diseases Department, University Hospitals of Geneva, Geneva, Switzerland. 2Epidemiology, University of Zurich, Zurich, Switzerland. 3Infectious Diseases Department, University of Bordeaux, Bordeaux, France. 4Infectious Diseases Department, University Hospital of Bern, Bern, Switzerland. 5Infectious Diseases Department, University Hospital of Basel, Basel, Switzerland. 6Infectious Diseases Department, University Hospital of Lausanne, Lausanne, Switzerland. 7Infectious Diseases Department, University Hospital of Zurich, Zurich, Switzerland. 8Infectious Diseases Department, University of Bern, Bern, Switzerland. 9Epidemiology, Basel University Hospital of Basel, Switzerland. 10Virology, Infectious Diseases Department, Zurich, Switzerland. 11Virology, University Hospital of Geneva, Geneva, Switzerland. 12Infectious Diseases Department, Azienda Ospedaliera‐Polo Universitario San Paolo, Milan, Italy. 13Molecular Biology, University of Siena, Siena, Italy. 14Virology, University Hospital of Bordeaux, Bordeaux, France. 15Infectious Diseases Department, University Hospital Erasmus MC, Rotterdam, Netherlands. 16Academic Medical Center, University of Amsterdam, Amsterdam, Netherlands. 17Infectious Diseases Department and Epidemiology, University Hospital of Zurich, Zurich, Switzerland. 18Infectious Diseases Department, University Hospital of Geneva, Geneva, Switzerland Background: The impact of archived resistance mutation M184V/I on virological success remains unclear in treatment‐experienced patients switching to the fixed dose combination abacavir (ABC)/lamivudine (3TC)/dolutegravir (DTG). Considering the possible role of this mutation in impairing the efficacy of both ABC and 3TC, we aimed to determine its impact on the virological failure (VF) rate in patients with suppressed viraemia on cART switching to an ABC/3TC/DTG regimen. Materials and methods: This prospective study included treatment‐experienced adults from five European HIV cohorts (ARCA, Aquitaine, ATHENA, ICONA and SHCS) who switched to ABC/3TC/DTG between 2012 and 2016, with ≤50 copies/mL of HIV‐RNA at the time of switch and at least one pre‐existing genotypic resistance test from plasma (when drug‐naïve or during previous treatment failure). The primary outcome was the time to first VF (defined as two consecutive HIV‐RNA measurements >50 copies/mL or one HIV‐RNA measurement >50 copies/mL accompanied by a change in ART). We further considered a composite outcome considering the presence of VF or virological blips, defined as an isolated detectable HIV‐RNA >50 copies/mL followed by a return to virological suppression. A secondary outcome was discontinuation due to adverse events. Multivariate Cox proportional hazard models were used to quantify the effect of the M184V/I on outcomes. Results: One thousand six hundred and twenty‐six patients were included in the analysis (median follow‐up 289 days; IQR 154 to 441). Patients with archived M184V/I (n = 137) were older, more likely to have a history of previous injection drug use and with a longer duration of virological suppression before the switch. The VF rate was 15.1 per 1000 person‐years (95% CI 9.9 to 23.2) and higher in patients harbouring an archived M184V/I mutation, although not statistically significant (29.80 [11.17 to 79.39] vs. 13.56 [8.43 to 21.83]; p = 0.093) (Table 1). Figure 1 shows the estimated probability of being free from VF according to the presence of M184V/I mutations. In the multivariable model, M184I/V was not associated with VF (adjustment for VL zenith) or the composite endpoint (HR 3.03, CI 0.84 to 10.82; HR 2.22, CI 0.8 to 5.6, respectively). There were no differences in treatment discontinuation for reasons other than VF between patients with and without documented M184V/I (10.22% vs. 15.58%, respectively; p = 0.12). Abstract O214 – Figure 1. Estimated probability of being free from virological failure according to the presence of M184V/I mutations. Abstract O214 – Table 1. Virological primary outcomes Without M184V/I, N = 1489 (%) With M184 V/I, N = 137 (%) Virological failure (VF) 17 (1.14) 4 (3) a) 1st definition: 2x HIV‐RNA >50 copies/mL 10 (0.67) 1 (0.73) b) 2nd definition: 1x HIV‐RNA >50 copies/mL + ABC/3TC/DTG stop 7 (0.47) 3 (2.1) Treatment (ABC/3TC/DTG) discontinued for reasons other than VF 232 (15.58) 14 (10.22) Virological blipsa (VB) had at least one blip during ABC/3TC/DTG 63 (4.2) 12 (8.8) VB, median copies/mL (IQR) 79 (62 to 122) 72 (58 to 154) VF incidence (per 1000 person‐years) 13.56 (8.43 to 21.83) 29.8 (11.17 to 79.39) avirological blips: any viral load measurement >50 copies/mL, with viral load subsequently undetectable. Conclusions: The VF rate was very low among treatment‐experienced patients with or without an M184V/I archived mutation who switched to ABC/3TC/DTG and starting this regimen in these patients is safe and well tolerated. However, data over a longer period are needed to confirm the total absence of any impact of M184V/I on the risk of VF. O215 Approaches towards a cure for HIV S Fidler Department of GUM and HIV Medicine, Imperial College and Imperial College NHS Trust, London, UK Whilst ART has dramatically improved survival for PLWHIV, maintaining lifelong health through viral suppression requires sustainable global access to ART, lifelong daily adherence to medication, which is untenable for both the individual and implementers. An HIV cure or remission is therefore highly desirable. The accepted definition of an HIV cure is the goal of a significant period of maintained viral suppression off ART (post‐treatment viral control), maintaining zero risk of onward viral transmission as well as individual health. The key barrier to curing HIV is the persistence of virus within a pool of latently infected cells, the so‐called HIV reservoir. These cells appear to persist for the lifetime of the individual and despite years of suppressed viraemia, viral replication re‐emerges, quite rapidly on stopping ART for the vast majority of individuals. The research field of HIV cure has moved rapidly towards new innovations which are currently under trial in addition to ART to explore how to eliminate or significantly reduce the size of the measured HIV reservoir, with a view towards HIV remission. In this presentation I will explore the results of recent studies that are investigating the different approaches to HIV cure. I will present data from some of the UK studies that have recently completed and discuss the next planned studies and approaches with a discussion of what we have learnt along the way. O216 HIV cure and cancer immunotherapy: cross‐disciplinary research at its best S Deeks Department of Medicine, University of California, San Francisco, CA, USA Given the challenge of delivering complex, expensive and potentially harmful ART on a global level, there is intense interest in the development of short‐term, well‐tolerated regimens that will either fully eradicate all HIV (a “cure”) or durably prevent HIV replication in the absence of any therapy (a “remission”). Most experts agree that a remission will be easier to achieve than a complete cure. Enthusiasm for this approach is driven in part by recent advances in using novel immunotherapies to reduce and control cancer cells. Cancer and HIV persistence share a number of similarities. In each case, a rare population of cells with the capacity to cause harm becomes established in difficult‐to‐reach tissues. The local environment in cancer and perhaps HIV reshaped to prevent immune mechanisms from clearing the diseased cell. Specifically, a chronic inflammatory environment is often present, resulting in upregulation of a number of pathways which prevent effective immune responses. Therapies that target these immune pathways have either been very successful (in cancer) or now entering the clinic (in HIV disease). Recent observations in HIV‐infected adults with cancer suggest that these approaches are unlikely to cure HIV alone and might have an unacceptable safety profile. Novel approaches to enhance their efficacy and reduce risk are being developed and will be discussed. O22 – #Adolescent Lives Matter O221 Living with it: complications of long‐term HIV R Ferrand London School of Hygiene and Tropical Medicine, London, UK, and Harare, Zimbabwe The global scale‐up of ART has dramatically increased survival of people with HIV and turned the infection from an invariably fatal disease into a chronic condition. Hence, increasing numbers of children who would previously have died in childhood are now reaching adolescence and adulthood. However, despite ART, HIV‐infected children commonly experience chronic multisystem complications that result in considerable morbidity and increased mortality risk. These include cardiorespiratory, musculoskeletal, neurocognitive and skin disease. These complications are likely a consequence of HIV infection itself, a sequelae of HIV‐associated infections and/or HIV treatment. As coverage of ART increases, it is increasingly apparent that ART alone is insufficient to maintain health and quality of life of children living with HIV. Research to understand the pathogenesis of these complications and to develop therapeutic strategies is needed. Looking ahead, HIV programmes will need to focus not only on delivery of ART but on management of these complications to ensure that children reaching adulthood achieve optimum health outcomes. O222 Taking it: PrEP experiences among adolescent MSM S Hosek John Stroger Hospital of Cook County, Chicago, IL, USA While oral HIV pre‐exposure prophylaxis (PrEP) has demonstrated safety and efficacy across populations, uptake and persistent use have been suboptimal among young adult MSM. Furthermore, disparities in race and health care access are emerging that may inhibit the true potential of PrEP as an HIV prevention strategy for young MSM. Prescription rates among adolescent MSM under the age of majority fall even further behind adults, due in part to ethical and regulatory interpretations of consent laws as well as limited knowledge about PrEP and subsequent discomfort of prescribers. This presentation will review the current regulatory landscape for adolescent PrEP access as well as provide data from several recent studies of PrEP knowledge, uptake, adherence and persistence among adolescent and young adult MSM. O223 Staying with it: novel ways to increase adolescent adherence C Foster Imperial College NHS Trust, London, UK Is lifelong adherence to ART, or to any medication, potentially for 80+ years, really possible? Why is adherence poorer in adolescents when compared to younger children or older adults? What is the impact of adolescent cognitive development and of mental health on adherence? Are these issues specific to adolescents living with HIV and what can we learn from other chronic diseases? With these questions in mind, how do healthcare professionals, families, peers and the wider community best support adherence for adolescents living with HIV? The existing evidence of the impact of technology, peer mentors, disclosure, education, ART simplification, economic strengthening and cash transfers will be explored, highlighting examples of best practice, yet considerable data gaps persist. The needs of the research community to move from small pilot interventions to well‐powered randomised controlled trials for adolescents struggling with adherence to ART, frequently not a favoured population for research funding, is critical. O224 Dealing with it: mental health and stigma. Report from a study of YPLHIV in the Ukraine M Durteste1, G Kyselyova2, A Volokha2, A Judd3, C Thorne1, R Malyuta4, V Martsynovska5,6, N Nizova6, H Bailey 1, the Study of Young People Living with HIV in Ukraine 1UCL Great Ormond Street Institute of Child Health, University College London, London, UK. 2Shupyk National Medical Academy of Postgraduate Education, Kiev, Ukraine. 3MRC Clinical Trials Unit at UCL, Institute of Clinical Trials & Methodology, University College London, London, UK. 4Perinatal Prevention of AIDS Initiative, Odessa, Ukraine. 5Institute of Epidemiology and Infectious Diseases, The Public Health Center of the Ministry of Health of Ukraine, Kiev, Ukraine. 6The Public Health Center of the Ministry of Health of Ukraine, Ukraine. Background: Ukraine has the second largest European HIV epidemic. This study aimed to describe stigma, demographic and social factors and their association with anxiety among young people with perinatally‐acquired HIV (PHIV) or behaviourally‐acquired HIV (BHIV) in Kiev and Odessa. Methods: One hundred and four young people with PHIV and 100 with BHIV aged 13 to 25 years confidentially completed a tablet‐based survey. Tools included the Hospital Anxiety and Depression Scale (HADS) (score of 8 to 10 on anxiety sub‐scale indicating mild and ≥11 indicating moderate/severe symptoms in last seven days), Rosenberg Self‐Esteem Scale (RSES) and HIV Stigma Scale (HSS, short version). Unadjusted Poisson regression models were fitted to explore factors associated with moderate/severe anxiety symptoms. Results: PHIV and BHIV young people had median age 15.5 [IQR 13.9 to 17.1] years and 23.0 [21.0 to 24.3] years respectively and had registered for HIV care a median 12.3 [10.3 to 14.4] years and 0.9 [0.2 to 2.4] years previously; 97% (97/100) and 66% (65/99) respectively were on ART. Overall 30% reported mild and 13% (25/188) moderate/severe anxiety symptoms, with no difference by mode of HIV acquisition (p = 0.405) or sex (p = 0.700). Forty‐two percent (75/180) reported history of an emotional health problem for which they had not been referred/attended for care, or were unsure regarding care. Higher risk of moderate/severe anxiety symptoms was found with higher HIV‐related stigma (prevalence ratio [PR] 1.24, 95% CI 1.14 to 1.34 per HSS unit increase), lower self‐esteem (PR 0.83, 95% CI 0.78 to 0.90 per RSES point increase), less stable living situation (PR 3.13, 95% CI 1.31 to 7.47 for ≥2 vs. no home moves in last three years), CD4 ≤350 cells/mm3 (PR 2.29, 95% CI 1.06 to 4.97), having no‐one at home who knew the respondent's HIV status (PR 9.15, 95% CI 3.40 to 24.66 vs. all know) and, among BHIV, history of drug use (PR 4.64, 95% CI 1.83 to 11.85). Conclusions: Results indicated unmet need for support. Future work is needed to explore strategies for mental health support, particularly around disclosure, self‐esteem and stigma. O23 – Mental Health and HIV: What We All Need to Know O23 Mental health and HIV: what we all need to know C Orkin1, F Lampe2, C Izambert3, L Waters4 1Barts Health NHS Trust and Queen Mary University of London, London, UK; Chair, British HIV Association (BHIVA). 2Institute for Global Health, University College London, London, UK. 3AIDES, France. 4Mortimer Market Centre, University College London Hospitals, London, UK. People living with HIV experience a higher prevalence of mental health conditions than the general population. Mental health, like HIV, attracts unacceptable levels of stigma; mental ill health negatively impacts quality of life, morbidity & mortality and adherence to medication; neuropsychiatric symptoms, including insomnia, are common and can be associated with HIV medication; medications to treat anxiety, depression and other psychiatric conditions are associated with numerous drug‐drug interactions, requiring prescribers, and patients, to understand how to manage, and mitigate, the risk of related complications. This BHIVA‐led symposium will, with BHIVA and European speakers, describe the epidemiological, stigma and drug interaction challenges related to HIV and mental health and will summarise the management of people reporting insomnia with a view to equipping the community and clinicians to improve mental health care. O31 – HIV and Migration: a Renewed Challenge O311 HIV and migration: a renewed challenge J del Amo Valero National Center for Epidemiology, Institute of Health Carlos III, Madrid, Spain Migrants are heterogeneous and dynamic populations originating from a variety of countries. In Europe, migrants are exposed to multiple risk‐contexts for HIV infection, which are also related to migration drivers such as poverty and homophobia. This heterogeneity hampers the concept of “migrant” as a single category for analyses. The contribution of migrants to national epidemics varies globally but is the highest in Europe. In the European Union and Economic Area, over half of HIV reports in persons born in a different country to that of residence originate from Sub‐Saharan Africa (SSA), but HIV‐positive persons from Latin America and the Caribbean and from Western, Central and Eastern Europe account for large numbers. These different geographical origins are associated with different epidemiological characteristics, and thus require distinct interventions. The epidemiological patterns largely resemble those of the countries of origin; with a fundamentally heterosexually acquired epidemic in migrants from SSA, a very high proportion of MSM among cases from Latin America and the highest proportion of persons who inject drugs among HIV‐positive European migrants. Time trends are also different; for migrants from SSA sustained declines in new HIV reports have been observed from 2003 onwards whereas steady increases in HIV diagnoses in MSM from Latin America and the Caribbean have been reported. There is solid evidence that HIV acquisition among migrant MSM takes place largely after migration into European cities and accounts for a larger than previously thought proportion among heterosexual migrants from SSA. For most migrant groups, but particularly for undocumented migrants, difficulties to access HIV testing and health care are issues in many European countries and in some, undocumented migrants are not entitled to universal antiretroviral treatment. This talk will address how the dynamism and heterogeneity of migrant populations in Europe demands renewed answers for renewed challenges. O32 – Coinfections: TB and Viral Hepatitis O321 HIV‐associated tuberculosis: diagnosis, management and prevention G Meintjes University of Cape Town, Cape Town, South Africa Tuberculosis (TB) is the leading cause of death (40%), hospitalisation (18%) and in‐hospital death (25%) among HIV‐infected people globally. The World Health Organization estimates that in 2016, HIV‐associated tuberculosis caused 374,000 deaths. The diagnosis of tuberculosis in patients with advanced HIV can be challenging due to several factors: patients not producing sputum or sputum with low bacillary quantities, and extrapulmonary presentations. Recent advances that have improved diagnostic yield in HIV‐infected people include the Xpert MTB/RIF, Xpert MTB/RIF Ultra and the urine lipo‐arabinomannan (LAM) assay. New versions of the LAM assay promise higher sensitivity, but further field evaluation is required. Co‐treatment of HIV and tuberculosis is complicated by pill burden, drug interactions, shared drug toxicities and the tuberculosis‐associated immune reconstitution inflammatory syndrome (TB‐IRIS). A series of clinical trials published over the last decade have clarified the optimal time to initiate ART in patients being treated for tuberculosis; priority should be given to patients with CD4 3.25, liver decompensation, hepato‐carcinoma or liver‐related death) was calculated as number of events per person‐years of follow‐up (PYFU). A Poisson regression framework was used to estimate relative rates (RR [95% CI]) of the described endpoints according to current anti‐HBV treatment. A linear mixed model with random intercept and slope was used to compare ALT changes according to anti‐HBV treatments. Results: We included 624 patients. They were mostly male (501, 80.3%), born in Italy (468, 75%), had acquired HIV through heterosexual contact (248, 38.4%) and showed HCV‐ (122, 18.9%) or HDV‐positivity (98, 22.2%). Median age at ART initiation was 38 years (32 to 45); CD4 and HIV RNA were 373 cells/uL (190 to 591) and 4.58 log10 copies/mL (3.80 to 5.12). XTC‐containing regimens were less used in recent calendar periods (48.5% [2000 to 2003], 9.51% [2012 to 2015]) with the exception of the most recent period (18.64% [2016 to 2018]). Over 2203 PYFU we observed 511 ESLD events in treated patients; they were more frequent in subjects receiving XTC (32.7/100 PYFU), non‐anti‐HBV antiretrovirals (30.4/100 PYFU) and less common in those on TDF/XTC (17.1/100 PYFU) or TDF (10.5/100 PYFU). In the unadjusted model patients on TDF/XTC or TDF had lower risk of ESLD compared to those on XTC (RR 0.54 [0.43 to 0.66] and 0.33 [0.12 to 0.89]); in the fully adjusted model the RR was 0.14 for TDF/XTC [0.07 to 0.28, p 1000 copies/mL were randomised (1:1) to DTG or EFV‐400, both with tenofovir (TDF)/lamivudine (3TC). Randomisation was stratified by screening VL and by site. The primary endpoint was the proportion of patients with VL 100,000 copies/mL). In the ITT analysis snapshot, the proportion of patients with HIV RNA 100,000 copies at baseline, the proportion was 66.2% (137/207) and 61.5% (123/200), respectively (difference, 4.7%; 95% CI ‐4.6 to 14.0). In ITT analysis for VL 100,000 copies, suboptimal VL suppression at the threshold of 50 copies was observed, but with no statistical differences between the two arms. Particular attention has to be given to patients with persistent low level viraemia, and it is essential to ensure a long‐term follow‐up. O343 Efficacy of MK‐8591 against diverse HIV‐1 subtypes and NRTI‐resistant clinical isolates J Grobler, Q Huang, D Hazuda, M Lai Infectious Disease Biology, Merck Sharp & Dohme, Kenilworth, NJ, USA Background: MK‐8591 is a novel and potent nucleoside reverse transcriptase translocation inhibitor (NRTTI) currently in Phase II clinical development. Single oral doses of MK‐8591 as low as 0.5 mg have demonstrated HIV‐1 suppression in patients out to seven days. MK‐8591 inhibits HIV‐1 reverse transcriptase (RT) by multiple mechanisms and makes unique interactions with RT. Here, we present a comprehensive evaluation of MK‐8591 antiviral activity against a broad panel of clinical isolates harboring mutations that confer resistance to approved NRTIs as well as 11 different wild‐type HIV‐1 subtypes. Materials and methods: Antiviral activity of MK‐8591, tenofovir (TFV)‐alafenamide (TAF), zidovudine (AZT) and lamivudine (3TC) was evaluated in human PBMCs with wild‐type HIV‐1. The Monogram PhenoSense assay was employed to evaluate the susceptibility of 50 wild‐type (wt) isolates from 11 HIV‐1 subtypes and 94 NRTI‐resistant clinical isolates to MK‐8591, TFV, AZT and 3TC. Susceptibilities were determined using IC50s of wild‐type HIV in PBMCs and fold‐shifts in potencies compared to wild‐type virus. Results: Against wild‐type HIV‐1 in human PBMCs, MK‐8591 (IC50 = 0.2 nM) was over 10‐fold more potent than TAF, AZT and 3TC (IC50s of 3 nM, 10 nM and 144 nM, respectively). Against the common NRTI resistance mutations M184I and M184V, MK‐8591 exhibited IC50s of 0.8 nM (3.9 fold change [FC] vs. wt) and 1.1 nM (5.0 FC), respectively. The mutations K65R, L74V and Q151M displayed hypersusceptibility to inhibition by MK‐8591. Thymidine analog mutations and 69 ins mutations decreased susceptibility to MK‐8591 by less than 4‐fold and 10‐fold, respectively. The susceptibility of these mutants was further reduced 2‐fold with M184I/V. MK‐8591 IC50s against resistant isolates were lower than the IC50 of TAF versus wild‐type HIV. All 11 different HIV‐1 subtype viruses displayed similar susceptibility to MK‐8591. Conclusions: MK‐8591 is more potent against common NRTI resistant HIV‐1 isolates than any marketed NRTI is against wt HIV‐1. MK‐8591 should provide broad mutant and subtype coverage as a component of an HIV‐1 treatment regimen. O344A Week 48 safety and efficacy of the HIV‐1 attachment inhibitor prodrug fostemsavir in heavily treatment‐experienced participants (BRIGHTE study) J Aberg1, J Molina2, M Kozal3, P Cahn4, J Lalezari5, M Thompson6, R Diaz7, A Castagna8, G Pialoux9, M Gummel10, A Pierce11, P Ackerman 12, C Llamoso12 and M Lataillade12 1Infectious Diseases, Icahn School of Medicine at Mount Sinai, New York, NY, USA. 2Infectious Diseases, Hopital Saint‐Louis, APHP and University of Paris Diderot, Paris, France. 3Infectious Diseases, Yale University School of Medicine, New Haven, CT, USA. 4Infectious Diseases, Fundación Huesped, Buenos Aires, Argentina. 5Internal Medicine, Quest Research, New Haven, CT, USA. 6Infectious Diseases, AIDS Research Consortium of Atlanta, Atlanta, GA, USA. 7Infectious Diseases, Federal University of São Paulo, São Paulo, Brazil. 8Infectious Diseases, Ospedale San Raffaele, Milan, Italy. 9Infectious Diseases, Hopital Tenon, Paris, France. 10Statistics, GlaxoSmithKline, Upper Providence, PA, USA. 11Clinical Development, ViiV Healthcare, Research Triangle Park, NC, USA. 12Clinical Development, ViiV Healthcare, Branford, CT, USA Objectives: Fostemsavir (FTR) is an investigational first‐in‐class prodrug of the active moiety temsavir, which binds to HIV‐1 gp120 and prevents attachment to CD4 receptor on T cells, and other immune cells, thereby blocking virus infection. BRIGHTE is an ongoing Phase III study evaluating FTR in heavily treatment‐experienced (HTE) patients infected with multi‐drug resistant HIV‐1 who are unable to form a viable ARV regimen. Methods: HTE participants failing their current ARV regimen (confirmed HIV‐1 RNA >400 copies/mL) were assigned to the randomized cohort (RC) or non‐randomized cohort (NonRC), depending on if they had 1 to 2 or zero remaining currently approved ARV classes, respectively. Week 24 results, which included the primary efficacy endpoint (mean change in log10 HIV‐1 RNA from Day 1 at Day 8), were presented previously [1]. Week 48 results are presented here. Results: Week 48 rates of virologic suppression (HIV‐1 RNA pregnancy week 28, n (%) 11.5 7.2 8.0 Pregnancy week of delivery (mean), n 38.9 38.8 38.8 HIV transmission, n (%) 0 1 (0.5) 1 (0.4) Congenital malformations, n 1 6 7 Conclusions: We observed 274 pregnancies of HIV‐positive women, resulting in 281 children. Fifty‐two (19%) of the pregnant women received INSTI; 48 raltegravir and four dolutegravir. Despite the significant shorter in utero exposure to ART in the INSTI arm there were no differences between the two groups according to virological response, vertical HIV transmission and congenital malformations. P003 Successes and emerging challenges in prevention of vertical HIV transmission in the UK and Ireland H Peters, R Sconza, K Francis, A Horn and C Thorne University College London, Great Ormond Street Institute of Child Health, London, UK Background: In 2012–2014, the vertical HIV transmission rate (VTR) was 0.27% among diagnosed women living with HIV (WLHIV) in the UK/Ireland. The British HIV Association (BHIVA) currently recommends formula‐feeding infants born to WLHIV, eliminating postnatal transmission risk; however, BHIVA states that virologically suppressed treated women with good adherence to ART who choose/plan to breastfeed may be clinically supported to do so. Materials and methods: The National Study of HIV in Pregnancy and Childhood (NSHPC) conducts comprehensive surveillance of all pregnancies to diagnosed pregnant WLHIV in the UK/Ireland. HIV‐infected children 0.1). Conclusion: Our findings show that there was a three‐fold increase in proportion of pregnant women receiving ART prior to first ANC post‐intervention. This was accompanied with a drop in HIV‐positivity rates among HEIs, though not statistically significant. These findings suggest that efforts to increase uptake of antiretrovirals among HIV‐positive women of reproductive age will not only benefit individuals but may also contribute to reducing new paediatric HIV infections. P007 Elimination of pediatric HIV in New York City, 2007 to 2018: missed chances to prevent mother‐to‐child transmission (MTCT) in the era of U = U K Beckerman Obstetrics & Gynecology, BronxCare Health System, New Rochelle, NY, USA Background: More than 4000 children have been perinatally HIV‐infected in New York City. MTCT peaked in 1990 and fell steady to single digits by 2003. Since then, MTCT reached zero only once, in 2015, and incident pediatric HIV, although rare, persists. Material and methods We performed a retrospective review of hospital, local and New York State records to identify missed opportunities of transmission prevention among women and children in our region. In our hospital, in addition to universal maternal and newborn HIV antibody screening, all women registering for prenatal care have been asked about their partner's HIV status since 2015. Results: The Bronx continues to have the lowest life expectancy and highest HIV seroprevalence in New York State. While overall HIV incidence is declining, absolute numbers of PLWHIV increase every year. We met USPHS/CDC criteria for elimination of MTCT during only one year, 2015. We identified five missed opportunities of prevention of maternal and/or pediatric HIV infection in our community (Table 1). Abstract P007 – Table 1. Missed opportunities for prevention of HIV infection of women and newborns, Bronx, New York, 2007 to 2018 Year Maternal Hx Delivery Infant Del VL (copies/mL) Infant Dx 2007 28yo G3P2002 term labor, outside prenatal care, on cART, stated VL undetectable NSVD With IV ZDV ZDV to 6w 40,000 HIV‐1 PCR positive at 1 mo 2009 24yo G2P0 at 37w. s/p 1w AP admission for FUO at 34w, HIVAb neg. Expedited HIV not done. NSVD No IV ZDV No ZDV. HIVAbPos neonatal screen Unknown HIV‐1 PCR positive at 2 w 2013 31yo G5P2022 a 34w. Complete PNC HIVAb/Ag neg 1st tri & 32w. NSVD. No IV ZDV No ZDV. HIVAbPos late NICU screen Unknown HIV‐1 PCR positive at 2 w 2017 30yo G6P5004, late to PNC at outside clini. Denied HIV exposure. HIVAb/Ag neg, L&D expedited HIV neg. NSVD Normal neonatal screen Presumed undetectable P Jiroveci positive ET aspirate at 4 mo 2018 nPEP too late Still Pregnant In case #1, providers had to make delivery recommendations to their client, who presented in active labor, without access to prenatal and laboratory records. The other four cases involved acute maternal seroconversion during pregnancy and the peripartum period. In case #5, as yet undelivered, maternal seroconversion was documented during the first trimester while she was receiving nPEP. The other three occurred among women who had attended prenatal care and had both early and late negative HIV antibody screens. In retrospect, two occurred very late in pregnancy, and one during breast‐feeding. Of these three, two infected infants were identified by routine neonatal heel‐stick HIV screening. The most recently infected infant, as well as her infected parents, was identified by her endotracheal aspirate at 4 months of age when she was admitted to hospital in respiratory failure. Conclusions: Elimination of MTCT will require new strategies that go beyond universal HIV antibody screening in pregnant women and newborns. When women commonly deliver at hospitals other than where they receive prenatal and/or HIV care, inter‐institutional access to records will be essential. Other strategies must include aggressive maternal/infant antibody and viral testing following recognized HIV exposures (already done in some clinics but not codified in guidelines) as well as expanded, universal, on‐site, routine couples testing as a standard part of prenatal care. Only then will we be able to fully exploit the benefits of PrEP, nPEP and HIV treatment during pregnancy. P008 Efficiency of provider‐initiated HIV testing and counselling in Odessa Regional Hospital, Ukraine Y Lopatina 1, Y Kovalenko2, A Chuykov3, G Tyapkin1, S Esipenko4, Y Kvasnevska1, A Davies3 and A Zakowicz3 1AIDS Healthcare Foundation, Ukraine, Kiev, Ukraine. 2Odessa Regional Hospital, Odessa, Ukraine. 3AIDS Healthcare Foundation, Europe, Amsterdam, Netherlands. 4Odessa Regional AIDS Centre, Odessa, Ukraine Background: According to Ukrainian Public Health Center, HIV prevalence in Odessa Region has reached 860,1 per 100,000 population in 2017. Two thousand three hundred and thirty‐four new HIV cases have been identified in Odessa Region in 2017, 74% of them were cases of late HIV diagnosis. According to the international guidelines (including World Health Organization, HIV in Europe), health care providers should recommend HIV testing and counselling to all patients who present with HIV‐related disease. Description: The programme consisted of two phases. In the first phase (2016) we tested for HIV using AHF Rapid Testing Program model (RTP) all patients of Odessa Regional Hospital presenting with conditions listed in “HIV Indicator Conditions” tool by HIV in Europe. Rapid HIV tests and linkage to care support were provided by AHF. The departments in which HIV seropositivity rate was the highest were identified: pulmonology, intensive care and infectious disease unit. In the second phase HIV rapid testing and counselling were offered to patients with HIV indicator diseases in the departments chosen in the first phase. Lessons learned In 2017 a total of 699 HIV rapid tests were performed in Odessa Regional Hospital with 131 (19%) positive results in comparison to 2016 where total 2988 patients were tested with 94 (3%) positive results. In 2017 highest positivity rate was registered in the pulmonology department, 12 cases among 55 tested (22% positivity rate), in intensive care unit, 27 among 123 (22%) and in infectious diseases unit, 21 among 94 (22%). All cases were not registered previously. For the period 2016 to 2017 total 24 HIV‐positive patients passed away after several days of HIV diagnosis because of critical conditions. Eighty‐three percent of diagnosed were linked to care within 3 months. Conclusions: Provider‐initiated HIV testing for patients admitted to hospitals is an effective strategy to identify people who do not know their HIV status. Pulmonology, intensive care and infectious disease departments were the units with the highest rate of new HIV cases identified in Odessa Regional Hospital. Hospitals need to put special attention to the departments, where case detection is the highest. This intervention will contribute to closing the gap in new case detection in Ukraine. P009 Perinatal HIV‐1C transmitted drug resistance mutations in newly diagnosed antiretroviral‐naïve infants in Botswana M Mogwele 1, K Seatla1, S Gaseitsewe1, M Leteane2 and S Moyo1 1Research, Botswana Harvard AIDS Institute Partnership, Gaborone, Botswana. 2Biological Sciences, University of Botswana, Gaborone, Botswana Background: Low and middle income countries (LMICS) continue to lag behind in preventing mother‐to‐child transmission (PMTCT) of HIV‐1 to children; Botswana has a transmission prevalence of 1.2%. More often than not, these children end up being infected with a virus harbouring drug‐resistant mutations precluding them from most readily available liquid formulated ART. In Botswana, there is scarcity of data on pre‐treatment drug resistance mutations (PDR) in antiretroviral‐naïve HIV‐1 infected infants. We sought to determine the prevalence of HIV‐1 drug resistance mutations in 27 newly diagnosed HIV‐1 infected infants prior to ART. Methods: Stored HIV‐1C infected infant plasma samples of previously completed studies collected between 2002 and 2015 were analysed. In‐house protocols were used to extract total nucleic acids and a commercial genotyping assay was used to amplify protease and reverse transcriptase fragments. Sanger sequencing was done using big dye technology and drug resistance mutations were analysed using Stanford HIV drug resistance database and the International Antiviral Society‐USA (IAS‐USA) 2017 mutational list. Results: The overall prevalence of HIV‐1 drug resistance mutations in analysed was 18.5% (6/27). Where the predominant mutations were conferring resistance to NNRTI a drug which was used by most HIV‐1 infected mothers in Botswana. The frequencies of mutations analysed were: Y181C n = 1, Y181YC n = 1, M230ML n = 1, K103N n = 2, E138A n = 2, V179D n = 1, Y115F n = 1, L24LF n = 1, M46L n = 1 and I47R n = 1. One infant had multiple drug resistance mutations for NNRTI, protease inhibitor and the mutations were as follows: M46L, Y181C, M230L, M46L, I47I; while another infant had K103N and E138A mutations. Most infants with resistance mutations were resistant to efavirenz, etravirine and nevirapine. Conclusion: There is high prevalence of pre‐treatment drug resistance mutation in infants who become HIV‐1 infected despite the use of and this is a problem especially in consideration that ART formulation for infants is limited in cases of treatment failure. Therefore, transmission of perinatal HIV‐1 drug resistance can be prevented by strengthening the PMTCT framework. ARV‐BASED PREVENTION – PREP/TASP P010 Generic tenofovir disoproxil fumarate and emtricitabine tablets obtained from the internet: are they what they say they are? X Wang 1, W Nutland2, M Brady3, I Green3, M McClure1 and M Boffito4 1Imperial College London, London, UK. 2PrEPster, London, UK. 3Terrence Higgins Trust, London, UK. 4Chelsea & Westminster Hospital, London, UK Background: Pre‐exposure prophylaxis (PrEP) with tenofovir disoproxil fumarate (TDF)/emtricitabine (FTC) has been shown to reduce dramatically the risk of HIV acquisition. However, the National Health Service (NHS) in England has declined to routinely commission PrEP, leading high‐risk individuals to purchase generic versions on‐line. Our team provided therapeutic drug monitoring for tenofovir (TFV) and FTC in 2016 to 2017 for 293 individuals taking generic PrEP and reported that plasma concentrations were all above the target values. More recently concerns have arisen over the authenticity of generic PrEP purchased online. In the present study, we sampled generic PrEP from mainstream brands and suppliers and measured TDF and FTC in the purchased tablet form. Methods: Generic PrEP tablets in sealed bottles of different brands and from different suppliers were obtained from the internet through test purchases. Truvada from Gilead was purchased from Imperial College Healthcare NHS Trust. The brand and supplier of the generic PrEP samples were blinded from the researcher carrying out the analysis. The active pharmaceutical ingredient TDF and FTC was quantified using ultra‐performance liquid chromatography. Results: A total of nine samples were obtained for analysis, including Truvada from Gilead, PrEP used in the IMPACT trial and seven generic PrEP tablets obtained from Mylan, Cipla, Hetero Healthcare and Emcure. The suppliers were Dynamix International, In House Pharmacy, Green Cross Pharmacy and United Pharmacy. As summarised in the table, all the PrEP tablets contained 94.3% to 104.9% of the 300 mg of TDF claimed on the label and 97.3% to 104.4% of the 200 mg FTC claimed on the label. Abstract P010 – Table 1. Analysis of nine samples Tenofovir disproxil fumarate Emtricitabine Drug no. Drug name Manufacturer Lot no. Expiration date Supplier source Measured amount in mg % of label claim (300 mg) Measured amount in mg % of label claim (200 mg) 0 Truvada Gilead 5595307D Nov‐19 Imperial Healthcare NHS Trust 300.1 100.0 201.2 100.6 1 Emtricitabine/ tenofovir disoproxil Mylan 1091973 not available IMPACT trial 299.2 99.7 198.3 99.2 2 RICOVIR‐EM Mylan 3067417 Apr‐20 Dynamix International 307.4 102.5 201.6 100.8 3 TENVIR‐RM Cipla GG80516 Feb‐20 In House Pharmacy 314.8 104.9 208.4 104.2 4 TENOF‐EM Hetero Healthcare 31171625 Apr‐20 Dynamix International 306.4 102.1 205.2 102.6 5 TENOF‐EM Hetero Healthcare 31171625 Apr‐20 Green Cross Pharmacy 313.2 104.4 207.2 103.6 6 TENVIR‐EM Cipla GG80516 Feb‐20 Dynamix International 312.0 104.0 208.7 104.4 7 TAVIN‐EM Emcure E16HX18001 Dec‐20 United Pharmacy 283.0 94.3 194.7 97.3 8 TENVIR‐EM Cipla GG80114 Dec‐19 United Pharmacy 307.5 102.5 203.5 101.8 Conclusions: All the PrEP tablets sampled in this study contained the claimed amount of TDF and FTC. We were able to confirm the claimed content of the PrEP tablets from various manufacturers and suppliers. Further testing has been planned with additional PrEP tablets being sourced from different manufacturers and suppliers. This study provides reassurance to the community purchasing generic PrEP online and is a good example of a close collaboration between academics, clinicians, HIV charities and PrEP advocates. P011 Geographic barriers result in HIV pre‐exposure prophylaxis discontinuation: how to improve retention in care Z Greenwald1, K Card2, N Niaki, N Lachowsky2 and R Thomas3 1Epidemiology, Clinique Médicale l'Actuel, Montreal, Canada. 2School of Public Health & Social Policy, University of Victoria, Victoria, Canada. 3Clinical, Clinique Médicale l'Actuel, Montreal, Canada Background: The potential for pre‐exposure prophylaxis (PrEP) to reduce HIV incidence relies on equitable access to PrEP and retention in care. Despite the availability of low‐cost PrEP through high‐volume clinics in Quebec, these services are centralized in Montreal's downtown gay village – potentially threatening the effectiveness of PrEP for HIV elimination in suburban and rural areas. We aim to investigate barriers to PrEP retention including clinic access. Methods: We examined factors associated with time to PrEP discontinuance using clinical data collected between January 2011 and April 2018 at Canada's largest PrEP clinic (l'Actuel). Cox proportional hazard models estimated adjusted hazard ratios (aHRs) for risk of PrEP discontinuation with censoring of patients maintained in care as of 1 December 2017. Our primary explanatory factor, clinic access, was measured by assessing driving distance from residential postal code centroids to l'Actuel. Other covariates included baseline PrEP regimen, age, income, education and behavioral risk factors (i.e. number of sexual partners within 12 months, antecedent STIs and chemsex use). Results: In total, 1473 clients (median age 36, IQR 29 to 45; 98% MSM) initiated PrEP (82% daily, 18% intermittent) – providing 1460 person‐years of observation. Twelve‐month retention rate was 52%. In April 2018, 662 individuals (45%) were actively maintained in PrEP care. Half (49%) of all PrEP users resided within 5 km of l'Actuel, 28% resided 5 to 9 km away, 10% resided 10 to 19 km away, 7% resided 20 to 49 km away and 6% resided over 50 km away. In multivariate modeling, only greater distance from l'Actuel (aHR 1.002, 95% CI 1.000 to 1.003) and younger age (aHR 0.979, 95% CI 0.970 to 0.989) were associated with increased risk of PrEP discontinuation. Discussion: Few significant findings predict PrEP discontinuance, only greater distance to our clinic and younger age – highlighting the need for additional research regarding patterns of clinical retention among PrEP users. In our setting, young MSM are the highest risk group for HIV acquisition, and greater efforts to initiate and retain young patients on PrEP are essential. Improved public health messages, provider training and alternative PrEP delivery options are needed to expand spatial coverage beyond that of downtown urban cores and to younger clients who may be less likely to live in urban centers or gay neighborhoods. P012 The HIV continuum of care in Austria from 2010 to 2016: data and challenges G Leierer1, A van Sighem2, A Rieger3, B Schmied4, M Sarcletti1, A Öllinger5, B Haas6, A Egle7, M Rappold1 and R Zangerle 1 1Department of Dermatology and Venereology, Medical University of Innsbruck, Innsbruck, Austria. 2Stichting HIV Monitoring, Amsterdam, Netherlands. 3Department of Dermatology, Medical University of Vienna, Vienna, Austria. 4Otto‐Wagner Hospital, Vienna, Austria. 5Department of Dermatology and Venereology, Kepler University Hospital, Med Campus III, Linz, Austria. 6Department of Internal Medicine, General Hospital Graz Sued‐West, Graz, Austria. 7Department of Internal Medicine III, Paracelsus Medical University, Salzburg, Austria Background: UNAIDS has set a 90‐90‐90 target to curb the HIV epidemic by 2020, but methods used are not standardised. HIV surveillance in Austria relies on a hospital‐based cohort not taking into account the transfer of care to private physicians in Vienna, which comprised 24% of all patients with cART in 2016. Methods: Data from the Austrian HIV Cohort Study were used to derive the four‐stage continuum of HIV care. PLWHIV estimates were obtained using back‐calculation models (ECDC tool 1.3.0) to estimate HIV incidence and the undiagnosed fraction. The proportion ever diagnosed who ever initiated ART and the proportion of them who were virally suppressed (≤200 copies/mL) were assessed for all patients and for MSM for the years 2010 to 2016. For high estimates patients lost to follow‐up (LTFU, no contact 1.5 years before the end of the respective year) were excluded and for low estimates they were included. The preferred estimate was the mid‐point between the high and low estimate. Missing HIV‐RNA was considered as unsuppressed. Logistic regression was used to identify factors associated with LTFU. Results: The fraction undiagnosed decreased from 19% (95% CI 18% to 21%) in 2010 to 10% (95% CI 9% to 13%) in 2016, among MSM from 20% (95% CI 18% to 22%) in 2010 to 8% (95% CI 6% to 11%) in 2016. The proportion of diagnosed patients who have ever started ART increased from 81% (79% among MSM) in 2010 to 93% (93% among MSM) in 2016. The proportion of individuals virally suppressed improved from 77% to 84% (76% to 85% among MSM). The fraction of the virally suppressed among PLWHIV increased from 51% to 70%, among MSM from 48% to 73%. Estimates of the number of new HIV infections decreased from 258 (95% CI 237 to 282) to 171 (95% CI 107 to 262), among MSM from 157 (95% CI 143 to 168) to 70 (95% CI 39 to 114). The overall rate of LTFU was 37.8%, multivariable logistic regression revealed three significant factors for LTFU: younger age, residency in Vienna and non‐Austrian origin. Conclusions: Austria is nearing the 90‐90‐90 target. Viral suppression was comparatively low and maybe explained substantially by transfer of care in Vienna and out‐migration. This and the decrease in HIV incidence supports the hypothesis that the high estimate of being on ART and virally suppressed is the more likely scenario. For more reliable nationwide estimates there is urgent need to include data from private physicians. P013 Behavioural, psychological and network characteristics of MSM eligible for PrEP enrolled by respondent‐driven sampling network strategy M Psichogiou 1, M Papadopoulou1, S Chanos2, V Sypsa3, S Roussos3, D Paraskevis3, N Dedes2, G Daikos1, J Schneider4 and A Hatzakis3 11st Internal Medicine Department, Laiko General Hospital, National and Kapodistrian University of Athens, Athens, Greece. 2Positive Voice, Athens Check Point, Athens, Greece. 3Department of Hygiene, Epidemiology and Medical Statistics, Medical School, National and Kapodistrian University of Athens, Athens, Greece. 4Public Health, University of Chicago, Chicago, IL, USA Background: The benefits of pre‐exposure prophylaxis (PrEP) in HIV prevention are well established. Elimination of HIV transmission could be achieved through a combination of preventive strategies taking into consideration specific population characteristics and the potential for high population coverage. We investigated demographic, behavioural, psychosocial and network characteristics of MSM at ongoing high risk for acquiring HIV infection who were willing to take PrEP. Materials and methods: Sophocles‐P4G, a pilot PrEP study in Athens, Greece, was designed to identify, within MSM networks, the population at highest risk for HIV, based on specified criteria, who were eligible to be treated with PrEP. It was based on recruitment through respondent‐driven sampling (RDS) and facilitated by a community organisation (Positive Voice). The programme included rapid HIV testing and interview with a structured questionnaire. Results: Between 2016 and 2018, of the 308 MSM enrolled, 21 were already known to be HIV+ seeds, five were newly HIV diagnosed and 282 were confirmed to be HIV negative. The 282 MSM at risk had a mean (SD) age of 28.3 (8.2) years, a mean (SD) of 15.7 (2.5) years of education; 54 (19.4%) were unemployed; 40 (14.2%) identified as bisexual; 211 (74.82%) reported having tested for HIV during the previous year; 71 (25.36%) had a sexually transmitted infection (STI) during the previous year. The median (25th, 75th) size in their sexual network was 10 (3, 30) people; 165 (58.5%) reported using drugs associated with chemsex and 84 (31.2%) reported symptoms of depression. PrEP enrolment criteria were reported by 129/282 (41.9%) MSM. Specifically, compared to the low‐risk MSM, they reported more frequent condomless sex (66.3% vs. 25%; p = 0.005), higher participation in group sex (90.7% vs. 41.2%; p 100,000 copies/mL were high and similar between arms. Across both studies, six participants on DTG + 3TC and four on DTG + TDF/FTC met protocol‐defined virologic‐withdrawal criteria through Week 48; none had treatment‐emergent integrase‐strand‐transfer‐inhibitor or NRTI resistance mutations. Overall rates of AEs were similar between arms, with low rates of withdrawals due to AEs in both arms [GEMINI‐1&2 pooled: DTG + 3TC 15/716 (2%) vs. DTG + TDF/FTC 16/717 (2%)]. More drug‐related AEs were reported with DTG + TDF/FTC [GEMINI‐1&2 pooled: DTG + 3TC 126/716 (18%) vs. DTG + TDF/FTC 169/717 (24%)]. The frequency of AEs was generally similar across subgroups. Abstract P021 – Table 1. Proportion of participants with plasma HIV‐1 RNA 100,000 65/74 (88%) 69/76 (91%) 64/66 (97%) 69/77 (90%) 129/140 (92%) 138/153 (90%) Baseline CD4 + (cells/mm3) ≤200 25/31 (81%) 26/29 (90%) 25/32 (78%) 25/26 (96%) 50/63 (79%) 51/55 (93%) >200 295/325 (91%) 306/329 (93%) 310/328 (95%) 312/333 (94%) 605/653 (93%) 618/662 (93%) Conclusions: In GEMINI‐1&2, DTG + 3TC demonstrated non‐inferior efficacy to DTG + TDF/FTC in treatment‐naïve adults with screening HIV‐1 RNA ≤500,000 copies/mL at Week 48. Both regimens were well tolerated. Subgroup analyses of efficacy and safety performed based on baseline disease and demographic characteristics were generally consistent with overall study results. These results further demonstrate DTG + 3TC is an option for initial treatment of HIV‐infected patients across a spectrum of disease characteristics and patient populations. The studies are ongoing to explore long‐term durability and safety. Reference: [1] Cahn P, Sierra Madero J, Arribas J, et al. Non‐inferior efficacy of dolutegravir (DTG) plus lamivudine (3TC) versus DTG plus tenofovir/emtricitabine (TDF/FTC) fixed‐dose combination in antiretroviral treatment‐naïve adults with HIV‐1 infection – 48‐week results from the GEMINI studies [abstract TUAB0106LB]. 22nd International AIDS Conference; 2018 Jul 23‐27; Amsterdam, Netherlands. P022 Cerebrospinal fluid exposure of cenicriviroc in HIV‐positive individuals with cognitive impairment J Alagaratnam 1, L Else2, S Dilly Penchala2, E Challenger2, K Legg1, C Petersen3, B Jones4, R Kulasegaram5, S Seyedkazemi6, E Lefebvre7, S Khoo2 and A Winston1 1Department of Infectious Diseases, Imperial College London, London, UK. 2Department of Pharmacology, University of Liverpool, Liverpool, UK. 3Clinical Trials Centre, Imperial College London, London, UK. 4Department of Radiology, Imperial College Healthcare NHS Trust, London, UK. 5Department of GU/HIV, St Thomas’ Hospital, London, UK. 6Clinical Development, Allergan plc, South San Francisco, CA, USA. 7former employee of R&D, Allergan plc, South San Francisco, CA, USA Background: Cenicriviroc, a dual C‐C chemokine receptor type 2 (CCR2) and type 5 (CCR5) antagonist, is a potential adjunctive therapy, along with ART, for the management of HIV‐associated cognitive disorders. Materials and methods: Virologically suppressed PLWHIV with a clinical diagnosis of HIV‐related cognitive impairment intensified ART with cenicriviroc once daily, dose dependent on current ART, for 8 weeks. Subjects with current or previous use of CCR5 inhibitors were not eligible. We assessed cerebrospinal fluid (CSF) exposure of cenicriviroc and CSF albumin at Week 8, and changes in cognitive function over 8 weeks. Cenicriviroc concentration was determined using reverse phase high‐performance liquid chromatography (HPLC) with geometric mean (GM) and 95% CI calculated. The proposed cenicriviroc target concentration was above the 90% effective concentration (EC90) for cenicriviroc (0.17 ng/mL), with the lower limit of quantification (LLQ) 0.24 ng/mL taken as target concentration. Cognitive function assessment comprised of seven domains with composite Z‐scores reported. Results: Four of seven enrolled participants completed all study procedures. Median age was 43 years (interquartile range [IQR] 39–47), current CD4 + count 375 cells/µL (IQR 315–555) and plasma HIV RNA undetectable in all. At Week 8, CSF exposure was detected in two subjects (0.82 and 0.40 ng/mL) and below the LLQ in two (Table 1). Mean CSF: plasma cenicriviroc concentration ratio was no more than 0.18% (95% CI of the upper estimate 0.09%–0.28%). Median CSF: serum albumin ratio was 10.1 (IQR 7.2–19.5) and CSF albumin was higher in those with detectable cenicriviroc in the CSF. Overall cognitive performance Z‐score was −0.14 (95% CI −1.35 to 1.07) at baseline, and −0.27 (95% CI −1.70 to 1.17) at Week 8. Abstract P022 – Table 1. Individual subject blood and cerebrospinal fluid parameters at Week 8 Subject 1 2 3 4 CSF cenicriviroc concentration, ng/mL 0.82 0.40 0.24 (less than LLQ) 0.24 (less than LLQ) Plasma cenicriviroc concentration, ng/mL 718.6 211.1 411.9 70.5 CSF: plasma cenicriviroc concentration ratio (%) 0.11 0.19 0.06 0.34 CSF albumin concentration, mg/mL 1070 453 374 202 Serum albumin concentration, g/L 38 42 40 40 CSF: serum albumin ratio 28.2 10.8 9.4 5.1 Antiretroviral therapy abacavir, lamivudine, raltegravir lamivudine, atazanavir, ritonavir tenofovir DF, emtricitabine, dolutegravir tenofovir DF, emtricitabine, raltegravir Cenicriviroc dose 150 mg 50 mg 150 mg 150 mg tenofovir DF = tenofovir disoproxil fumarate. Conclusion: In PLWHIV with cognitive impairment, cenicriviroc CSF exposure is close to the EC90. P023 Abstract withdrawn P024 Abstract withdrawn P025 Pharmacokinetics of MK‐8591, dolutegravir and tenofovir disoproxil fumarate are not altered after co‐administration when compared to single agent administration D Rudd 1, S Zhang1, K Fillgrove1, S Fox‐Bosetti1, R Matthews1, E Friedman1, D Armas2, S Stoch1 and M Iwamoto1 1Merck Research Laboratories, Merck Sharp & Dohme Corp., Kenilworth, NJ, USA. 2Preclinical Research, Celerion, Tempe, AZ, USA Background: MK‐8591 is a highly potent nucleoside reverse transcriptase translocation inhibitor that is in development for the treatment of HIV‐1 infection. As MK‐8591 may be co‐administered with other antiretrovirals, a drug interaction study with dolutegravir (DTG) and tenofovir (TFV) disoproxil fumarate (TDF) was conducted. MK‐8591 is primarily eliminated by renal excretion and may be affected by drugs that inhibit renal transporters. DTG is primarily metabolized through glucuronidation (UGT1A1). DTG is known to inhibit renal transporters OCT2 and MATE1. TDF is a commonly used NRTI that is eliminated (as TFV) by glomerular filtration and active tubular secretion. TFV may compete with other compounds that are also renally eliminated. Materials and methods: The two‐way interaction between MK‐8591 and DTG+TDF was investigated in a two‐period, two‐way DDI study in 12 healthy adult subjects. Subjects received a single oral dose of 20 mg MK‐8591 on Day 1, followed by a washout of seven days. In Period 2, subjects received 11 days of 50 mg DTG/300 mg TDF once daily, with 20 mg MK‐8591 co‐administered on Day 8. PK for DTG and TFV was taken on Days 7 and 8, while PK for MK‐8591 was taken on Day 8. Results: Administration of MK‐8591 alone and in combination with DTG and TDF was generally well tolerated. MK‐8591 AUC and Cmax were not meaningfully affected with DTG+TDF co‐administration, and the DTG and TDF AUC0‐24, C24 and Cmax were similar with and without MK‐8591 (Table 1). The apparent terminal half‐life of MK‐8591, DTG and TFV were not meaningfully affected with co‐administration. Abstract P025 – Table 1. Comparing MK‐8591, DTG and TFV PK post‐administration of DTG and TDF with/without MK‐8591 in healthy adult subjects Geometric mean ratio (GMR) with 90% confidence interval, relative to single agent administration N MK‐8591 DTG TDFa AUC0‐24 12 1.08 (1.02 to 1.14) 1.05 (0.96 to 1.14) AUC0‐inf 12 1.28 (1.9 to 1.37) ‐ C24 12 1.10 (1.02 to 1.17) 1.05 (0.97 to 1.14) Cmax 12 1.07 (0.93 to 1.22) 1.02 (0.94 to 1.11) 0.98 (0.88 to 1.10) aPK parameters shown are for TFV in plasma. Conclusions: MK‐8591 had no clinically significant effect on the PK of DTG or TFV. While DTG is known to inhibit OCT2 and MATE1, there was no effect on the PK of MK‐8591. There also appears to be no competition between the renal elimination of MK‐8591 and TFV. These findings support co‐dosing of MK‐8591 and DTG+TDF, if indicated. P026 “Treatment access cascades”: effects of viral load, resistance testing and safety in pregnancy on access to dolutegravir in low‐ and middle‐income countries J Levi 1, P Clayden2 and A Hill3 1Medicine, Imperial College London, London, UK. 2Medicine, HIV i‐Base, London, UK. 3Department of Pharmacology and Therapeutics, Liverpool University, Liverpool, UK Objectives: Generic antiretroviral treatment with tenofovir disoproxil fumarate/lamivudine/dolutegravir (TDF/3TC/DTG; TLD) is available in eligible countries for $75/person‐year. There are plans to transition millions of HIV+ people to TLD in 2018 to 2019. The Phase III trial programme for dolutegravir excluded: pregnant women, ART‐naive with transmitted drug resistance and ART‐experienced without genotypic resistance testing. The DAWNING study excluded ART‐experienced people with no active NRTIs. Recent reports of neural tube defects (NTDs) after use of DTG at conception could limit use in women of childbearing potential. Methods: Using UNAIDS 2016 data we made four cascades for each low‐/middle‐income country (LMIC) with available data. Cascade 1 shows all diagnosed HIV+ people, split into those on ART, those achieving HIV‐RNA suppression and those on ART but unsuppressed. We modelled the impact on access to TLD if patients could not be started or switched due to: lack of access to viral load testing (Cascade 2); or to resistance testing (Cascade 3); or due to risks associated with pregnancy (Cascade 4). We defined lack of access to viral load testing as 70%. Multivariable logistic regression models were applied for the statistical analysis. Late presenters were defined as persons with initial CD4 count between 200 and 350 cells/μL; those with advanced disease had an initial CD4 count 38°C for ≤seven days) in emergency departments in Dar es Salaam between July 2013 and May 2014. Detailed medical history and clinical examination were done. Rapid diagnostic test for HIV was systematically performed and confirmed in case positivity following national recommendations. All patients with a negative HIV rapid test had an antigen p24 screening. Additional rapid, culture‐ and molecular‐based microbiological tests were performed according to pre‐defined algorithms to investigate the causes of fever. Results: Among 519 patients with an acute febrile illness, HIV prevalence was 25% (N = 128; prevalence in adults in Dar es Salaam is 12%). Among these 128 patients, 122 (95%) had a positive rapid test which was confirmed by a second test. Six patients had only a positive p24 antigen suggesting an acute HIV infection in 1.2% of febrile adults: four presented with fever without focus, one having a rash and two concomitant dengue; one patient presented with a pharyngitis and one with a bronchitis. Among patients diagnosed with chronic HIV infection, 62% (N = 76) had an advanced disease (CD4 100,000 copies/mL). Over one‐third (38%) had baseline antiretroviral resistance to at least one drug class. Among those previously diagnosed outside Canada, 31% (16/52) were late presenters and 15% (8/52) had advanced HIV; 31% (16/52) had detectable VLs at presentation. Opportunistic infections were rare among both groups. Among newly diagnosed refugees, time from IME screening to notification of diagnosis: 31 days [IQR 21 to 49]; notification to linkage to care: six days [IQR 1.5 to 17.5]; linkage to cART prescription: 11 days [IQR 5.5 to 16.5]; and cART prescription to first undetectable VL: 36 days [IQR 28 to 72]. Overall, 38% of newly diagnosed patients were linked to HIV care within 30 days, 74% within 60 days and 86% within 90 days from HIV screening. Median time from HIV screening to cART prescription was 62 days [IQR 46 to 89]. Abstract P041 – Table 1. Characteristics of the study sample Overall (n = 102) Diagnosed in Canada (n = 50) Diagnosed outside Canada (n = 52) Age (median [IQR]) 37 [32 to 44] 37 [32 to 44] 37 [33 to 43] Sex Female 68 (67%) 27 (54%) 41 (78%) Male 34 (33%) 23 (46%) 11 (21%) Country of origin Africa Nigeria 23 (23%) 9 (18%) 14 (27%) Other 32 (31%) 13 (26%) 19 (37%) Latin America Haiti 45 (44%) 28 (56%) 17 (33%) Other 2 (2%) 0 (0%) 2 (4%) CD4 at presentation in Canada, cells/µL (median, range, [IQR]) 361, 11 to 1136, [229 to 407] 309, 11 to 811, [210 to 386] 446, 14 to 1136, [271 to 674] CD4 nadir 1 million, [ 1 million, [5558‐97,534] 1 million, [ 500,000 copies/mL and three for HLAB5701 positivity. The snapshot analysis at 48 weeks showed a virological success of 77.3% in RAL and 66.7% in DRV. Time to starting treatment was 34.5 days in RAL and 53 days in DRV arms, respectively. At the as treated analyses the median increase in CD4 + was 297 in RAL and 239 in DRV. No difference in total cholesterol, while triglycerides were higher in DRV arm. No statistical analyses were performed due to the low number of patients enrolled. Conclusions: Patients: late presenters are frequent but very difficult to enrol in clinical trials. In these patients, the test and treat strategy is rarely applicable. The rate of virological success is similar to that described in the literature and very far from results of the recent trials in naive patients. P043 Impact of a training project for primary health‐care providers (FOCO project) in the HIV screening and HIV late diagnosis M Pérez Elías 1, G Sampériz2, D Dalmau3, A Romero4, B de la Fuente5, I de los Santos6, J Lopez7, P Arazo2, V Estrada8, F Lozano9, M Pastor10, A Ocampo11, A Arrillaga12, M Fuster‐Ruizdeapodaca13 and M Galindo14 1Infectious Diseases, Hospital Ramón y Cajal, IRYCIS, Madrid, Spain. 2Infectious Diseases, Hospital Miguel Servet, Zaragoza, Spain. 3Infectious Diseases, Hospital Mutua Terrassa, Barcelona, Spain. 4Infectious Diseases, Hospital de Puerto Real, Cádiz, Spain. 5Infectious Diseases, Hospital Cabueñes, Gijón, Spain. 6Infectious Diseases, Hospital La Princesa, Madrid, Spain. 7Infectious Diseases, Hospital Gregorio Marañón, Madrid, Spain. 8Infectious Diseases, Hospital Clínico San Carlos, Madrid, Spain. 9Infectious Diseases, Hospital Valme, Sevilla, Spain. 10Management, Bizkaisida, Bilbao, Spain. 11Infectious Diseases, Hospital Alvaro Cunqueiro, Vigo, Spain. 12Coordinator, Plan del Sida del País Vasco, San Sebastian, Spain. 13Management, Sociedad Española Interdisciplinaria del SIDA (SEISIDA), Madrid, Spain. 14Infectious Diseases, Hospital Clínico Universitario, Valencia, Spain Background: Reducing HIV late diagnosis remains an epidemiological challenge [1]. The objective of this project was to promote early HIV diagnosis through the training of primary health‐care providers (PHCP). Materials and methods: HIV specialists conducted training sessions in 108 primary care centres (PCC) from six Spanish regions during 2016 and 2017, and with 1804 PHCP involved. The intervention was evaluated using a pre‐experimental design collecting the dependent variables both in the six months before and after the intervention. Number of requests for HIV tests from the PCC trained and clinical data of new HIV diagnosed patients were collected. Parametric and non‐parametric tests were used to assess differences between pre‐ and post‐intervention data. Results: Number of HIV tests performed was higher after the intervention (16,833 vs. 19,793, p 20% of post‐2010 subjects failed over 3 years. Few demographic factors predicted failure. Phase III trials yield overly optimistic impression of real‐world efficacy. All guidelines should list INSTI‐based initial ART as preferred. Strategies are needed to improve access to pre‐ART genotyping and to increase early initiation of once‐daily ART. P052 Effectiveness, persistence and safety of E/C/F/TAF, F/TAF+3rd agent or R/F/TAF use in treatment‐naive HIV‐1 infected patients: 12‐month results from the German TAFNES cohort study T Heuchel1, H Hillenbrand2, H Jessen3, R Pauli4, N Postel5, R Haubrich6, M Heinzkill 7, K Goerner7 and H Stellbrink8 1Clinical Care, Praxis Heuchel, Chemnitz, Germany. 2Clinical Care, MVZ Praxis City Ost, Berlin, Germany. 3Clinical Care, Gemeinschaftspraxis Jessen, Berlin, Germany. 4Clinical Care, Gemeinschaftspraxis Becker/Pauli, Munich, Germany. 5Clinical Care, Prinzmed Private Practice, Munich, Germany. 6Clinical Science, Gilead Sciences, Foster City, CA, USA. 7Clinical Science, Gilead Sciences, Munich, Germany. 8Clinical Care, ICH Study Center, Hamburg, Germany Background: Based on controlled clinical trials, tenofovir alafenamide (TAF)‐based regimens are among recommended regimens for first‐ and further‐line ART of HIV infection in Germany. To evaluate the effectiveness and safety of TAF‐based single‐tablet (STR) or multi‐tablet regimens (MTR) when used in treatment‐naïve (TN) or treatment‐experienced adult HIV‐infected patients in a real‐life setting, the non‐interventional 24‐month prospective TAFNES cohort study was initiated. Methods: Month‐12 (M12) evaluation of using TAF‐based regimens, i.e. E/C/F/TAF, F/TAF+3rd agent or R/F/TAF, in TN patients of the TAFNES cohort. The analysis population consisted of patients starting treatment at least 9 months prior to data‐cut (May 2018). Outcome measures included ART persistence (using Kaplan‐Meier analyses), virological effectiveness (HIV‐RNA 100,000 copies/mL, n (%) 67 (28) 31 (22) 36 (55) 0 (0) Reason for discontinuation, n (%) ADRs 4 (1.7) 2 (1.4) 2 (3.1) 0 (0.0) Drug‐drug interaction 4 (1.7) 3 (2.1) 0 (0.0) 1 (3.2) Virological failure 2 (0.8) 1 (0.7) 0 (0.0) 1 (3.2) Patient decision 2 (0.8) 0 (0.0) 2 (3.1) 0 (0.0) Other 3 (1.3) 1 (0.7) 2 (3.1) 0 (0.0) Lost to follow‐up 17 (7.1) 8 (5.6) 7 (10.8) 2 (6.5) adefined as CD4 cell count 90% probability that DTG has greater VS at 48 weeks versus all core agents analysed. Abstract P054 – Table 1. Odds ratios (OR) of virologic suppression 1 for ORs of viral suppression and >0 for mean increase in CD4 cells. Conclusions: Study results showed higher probabilities of VS and larger CD4 increases for DTG compared to some treatments and similar outcomes relative to others. These results suggest DTG is among the most effective treatments available for the initial treatment of HIV‐1 infection. P055 Trends in modification and discontinuation of initial antiretroviral treatment in Turkish HIV‐TR cohort, 2011‐2017 V Korten 1, D Gökengin2, M Fincancı3, T Yıldırmak4, S Gencer5, H Eraksoy6, D İnan7, F Kaptan8, B Dokuzoğuz9, Karaoğlan10, A Willke11, Ö Ergönül12 and HIV‐TR study group13 1Infectious Diseases, Marmara University Hospital, Istanbul, Turkey. 2Infectious Diseases, Ege University Hospital, Izmir, Turkey. 3Infectious Diseases, Istanbul Education and Research Hospital, Istanbul, Turkey. 4Infectious Diseases, Okmeydanı Education and Research Hospital, Istanbul, Turkey. 5Infectious Diseases, Lütfi Kırdar Education and Research Hospital, Istanbul, Turkey. 6Infectious Diseases, Istanbul University, Istanbul School of Medicine, Istanbul, Turkey. 7Infectious Diseases, Akdeniz University Hospital, Antalya, Turkey. 8Infectious Diseases, Izmir Katip Çelebi University Atatürk Education Training and Research Hospital, Izmir, Turkey. 9Infectious Diseases, Ankara Numune Education and Research Hospital, Ankara, Turkey. 10Infectious Diseases, Gaziantep University Hospital, Gaziantep, Turkey. 11Infectious Diseases, Kocaeli University Hospital, İzmit, Turkey. 12Infectious Diseases, Koç University School of Medicine, Istanbul, Turkey. 13HIV‐TR Study Group Background: The aim of this study was to compare the frequency, reasons and the predictors for discontinuation and modification of ART before and after the availability of better tolerated and less complex regimens. Methods: A total of 3019 antiretroviral‐naive patients registered in the HIV‐TR cohort who started ART between January 2011 and February 2017 were studied. Only the first modification for each patient within 1 year was included in the analyses. Reasons were classified as listed in the coded form in the web‐based database. Time to treatment modification was analysed using Kaplan‐Meier curves and log‐rank tests, and factors associated with regimen modification were examined using Cox proportional hazards models. Results: Of 3019 patients, initial regimen was modified or discontinued in 379 patients (12.6%) within the first year. The main reason for modification or discontinuation was intolerance/toxicity (41.7%), followed by treatment simplification (9%), death (7.4%), patient's willingness (6.9%), poor compliance (6.6%), to prevent future toxicities (5.5%), virological failure (5%) and provider's preference (5%). Median time to treatment modification was shorter in patients treated with NNRTI‐ compared with integrase strand transfer inhibitor (InSTI)‐ and protease inhibitor (PI)‐based regimens (3.9, 4.9 and 5.6 months, respectively; p = 0.049). In a multivariable Cox model, only predictor of discontinuation was baseline AIDS diagnosis, (aHR 1.4, 95% CI 1.1 to 1.8; p = 0.01). Discontinuation rate was higher among PI (17.8%) and NNRTI (14.3%) when compared to InSTI‐based regimens (6.3%) (p 0.05). Conclusion: There was a relatively low rate of modification and discontinuation of ART regimens within the first 12 months as compared with other countries [1–4]. InSTI‐based regimens were less likely to be modified than PI‐ and NNRTI‐based ART. References [1] Vo TT, Ledergerber B, Keiser O, Hirschel B, Furrer H, Battegay M, et al. Durability and outcome of initial antiretroviral treatments received during 2000‐‐2005 by patients in the Swiss HIV Cohort Study. J Infect Dis. 2008;197:1685‐94. [2] Di Biagio A, Cozzi‐Lepri A, Prinapori R, Angarano G, Gori A, Quirino T, et al. Discontinuation of initial antiretroviral therapy in clinical practice: moving toward individualized therapy. J Acquir Immune Defic Syndr. 2016;71:263‐71. [3] Sun J, Liu L, Shen J, Qi T, Wang Z, Song W, et al. Reasons and risk factors for the initial regimen modification in Chinese treatment‐naïve patients with HIV infection: a retrospective cohort analysis. PLoS One. 2015;10:e0133242. [4] Sheth AN, Ofotokun I, Buchacz K, Armon C, Chmiel JS, Hart RL, et al. Antiretroviral regimen durability and success in treatment‐naive and treatment‐experienced patients by year of treatment initiation, United States, 1996‐2011. JAIDS. 2016;71:47‐56. P056 Reasons for choosing darunavir/ritonavir 600/100 mg BID versus 800/100 mg QD in ART‐naïve patients A Tavelli 1, M Palma2, S Lo Caputo3, G Madeddu4, P Bonfanti5, B Menzaghi6, S Nozza7, A Antinori8, R Termini2, A d'Arminio Monforte9, on behalf of Icona Foundation Study Group 1Icona Foundation, Milan, Italy. 2Janssen‐Cilag SpA, Cologno Monzese, Italy. 3Clinic of Infectious Diseases, Policlinic of Bari, Bari, Italy. 4Unit of Infectious Diseases, University of Sassari, Sassari, Italy. 5Unit of Infectious Diseases, A Manzoni Hospital, Lecco, Italy. 6Unit of Infectious Diseases, ASST della Valle Olona, Busto Arsizio, Italy. 7Clinic of Infectious Diseases, San Raffaele Scientific Institute, Milan, Italy. 8HIV/AIDS Unit, INMI L. Spallanzani IRCCS, Rome, Italy. 9Clinic of Infectious and Tropical Diseases, ASST Santi Paolo e Carlo University of Milan, Milan, Italy Background: Ritonavir‐boosted darunavir (DRV/r) is recommended in ART‐naïve, specifically in those with perceived low adherence and before results of resistance test. DRV/r 600/100 mg twice daily (BID) is licensed for ART‐experienced, while 800/100 mg once daily formulation (QD) is recommended in first‐line and in experienced patient without DRV RAMs. In clinical practice, however, a non‐negligible proportion of ART‐naïve subjects started with DRV/r BID. The aim of this study is to identify patterns of prescription of DRV/r BID in ART‐naive, analysing predictors of DRV/r BID start versus QD. Materials and methods: All patients from Icona cohort that started a DRV/r‐based regimen from ART‐naïve in 2008 to 2017 were included. A cross‐sectional analysis was performed comparing demographics, clinical and lifestyle factors at the time of starting DRV/r in BID or QD, using chi‐square and Wilcoxon test as appropriate. Univariable and multivariable logistic regression models were used to identify predictors of DRV/r BID start. Results: One thousand five hundred and three ART‐naive patients were included: 1297 started DRV/r QD (86%) and 206 DRV/r BID (14%). Eighty‐two percent male, median age 40 years, 80% Italian, 80% in triple therapy, 6% in two‐drug regimen and 14% in combination with >3 ARV drugs (69% with raltegravir and 13% with maraviroc). Median HIV‐RNA in DRV/r QD and BID groups were 5.1 and 5.3 log10 copies/mL (p = 0.02), with 31% DRV/r BID with HIV‐RNA >500,000 copies/mL (vs. 22% QD; p 40 years: adjusted OR [aOR] 1.44; p = 0.019), subjects with CD4 3 ARV drugs (aOR 2.11; p = 0.001), centres from southern Italy (aOR 2.57; p 3 ARV was greater in people with HIV‐RNA >500,000 than in those with HIV‐RNA ≤500,000 (OR 5.9; p 900 cells/µL and CD4/CD8 ratio > 1) at 1 and 3 years. Secondary endpoints were adverse events (AEs) leading to ART discontinuation at 1 and 3 years. ITT and PP analysis were performed. Results: Viral suppression (overall suppression of 96% at 1 year and 99% at 3 years) was comparable in all ART regimens. Among the InSTI group, levels of viral suppression were comparable for dolutegravir‐ and elvitegravir‐based regimens. At 1 year there was an increment of 350 CD4+T cells/µL, which was comparable in all ART regimens. Overall 36% and 39% achieved CD4 >900 cells/µL and 43% and 66% a CD4/CD8 >1 at 1 and 3 years, respectively. In a subanalysis of immune recovery comparing Fiebig stages I to V with Fiebig stage VI, starting ART during the earliest Fiebig stages was associated with higher rates of CD4 >900 cells/µL at 3 years (p = 0.027). Discontinuation due to AEs was more frequent with NNRTI compared to other ART families (p = 0.036 at 1 year, p = 0.040 at 3 years) with high rates of neuropsychiatric AEs (Table 1). Conclusions: Viral suppression and immunological recovery were excellent in acute/recent patients, with no differences between ART regimens. Earlier ART initiation ( 900 at 1 year, n (%) N = 114 41 (36%) 6 (29%) 8 (31%) 27 (40%) 0.419 CD4+ T cell >900 at 3 years, n (%) N = 57 22 (39%) 6 (40%) 7 (33%) 9 (43%) 0.745 CD4/CD8 >1 at 1 year, n (%) N = 114 52 (46%) 10 (48%) 11 (42%) 31 (46%) 0.986 CD4/CD8 >1 at 3 years, n (%) N = 57 36 (63%) 8 (53%) 13 (62%) 15 (71%) 0.518 Toxicity At least one adverse event at 1 year, n (%) 36 (26%) 8 (36%) 10 (36%) 18 (21%) 0.146 At least one adverse event at 3 years, n (%) 49 (36%) 11 (50%) 12 (43%) 26 (30%) 0.145 Discontinuation rateb at 1 year, n (%) 13 (9%) 5 (23%) 3 (11%) 5 (6%) 0.036 Discontinuation rateb at 3 years, n (%) 20 (15%) 6 (27%) 6 (31%) 8 (9%) 0.040 Kruskal‐Wallis test was used to assess for quantitative variables; Fisher's exact test and chi‐squared for categorical variables. aboosted with either ritonavir or cobicistat. bdiscontinuation rate applies to any drug, even replaced by another drug of the same family. HTSX = heterosexual; IDU = intravenous drug user. P059 Comparison of raltegravir (RAL) and boosted darunavir (DRV/b) versus dolutegravir (DTG) both associated with tenofovir/emtricitabine (TDF/FTC) in primary HIV infection (PHI): viro‐immunological outcomes of two different integrase inhibitor (INSTI)‐based strategies A Mondi 1, C Pinnetti1, P Lorenzini1, M Plazzi1, I Abbate2, G Rozera2, C Agrati3, R Libertone1, S Menichetti1, I Mastrorosa1, A Ammassari1 and A Antinori1 1HIV/AIDS Department, National Institute for Infectious Diseases, Lazzaro Spallanzani IRCCS, Rome, Italy. 2Laboratory of Virology, National Institute for Infectious Diseases, Lazzaro Spallanzani IRCCS, Rome, Italy. 3Cellular Immunology and Pharmacology Laboratory, National Institute for Infectious Diseases, Lazzaro Spallanzani IRCCS, Rome, Italy Background: The aim of this study was to compare the viro‐immunological response to a four‐ versus three‐drug, both INSTI‐based, regimens in the setting of PHI. Material and methods Monocentric, prospective, observational study including all patients (pts) diagnosed with PHI from July 2013 to April 2018. ART was started, before GRT results, with TDF/FTC and either RAL+DRV/b or DTG. Data were collected from ART start (baseline, BL) until the last observation. Probability and predictors of achieving virological suppression (VS) (HIV‐RNA 1 were assessed by Kaplan‐Meier curves and Cox regression analysis, respectively. Results: One hundred and forty‐four pts included. At BL: median HIV‐RNA and HIV‐DNA were 5.6 log10 copies/mL (IQR 4.8 to 6.6) and 4.4 log10 106 PBMC/mL (IQR 3.8 to 4.8), respectively; 58% pts had CD4 >500/mm3 and 25% CD4/CD8 >1. Fiebig stage was: II/III in 15%, IV in 31%, V in 29% and VI in 23% pts. ART was started, within a median of 5 days (IQR 3 to 9) from diagnosis, with four drugs in 110 pts (76%) or three drugs in 34 pts (24%). BL characteristics did not substantially differ between the two treatment groups. Over a median follow‐up of 18 months (IQR 8 to 23), 139 pts (97%) reached VS. The 12‐month probability of achieving VS was high and similar between the four‐drug and three‐drug groups (98% vs. 96%, p = 0.232). After stratification by pre‐ART viraemia, the three‐drug group compared to the four‐drug group showed an increased probability of achieving VS in the stratum with BL HIV‐RNA 500/mm3 (aHR 1.73, p = 0.021) and CD4/CD8 >1 (aHR 1.95, p = 0.014) predicted VS achievement. The 12‐month probability of reaching CD4/CD8 >1 did not significantly differ between four‐drug and three‐drug groups (58% vs. 74%, p = 0.169). At multivariable analysis, a higher BL CD4/CD8 (aHR 5.89 for each point higher; p = 0.003) and BL CD4 >500/mm3 (aHR 2.25; p = 0.003) predicted the achievement of CD4/CD8 >1. Abstract P059 – Figure 1. Kaplan‐Meier curves estimating the cumulative probability of achieving virological suppression according to pre‐ART viral load. Conclusions: In PHI, INSTI‐including regimens achieved VS in almost all treated patients. In subjects with HIV‐RNA 50 copies/mL). Results: Data from 43 patients with AHI were retrospectively collected: 20 on DTG, 23 not in NODTG. No difference in the follow‐up in the two groups was observed (median 1.8 for both groups). Overall in the cohort 81.4% were Italian and 83.7% males with a median age of 41 years (IQR 31 to 48). Thirty‐one (72.1%) were MSM. The median time from diagnosis to treatment initiation was 12 days [IQR 5 to 28]. Median days from diagnosis to start therapy was 12 [IQR 5 to 28]. Differences between the two groups were reported in Table 1. Three patients (7.0%) had detectable viraemia at the end of follow‐up (EOF) with no difference between the two groups (p = 0.468). Nineteen subjects modified, 15 for simplification, four for toxicity (two on DTG for neurological toxicity, two on elvitegravir for gastrointestinal toxicity). Six patients had transmitted mutations at baseline (none for INI) all in DTG group (p = 0.005). The 184V mutation was detected in two patients on 3TC/ABC, both undetectable at the end of follow‐up. The probability of achieving virological suppression during the follow‐up is shown in Figure 1 (log rank: p = 0.5672). One patient on DTG with 184V achieved virological suppression after 2 years. CD4+ cell count, and CD4+/CD8+ ratio increased significantly within groups at 3, 6, 12, 24 and 36 months (p 38°, n (%) 10 (50.0) 9 (39.1) 0.430 Lymphoadenopathy, n (%) 10 (50.0) 12 (52.2) 0.887 GI symptoms (diarrhoea/vomit), n (%) 5 (25.0) 2 (8.7) 0.149 CD4 0.4 and explained 50% of the variance in the data. The six factors were neuromuscular (Cronbach's α = 0.84), sexual problems (α = 0.82), emotional/mood (α = 0.85), minor illnesses (α = 0.66), skin problems (α = 0.73) and gastrointestinal symptoms (α = 0.75). Recognising the pragmatic benefits of a single symptom scale a forced one‐factor EFA was also run. Stepwise removal of low‐loading items resulted in a 38‐item composite subscale. All items loaded >0.4 and explained 28% of data variance. Preliminary analyses found UK participants reported significantly more bother on the composite score (p 350 cells/mm3), those on DTG‐2DC (3%, 6% and 91%, respectively) and PI‐2DC (1%, 9% and 90%, respectively) were switched with higher CD4 (p 90 min/mL/1.73 m2 in TT vs. 15%, 47% and 38% in DTG‐2DC; 10%, 47% and 43% in PI‐2DC; p 50 copies/mL in 39.4% of patients and >200 copies/mL in 21.1%. The reason for switch was simplification/optimisation in 61.5%, VF in 23.9%, toxicity/intolerance 7.3% and other 6.4%. Eighty of 109 have completed 24 weeks of follow up and of these, 75 (93.75%) had VL 200 copies/mL or two consecutive pVL >50 copies/mL) at Week (W)24. The secondary endpoint was the proportion of patients without virological failure at W48 for patients having completed at least 48 weeks from the switch. Results: Twenty patients were included, including 13 having completed at least 48 weeks from the switch. Fifteen of 20 (75%) were men. The median (IQR) characteristics at baseline were: age, 58 years (52 to 64); time since HIV diagnosis, 26 years (20 to 28) with CD4 nadir, 154 cells/mm3 (56 to 197); CD4 count, 450 cells/mm3 (350 to 551); CD4/CD8 ratio, 0.55 (0.38 to 0.70). All patients had pVL 200 copies/mL at admission, respectively (p = NS). According to current hospital fares in Spain, the hospital stays generated a whole cost of 880.402 €. Conclusions: Non‐AIDS defining infectious diseases were the main reason for admission due to medical conditions. Most subjects initiating or changing their ART during the hospital stay were placed on an INI‐based regimen. Despite having access to free medical attention, most infected subjects admitted their lack to link to care. P075 HIV‐EVOL: changes in ART during hospitalisation from 2009 to 2017 in a tertiary hospital in Madrid (Spain) A Díaz‐de Santiago 1, S De La Fuente1, L Biscari2, P Martin2, C Folguera3 and A Ángel‐Moreno1 1Internal Medicine, Hospital Universitario Puerta de Hierro Majadahonda, Madrid, Spain. 2Medicine, Autónoma de Madrid University, Madrid, Spain. 3Pharmacy, Hospital Universitario Puerta de Hierro Majadahonda, Madrid, Spain Background: Our main objective is to evaluate HIV treatment during hospital stay for any reason at University Hospital Puerta de Hierro (tertiary Public Health System centre) during 2009 to 2017. Materials and methods: Observational retrospective and descriptive study (AEMPS code: EVOL‐VIH. ADS‐TEN‐2018‐01). We used Stata programme (v.12.0) for statistical analyses. Results: We observed 597 admissions from 260 patients in last nine years (840 patients followed up in our HIV unit, admission rate 4.85 per 100 patient‐years). 74.6% were male, with median age of 48 years. Eighty‐five percent Spanish, IDU 38%, MSM 21%, heterosexual 19%. HIV infection median time was 18 years. Mean time since first ART was 16.2 years and mean time since last ART regimen before admission was 2.5 years. Fifty‐one percent AIDS CDC stage. Median time with undetectable HIV plasma load was 11 years. Mean nadir and baseline TCD4 + cell count were 164 and 364, respectively. Baseline CD4/CD8 ratio mean was 0.54. Forty‐five percent showed HCV co‐infection (58% if admissions were analysed instead of patients). Sixty‐one percent had undetectable HIV load before hospitalisation (72% on ART). Twenty‐one percent (126/597) changed ART regimen during hospital stay. Reasons for change were: virological failure (47.2%), naive (29.6%), toxicity (11.2%), drug‐drug interactions (6.4%), simplification (5.6%). ART regimen types were: monotherapy 0%, dual therapy 8%, triple therapy 82.4%, quadruple therapy 9.6%. Dual therapies were used between 2015 and 2017 (from 18.7% in 2015 to 14.3% in 2017). Triple treatments decayed (from 100% in 2009 to 78.6% in 2017, p = 0.040). Four‐drug regimens diminished too (from 25% in 2010 to 7.1% in 2017, p = 0.040). Twenty‐eight percent of patients changed to a single‐tablet regimen (STR) during admission. In triple standard therapies, third drug represented: NNRTI 25%, boosted PI 36%, INSTI 38%, but this proportions varied among study period: NNRTI decreased from 44% in 2009 to 10% in 2017; boosted PI decayed from 55% in 2009 to 20% in 2017; and INSTI increased its proportion from 0% in 2009 to 70% in 2017 (all p values 50% of the transmission events inferred, highlighting the potential of source attribution methods for helping to guide real‐time phylogenetic interventions towards those most likely to transmit. Conclusions: Individuals could be prioritized based on the number of transmissions they are associated with (high outdegree); however, we do not currently know how outdegree changes over time. We are currently identifying whether the key set of transmitters in the outbreak changed over time, and estimating the proportion of infections coming from unsampled individuals. Importantly, the ethics and acceptability of this type of intervention remain to be established. P078 An outbreak of HIV amongst homeless people who inject drugs (PWIDs): an innovative HIV clinical service model adaptation leading to successful clinical outcomes R Metcalfe, C Glover, K Brown, E Peters Brownlee Centre, Gartnaval General Hospital, NHS Greater Glasgow and Clyde, Glasgow, UK Background: Since November 2014, Glasgow has witnessed a significant rise in HIV diagnoses amongst homeless PWIDs, almost exclusively using heroin and cocaine, with 125 new diagnoses. In Glasgow, injecting equipment provision (IEP) is free and provided at multiple sites across the city alongside a comprehensive and accessible addictions service providing free opiate substitution therapy. The existing model of hospital‐based HIV care is not suitable for this cohort and a new service model to target this under‐served population has been implemented. Methods: We reviewed the cohort to describe the epidemic and measured effectiveness of the new service with clinical outcome measures. The model includes: a weekly consultant‐led HIV clinic within the homeless health facility, also providing sexual and reproductive health and soft tissue infection expertise; a dedicated BBV specialist outreach nurse; ART dispensed via community pharmacies. Results: One hundred and twenty‐five PWIDs have been diagnosed with HIV, of whom 112 are confirmed clade C virus with primary NNRTI mutations. There were more new diagnoses in 2017 than 2016 (37 vs. 30), with the majority, 54/125 (43.2%), diagnosed in hospital settings. The mean age is 42 and 46/125 (36.8%) are female. Seventeen of 125 (13.6%) are deceased, 4/125 (3.2%) have moved. Of those with results, 38/117 (32.4%) had avidity 50 11/101 11 Genotype 1A 51/101 51 2 2/101 2 3 17/101 17 Unknown 30/101 29 Previous hepatitis C treatment 1/101 1 Patients on antiretroviral therapy 97/101 96 Viral suppression (viral load 18 years old, to be under ART for at least six months. A semi‐structured questionnaire before and after Doctor Apollo use to evaluate the current interaction modalities with the clinical centre and the degree of satisfaction was administered. Results: The needs of HIV people identified were: a diversified level of required privacy, the possibility to receive the results of blood test in real time, have a contact with the centre even when it is closed, and a leaner appointment management. The instant messaging platform chosen to develop and deliver Doctor Apollo has been Telegram. Thirty‐four patients have been enrolled into the study (four declined to sign informed consent for privacy reason). Characteristics of the study population are listed in Table 1. In one month period 216 messages between patients and Doctor Apollo were exchanged, many of those outside working hours. The most used function was “Results” (20.3%), followed by “Book appointment” (18.8%) and “Start” (15.6%) (Figure 1). Unexpectedly, the “Medication reminder” function has been used in only 10% of days. Abstract P086 –Table 1. Characteristics of the study population Sex 17F, 17M Age, median (min‐max) 46 (21 to 65) Heterosexual, n (%) 15 (44%) Men who have sex with men, n (%) 19 (56%) CD4 mm3, median (min‐max) 527 (170 to 1207) HIV‐RNA copies/mL 18 years old who entered care and started antiretroviral treatment from 2004 to 2016 and followed at least 12 months. Retention to antiretroviral therapy care was defined as having at least one visit each three months of care throughout the entire follow‐up period. Potential risk factors associated with dropout during treatment have been identified by using multivariate logistic regression models. SPSS Statistics v.22.0 was used for statistical analysis. Results: It has been found that the proportion of HIV‐infected patients being under antiretroviral therapy in the cohort during the observation period was maintained at 66% to 85%. Thus, the analysis of ART monitoring has shown that there was a decrease of staying in antiretroviral care among the patients within 2004 to 2016 in Poltava region (from 85% among patients under treatment during one year to 66% among patients receiving ART 12 years), and the AIDS‐death level in patients receiving ART increased (from 7% among patients receiving ART during one year to 20% among patients being under ART during 12 years). Retention in care during ART treatment was best predicted by attending clinic of integrated services with access to opioid replacement therapy (OR 1.1, 95% CI 1.0 to 1.1), social support (OR 1.4, 95% CI 1.0 to 2.1) and evidence of previous tuberculosis (TB) treatment (OR 1.2, 95% CI 1.1 to 2.8). The risk for discontinuing in care was significantly higher in HIV‐infected patients who injected drugs and women who diagnosed during the pregnancy (OR 1.6, 95% CI 1.1 to 1.8; OR 1.9, 95% CI 1.5 to 2.3). Engaged HIV‐positive patients had significantly better mean quality of life scores than unengaged patients in the physical (72.08 ± 21.13 vs. 47 ± 23.14, p = 0.013) and psychological domains (77.12 ± 13.03 vs. 52.22 ± 17.12, p = 0.023). Conclusion: The main factors associated with complete retention to care among HIV‐positive patients under ART in Ukraine were good access to integrated services clinics with opioid replacement therapy, social support, high quality of life scores in the physical and psychological domains, evidence of previous TB treatment. P091 The influence of education and employment on ART adherence among PLWHIV aged 25 to 44 in Tashkent: an exploratory research D Karimov Infectious and Pediatric Infectious Diseases, Tashkent Medical Academy, Tashkent, Uzbekistan Background: At present, there are no reliable data on the level of ART adherence among PLWHIV in Tashkent. Because of unemployment and a low level of education, the influence of social factors on ART adherence is highly relevant. This study was conducted to analyse the relationship between social factors such as employment, education and ART adherence. The study aims to develop recommendations for improving ART adherence in key populations. Materials and methods: The study was conducted at the AIDS Center of Tashkent from August to December 2017 and included 146 HIV‐positive patients. A cluster method for classifying patients by adherence, based on questionnaires, and a retrospective analysis of outpatient card data were used. The patients were surveyed on the basis of the questionnaire which included questions about missed doses, compliance with the time of medicine intake, the connection of taking medications with food intake. Patients had to answer each question by writing “yes” or “no”. Each answer “yes” was estimated from 1 to 3 points. Then the points were summed up and all the respondents were categorised according to the World Health Organization classification of ART adherence: 24 to 23 points (≥95%), good adherence (GA); 22 to 21 (85 to 94%), insufficient adherence (IA); 20 and below (≤84%), low adherence (LA). The participants were divided by the level of education into two groups: having higher education/students in the university (group 1A) and without education (group 1B); and by their employment status: officially working (group 2A) and officially unemployed patients (group 2B). Participants in each group were divided into clusters. Results: The average age was 29.7 years. Seventy‐nine (54.1%) patients were female and 67 (45.9%) male. In the group 1A (n = 60): people with GA 75.0% (n = 45), IA 16.6% (n = 10), LA 8.3% (n = 5); and in the group 1B (n = 86): GA 55.8% (n = 48), IA 24.4% (n = 21), LA 19.8% (n = 17), respectively (p 18 years old, on ART with VL 50 copies/mL. Results: A total of 49 patients were included between January 2008 and December 2017, with the following baseline characteristics at time of switch: 35 men (71%), median (IQR) age 50 (44 to 55) years, time since HIV diagnosis 13 (8 to 22) years, CD4 count nadir 214/mm3 (77 to 312), ART duration 10 (6 to 17) years, duration of viral suppression with VL 50 copies/mL. Of the seven with detectable VL, three were confirmed as virological failures (one at Week 4 and the other two at Week 24) and the other four were classified as blips. Overall, the efficacy of 3TC+DTG at 24 weeks was 98.1%. Of the three patients who presented virological failure, only one drug resistance test was carried out successfully, detecting the mutations K103R and S147G (both mutations that do not limit integrase inhibitor activity). Conclusion: These data suggest that the combination of lamivudine plus dolutegravir in treatment‐experienced and virologically stable patients is a new and attractive simplification strategy for antiretroviral treatment, as effective as the triple therapy. Reference: [1] Cuchetto G, et al. J Antimicrob Chemother. 2017. P095 Long‐term follow‐up of dolutegravir as single antiretroviral agent in patients with suppressed HIV viraemia G Tebano1, L Schneider1, C Soulié2, C Blanc1, S Seang1, M Valantin1, R Tubiana1, A Marcelin2 and C Katlama 1 1Infectious Diseases Department, Pitié‐Salpêtrière Hospital, AP‐PH, Sorbonne University, UPMC Univ Paris 06, INSERM, IPLESP (UMRS 1136), Paris, France. 2Department of Virology, Pitié‐Salpêtrière Hospital, AP‐PH, Sorbonne University, UPMC Univ Paris 06, INSERM, IPLESP (UMRS 1136), Paris, France Background: HIV‐infected patients are exposed to decades of antiretroviral treatment. Drug‐reduced regimens (dual therapy/monotherapy) are increasingly investigated to reduce drug exposure and long‐term toxicity. Dolutegravir (DTG) has been evaluated in monotherapy, due to its potency, long half‐life and high genetic barrier to resistance. However, concerns have been raised about emergence of resistance in case of virological failure. We report here our experience with long‐term follow‐up in patients receiving DTG monotherapy. Materials and methods: This is an ongoing single centre, observational cohort study, that included all HIV‐1 positive/HBV‐negative patients who had been switched to monotherapy with DTG 50 mg once daily (mono‐DTG), after a period of viral suppression ≥six months (VS, 92% efficacy and the few viral failures (12 overall) were not accompanied by the selection of new mutations. Abstract P098 – Table 1. Demographic and epidemiologic determinants of the treatment groups Companion drug(s) ABC/3TC 3TC RPV bDRV b/uATV TFV/FTC Sex M 485 (79%) 34 (72%) 85 (64%) 61 (64%) 34 (58%) 73 (79%) F 128 (21%) 13 (28%) 47 (36%) 44 (36%) 25 (42%) 19 (21%) Ethnicity Caucasian 500 (81%) 42 (89.4%) 121 (91.7%) 80 (84.2%) 47 (79.7%) 82 (89.1%) African 35 (5.7%) 3 (6.4%) 9 (6.8%) 4 (4.2%) 5 (8.5%) 2 (2.2%) Asian 11 (1.8%) 2 (4.2%) 2 (1.5%) 0 0 1 (1.1%) Hispanic/Latino 67 (10.9%) 0 0 6 (6.3%) 6 (10.1%) Other 4 (0.6%) 0 0 5 (5.3%) 1 (1.7%) 1 (1.1%) Risk factor Heterosexual 215 (35%) 19 (40.5%) 46 (34.8%) 37 (38.9%) 22 (37.3%) 20 (21.6%) Male having sex with males 264 (43%) 20 (42.5%) 43 (32.7%) 19 (20%) 18 (30.5%) 35 (38.3%) Intravenous drug addiction 129 (21%) 8 (17%) 41 (31%) 38 (40%) 19 (32.2%) 37 (40.1%) Other 6 (1%) 0 2 (1.5%) 1 (1.1%) 0 0 Mean age, years 49.7 39 47 56.5 54.5 50.3 Abstract P098 – Table 2. Reasons to switch to each specific treatment group ABC/3TC/DTG vs DTG+3TC DTG+RPV DTG+bDRV DTG+b/uATV DTG+TFV/FTC Simplification p 90 (ref.) 1.00 1.00 1.00 60 to 90 1.81 (1.53 to 2.13) 1.53 (1.33 to 1.75) 1.20 (1.04 to 1.39) 10%, suppressed on PI/r‐based cART, were randomised: 205 to immediate switch to DTG (DTG‐I), 210 to deferred switch to DTG at Week (W)48 (DTG‐D). After 96 weeks, DTG‐I and DTG‐D arm participants had received DTG for 96 and 48 weeks, respectively. Primary endpoints (published) demonstrated non‐inferior virological efficacy and significant lipid improvements in the DTG‐I arm at W48. We calculated mean body mass index (BMI, kg/m2) and weight (kg) change over 48/96 weeks (DTG‐I vs. DTG‐D arm) and change within arms. BMI/weight change over time and associated baseline factors were estimated using mixed models with random effects; mixed models were used to compare BMI slopes between arms. Results: Baseline median BMI was 25.9 (IQR 23.7 to 28.3). Mean BMI change from W0 to W48 was +0.272/+0.064 in the DTG‐I/DTG‐D arms, significant for DTG‐I but not DTG‐D (p = 0.003/0.471 respectively); difference between arms was statistically significant (p = 0.008). From W48 to W96 mean BMI change was −0.002/+0.332 on DTG‐I/DTG‐D arms, significant for DTG‐D but not DTG‐I (p = 0.984/0.004 respectively); difference between arms was statistically significant (p = 0.002). Median weight change was statistically significant for W0 to 48 (+0.82 kg in DTG‐I vs. +0.25 kg in DTG‐D; p = 0.008) and W48 to 96 (+0.03 kg in DTG‐I vs. +0.98 kg in DTG‐D; p = 0.002). In multivariable analysis, baseline factors associated with higher BMI gain on DTG were Framingham risk >15% (p = 0.042) and high blood pressure (p = 0.035), while protective factors were switching from PIs other than darunavir or atazanavir (p = 0.032), current smoking (p = 0.006), daily exercise (p = 0.036) and HDL‐cholesterol (p 6 months) viraemia 1000 copies/mL after 16 weeks of therapy, or VL >400 copies/mL after 24 weeks or confirmed (two consecutive) VL >50 copies/mL after 48 weeks or after having been suppressed. Blips were excluded. Suboptimal therapies consisted of a single‐NRTI or 2NRTIs. Cox models were used to estimate the effect of previous virologic failure or suboptimal therapies on post‐switch virologic failure (defined as any confirmed VL >50 copies, or the last VL available >50 copies/mL). Hazard ratios (HR) were adjusted for age at switch. Results: A switch to DTG + 2NRTIs was observed among 1209 patients who had undetectable VLs for ≥6 months before treatment change. These patients had a median age of 50.9 years (IQR 43.4 to 57.9) and CD4 count of 660 cells/µL (IQR 500 to 843), respectively. The median number of previous regimens before switch was 5 (IQR 3 to 9). Among these patients, 496 (41.0%) had pre‐switch experienced at least one virologic failure or suboptimal therapy in their prior antiretroviral treatment history whereas 713 (58.9%) did not. Mean follow‐up time after DTG switch was 531.0 days (SD 321.8) and 592.8 days (SD 316.4) among patients with and without previous virologic failure or suboptimal therapy, respectively. Post‐switch virologic failure was observed in 13 (2.8%) patients with previous virologic failure or suboptimal treatment and 18 (2.6%) patients without previous virologic failure or suboptimal therapies. The crude and adjusted HRs for the association between post‐switch virologic outcomes and previous suboptimal therapies or virologic failure were respectively 1.21 (95% CI 0.59 to 2.48) and 1.57 (95% CI 0.74 to 3.30). Conclusion: In this study, patients with a history of virologic failure or suboptimal therapies did not experience significantly increased risks of virologic failures when switched to DTG + 2NRTIs. P107 Efficacy of rilpivirine‐based regimens as switch therapy from nevirapine‐based regimens in HIV‐infected patients with complete virological suppression: a randomised controlled trial P Petchkum 1, S Sungkanuparp2, S Kiertiburanakul1 and A Phuphuakrat1 1Dept of Medicine, Division of Infectious Diseases, Ramathibodi Hospital, Faculty of Medicine, Bangkok, Thailand. 2Faculty of Medicine, Chakri Naruebodindra Medical Institute, Ramathibodi Hospital, Samut Prakan, Thailand Background: Nevirapine (NVP)‐based ART remains to be used in HIV‐infected patients in resource‐limited countries despite its compliance and adverse effect concerns. Rilpivirine (RPV), a newer non‐nucleoside reverse transcriptase inhibitor, could be used as an alternative to NVP in virologically suppressed patients. However, there has been limited experience with switching from NVP‐based to RPV‐based regimens. We aimed to study efficacy and adverse events after ART switching from NVP‐based to RPV‐based regimens. Methods: A randomised controlled non‐inferiority trial was conducted in HIV‐infected patients who received NVP‐based regimens and had undetectable plasma HIV RNA for more than 6 months. Patients were randomised 1:1 to continuation arm (NVP‐based regimens were continued) or switch arm (NVP‐based regimens were switched to RPV‐based regimens). Tenofovir disoproxil fumarate (TDF) plus lamivudine (3TC) or emtricitabine (FTC) remained as backbone of the regimens. Primary endpoint was HIV RNA 0.05). At 24 weeks, 53 patients (96.36%) in the continuation arm and 49 patients (96.07%) in the switch arm had virological success. The switch arm was non‐inferior to the continuation arm [efficacy difference 0.29%, 95% CI −6.98% to 7.56%, p > 0.999]. Both regimens were generally well tolerated, although one patient developed gastrointestinal adverse events and resumed NVP‐based regimen. A significant decrease in mean total cholesterol was observed in the switch arm (−15.33 mg/dL, 95% CI −24.07 to −0.60, p = 0.001). Conclusions: In HIV‐infected patients virologically suppressed with NVP‐based regimens, once‐daily RPV‐based regimens are an alternative switch option. These regimens can maintain virological suppression and decrease total cholesterol. Further study of long‐term efficacy of this switching strategy should be pursued. P108 Patient‐reported outcomes in an observational cohort of adult HIV‐1 positive patients after 48 weeks of treatment of darunavir/cobicistat‐based regimens (TMC114FD1HTX4003: ST.O.RE. study) A Antinori 1, A Gori2, D Ripamonti3, S Rusconi4, N Gianotti5, R Maserati6, A Muscatello7, V Di Cristo8, A Castagna9, G Rizzardini10, A Cattelan11, B Menzaghi12, G Sterrantino13, S Kiros14, F Castelli15, E Focà15, B Saccani15, G Orofino16, M Farenga16, R Cauda17, S La Monica17, V Vullo18, A De Luca19, B Rossetti20, E Manzillo21, C Gioè22, B Celesia23, M Locatelli23, G Madeddu24, P Bagella24, T Santantonio25, S Ferrara25, L Cosco26, E Pontali27, A d'Arminio Monforte28, R Curetti29, M Andreoni30, C Stingone30, A Uglietti31, R Termini31 and D Mancusi31 1HIV/AIDS Department, National Institute for Infectious Diseases Lazzaro Spallanzani IRCCS, Roma, Italy. 2Infectious Diseases Unit, Fondazione IRCCS Ca’ Granda, Ospedale Maggiore Policlinico, University of Milan, Milano, Italy. 3Division of Infectious Diseases, ASST Papa Giovanni XXIII, Bergamo, Italy. 4Divisione Malattie Infettive, DIBIC Luigi Sacco, Università degli Studi di Milano, Milano, Italy. 5Dipartimento di Malattie Infettive, San Raffaele Scientific Institute, Milano, Italy. 6Clinica Malattie Infettive, Fondazione Policlinico San Matteo, Pavia, Italy. 7Infectious Diseases, Fondazione IRCCS Ca’ Granda, Ospedale Maggiore Policlinico, Milano, Italy. 8Infectious Diseases, DIBIC Luigi Sacco and University of Milan, Milano, Italy. 9Unit of Management & ARV Treatment of HIV Infection, Faculty of Medicine and Surgery, Vita‐Salute San Raffaele University and IRCCS San Raffaele Hospital, Milano, Italy. 101st Division of Infectious Diseases, ASST Fatebenefratelli‐Sacco, Milano, Italy. 11Division of Infectious and Tropical Diseases, Azienda Ospedaliero‐Universitaria di Padova, Padova, Italy. 12Infectious Diseases, Azienda Socio Sanitaria Territoriale della Valle Olona‐Busto Arsizio, Busto Arsizio, Italy. 13Division of Tropical and Infectious Diseases, Azienda Ospedaliero‐Universitaria Careggi, Firenze, Italy. 14Infectious Diseases, University of Firenze, Firenze, Italy. 15Department of Infectious and Tropical Diseases, University of Brescia and Spedali Civili General Hospital, Brescia, Italy. 16Unit of Infectious Diseases, Amedeo di Savoia Hospital, Torino, Italy. 17Institute of Clinical Infectious Diseases, Catholic University of the Sacred Heart of Roma, Roma, Italy. 18Department of Public Health and Infectious Disease, Sapienza University, Roma, Italy. 19Department of Medical Biotechnologies, University of Siena, Siena, Italy. 20University Division of Infectious Diseases, Siena University Hospital, Siena, Italy. 21VIII Divisione di Malattie Infettive, A.O.R.N. Cotugno, Napoli, Italy. 22Infectious Diseases Division, Policlinico Universitario, Palermo, Italy. 23UOC Malattie Infettive, ARNAS “Garibaldi”, Catania, Italy. 24Unit of Infectious Diseases, University of Sassari, Sassari, Italy. 25Department of Clinical and Experimental Medicine, University of Foggia, Foggia, Italy. 26Unit of Infectious Diseases, Pugliese‐Ciaccio Hospital, Catanzaro, Italy. 27Infectious Diseases, Ospedali Galliera, Genova, Italy. 28Dipartimento di Scienze della Salute, ASST Santi Paolo e Carlo, Milano, Italy. 29Clinica Malattie Infettive e Tropicali, ASST Santi Paolo e Carlo, Milano, Italy. 30Department of Medicine of Systems, University of Roma “Tor Vergata”, Roma, Italy. 31Medical Affairs, Janssen‐Cilag SpA, Cologno Monzese, Italy Background: Current ART for the treatment of HIV‐1‐infected patients has the objective to provide control of viral load while simplifying drugs’ administration. Darunavir/cobicistat (DRV/c) is a fixed‐dose combination including the protease inhibitor (PI) DRV and its booster cobicistat, developed to reduce pill burden, making the PI easier to take and possibly avoiding mistakes in drug administration. The ST.O.RE. study was designed to prospectively collect clinical practice data on effectiveness of ART DRV/c‐based and to assess also the patients’ satisfaction and actual symptoms by means of two validated patient‐reported outcomes (PROs) questionnaires: HIV Treatment Satisfaction (HIV‐TSQ) and HIV Symptoms Distress Module (HIV‐SDM). Materials and methods: Three hundred and forty‐eight virosuppressed patients, coming from a PI/r‐based ART and switching to DRV/c were enrolled in this single‐arm, prospective, non‐interventional study. Of them, 336 were evaluable. The questionnaires were administered to patients at fixed study timepoints: baseline (Visit [V]1); after 4 to 8 weeks from baseline (V2); and at last visit (V4) 48 ± 6 weeks from baseline. For HIV‐TSQ, an improvement is shown by an increased score, while for HIV‐SDM questionnaire, symptoms improvement is connected to a decreased score. Here we show the results obtained comparing the questionnaires scores registered at study baseline (V1) with the scores at last visit (48 weeks of treatment, V4). Results: In total, 250 patients (174 males and 76 females) filled both the HIV‐TS and the HIV‐SDM questionnaires at baseline and at V4. Their demographic characteristics are shown in Table 1. Regarding the HIV‐TSQ, the patients’ satisfaction for their treatment was good at enrolment (67% of patients were very satisfied) but it further improved in 87.4% of patients at V4. Regarding the HIV‐SDM scores, the overall burden of symptoms decreased from V1 to V4: the mean value (SD) was 10.1 (9.9) at baseline, while it was 9.3 (10.4) at V4 (overall). Gender (p = 0.0055) and SDM baseline score (p 190 mg/dL and/or LDL cholesterol >115 mg/dL, hypertriglyceridaemia as TG >180 mg/dL. Plasma concentrations were measured by a validated HPLC methods. FIB‐4 score was calculated as (age*AST)/(platelets*SQR(ALT)). Variables were described as medians (interquartile range [IQR]) or number (percentage) and were compared using paired non‐parametric tests (Wilcoxon and McNemar's). Results: We included 121 patients: baseline characteristics are shown in Table 1. Fifteen (12.3%) patients discontinued. BL and W24 HIV RNA was 5%) were prevention of renal/bone toxicity (51%), use of TAF as TDF successor drug (12%), adverse drug reactions (ADRs) on prior ART (10%), ART simplification (9%) and patient request (7%). Until study end, seven patients (5%) discontinued TAF‐based regimens (four patients due to ADRs (3%)); in addition, four study discontinuations (unrelated to the use of TAF‐based ART) were reported. Kaplan‐Meier estimates of persistence on TAF at months 6, 12 and 18 were 96%, 92% and 92%, respectively. At last follow‐up, after a median of 7.6 months on TAF‐based ART (IQR 4.9 to 11.9, max. 20.3), HIV‐RNA levels were 100,000 prior to ART initiation, N (%) 54 (36) 32 (55) 9 (18) 13 (32) Late presentation at ART initiationb, N (%) 70 (47) 36 (62) 14 (27) 20 (49) aother than RPV/FTC/TAF or EVG/COBI/FTC/TAF. bCD4 100,000 copies/mL HIV‐1 RNA. Two hundred and eighty‐six patients were randomized into the MP, 258 completed MP with 252 entering EP. At W160, 90% (104/115; Q8W) and 83% (95/115; Q4W) of randomized IM patients remained suppressed 100,000 copies (c)/mL or CD4 100,000 c/mL or CD4 count 100,000 c/mL (B/F/TAF n = 95/634 [15%], DTG/ABC/3TC n = 43/315 [14%], DTG + F/TAF n = 46/325 [14%]), and 122 (B/F/TAF n = 65/634 [10%], DTG/ABC/3TC n = 26/315 [8%], DTG + F/TAF n = 31/325 [10%]) had baseline CD4 count 200 copies/mL in patients with a prior VL 20 copies/mL after six months in 4/73 cases (5%). In situation of virological failure, change guided by DNA genotyping led to VL 100,000 copies/mL, n (%) 17 (27.4) NA HIV‐RNA level 500 for seven years off ART) and EC (undetectable VL off ART, ≥2 readings in 12 months). Data were collected by medical staff using a proforma, including demographics, disease progression, ART start and reasons for starting. Analysis is descriptive, using Excel and Stata. Results: We identified 52 individuals; median age at diagnosis 32 years (IQR 28 to 38), 62% male. HIV was diagnosed between 1982 and 2012, with median baseline VL 779 copies/mL (IQR 326 to 10,300) and CD4 count 680 cells/µL (IQR 521 to 919). Available follow‐up time was median 13 years (range 7 to 35). The latest median VL was 500 copies/mL 1 (1.7%) 0 Lost 1 (1.7%) 2 (7.7%) AIDS event 0 0 Non‐AIDS event 4 (6.9%) 4 (15.4%) Death 0 1 (3.8%) References [1] Antiretroviral Therapy Cohort Collaboration (ART‐CC); Vandenhende MA, Ingle S, May M, Chene G, Zangerle R, Van Sighem A, et al. Impact of low‐level viremia on clinical and virological outcomes in treated HIV‐1‐infected patients. AIDS. 2015;29:373‐83. [2] AIDSinfo (Department of Health and Human Services). Guidelines for the use of antiretroviral agents in HIV‐1‐infected adults and adolescents [Internet]. Available from: http://www.aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf. [3] European AIDS Clinical Society (EACS). Guía Clínica EACS [EACS Guidelines] versión 9.0 [Internet]. 2017. Available from: http://www.eacsociety.org/files/guidlines‐9.0‐spanish.pdf. [4] Bernal E, Gómez JM, Jarrín I, Cano A, Muñoz A, Alcaraz A, et al. Low level viremia is associated with clinical progression in HIV‐infected patients receiving antiretroviral treatment. J Acquir Immune Defic Syndr. 2018;78:329‐37. P131 When to start antiretroviral therapy in HIV‐2: the challenge remains M Cardoso, B Pimentel, J Granado, J Vasconcelos, A Miranda, S Peres, T Baptista, K Mansinho Serviço de Infeciologia e Medicina Tropical, Hospital de Egas Moniz, Centro Hospitalar de Lisboa Ocidental, Lisbon, Portugal Background: The prevalence of HIV‐2 in Portugal is 3.3% [1]. HIV‐2 treatment is limited, facing intrinsic resistance to NNRTI and fusion inhibitors, and different response to protease inhibitors. While current guidelines for HIV‐1 recommend treatment for all, that endpoint is not properly defined for HIV‐2 [2,3]. Materials and methods: Retrospective observational study of HIV‐2 patients diagnosed between 1985 and 2017 followed at an infectious disease clinic. Statistical analysis processed by Microsoft Excel. Results: This cohort included 121 patients. Female predominance (66%) was observed and mean age was 58 years. Most patients (69%) are originally from west Africa and 29% are Portuguese. Mean time since diagnosis was 15 years and main reasons that led to diagnosis were: routine blood screen (41%) and pregnancy (21%). The most frequent transmission route was heterosexual contact (88%). At diagnosis, 79% were asymptomatic, mean TCD4 + count was 515 cells/mm3 and 53% had undetectable HIV‐2 RNA ( 80% and low discontinuation rates at primary endpoint, DTG‐based regimens (DBR) are a recommended first‐line treatment option in international HIV treatment guidelines. Climate‐HIV is an electronic patient record designed to support the management of PLWHIV. The retrospective study aims to describe the real‐world use, effectiveness and safety of DTG in routine clinical practice in several UK HIV services. Methods: Patients prescribed a DBR were included if they were aged ≥18 years and attended one of four large UK HIV units using Climate‐HIV between December 2012 and February 2018. Data for these patients were included from DBR initiation until their last recorded visit, i.e. until subjects were lost to follow‐up (LTFU) or switched from or discontinued DTG. Data regarding demographics, ARV regimens, virological outcomes, adverse events (AEs) and reasons for switch were collected. Results: Nine hundred and thirty‐four of 5590 (17%) patients were prescribed a DBR, 880 (94%) once daily. Two hundred and seventeen (23%) were naive and 717 (77%) were treatment experienced (TE) at DBR initiation. Baseline characteristics are shown in Table 1. The most commonly prescribed DBR was DTG/ABC/3TC (Triumeq®) in 533 (61%) of 880. Of the 934 patients initiating DBR, 809 (87%) remained on DBR at last visit; the median duration was 12.6 months [interquartile range (IQR) 4.4 to 22.8]. Ninety‐four (10%) patients switched to a non‐DBR and 31 (3%) patients discontinued ARV treatment without restarting and were classed as LTFU. Median time to switching off DBR was 101 days (IQR 30 to 245); common reasons for switching included adverse events [16/934 (1.7%)], clinician's decision [15/934 (1.6%)] and cost reduction [13/934 (1.4%)]. At last visit, 797/934 (85.3%) patients had an undetectable viral load. Abstract P134 – Table 1. Baseline characteristics Overall (N = 924) DTG as first regimen N = 217 (23.5%) Subjects who switched to a DTG regimen N = 717 (77.5%) Sex, male, n (%) 597 (63.9) 159 (27) 438 (73) Age, years, median (IQR) 43 (34 to 51) 37 (30 to 45) 44 (35 to 52) Median time (m) since diagnosis to initiation of DTG 90.1 (21 to 161) 8.5 (0.8 to 57) 119.5 (44 to 177) Proportion of black and minority ethnicities, n (%) 520 (57) 107 (21) 413 (79) First database recorded HIV viral load 18278.5 (589 to 100509) 24945 (3637.5 to 123800) 14648 (168 to 96152) Median HIV‐1 RNA at start of DTG regimen (IQR) (copies/mL) 40 (40 to 6868) 23801 (4177 to 115848) 40 (40 to 107) First database recorded CD4 count (IQR) 258 (20 to 491) 309 (1 to 564) 240 (39 to 459) Median CD4 at start of DTG regimen (IQR) (cells/mm3) 258 (20 to 491) 309 (0.8 to 564) 240 (39 to 459) Age 200 copies/mL) 88 (9.42%) 34 (15.67%) 54 (7.53%) Most common DTG ART‐combination partners Abacavir/lamivudine, n (%) 533 (61) 136 (26) 397 (74) Emtricitabine/tenofovir, n (%) 178 (20) 43 (24) 135 (76) Abstract P134 – Figure 1. Kaplan‐Meier analysis for time to switch or discontinue for DTG based regimens. Conclusion: This study observed low discontinuation and high effectiveness of DBR in a diverse population within the UK. These real‐world data are broadly consistent with the safety and efficacy data from Phase III clinical trials. Overall persistence of DTG was 87% over a median duration of 12.6 months. P135 NNRTI raltegravir lamivudine (NRL): the NatuRAL choice for ageing patients I Cormack HIV Department Heath Clinic, Croydon University Hospital, Croydon, UK Background: Ageing patients with HIV may not be suitable for abacavir (ABC)‐ or tenofovir (TDF)‐based HAART. There is a growing need for new treatment strategies. Methods: The joint HIV renal clinic identified 23 HIV‐positive patients on NNRTI raltegravir lamivudine (3TC) HAART and one patient who was on NNRTI + raltegravir (no 3TC). We calculated eGFR using CKD‐EPI on all patients pre‐starting NRL and compared this to current eGFR or eGFR before switching off NRL. Demographic, clinical and baseline data were also collected from patient records. There are 11 men and 13 women average age 64 years (31 to 94) 88% over age 50 years. Ten of 24 (42%) diabetic, 14/24 (58%) hypertensive, 4/24 (16%) HIVAN and 11/24 (46%) had CKD3 or worse pre‐switch to NRL. Fifteen of 24 (58%) had a baseline HIV VL >100,000. Results: All 24 patients are currently virologically suppressed on NRL including one patient who is not on 3TC. Total length of time on NRL is 3121 weeks, average 130 weeks (19 to 375) 18/24 > 48 weeks (75%). Three patients were new starters with pre‐treatment VLs 12,436, 22,000, 60,500 and have been on NRL for 92, 213 and 251 weeks respectively. Ten switched to NRL from TDF and 11 switched from ABC or protease inhibitors (PIs). NNRTIs: efavirenz (14) nevirapine (five) rilpivirine (three) etravirine (two). Pre‐switch 11/24 (46%) had eGFR 60 mL/min on NRL. Post‐TAF STR switch all showed loss in eGFR on average 18 mL/min (9 to 27) with average length TAF STR treatment 55 weeks (33 to 84 weeks). Three of four (75%) of these patients now have an eGFR 20 copies/mL (5.4% >50 copies/mL) and 6.7% maintained a LLV >20 copies/mL (1.8% >50 copies/mL). A mean VL of 87,977 copies/mL was observed in the eight (3.5%) experienced non‐suppressed patients. Virological suppression below 50 copies/mL at Week 48 was seen in 75.0% and 62.5% under 20 copies/mL. Of the patients with a VL >100,000 copies/mL (24 naïve and two experienced; 65% under regimens with ABC/3TC), six (23.1%) were suppressed 100,000 (copies/mL), n (%) 201/250 (74) 84/124 (68) 117/126 (93) HIV‐RNA >500,000 (copies/mL), n (%) 60/250 (22) 32/124 (26) 28/126 (22) avalues are expressed as n (%) except for median (IQR). Abstract P139 – Table 2. Diagnosed AIDS‐defining events, IRIS and deaths according to antiretroviral regimen Overall (n = 272) (%) ART with DTG (n = 132) (%) ART without DTG (n = 140) (%) Pneumocystis jirovecii pneumonia 23 (8) 22 (17) 1 (1) Pulmonary tuberculosis 22 (8) 5 (4) 17 (12) Progressive multifocal leukoencephalopathy 18 (7) 3 (2) 15 (11) Wasting syndrome 16 (6) 5 (4) 11 (8) Candida oesophagitis 15 (5) 12 (9) 3 (2) Cytomegalovirus symptomatic infection 12 (4) 12 (9) 0 Non‐Hodgkin lymphoma 12 (4) 1 (1) 11 (8) Kaposi's sarcoma 11 (4) 7 (5) 4 (3) Cerebral toxoplasmosis 9 (3) 6 (4) 3 (2) Cryptosporidium infection 7 (3) 2 (1) 5 (3) Cervical cancer 5 (2) 1 (1) 4 (3) AIDS dementia complex 4 (2) 3 (2) 1 (1) Disseminated herpetis virus infection 4 (2) 0 4 (3) Disseminated cryptococcosis 3 (1) 1 (1) 2 (2) Non‐tubercular mycobacteriosis 2 (1) 1 (1) 1 (1) IRIS 13 (5) 5 (4) 8 (6) Deaths 10 (4) 5 (4) 5 (4) Conclusions: The results confirm the high potency, the good tolerability and safety of DTG in advanced naïve patients, with a low risk for IRIS. P140 Malaria in HIV‐infected patients: a matched case‐control study in a non‐endemic setting E Lam 1, S De Wit2, M Delforge2 and C Martin2 1Internal Medicine, ULB, Brussels, Belgium. 2Infectious Diseases, CHU Saint‐Pierre, Brussels, Belgium Background: The impact of HIV infection on malaria is unclear in malaria non‐endemic areas. In endemic territories, it has been reported to be a risk factor for acquisition of malaria higher morbidity and parasitaemia and malaria treatment failure. In the context of HIV‐infected patients having a better quality of life and travelling more, in particular as VFR (visiting friends and relatives) represent a large proportion of travelling HIV‐infected patients, it is important to assess the impact of HIV on imported malaria. Material and methods This unicentric retrospective case‐control study collected data on HIV‐infected patients with malaria defined as positive thick smears with P. falciparum identification on microscopy and matched them with two controls based on age, sex and ethnicity. Clinical and biological parameters were collected and compared and different severity scores were applied on cases and controls. Results: We identified 47 cases and 94 controls. Malaria prophylaxis use and delay before medical contact did not differ between cases and controls. Comparing each of the World Health Organization (WHO) 2014 severity criteria, hyperparasitaemia above 10% (p = 0.006), icterus (p = 0.042), acute renal failure (p = 0.022) and bacteraemia (p = 0.014) were significantly more present in HIV‐infected patients than in controls, with a trend to more neuromalaria (12.8% vs. 6.4%). The severity of malaria (defined by WHO 2014) was inversely related to CD4 T‐cell count. HIV‐infected patients were hospitalised more frequently and for longer periods and tended to stay more in intensive care unit. Death rate was significantly higher in cases (6.4% vs. 0). De novo HIV diagnosis was obtained in 17% of cases during the malaria episode. These differences in occurrence of severe malaria disappeared when patients with CD4 T‐cells count >500/µL (n = 9) were compared to controls but small size of this subgroup does not allow to draw firm conclusion. Conclusion: HIV infection has an impact on the imported malaria profile, as well as in endemic areas. It is unclear if well‐controlled HIV‐infected patients have a higher risk of developing severe malaria. HIV‐infected patients should be particularly targeted for pre‐travel advice and physicians should always perform HIV testing during a malaria episode. P141 Concomitant syphilis infection in patients with diagnosed HIV/AIDS: a retrospective multicentre study F Sarigul 1, M Sayan2,3, D Inan4, A Deveci5, N Ceran6, M Celen7, A Cagatay8, H Ozkan Ozdemir9, F Kuscu10, G Karagoz11, Y Heper12, O Karabay13, B Dokuzoguz14, S Kaya15, N Erben16, I Karaoglan17, G Munis Ersoz18, O Gunal19, C Hatipoglu20, S Sayın Kutlu21, A Akbulut22, R Saba23, A Sener24 and S Buyuktuna25 1Infectious Disease and Clinical Microbiology, Health Sciences University, Antalya Education and Research Hospital, Antalya, Turkey. 2Clinical Laboratory, PCR Unit, Kocaeli University, Faculty of Medicine, Kocaeli, Turkey. 3Research Center of Experimental Health Sciences, Near East University, Nicosia, Northern Cyprus. 4Infectious Disease and Clinical Microbiology, Akdeniz University, Faculty of Medicine, Antalya, Turkey. 5Infectious Disease and Clinical Microbiology, Samsun 19 Mayıs University, Faculty of Medicine, Samsun, Turkey. 6Infectious Disease and Clinical Microbiology, Health Sciences University, Istanbul Haydarpasa Numune Education and Research Hospital, Istanbul, Turkey. 7Infectious Disease and Clinical Microbiology, Diyarbakir University, Faculty of Medicine, Diyarbakir, Turkey. 8Infectious Disease and Clinical Microbiology, Istanbul Unıversity, Faculty of Medicine, Istanbul, Turkey. 9Infectious Disease and Clinical Microbiology, Health Sciences University, Izmir, Education and Research Hospital, Izmir, Turkey. 10Infectious Disease and Clinical Microbiology, Cukurova University, Faculty of Medicine, Adana, Turkey. 11Infectious Disease and Clinical Microbiology, Health Sciences University, Istanbul Umraniye Education and Research Hospital, Istanbul, Turkey. 12Infectious Disease and Clinical Microbiology, Bursa University, Faculty of Medicine, Bursa, Turkey. 13Infectious Disease and Clinical Microbiology, Sakarya University, Faculty of Medicine, Sakarya, Turkey. 14Infectious Disease and Clinical Microbiology, Health Sciences University, Ankara Numune Education and Research Hospital, Ankara, Turkey. 15Infectious Disease and Clinical Microbiology, Karadeniz Technical University, Faculty of Medicine, Trabzon, Turkey. 16Infectious Disease and Clinical Microbiology, Eskisehir University, Faculty of Medicine, Eskisehir, Turkey. 17Infectious Disease and Clinical Microbiology, Gaziantep University, Faculty of Medicine, Gaziantep, Turkey. 18Infectious Disease and Clinical Microbiology, Mersin University, Faculty of Medicine, Mersin, Turkey. 19Infectious Disease and Clinical Microbiology, Health Sciences University, Samsun Education and Research Hospital, Samsun, Turkey. 20Infectious Disease and Clinical Microbiology, Health Sciences University, Ankara Education and Research Hospital, Ankara, Turkey. 21Infectious Disease and Clinical Microbiology, Pamukkale University, Faculty of Medicine, Denizli, Turkey. 22Infectious Disease and Clinical Microbiology, Elazıg University, Faculty of Medicine, Elazıg, Turkey. 23Infectious Disease and Clinical Microbiology, Private Medstar Antalya Hospital, Antalya, Turkey. 24Infectious Disease and Clinical Microbiology, Onsekiz Mart University, Faculty of Medicine, Canakkale, Turkey. 25Infectious Disease and Clinical Microbiology, Cumhuriyet University, Faculty of Medicine, Sıvas, Turkey Background: Treponema pallidum and HIV are both sexually transmitted agents of these infectious diseases with epidemiological similarities, and therefore co‐infect the same host [1,2]. The current number of HIV‐infected cases in Turkey is increasing significantly. For this reason, we aimed to reveal the syphilis status in HIV‐1 positive cases. Materials and methods: A retrospective descriptive case series study, patients (aged ≥18 years) were followed up by 24 clinics from 16 cities of Turkey between January 2010 and April 2018. Those clinics are from all seven regions of Turkey. We examined the sociodemographic, including age, gender, nationality, marital status, partner HIV status, education, number of sexual partners, probable transmission route of HIV and syphilis infections, years of HIV seropositivity, clinical stage of syphilis and HIV/AIDS, use of ART and laboratory parameters such as results of CD4 + T lymphocyte count and HIV viral load, quantative venereal diseases laboratory (VDRL) and T. Pallidum haemagglutination assay (TPHA) results and neurosyphilis association. Results: A total of 3641 patients were followed with HIV‐1 infection, and 291 (8%) patients were diagnosed with syphilis co‐infection during eight years (Table 1). Most patients were older than 25 years old (92%) including 96% were males, 74% were working, 23% were unemployed and 3% were students. Laboratory characteristics of patients with HIV/AIDS/syphilis is shown in Table 2. The sexual predilections consisted of heterosexual (46%), homosexual (23%), bisexual (19%) and no data (36%). The three highest prevalence of syphilis were in Marmara (33%), Mediterranean (26%), Black Sea Regions (18%) (Figure 1). Abstract P141 – Table 1. Demographic characteristics of patients with HIV/AIDS/syphilis and probable routes of transmission Characteristic Study group Patient, HIV/AIDS/syphilis, n (%) 3641/291 (8) Gender, n (%) Female, Male 13 (4), 278 (96) Age, median years (range) 41 (18‐90) Ethnicity, n (%) Turkish, Othersa 279 (96), 12 (4) Socio‐economic status, n (%) Employed, Unemployed, Student 214 (74), 67 (23), 10 (3) Education, n (%) Illiterate, Primary School, Secondary School, High School, University 3 (1), 91 (31), 39 (13), 97 (33), 63 (22) Marital status, n (%) Married, Single, Widowed 106 (36), 179 (62), 6 (2) Partner HIV status, n (%) Negative, Positive, Unknown 28 (10), 250 (86), 13 (4) Number of sexual partners, n (%) Single, Multiple, Unknown 28 (10), 250 (86), 13 (4) Condom usage, n (%) Yes, No, Unknown 22 (8), 166 (57), 103 (35) aother nationality; Macedonia (n = 2), Turkmenistan (n = 2), Thailand (n = 1), Germany (n = 1), South Africa (n = 1), Kenya (n = 1), Brazil (n = 1), Egypt (n = 1), Russian Federation (n = 1), Afghanistan (n = 1). Abstract P141 – Table 2. Laboratory characteristics of patients with HIV/AIDS/syphilis Characteristic Study group Patient, HIV/AIDS/syphilis, n (%) 3641/291 (8) Duration of HIV and syphilis diagnosis, n (%) At the same time, 1 to 2 years, 3 to 4 years, 4 years 211 (73), 26 (9), 25 (8), 29 (10) Antiretroviral usage, n (%) Yes, No 50 (17), 241 (83) AIDS, n (%) Yes, No 45 (15), 246 (85) Basal HIV‐1 RNA, median copy/mL (range) 9.7 + E5 (58‐1.0 + E9) Basal CD4 + T lymphocyte count, median cell/mm3 (range) 384 (5 to 1520) Late presenter status, n (%) ≤350 cells/mm3, >350 cells/mm3 134 (46), 157 (54) Serum VDRL, median titer (range) 1/64 (1/2‐1024) Serum TPHA, median titer (range) 1/640 (1/40‐1/5240) Syphilis stage, n (%) Primary, Secondary, Latent 30 (10), 65 (22), 196 (68) Neurosyphilis, n (%) Yes, No 25 (9), 266 (91) P141 – Figure 1. Distribution by region of HIV/syphilis patients. Conclusion: Turkey poses a risk in terms of geographical position about sexually transmitted diseases. On the other hand, Turkey is on international travel and trade route. Some countries with a high incidence of HIV/AIDS are also close. Additionally, syphilis and HIV/AIDS were closely related [3,4]. HIV/AIDS could increase the incidence of syphilis and vice versa, the increasing of syphilis could elevate HIV/AIDS. References [1] Braxton J, Davis D, Flagg E, Grey J, Grier L, et al. US Department of Health and Human Services Centers for Disease Control and Prevention, National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention, Division of STD Prevention [Internet]. Atlanta, Georgia, 30329‐4027 [cited 2017 Sep]. Available from: http://https://www.cdc.gov/std/stats. [2] European Centre for Disease Prevention and Control. Syphilis. In: ECDC. Annual epidemiological report for 2016 [Internet]. Stockholm: ECDC; 2017 [cited 2018 Apr 19]. Available from: https://ecdc.europa.eu/en/annual‐epidemiological‐reports. [3] European AIDS Clinical Society (EACS) guidelines for the clinical management and treatment of HIV‐infected adults [Internet]. 2017 [cited 2018 Apr 30]. Available from: http://www.eacsociety.org/files/guidelines_9.0‐english revised. [4] Harnanti DV, Hidayati AN, Miftahussurur M. Concomitant sexually transmitted diseases in patients with diagnosed HIV/AIDS: a retrospective study. Afr J Infect Dis. 2018;12:83‐9. P142 Overview of epidemiological, clinical and therapeutic features of visceral leishmaniasis/HIV co‐infection in Albania A Gjataj1, A Harxhi 2, E Meta3 and D Kraja2 1University Hospital Center of Mother Theresa, Department of Infectious Diseases Service, Tirana, Albania2University Hospital Center of Tirana, Faculty of Medicine, Infectious Disease, Tirana, Albania3University Hospital Center of Tirana, Infectious Disease, Tirana, Albania Background: Cases of visceral leishmaniasis (VL) in the course of HIV have regularly been recorded especially in endemic areas such as the Mediterranean. It is a zoonotic infection developed in HIV patients who suffer from severe immunosuppression. Materials/methods This is a retrospective study which analyses epidemiological, clinical and therapeutic data of VL/HIV co‐infection from 2007 to 2017 in 12 HIV patients followed up at ambulatory HIV clinic at University Hospital Center of Tirana “Mother Theresa”. Results: We describe the occurrence of this co‐infection in 12 patients from Albania. In total there were 66% males and 34% females and the mean age was 42 years old (range 25–50 years old). In half number of patients VL was diagnosed at the same time with HIV infection and only one case was reported as VL before HIV diagnosis. Mean CD4 at the first episode of VL was 14.5%‐135.6 cells/mm3. The main clinical signs and symptoms at admission were: fever (75%), sweat (83%), weight loss (91%). In objective examination we observed lymphadenopathy (83%), splenomegaly in all cases and hepatomegaly in 91% of them. Laboratoric findings were: anaemia in 83% of patients, leukopenia (91%), thrombocytopenia (66%) and increased levels of gamma globulinaemia in 75%. In most of cases the diagnosis of VL is based on bone marrow aspiration and in four cases it is isolated Leishmania infantum. Treatment of patients consisted in amfotericine B liposomale (58% of cases) and Glucantime in 41% of them. The outcomes of treatment resulted in a good clinical course and only three cases or 24% of patients developed relapses of VL even taking regularly antiretroviral therapy (the highest number of relapses was 5). Conclusions: VL affecting HIV patients is considered a challenging condition because of its high levels of mortality and relapse rates. Even for middle income countries like Albania AmBisome should be regarded as the most effective first‐line therapy for co‐infected patients. P143 Reduction of mortality rate among HIV/TB patients as a result of comprehensive approach of HIV/TB integration in the Kyiv Regional Tuberculosis Hospital, Ukraine A Chuykov 1, Y Lopatina2, A Zakowicz1, G Tyapkin2, O Holub3 and O Shevchenko3 1Europe Bureau, AIDS Healthcare Foundation, Amsterdam, Netherlands. 2Europe Bureau, AIDS Healthcare Foundation, Kiev, Ukraine. 3HIV/TB, Kiev Regional TB Hospital, Kiev, Ukraine Background: In accordance with the World Health Organization and UNAIDS recommendations in the settings with high TB and HIV prevalence the HIV and TB services should be integrated. Studies also show that early ART initiation in TB/HIV co‐infected patients lowers mortality. We would like to present a model of full TB/HIV integration and estimate its effect on ART coverage, the time of ART initiation and mortality rate. Materials and methods: We retrospectively reviewed TB registers and medical records of 988 TB/HIV co‐infected adults, registered for TB inpatient treatment at the Kyiv Regional Tuberculosis Hospital in the period between 2011 and 2017. The first HIV services were introduced into the TB Clinic in 2012; in 2012 AHF supported introduction of HIV rapid testing, PCP prophylaxis and HIV physician services at the clinic; in 2014 the clinic started provision of ART on its premises, OI diagnostic and treatment and CD4 POC (PIMA); in 2015 GeneXpert technology was implemented; in 2017 mental health component was added. Results: The analysis included 988 patients. Baseline characteristics of patients have not changed in 2011 to 2017. Eighty‐six percent of patients had CD4 count less than 200 cells, 44% of patients had MRTB. Mortality rate decreased almost two times from 265.3 per 1000 patients in 2011 to 144.5 in 2017; coverage of ART increased from 43.5% in 2011 to 73.5% in 2017 and median time of ART initiation decreased almost five times from 170 days in 2011 to 26 days in 2017. Impact of the interventions is shown in Figure 1. Abstract P143 – Figure 1. ARV coverage and impact of clinical interventions on mortality rate among HIV/TB co‐infected patients in Kyiv Regional Tuberculosis Hospital. Conclusions: Full TB/HIV care integration in AHF‐supported clinic led to increased coverage of ART in 1.7 times, reduction of time of ART initiation by an average of 144 days and 1.8 times decreased mortality. The intervention, which resulted in patients increased access to testing and treatment, better patient management and earlier ARV initiation, has a potential to be successfully used as an effective HIV/TB service delivery model. P144 Impact of geneXpert MTB/RIF for rapid tuberculosis diagnosis and rifampicin resistance detection among PLWHIV in South‐Western Nigeria S Usman Laboratory Services, APIN Public Health Initiatives, Abuja, Nigeria Background: Xpert MTB/RIF is described as a game changer in tuberculosis control as tuberculosis due to Mycobacterium tuberculosis (MTB) still remains a major public health issue, with the disease estimated to cause approximately 9 million cases annually and 1.5 million deaths. This study thus aimed at assessing the impact of using Xpert for tuberculosis diagnosis and the efficiency in terms of tuberculosis treatment initiation in selected tuberculosis clinics in South‐Western Nigeria. Materials and methods: A total of 300 consenting patients between 15 and 60 years of age attending tuberculosis clinic were recruited into this prospective cohort study from January 2015 to June 2017 and evaluated for pulmonary or extrapulmonary tuberculosis. Written consent was obtained from every participant and ethical clearance was obtained from the Ethical Review Committee of the Federal Teaching Hospital Ido Ekiti, Nigeria. Data were analysed using SPSS Statistics, with significance fixed at p 50 years old, the long‐term safety in addition to efficacy of HIV treatment continues to be paramount. TAF is a tenofovir prodrug associated with 90% lower tenofovir plasma levels than TDF resulting in less renal and bone toxicity. We evaluated bone mineral density (BMD) changes after switching participants 60 years and older from a TDF‐ to a TAF‐containing regimen. Material and methods Virologically suppressed (HIV‐1 RNA 60 years on a TDF‐containing regimen, were randomized (2:1) to open‐label E/C/F/TAF or continued TDF‐based regimen (TDF). The primary endpoint was the percent change from baseline to Week (W) 48 in spine and hip BMD. Differences in percentage changes from baseline in spine and hip BMD were analyzed using ANCOVA models with treatment as fixed effect and baseline BMD and sex as covariates. Secondary endpoints were HIV‐1 RNA −1.0; at baseline 51% and at W48 58%), compared to continued TDF (at baseline 51% and at W48 46%). At W48, HIV RNA 50 copies/mL was found in one participant in each arm; neither had drug resistance. There were no study drug‐related Grade 3 to 4 AEs. AEs leading to premature study drug discontinuation were similar: 3.6% with E/C/F/TAF and 1.8% with TDF. Abstract P145 – Figure 1. Spine and hip BMD changes over 48 weeks. p values are comparison between randomized groups. Conclusion: Through W48, spine and hip BMD significantly increased in older participants who switched to E/C/F/TAF compared to those who continued a TDF‐containing regimen. HIV‐1 RNA suppression was maintained through W48. The W48 BMD, safety and efficacy data support the switch to E/C/F/TAF in suppressed HIV‐infected participants aged ≥60 years. P146 Effect of age on efficacy and safety of elvitegravir, cobicistat, emtricitabine and tenofovir alafenamide (E/C/F/TAF) in virologically‐suppressed, HIV‐1‐infected participants aged ≥65 years: pooled analysis of two Phase III trials F Maggiolo 1, G Rizzardini2, F Raffi3, F Pulido4, G Mateo Garcia5, J Molina6, E Ong7, Y Shao8, S Chuck9, I McNicholl9, D Piontkowsky9, M Das10 and R Haubrich9 1USC Malattie Infettive, Azienda Ospedaliera Papa Giovanni XXIII, Bergamo, Italy. 2Divisione Di Malattie Infettive, Ospedale Luigi Sacco, Milan, Italy. 3Infectiologie, CHU Hotel, Nantes, France. 4Servicio Enfermedades Infecciosas, Hospital Universitario 12 de Octubre, Madrid, Spain. 5Internal Medicine, Hospital de Dia Infecciosas, Barcelona, Spain. 6Department of Infectious Diseases, Saint‐Louis Hospital and University of Paris, Paris, France. 7Department of Genitourinary Medicine, Royal Victoria Infirmary, Newcastle Upon Tyne, UK. 8Biometrics, Gilead Sciences, Foster City, CA, USA. 9Medical Affairs, Gilead Sciences, Foster City, CA, USA. 10Clinical Research, Gilead Sciences, Foster City, CA, USA Background: As the HIV population ages, analyzing safety and efficacy data for ARV agents in older adults living with HIV is increasingly important. TAF is a tenofovir prodrug associated with 90% lower tenofovir plasma levels and greater renal and bone safety than tenofovir disoproxil fumarate (TDF). We evaluated the efficacy and safety of E/C/F/TAF in individuals 50). Most patients in both groups were males, but mode of transmission differed, with more patients reporting heterosexual contact as their only risk behaviour in the second group. Patients in the second group started ART significantly sooner. Additionally, patients diagnosed at older age were more frequently hospitalised, but had better retention in care. Furthermore, these adults were more likely to have hypertension, cardiovascular and renal disease; there was no difference with respect to bone disease. Hyperlipidaemia was more frequent in patients diagnosed at a younger age. Mortality rate was also significantly higher in the second group, while these patients were on one, two or three concomitant medications in addition to ART approximately twice as often compared to patients diagnosed at a younger age. Demographic and clinical data of both groups are summarised in Table 1. Abstract P148 – Table 1. Demographic and clinical data of patients enrolled Diagnosis 3 DDIs n = 15). These 61 patients were treated with a median of 7 comedications [3 to 9] and 3 ARVs [3 to 4] and 80% (49/61) had at least one comorbidity, of whom most frequent were hypertension (33.6%), cardiovascular disease (28.9%) and dyslipidaemia (19.3%). Most frequent suspected ARVs were boosted protease inhibitors (PIs) (n = 81; 63%), non‐nucleoside reverse transcriptase inhibitors (n = 22; 17%), integrase inhibitor (n = 9; 7%) and cobicistat (n = 8; 6%). The major suspected comedications were statins (n = 41; 31.8%), calcic inhibitor (n = 14; 10.9%), antithrombotics (n = 12; 9.3%), alpha‐1 blocker (n = 8; 6.2%) and metformin (n = 7; 5.4%). Rosuvastatin and darunavir/ritonavir coprescription was the most frequently identified DDI (n = 11). Ten of 129 (7.8%) DDIs in seven subjects were contra‐indicated in at least one of the three expert databases: darunavir/ritonavir + alfuzosin in two subjects, darunavir/ritonavir + ticagrelor (n = 1), darunavir/ritonavir + amiodarone (n = 1), elvitegravir/cobicistat + budesonide (n = 1), nevirapine + ketoconazole (n = 1) and ritonavir + flecainide (n = 1). Conclusions: At least one relevant DDI was evidenced in one out of four elderly patients receiving ARVs plus comedications. PIs and statins were the most frequent drugs involved in DDI. These results highlight the increased risk of polypharmacy and DDIs, the need to consistently collect comedications to pinpoint drug‐drug interactions and to strive for simplified regimens in the elderly HIV population. P150 Prevalence of polypharmacy and potential drug‐drug interactions in PLWHIV: a cross‐sectional study from the Modena HIV Metabolic Clinic cohort A Raimondi, S Zona, I Franconi, F Carli, M Menozzi, G Nardini, B Beghetto, M Mancini, V Masi, C Mussini and G Guaraldi Clinic of Infectious Diseases, Policlinico di Modena, University of Modena and Reggio Emilia, Modena, Italy Introduction: Comorbidities parallel the ageing epidemic affecting PLWHIV and implies an increasing number of comedications and polypharmacy (PP). PP increases the risk for potential drug‐drug interactions (DDIs), adverse drug reactions and grows pill burden. Our aim was to describe more frequently used comedications, assess PP prevalence and estimate the number of potential drug‐drug interactions in PLWHIV attending Modena HIV Metabolic Clinic (MHMC) in 2017. Secondary objective was to estimate independent predictors of potential DDIs. Materials and methods: Cross‐sectional study targeting HIV+ patients on ART, attending MHMC in 2017. Comprehensive drug therapy was collected from patient charts using the fifth level of Anatomical Therapeutic Chemical (ATC) classification system and divided into categories. PP was defined in patients with ≥5 medications not including ART. Potential DDIs were assessed according to University of Liverpool HIV drug interactions database. For each category of drugs, the highest degree of interaction with every ART classes was reported as critical; the presence of minor interaction was reported as potential. Factors associated with DDIs were identified using univariate X2‐test for categorical variables and t‐test or Mann‐Whitney U test for normally or non‐normally distributed continuous variables, respectively. Results: One thousand eight hundred and thirty‐four patients were included, 1349 (73.56%) were men, mean age was 53 (SD 8.3), 408 (22.5%) had PP of concomitant medications and 819 (44.66%) had potential interactions with ongoing ART regimen, 166 of them with critical interaction. As shown in Figure 1, the most represented classes of concomitant medications were vitamins such as vitamin D supplementation (1415 patients, 77.15%), lipid lowering (640, 34.90%) and antihypertensive (630, 34.35%). The difference in use by age was statistically significant for all classes of drugs except steroids and drugs for thyroid dysfunction. Potential DDIs, as illustrated in Figure 2, were significantly associated with age, comorbidities, type of ART and CD4 + nadir (all p 60‐year‐old patients (p 1/3 of PLWHIV accessing HIV care in the UK are now aged ≥50 years. Challenges specific to the management of older PLWHIV include polymorbidities, frailty, polypharmacy and drug‐drug interactions (DDI). Materials and methods: We aimed at determining the prevalence of polymorbidities (≥2 comorbidities) and polypharmacy (≥5 medications) in PLWHIV attending the HIV over50 clinic during a 12‐month period (1 May 2017 to 30 April 2018). PLWHIV were stratified into three groups based on polypharmacy assessment: no polypharmacy (G1), polypharmacy (≥5 medications) (G2), heavy polypharmacy (≥10 medications) (G3). Potential DDI between antiretrovirals and non‐antiretrovirals were assessed using the University of Liverpool HIV Drug Interactions website. Frailty assessment was performed using the Rockwood Clinical Frailty Scale (RCFS) from 1 February 2018 to 30 April 2018. Sub‐analyses were conducted to assess changes in patients’ interaction profiles and frailty scales with increasing polypharmacy. Results: Out of the 229 PLWHIV referred to the HIV over50 clinic, 120 were included in the analysis, the majority of them were Caucasian MSM. Polypharmacy and polymorbidities were recorded in >70%, e.g. hypertension in 52% and diabetes in 26%, with a significant correlation between being on polypharmacy and presenting with polymorbidities (p 75 years old 22.8% 28.9% Hypertension 66.7% 69.2% Diabetes mellitus type 2 37.2% 30.9% Cardiovascular disease 24.5% 25.3% Chronic kidney disease 32.4% 37.8% Chronic obstructive pulmonary disease 11.4% 7.8% Dyslipidaemia 69.3% 72.9% Cancer 21.4% 19.5% Cirrhosis 5.5% 7% Multimorbidity 59.8% 66.9% Polypharmacy 14.3% 31.7% Most representative ARV regimens INSTI + 1 PI 29% 22% INSTI + 1 NNRTI 19.2% 17.1% INSTI + 1 NRTI 31.8% 31.9% BMI = body mass index. Abstract P155 – Figure 1. Prescriptions of INSTIs in GEPPO cohort. Conclusions: These data could be interpreted as a personalisation of ARV in a cohort of ageing patients with comorbidities and concomitant therapies. The newest dual regimens could reduce toxicity and spare drug options for the future. Dual therapy DTG based have high genetic barrier and could save options and toxicity in patients with long ARV experience, multimorbidities and polypharmacy. P156 Disturbance in iron metabolism parameters: a possible marker of preterm ageing in HIV‐infected males G Dragovic 1, B Dimitrijevic1, B Toljic2, J Milasin2, S De Luka3, D Jevtovic4 and A Trbovich3 1Department of Clinical Pharmacology, Pharmacology, University of Belgrade, School of Medicine, Belgrade, Serbia. 2Department of Human Genetics, University of Belgrade, School of Dental Medicine, Belgrade, Serbia. 3Department of Pathological Physiology, University of Belgrade, School of Medicine, Belgrade, Serbia. 4HIV/AIDS Center/Infective/Tropical Disease Clinic, University of Belgrade, School of Medicine, Belgrade, Serbia Background: Number of studies have documented age‐related iron accumulation in animal models. There are still not enough data in humans, especially in HIV‐infected patients. The aim of this study is to estimate levels of iron metabolism parameters and their correlation with multimorbidity occurrence in HIV‐infected ageing subjects. Materials and methods: In this cross‐sectional study we included 50 non‐HIV infected blood donor volunteers and 50 HIV‐infected patients on cART: 2NRTIs+NNRTI or 2NRTIs+PI. Both groups of patients were at age of 50 years and older, all males, all Caucasians. In each group iron metabolism parameters: serum iron concentration, transferrin iron binding capacity (TIBC), transferrin saturation, serum transferrin and ferritin concentrations were determined. In both groups occurrence of multimorbidity was recorded. Multimorbidity was defined as the occurrence of two or more chronic conditions. Comparisons of the two cohorts were made using a chi‐square test or Fisher's exact test for categorical variables and using a Mann‐Whitney U test for continuous variables. All p values less than 0.05 were considered significant. Results: HIV‐infected patients’ characteristics were: mean age was 59.6 ± 1.1 years, median CD4 + T‐count 567.0 ± 317.7 cells/mm3, HIV RNA pVL was 0.5), borderline neurocognitively impaired (0.3 > GDS G2*; G2 Age, X (SD) 65.2 (3.8) 66.7 (5.5) 68.2 (6.9) G1 Years of education, X (SD) 14 (4.3) 12.3 (5.2) 10.1 (4.5) G1 > G3** Years living with HIV, X (SD) 18.7 (9) 16.9 (10.6) 18.4 (7) N/S Years on ART, X (SD) 16.8 (7.2) 15.9 (8.2) 19.8 (19.4) N/S AIDS status, N (%) 19 (45.2) 9 (64.3) 12 (54.5) N/S IDU HIV transmission, N (%) 5 (9.3) 2 (12.5%) 4 (13.3) N/S Nadir CD4 History of chronic HCV, N (%) 8 (15.4) 3 (20) 6 (21.4) N/S Positive depression screening, N (%) 6 (11.8) 0 (0) 5 (17.2) G2 PAOFI daily function alteration, N (%) 10 (19.6) 4 (26.7) 15 (50) G1 Conclusions: The prevalence of neurocognitive impairment in our cohort of HIV patients older than 60 years did not differ widely from rates we found in virologically suppressed patients of other age groups [3]. Years of education appear to protect of neurocognitive impairment in older HIV patients. We consider essential to conduct neurological and neurocognitive follow‐up, particularly in elderly patients neurocognitively impaired and those borderline neurocognitively impaired. References [1] Milanini B, Wendelken LA, Esmaeili‐Firidouni P, Chartier M, Crouch PC, Valcour V. The Montreal cognitive assessment to screen for cognitive impairment in HIV patients older than 60 years. J Acquir Immune Defic Syndr. 2014;67:67‐70. [2] Carey CL, Woods SP, Gonzalez R, Conover E, Marcotte TD, Grant I, et al. Predictive validity of global deficit scores in detecting neuropsychological impairment in HIV infection. J Clin Exp Neuropsychol. 2004;26:307‐19. [3] Pérez‐Valero I, González‐Baeza A, Estébanez M, Montes‐Ramírez ML, Bayón C, Pulido F, et al. Neurocognitive impairment in patients treated with protease inhibitor monotherapy or triple drug antiretroviral therapy. PLoS One. 2013;8:e69493. COMORBIDITIES AND COMPLICATIONS OF DISEASE AND/OR TREATMENT: BONE P158 Incidence and risk factors associated with osteoporosis‐related fractures (ORF) among PLWHIV in British Columbia (BC), Canada J Barletta 1, M Ye2, M Lu2, M Kibel2, R Hogg2, V Lima2 and S Guillemi3 1Infectious Diseases, Hospital Juan A. Fernández, Buenos Aires, Argentina. 2Epidemiology and Population Health, British Columbia Centre for Excellence in HIV/AIDS, Vancouver, Canada. 3Clinical Education and Training Program, British Columbia Centre for Excellence in HIV/AIDS, Vancouver, Canada Background: Increased longevity of PLWHIV receiving ART is accompanied by higher prevalence of age‐related comorbidities including bone disease [1]. HIV‐related factors and antiretroviral drugs have been found to play a role in pathogenesis of bone disease in PLWHIV. We investigated incidence and risk factors associated with ORF among PLWHIV in BC. Materials and methods: These analyses were based on the Comparative Outcomes and Service Utilization Trends (COAST) cohort study, a population‐based retrospective cohort study examining health outcomes and service use of PLWHIV and a 10% random sample of HIV‐negative individuals (HIV‐) in BC [2]. We examined incidence of ORF among PLWHIV from 1996 to 2013 considering wrist, vertebrae, humerus and hip fractures [3]; using physician and hospital‐based administrative data and International Classification of Diseases 9/10. Age‐ and sex‐adjusted incidence rates were calculated using 2011 Canada population as reference. The effect of the variables on the risk of ORF was assessed by logistic generalised estimating equation (GEE) model. For our GEE model, only ORFs occurring after ART initiation were considered; sex, age at ART initiation, previous injuries (including motor vehicle collision, land transportation injuries, self‐harm and assault), injection drug use (IDU), ART initiation era and length of time (LOT) on antiretroviral drug classes were covariates. Results: A total of 6846 PLWHIV and 514,619 HIV‐ were included in the incidence analysis. ORF occurred in 416 PLWHIV and 28,028 HIV‐ (6.08% vs 5.45%, p = 0.02). PLWHIV were younger at first ORF (median age 47 years Q1 to Q3 38 to 53 vs 62 years Q1 to Q3 43 to 79, p 3000). FRAX scores combined with BMD (FRAX/BMD) and without BMD (FRAX) were calculated. Subjects were separated on the basis of major risk factors for fragility fracture and age. Two groups were defined: one group of subjects who received identical treatment recommendations from FRAX and FRAX/BMD, and another group who received a different treatment recommendation when BMD was included in the FRAX calculation. Result: Three hundred and fifty‐nine HIV‐positive patients were enrolled: 76 at high risk for fracture (with major risk factors for fragility fracture), 154 intermediate risk (FRAX >10% 10‐year risk of major osteoporotic fracture) and 129 low risk (FRAX ≤10%). The subjects with distinct risk had different results fulfilling treatment criteria: high‐risk, 14 (FRAX) versus 29 (FRAX/BMD) (difference 19.7%, 95% CI 5.8 to 33.7%), and intermediate‐risk, 16 (FRAX) versus 58 (FRAX/BMD) (difference 27.3%, 95% CI 18.2 to 36.3%), and low‐risk, 0 (FRAX) versus 16 (FRAX/BMD) (difference 12.4%, 95% CI 6.7 to 18.1%). Body mass index (BMI) 60 years but without major risk factors can be higher on the priority list for investigations or therapeutic management. P164 Bone turnover markers evolution after treatment initiation with Atripla in comparison to Truvada raltegravir: another glimpse to osteoporosis in HIV Y Oster 1, M Cohen2, R Dresner Pollak3 and H Elinav1 1Clinical Microbiology and Infectious Diseases, Hadassah Medical Center, Hebrew University, Jerusalem, Israel. 2Jerusalem District, Clalit Health Services, Jerusalem, Israel. 3Endocrinology and Metabolism, Hadassah Medical Center, Hebrew University, Jerusalem, Israel Background: Accelerated osteoporosis, one of the comorbidities related to HIV, is multifactorial and was shown to be drug related as well. Bone turnover markers as P1NP, CTX, osteocalcin and P1CP can potentially improve our understanding of mechanisms of osteoporosis in HIV. As therapies targeting factors involved in osteoporosis are evolving, identifying factors that can potentially be blocked might improve or prevent this condition in people living with HIV. Aims: The aim of our study was to identify bone turnover markers that play a role in osteoporosis in people living with HIV, and to compare the dynamic of their levels after initiation of two different and commonly used antiretroviral regimens. Method: Male patients that initiated treatment with Truvada+raltegravir or Truvada+efavirenz and maintained the same regimen for at least 12 months were included in the study. Control group included HIV patients that did not receive antiretroviral treatment. Fifteen patients were included in each group. Levels of P1NP, CTX were measured in frozen serum samples using standard immunoassays at treatment initiation, and 1, 6 and 12 months following treatment initiation of the two study groups, and at 0, 6 and 12 months for the control group. Results: Mean age of patients was 42.2 years and similar in all groups. Mean CD4 counts were similar in the treatment groups (205, 216 cells/mL) but higher in the control group (516 cells/mL) As expected, patients in the two treatment regimen groups showed a decrease in HIV viral load and similar increase in CD4 levels, while the control group had stable parameters during the follow‐up period. Baseline levels of CTX and P1NP levels were similar in all groups. While both treatment groups had showed significant increase over time of both markers, these parameters were stable in the control group. Levels of P1NP were statistically higher in the 6‐ and 12‐month time points (p = 0.002, p = 0.004, respectively) while CTX levels were statistically higher only at the 6‐month time point and then plateaued (p = 0.039). Conclusions: Bone loss after HAART initiation is likely due to a high bone turnover state as both CTX and P1NP increase in the first 6 to 12 months of treatment. As high bone turnover often leads to bone loss, therapy with bisphosphonates or anti RANKL antibody which reduce turnover should be considered in the first year of HAART, thus reducing bone loss at this vulnerable period and potentially improving long‐term bone density. P165 Audit of use of IV bisphosphonates in people living with HIV S Shah1, M Bracchi1, W Hurt1, Q Zhang1 and A Milinkovic 1 1Department of HIV Medicine, Chelsea and Westminster Hospital NHS Foundation Trust, London, UK. 2School of Medicine, Imperial College, London, UK Introduction: PLWHIV are at increased risk of osteopenia, osteoporosis and subsequent fragility fractures. National and international guidelines help recognise and manage those at greatest risk and control comparative trials had shown that use of IV bisphosphonate therapy is more effective than switching tenofovir disoproxil fumarate (TDF) at increasing BMD in HIV‐positive adults with low bone mass. Clear prescribing policy for use of IV bisphosphonate in the setting of HIV has not yet been established. Method: Our audit consists of a retrospective data collection on PLWHIV receiving IV bisphosphonate therapy on the Gazzard Day Unit at Chelsea and Westminster Hospital. The search included all patients who had received at least one dose of IV ibandronic or zoledronic acid between January 2015 and December 2017. Result: Ninety patients were identified, of which 71 (79%) patients were male, with an average age of 56 years. Only two patients had a FRAX score documented before starting IV bisphosphonate therapy; the majority of patients (88/90) had a pre‐treatment dual X‐ray absorptiometry (DXA) scan, of which 67 (74%) patients had confirmed osteoporosis. Twenty‐three patients (26%) were started on IV bisphosphonate therapy for other reasons (e.g. myeloma, vertebral wedge compression). Twenty‐five of 90 (28%) patients had regular DXA scans performed during treatment. At the time of auditing, 39 patients (43%) were still on antiretrovirals associated with an increased risk of osteoporosis, including combinations containing TDF (n = 19) (21%), or protease inhibitors (PIs) (n = 31) (34%), and combinations containing both TDF and boosted PIs (n = 6) (7%). At the time of the audit 60% of patients were actively receiving IV bisphosphonate therapy, with average treatment duration of 24 months. Fifteen (28%) patients had exceeded the recommended 5‐year duration of therapy. Renal function, calcium and vitamin D levels were regularly monitored during treatment for all patients on IV bisphosphonate therapy. Half of the patients have been given vitamin D supplementation; 17% have been prescribed both calcium and vitamin D. In our patient population no patients were found to have experienced adverse effects secondary to IV bisphosphonate therapy, such as osteonecrosis of the jaw or atypical bone fractures. Conclusions: Our audit has shown that guidance is needed to ensure appropriate use of IV bisphosphonate in PLWHIV. As a result of our findings, in our centre a multidisciplinary approach was taken to implement strategies aimed at improving management and monitoring of IV bisphosphonate administration. COMORBIDITIES AND COMPLICATIONS OF DISEASE AND/OR TREATMENT: CARDIOVASCULAR P166 Immune and inflammatory biomarkers in naïve HIV‐infected patients starting antiretroviral therapy: a prospective study M Saumoy 1, S Di Yacovo1, J Sanchez‐Quesada2, D Sviridov3, R Vila4, B Garcia1, A Navarro1, A Vernet5, D Giralt1, J Ordoñez‐Llanos2 and D Podzamczer1 1HIV and STD Unit, Infectious Disease Service, Hospital Universitari de Bellvitge, Hospitalet de Llobregat, Spain. 2Biochemistry and Molecular Biology Department, Biomedical Research Institute IIB Sant Pau, Barcelona, Spain. 3Laboratory of Lipoproteins and Atherosclerosis, Baker Heart and Diabetes Institute, Melbourne, Australia. 4Vascular Surgery, Hospital Universitari de Bellvitge, Hospitalet de Llobregat, Spain. 5Department of Mechanical Engineering, Universitat Rovira i Virgili, Tarragona, Spain Background: Atherogenesis in HIV patients is multifactorial. Our aim was to assess the effect of HIV infection and ART on several proatherogenic biomarkers and its relationship with subclinical atherosclerosis in a cohort of HIV‐infected naive patients. Methods: Multicentre prospective comparative study. Two groups of naive HIV patients (group A: CD4 > 500, not starting ART at baseline; group B: CD4 20%) or from intermediate to high after TDF stop. The percentage of patients who became eligible for statin within one year from TDF discontinuation was estimated as 3.8% (95% CI 3.1 to 4.6). Abstract P167 – Table 1. Mean values and differences between two values of lipids before and before/after TDF discontinuation Biomarker N T0 T1 Difference p value N T1 T2 Difference p value Mean 1 (SD1) Mean 2 (SD2) Mean 1 (SD1) Mean 2 (SD2) LDL 1019 112.7 (33.0) 110.6 (33.0) −2.1 0.178 557 111.5 (56.8) 121.7 (37.1) +10.2 20%). The cIMT was performed by the same operator at 1 cm from carotid bifurcation as the average of three measurements (abnormal >0.9 mm). Polypharmacy was defined as ≥5 drugs other than ARV treatment. Data are expressed as medians (interquartile ranges). Results: One hundred and eighty patients were included: age, gender (% male) and ethnicity (% Caucasian) were 50 years (42.2 to 55.5), 80% and 92.7%, respectively. Current and nadir CD4+ cell count were 566/μL (389 to 756) and 201/uL (85 to 330); treatment duration was 9.6 years (3.6 to 16) with 159 patients (89.9%) showing HIV RNA 0. Non‐calcified atherosclerotic lesions were observed at 109 of 173 (35.16%) CAC = 0 sites in 56 (68.29%) patients. In 18F‐NaF cohort 92 (89.32%) of the non‐calcified sites presented 18F‐NaF uptake in 46 patients (92%) whereas in 18F‐FDG cohort 15 (21.42%) of the non‐calcified sites presented. 18F‐FDG uptake in 10 (31.25%). FDG uptake was observed 10 (31.25%) in CAC non‐progressor and 22 (68.75%) times in CAC progressor patients whereas 18F‐NaF was observed 47 (31.54%) in CAC non‐progressor and 102 (68.46%) times in CAC progressor patients. Spearman correlation coefficient rS between CAC non‐progression and positive 18F‐NaF was −0.56 (p = 0.01, in 19 patients) while the correlation between CAC non‐progression and positive FDG was not evaluated (all CAC values were zero). Spearman correlation coefficient rS between CAC progression and positive 18F‐NaF was 0.20 (p = 0.33, in 30 patients) while CAC progression and positive FDG was −0.65 (p = 0.02, in 14 patients). Conclusion: PET/CT with 18F‐FDG and 18F‐NaF may allow evaluation of distinct pathophysiological processes in atherosclerotic lesions and might provide information on the complex interactions involved in formation and progression of atherosclerotic plaque. P170 Comparison between cardiovascular disease risk scores and observed rates of cardiovascular disease in people living with HIV M Lee 1, C Smith2, J Mok1, A Duncan1, R Kulasegaram1 and A Wierzbicki3 1Harrison Wing Department, Guy's and St Thomas Hospital NHS Foundation Trust, London, UK. 2Institute of Epidemiology and Health Care, University College London, London, UK. 3Department of Chemical Pathology, Guy's and St Thomas Hospital NHS Foundation Trust, London, UK Background: It is unclear which risk calculator best predicts cardiovascular disease (CVD) risk in HIV‐positive patients; current UK guidelines recommend QRISK2 [1] but this has never been validated in HIV‐positive populations. We assessed commonly used risk calculators against the D:A:D equation, previously validated in US, European and Australian HIV‐positive cohorts [2], and observed CVD events during follow‐up. Materials and methods: A representative sample of 245 HIV‐positive patients attending a London hospital aged >40, without a previous history of CVD, had baseline 10‐year CVD risk scores calculated using QRISK3, QRISK2, Framingham CVD equation and D:A:D risk calculators. D:A:D was re‐scaled from five‐ to ten‐year risk, assuming same rates in both five‐year periods. Scores were classified into low ( 18 years, without cardiovascular (CV) disease, receiving combined antiretroviral therapy (cART) and undetectable viral load in the last six months. Common carotid intima‐media thickness (cc‐IMT) and presence of plaque were assessed. SA defined: carotid plaque or cc‐IMT > percentile 75 (P75) of a reference population. Demographic, ART history, CV risk (CVR), lipid profile and plasma biomarkers (sCD163, sCD14, IL‐6, hs‐CRP, D‐dimer, sVCAM and lipoprotein‐phospholipase A2[Lp‐PLA2]) were evaluated. Multivariate logistic regression including all variables with p = 0.1 in univariate analyses was used to assess factors related with SA. Continuous variables are expressed as mean (SD). Results: Four hundred and fifty patients were included: age 50.4 (10), men 81%, 78% ever smoker, 24% hypertension, 7% diabetes, 23% used lipid lowering drugs. Duration of cART 14.7 (6.6) years, CD4 cell count 693 cells/uL. Mean cc‐IMT was 0.63 (0.13) mm; 30.5% had cc‐IMT >P75 of a reference population; 34.1% had at least one carotid plaque (mainly in bulb). SA prevalence was 49.2%. An increased CD4 was associated with a lower sCD163 (p 20%). EACS targets were 0.9 mm). Data are expressed as medians (interquartile ranges). Results: Five hundred and three patients were enrolled: age, gender (% male) and ethnicity (% Caucasian) were 49.7 years (42.4 to 57.4), 78.1% and 93.8%, respectively. Four hundred and sixty patients (91.4%) were on HAART, 388 patients (80.3%) HIV RNA. Current and nadir CD4+ cell count were 554/uL (374 to 718) and 198/uL (87 to 316.7); treatment duration was 9.8 years (3 to 16.3); in 10yy ASCVD risk strata were: L 58.5%, I 27.3% and H 14.3%. Two hundred and forty‐four (49.6%) patients were active smokers: y29 (15%) were classified as having a H CVR. Despite 156 patients (31.5%) were receiving lipid‐lowering treatment TC and LDL‐C targets were not attained in 128 (82%) and 118 (77.6%) [p 60 years, BMI >23 kg/m2, high ASCVD risk, hypertension, diabetes, metabolic syndrome were associated with diastolic dysfunction while female, statin exposure and LAVI >34 mL/m2 were associated with pulmonary hypertension, p 440 ms for male and >460 ms for female) was evidenced in 2.3% (n = 6) with values ranging from 443 to 474 ms. Of the 165 individuals with an ECG on‐ART (median duration on ART of 32.5 months, HIV RNA 1 episode of recurrence. The KM‐estimated time free from recurrences is shown in Figure 1. Abstract P181 – Figure 1. Kaplan‐Meier‐estimated time free from recurrences. Conclusions: KS remains a relevant AIDS‐related event and new diagnoses are still frequent even in recent years. In our experience recurrences are rare and different strategies for its management appeared to be effective and safe. P182 HPV infection among HIV‐positive women in some countries of Eastern Europe and Central Asia A Popova 1, O Shipulina2, M Deulina1, E Almamedova3, A Rzayeva4, A Kadyrova5, S Grigoryan6, A Asmaryan7, A Pepanyan7, A Davidyan7, L Ermolenko8, T Nevmerzhitskaya8, I Tavtyn9, U Kadyrbekov10, N Abylgazieva11, Z Zhaanbaeva11, A Spirin12, O Agafonova13, M Chesnokov14, L Kalenik14, S Karimov15, R Rahimova16, M Rustamova17, Z Nurlyaminova18, M Dmitryukova2 and V Pokrovsky1 1Russian Federal AIDS Center, Central Research Institute of Epidemiology, Moscow, Russian Federation. 2Department of Molecular Diagnostics, Central Research Institute of Epidemiology, Moscow, Russian Federation. 3Director, Republic Center of the Struggle Against AIDS, Baku, Azerbaijan. 4Reception Department, Republic Center of the Struggle Against AIDS, Baku, Azerbaijan. 5Director, Scientific Research Institute of Lung Diseases, Baku, Azerbaijan. 6Director, National Center for AIDS Prevention, Erevan, Armenia. 7Medical Care Department, National Center for AIDS Prevention, Erevan, Armenia. 8Clinical Department, Svetlogorsk Central Region Hospital, Svetlogorsk, Belarus. 9Director, Svetlogorsk Central Region Hospital, Svetlogorsk, Belarus. 10Director, Republican AIDS Center, Bishkek, Kyrgyztan. 11Department of Dispensary Observation, Republican AIDS Center, Bishkek, Kyrgyztan. 12Director, Samara Regional Clinical AIDS Centre, Samara, Russian Federation. 13Deputy Chief, Samara Regional Clinical AIDS Centre, Samara, Russian Federation. 14Outpatient Department, Samara Regional Clinical AIDS Centre, Samara, Russian Federation. 15Director, Republican Center of Prevention and Control HIV, Dushanbe, Tajikistan. 16Head, Ministry of Health and Social Protection of Population of Tajikistan, Dushanbe, Tajikistan. 17Head, Clinical Medicine of Academy of Medical Sciences of the Ministry of Health and Social Protection of Population, Dushanbe, Tajikistan. 18Dispensary Department, Republican Center of Prevention and Control HIV, Dushanbe, Tajikistan Background: The frequency of detection of HPV varies greatly depending on the region of residence of the woman. Also HIV‐infected women have a higher risk of HPV infection than HIV‐negative women, and a higher risk of persistence and malignancy. The aim of the study was: to study the prevalence of human papillomavirus of high carcinogenic risk (HPV HCR) in HIV‐infected women in some countries of Eastern Europe and Central Asia. Materials and methods: Six hundred and forty‐seven HIV‐infected women from Russia, Belorussia, Armenia, Azerbaijan, Tajikistan and Kyrgyzstan were examined from September 2017 to December 2017. All women underwent HPV‐test with the determination of 14 types of HPV HCR (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68). Results: Among the 647 women surveyed, mostly young people (under 40 years) predominated. As a result of the HPV‐test, 265 (41%) of HIV‐infected women were diagnosed with HPV HCR. The percentage of HPV detection ranged from 28% to 48%: Armenia 39%, Azerbaijan 43%, Belarus 28%, Kyrgyzstan 46.5%, Tajikistan 37.8%, Russia (Samara) 48%. All 14 HPV HCR genotypes were diagnosed in HIV‐positive women in the region. The distribution of HPV genotypes is different for these countries: 16 and 68 HPV genotypes are registered in Armenia (33.3% and 23%, respectively), in Azerbaijan 16, 18 and 56 HPV genotypes (32.6%, 20.9%, 20.9%), in the Republic of Belarus 16 (28.6%) and 56 (28.6%), in the Republic of Kirghizia 16, 31 and 68 (23.9%, 21.7% and 21.7% respectively), in Tajikistan 16, 31, 56 (29.7%, 21.6%, 21.6%), in Samara (Russian Federation) 52 and 16 (23.6% and 22.2%). HPV infection was caused by a combination of several genotypes in 49.1%. The leading genotypes amond 265 HIV‐infected women with HPV were: 16 genotype 26.4%, 31 genotype 13.6% and 18 genotype 9.4%. Conclusion: There is a high incidence of HPV infection in HIV‐infected women. Given the high risk of developing cervical cancer and the wide spectrum of detectable genotypes of HPV HCR in this group, it is necessary to use a test system to diagnose the 14 genotypes of HPV. The results should be taken into account when planning vaccination in the region. P183 Characteristics of AIDS‐related and non‐AIDS‐related cancers in an Italian cohort of HIV patients in the period 1996 to 2018 S Nicolè 1, C Mengoli2, G Marini1, D Coletto1, S Cavinato1, S Marinello1 and A Cattelan1 1Infectious Diseases Department, University Hospital of Padova, Padova, Italy. 2Infectious Diseases Department, University of Padova, Padova, Italy Background: The advent of the combined antiretroviral therapy (cART) led to a strong reduction of the incidence of AIDS‐defining cancers (ADCs) and raised life expectancy among PLWHIV [1]. At the same time, non‐AIDS‐defining cancers (NADCs) increased [2,3]. The aim of this study was to define from an epidemiological and prognostic point of view the neoplastic pathology in PLWHIV. Materials and methods: This retrospective cohort study has been carried out at the Infectious Diseases Department of Padova, on PLWHIV with a cancer diagnosis occurred between January 1996 and March 2018. The clinical and immunovirological characteristics and survival rates were analysed comparing two periods of observation (1996 to 2006 vs. 2007 to 2018). Results: One hundred and eighty‐eight patients (35F/153M) were enrolled, for a total of 204 cancer diagnoses; 15 patients had more than one diagnosis. We had 104 ADCs diagnosis and 100 NADCs: 97 cancers from 1996 to 2006, 107 from 2007 to 2018. The most common cancers were NHL (26.0%) and KS (25.0%); among the NADCs, the most frequent were hepatocellular carcinoma (HCC) (8.3%), anal cancer (5.9%) and lung tumour (4.9%). During the 20 years of observation, the ADCs incidence showed a significantly decreasing trend (p 50 years, n, % 44 (36.4) Previous syphilis, % 53 (43.8) Previous hepatitis A, % 18 (14.8) Previous condylomatosis, % 33 (27.2) Previous AIDS‐defining events, % 15 (12.3) Co‐infections (HBV, HCV), % 8 (6.6) Nadir CD4, cells/mm3 ± SD 326 ± 199 Nadir CD4 40 copies/mL, % 4 (3.3) CD/CD8 ratio 140 mm Hg and diastolic >90 mm Hg; and (2) DM: blood sugar 50 years (OR 4.54, CI 3.85 to 5.35), having CD4 >500 compared to CD4 50 years (OR 4.68, CI 4.38 to 5.00), longer duration on ART (OR 1.11, CI 1.10 to 1.12), having CD4 500 (OR 1.05, CI 0.92 to 1.19), being on second‐line ART regimen (OR 1.10, CI 1.02 to 1.18) and third‐line ART regimen (OR 2.12, CI 1.53 to 2.93) compared to first‐line ART regimen. One hundred and forty‐nine (1.8%) patients had both HTN and DM. Conclusion: The results demonstrate that hypertension and diabetes mellitus and their risk factors are a growing public health problem especially among male elderly HIV‐positive patients. The findings emphasise need for health care intervention to reduce the growing burden of chronic comorbidities. Reference: [1] Kharsany AB, Karim QA. HIV infection and AIDS in Sub‐Saharan Africa: current status, challenges and opportunities. Open AIDS J. 2016;10:34‐48. COMORBIDITIES AND COMPLICATIONS OF DISEASE AND/OR TREATMENT: METABOLIC P187 Increases in lipid profile after switch from TDF to TAF‐based HAART regimens in a cohort of HIV‐positive patients: is it clinically relevant? L Gazzola 1, G Tagliaferri1, D Mondatore1, A De Bona1, C Borsino2, T Bini1, G Marchetti1 and A d'Arminio Monforte1 1Infectious Disease, ASST Santi Paolo e Carlo, Milan, Italy. 2Pharmacy Unit, ASST Santi Paolo e Carlo, Milan, Italy Background: Switching from TDF to TAF increases lipid profiles; aim of our study is to evaluate whether this is clinically relevant and results in an increased frequency of patients out‐of‐target‐LDL, according to their CV risk score. Materials and methods: All HIV patients switching from TDF to TAF, with no changes of the third drug, and plasma lipids available within six months before and after the switch, were included. Ongoing statin therapy was an exclusion criterion. Demographic, HIV‐related parameters, CV risk factors and lipid profile on TDF and on TAF were collected for each patient. The CV risk SCORE and the target of LDL for each SCORE strata were calculated according to 2016 ESC/EAS guidelines for the management of dyslipidaemias. Modifications in lipid profiles and in the frequency of patients out‐of‐target‐LDL were evaluated after switch to TAF. Results: Two hundred and twenty‐one patients were included: 13 on PI/cobi, 101 on INSTI/cobi, 11 on INSTI, 96 on NNRTI. Median CD4 count at switch was 640 cells/mm3 (IQR 493 to 849); HIV‐RNA was 0.9 for males and >0.85 for females. Univariate analysis and binary logistic regression estimated the contribution of a range of factors to hepatic steatosis risk. Receiver operator characteristic (ROC) curves were used to estimate the sensitivity and specificity of using clinical measures to identify hepatic steatosis. Results: Of 338 patients sampled, 71 (21%) had a confirmed diagnosis of hepatic steatosis, with age, non‐Caucasian ethnicity, dysglycaemia and body mass index category all significantly associated with risk (p 1.0 for males and >0.95 for females identifying hepatic steatosis with a sensitivity of 0.761 and specificity of 0.723. ROC curve analysis for waist calculated an AUC of 0.738 (95% CI 0.662 to 0.813, p 101 cm and 105 cm for men of South or East Asian/Central or South American and other ethnicities respectively, and >91 cm for women identifying hepatic steatosis with a sensitivity of 0.800 and a specificity of 0.526. Conclusion: Waist:hip ratio can identify hepatic steatosis risk in PLWHIV. Waist circumference is less specific and sensitive than waist:hip ratio but still has validity. We recognise that in clinical practice measuring waist alone may be pragmatic, but recommend using the waist:hip ratio wherever possible to identify hepatic steatosis risk, for example in research. P191 Dysglycaemia is prevalent in HIV patients over 40 and may be detected using routine screening for cardiovascular risk J Mok 1, L Goff2 and A Duncan3 1School of Medical Education, King's College London, London, UK. 2Department of Nutritional Sciences, King's College London, London, UK. 3Harrison Wing, Guy's and St Thomas’ NHS Foundation Trust, London, UK Background: Current UK guidelines recommend that all PLWHIV over the age of 40 are screened annually for metabolic comorbidities including glycated haemoglobin (HbA1c) for diabetes risk [1]. However, it is unknown whether these guidelines are best suited for all cohorts with a diversity of age and ethnicity. Secondly, access to HbA1c screening may not be universal or may not be routinely measured in clinical practice. We aimed to investigate the prevalence of pre‐diabetes and type 2 diabetes (T2DM) in PLWHIV, and to assess whether the cardiovascular risk tool QRISK2 can identify PLWHIV at risk of diabetes. Materials and methods: A cohort of HIV‐positive adults was purposively sampled in order to represent the demographic of three large South London clinics. Baseline demographic, anthropometric and routine clinical data were gathered and analysed. Glycaemic status was established using World Health Organization criteria for normoglycaemia, pre‐diabetes and T2DM ( 200 mg/dL were reported in 21% of patients, mean total cholesterol (TC) was 186 mg/dL (IQR 162 to 210) and 7% of patients had TC >240 mg/dL, mean LDL‐cholesterol was 71 mg/dL (IQR 95 to 139). At 12 and 24 weeks, mean triglycerides had increased by +25.75 mg/dL (SD 100.54) and +13.09 mg/dL (SD 44.26), and TC had increased by +20.33 mg/dL (SD 26.38) and +13.09 mg/dL (SD 44.26) after switch to TAF, respectively, and differences in mean changes were significant (p 240 mg/dL after switching was associated with older age (increased by 2% per year; p = 0.027), BMI >25 kg/m2 (p = 0.020) and elevated baseline LDL‐cholesterol (p 50 and BMI >25 kg/m2 remained independently associated with TC >240 mg/dL (OR 1.58 and 2.08 respectively). Conclusions: Results of our study show that patients with pre‐existing higher CV risk are more likely to show increases in lipids after switching from TDF which may worsen the cardiovascular risk profile. In these patients risk and benefit of switching TDF to TAF should be carefully accessed, taking lipids in consideration. P196 Correlation between PAI‐1, leptin and ferritin with HOMA in HIV/AIDS patients G Dragovic 1, M Sumarac Dumanovic2, K Al Musalhi3, I Soldatovic4, B Dimitrijevic1, D Jevtovic5 and D Nair3 1Department of Clinical Pharmacology, University of Belgrade, School of Medicine, Belgrade, Serbia. 2Endocrinology, Diabetes, Metabolism Disease, University of Belgrade, School of Medicine, Belgrade, Serbia. 3Department of Clinical Biochemistry, Royal Free Hospital and University College London, London, UK. 4Department of Biomedical Statistics, University of Belgrade, School of Medicine, Belgrade, Serbia. 5HIV/AIDS Center/Infective/Tropical Disease Clinic, University of Belgrade, School of Medicine, Belgrade, Serbia Background: Data about correlation of interleukins (IL‐1α, IL‐1β, IFNγ, IL‐2, IL‐4, IL‐6, IL‐8, IL‐10), adipocytokines (leptin, adiponectin, monocyte chemoattractant protein‐1 [MCP‐1], resistin, plasminogen activator inhibitor‐1 [PAI‐1], tumour necrosis factor alpha [TNFα]), ferritin, C reactive protein (CRP) and vascular endothelial growth factor (VEGF)) with homeostasis model assessment (HOMA) in HIV/AIDS patients are still limited. Therefore, the aim of this study was to evaluate the possible correlations of serum interleukins (IL‐1α, IL‐1β, IFNγ, IL‐2, IL‐4, IL‐6, IL‐8, IL‐10) levels, adipocytokines (leptin, adiponectin, MCP‐1, resistin, PAI‐1, TNF α), ferritin, CRP and VEGF with HOMA in HIV/AIDS patients treated with combined ART (cART). Materials and methods: This cross‐sectional study included 64 HIV/AIDS patients, all Caucasians, receiving cART at the HIV/AIDS Centre. PAI‐1, leptin, ferritin and insulin levels were measured using the Metabolic Syndrome Array I (Randox Laboratories Ltd., London, UK), while adiponectin and resistin levels were measured using Metabolic Syndrome Array II (Randox Laboratories Ltd., London, UK), interleukins (IL‐1α, IL‐1β, IFNγ, IL‐2, IL‐4, IL‐6, IL‐8, IL‐10), MCP‐1, TNF‐α as well as VEGF was measured using Cytokine Array I (Randox Laboratories Ltd., London, UK). Insulin resistance was determined using the homeostasis model assessment index (HOMA). All biochemical analyses were performed at the Department of Clinical Biochemistry, Royal Free Hospital and University College London, UK. Multicollinearity of independent variables in multivariate model was analysed using variance inflation factor. Results: Correlation analysis revealed significant correlations between HOMA and waist circumference, body mass index, patients’ age, number of cART combinations and triglycerides (p = 0.001, p = 0.001, p = 0.050, p = 0.044, p = 0.002, respectively). HOMA negatively correlated with levels of high density lipoprotein (Rho=−0.282, p = 0.025). PAI‐1 (Rho = 0.334, p = 0.007) and leptin (Rho = 0.492, p = 0.001) together with ferritin (Rho = 0.396, p = 0.001) levels positively and significantly correlated with HOMA. Levels of IL‐1α, IL‐1β , IFNγ, IL‐2, IL‐4, IL‐6, IL‐8, IL‐10, adiponectin, MCP‐1, resistin, TNFα, CRP and VEGF did not significantly correlate with HOMA. Further, multiple logistic regression showed that there is a statistically significant correlation between PAI, leptin and ferritin with HOMA levels (p = 0.042, p three years EFV LPV/r Male gender Smoking Increase in TC level 7.39 (6.47 to 8.31) No influence No influence No influence No influence Increase in TG level 7.24 (6.02 to 8.46) No influence 3.8 (3.03 to 4.57) 2.4 (1.33 to 3.47) No influence Decrease in HDL level No influence 2.01 (0.88 to 3.14) No influence 1.55 (0.11 to 2.99) 1.56 (0.85 to 2.27) Increase in LDL level 1.88 (1.18 to 2.58) No influence No influence No influence No influence Increase in VLDL level 3.26 (1.16 to 5.36) No influence 3.53 (2.12 to 4.94) 5.51 (3.41 to 7.61) 3.46 (2.26 to 4.66) Increase in AIP 1.57 (0.9 to 2.24) 2.26 (1.41 to 3.11) No influence 1.84 (1 to 2.68) 1.67 (0.97 to 2.37) Increase in Apo B level No influence No influence No influence No influence 6.64 (4.47 to 8.81) Conclusions: It is necessary to choose ART scheme with the least impact on lipid levels, and to advise the patients to give up smoking. In case of significant abnormalities in the course of therapy, the regime of ART can be corrected and/or lipid‐lowering medicinal products can be administered. Routine screening of the patients is important for timely diagnostics of impaired glucose tolerance, diabetes mellitus, insulin resistance, and for conduction of corrective measures. References [1] Calza L, Manfredi R, Chiodo F. Hyperlipidemia in patients with HIV‐1 infection receiving highly active antiretroviral therapy: epidemiology, pathogenesis, clinical course and management. Int J Antimicrob Agents. 2003;22:89‐99. [2] Dube M, Fenton M. Lipid abnormalities. Clin Infect Dis. 2003;36 Suppl 2:79‐83. [3] Gelato M. Insulin and carbohydrate dysregulation. Clin Infect Dis. 2003;36 Suppl 2:91‐5. [4] Data Collection on Adverse Events of Anti‐HIV drugs (D:A:D) Study Group. Factors associated with specific causes of death amongst HIV‐positive individuals in the D:A:D study. AIDS. 2010;24:1537‐48. COMORBIDITIES AND COMPLICATIONS OF DISEASE AND/OR TREATMENT: NEUROLOGICAL P200 Factors associated with HIV‐associated neurocognitive disorder in an unselected cohort in East South London: the HAND study S Rackstraw 1, O Davies2, A Thiyagarajan1, A Sharp2, N Patel2, R Szydlo3 and R Kulasegaram2 1HIV, Barts Health NHS Trust, London, UK. 2HIV, Guy's & St Thomas’ Hospitals NHS Trust, London, UK. 3Statistics, Imperial College London, London, UK Background: UK studies have shown low levels of neurocognitive impairment (NCI) in selected cohorts. Several clinical factors including ART CNS penetration effectiveness (CPE) score have been linked with HIV‐associated neurocognitive disorder (HAND). In the first study of HAND in an unselected cohort in London, we aimed to: (1) determine the extent of NCI in this cohort; (2) establish correlation with HIV‐related factors and medical comorbidities. Methods: Seven hundred and eighty‐six HIV+ participants aged >18 were prospectively recruited from four HIV clinics in East and South London. Medical and ART history and mental state assessment were completed. Computerised assessment of neurocognitive function was performed using Cogstate tests. Participants had NCI if they were >1 standard deviation (SD) outside of the population mean in two or more cognitive domains. Results: The median age was 46. Sixty‐five percent were Caucasian. Eighty‐one percent had HIV VL 8 35 3.22 0.037 Conclusion: NCI was observed in 37.2% of this cohort. Anxiety, abnormal IHDS, being black African, below college level education, non‐MSM transmission route and CPE score >8 were associated with HAND. The CPE score effect is at variance with that observed in other studies. This could be due to intensification of ART in those with prior NCI or even a degree of drug toxicity. More research is needed on the effects of ART on HAND. P201 Efficacy of a computerised cognitive rehabilitation training in improving HIV‐associated neurocognitive disorders F Bai 1, M Allegrini1, C Falcinella1, I Strada2, L Borghi2, E Merlini1, A d'Arminio Monforte1 and G Marchetti1 1Dept Health Sciences/Clinic of Infectious Diseases, San Paolo Hospital, University of Milan, Milan, Italy. 2Dept Health Sciences/Unit of Clinical Psychology, San Paolo Hospital, University of Milan, Milan, Italy Background: We aimed to investigate the efficacy of a new computer‐based cognitive rehabilitation protocol (restorative approach) in improving the cognitive performance of patients affected by HIV‐associated neurocognitive disorders (HAND). Materials and methods: Pilot unblinded randomised controlled trial (parallel allocation 1:1) enrolling HIV‐infected patients on combination antiretroviral treatment (cART) in follow‐up at the San Paolo Infectious Diseases (SPID) Cohort, Milan, Italy. At screening, patients underwent a neuropsychological battery (11 tests, seven cognitive domains plus assessment of mental health and quality of life by Beck Anxiety Inventory, Beck Depression Inventory and Medical Outcome Study HIV Health Survey) to diagnose HAND (Frascati criteria). Patients diagnosed with HAND and meeting the eligibility criteria (exclusion criteria: AIDS‐defining illnesses, not adequately managed depression, neurological/psychiatric comorbidities, active alcohol/substance abuse, cirrhosis/severe comorbidities requiring hospitalisation, not comprehension of Italian language, detectable HIV‐RNA for cART‐treated patients) were randomly allocated to continue cART (control group) or to the cognitive rehabilitation training in association with cART (intervention group: ERICA software, 12 weekly sessions of nine computer‐based exercises lasting one hour under the supervision of a physician aimed at improving five cognitive domains). After completion of the protocol (t12), the intervention group was reassessed by the same neuropsychological battery; both control and intervention groups were also planned to be tested at Week 24. In the intervention group, repeated measures ANOVA was used to compare pre‐ and post‐rehabilitation mean T scores on each cognitive domain. Results: We screened 55 patients; 28 patients with HAND were randomised (14 intervention group, 14 control group). The two groups did not differ at randomisation. The raw scores obtained at the neuropsychological tests are corrected for age, educational level and gender and converted to normative T scores. In the intervention group, four patients declined and three patients were lost to follow‐up; seven patients completed the cognitive rehabilitation programme. At t12, the proportion of HAND has declined with no cognitive impairment in 2/7 (28%) patients. The mean T scores in two cognitive domains (attention/working memory and abstraction/executive functions), and five neuropsychological tests (Digit Span Test‐Backward, Rey Auditory Verbal Learning Test‐Delayed Recall, Rey‐Osterrieth Complex Figure Test‐Delayed Recall, Stroop Color and Word Test‐Errors and Phonemic Fluency Task), significantly improved from baseline to t12. Abstract P201 – Figure 1. Mean T scores for each cognitive domain at t0 (screening) and t12 (after 12 weeks of rehabilitation training) in the intervention group. Conclusions: While not recovering HAND, 12 weeks of this new computer‐based rehabilitation protocol was effective in improving specific cognitive domains. P202 The effect of drug use on specific neuropsychological domains among patients enrolled in the Neurocognitive Assessment in the Metabolic and Aging COhort (NAMACO) study K Darling 1, I Locatelli2, N Benghalem1, I Nadin3, P Brugger4, K Gutbrod5, S Frueh6, S Rossi7, U Kunze8, T Lecompte9, C Hauser10, B Hasse11, H Kovari11, P Tarr12, M Stoeckle13, C Di Benedetto14, P Schmid15, R Du Pasquier16 and M Cavassini1 1Infectious Diseases Service, Lausanne University Hospital, Lausanne, Switzerland. 2Division of Biostatistics and Quantitative Methods, Lausanne University Hospital, Lausanne, Switzerland. 3Department of Clinical Neurosciences, Lausanne University Hospital, Lausanne, Switzerland. 4Department of Neurology, University Hospital Zürich, Zürich, Switzerland. 5Department of Neurology, Inselspital Bern, Bern, Switzerland. 6Department of Neurology, Kantonsspital St. Gallen, St. Gallen, Switzerland. 7Unità di Neuropsicologia e Logopedia, Ospedale Civico Lugano, Lugano, Switzerland. 8Memory Clinic, Felix Platter Hospital, Basel, Switzerland. 9Infectious Diseases Division, Hôpitaux Universitaires de Genève, Geneva, Switzerland. 10Department of Infectious Diseases, Bern University Hospital, Bern, Switzerland. 11Department of Infectious Diseases, Universitätsspital Zurich, Zurich, Switzerland. 12University Department of Medicine, Kantonsspital Bruderholz, Bruderholz, Switzerland. 13Infectious Diseases, University Hospital Basel, Basel, Switzerland. 14Infectious Diseases, Ospedale Regionale di Lugano, Lugano, Switzerland. 15Infectious Diseases, Kantonsspital St. Gallen, St. Gallen, Switzerland. 16Department of Clinical Neurosciences, Lausanne University Hospital, Lausanne, Switzerland Background: HIV‐associated neurocognitive impairment (NCI) is diagnosed based on neuropsychological (NP) testing and functional assessment, after excluding non‐HIV‐associated confounders. The effect of active or previous drug use on the development of NCI among people living with HIV has been incompletely characterised. The aim of this study was to examine the influence of drugs on NP tests in a large cohort of HIV‐positive patients with well‐controlled infection. Methods: We analysed patients enrolled in the Neurocognitive Assessment in the Metabolic and Aging COhort (NAMACO) study, an ongoing, prospective, longitudinal, multicentre and multilingual study within the Swiss HIV Cohort Study (SHCS). NAMACO patients had to be aged ≥45 years old and speak a national language (French, German, Italian) to participate, and all underwent baseline NP tests. Drug use data, which are collected twice yearly at clinic visits, were obtained from the SHCS database. We defined binary (presence/absence of impairment) and continuous (mean z‐score) outcomes for NCI and five specific NP domains: motor skills, speed of information processing, executive function, attention/working memory and verbal learning/memory. The effects of previous injecting drug use (IDU) (heroin) and current non‐IDU (cocaine, cannabis) were examined with uni‐ and multivariable logistic and linear regression models adjusted for age, sex, ethnicity, education level and alcohol consumption. Results: Among 981 included patients, mean age was 54.5 years (SD 7.5); 782 patients (79.7%) were male, 899 (91.6%) were Caucasian and 942 (96.2%) had undetectable viral loads ( 9 per 10,000 ppy after 72 months of exposure. Compared to the controls, in univariate analyses, cases were more likely to have advanced disease (WHO stage III to IV) and have lower CD4 counts but other baseline characteristics were similar (Table 1). In multivariate analyses, patients at WHO stage III and IV at ART initiation had greater odds of switching from TDF‐based regimen due to toxicity compared to those who were in WHO stage I and II (OR 2.12, 95% CI 1.25 to 3.55, p = 0.004). Higher CD4 at ART start was protective of switching (OR 0.94, 95% CI 0.89 to 0.98, p = 0.017). However, we did not find any association between odds of substitution and other baseline characteristics (Table 1). Abstract P205 – Table 1. Patient characteristics at baseline Characteristics Cases (n = 113) Controls (n = 339) p value Male sex, % 42.48 44.31 0.736 Age, median (IQR) 42.1 (35.00 to 51.20) 41.50 (34.8 to 49.10) 0.57 BMI, % 0.584 Underweight ( 25) 19.23 24.18 WHO stage, % 140/90 mmHg) there was not significant difference between the groups (Table 1.) Abstract P207 – Table 1. Distribution on PCR and high blood pressure PCR mg/mmol N (%) High BP % Normal 50 5 (5.3) 1 20.0 Total 94 40 So using the modified KDIGO classification (Table 2) we had a prevalance rate of low 69.1%, moderate 18.1%, high 11.7%, very high 1.1%. Compared to a Japanese cohort we have higher attrition rate of CKD in our HIV population [3]. They analysed 1447 PLWHIV (97% male; 3% female), with an average age of 44.4 years, and had prevalance rates of low 85.9%, moderate 11.0%, high 2.1% and very high 1.0%. So although the groups demographics are slightly different we still believe there are further multifactorial causes for the higher prevalence of more severe renal disease, a reason for the higher prevalence of CKD between the populations is likely due to the eGFR not accounting for Afro‐Caribbean ethnicity. Abstract P207 – Table 2. Distribution of PLWHIV determined by the modified KDIGO 12 classification CKD grading eGFR (mL/min) P1 (%) P2 (%) P3 (%) Total (%) G1 >90 28.7 9.6 1.1 39.4 G2 60 to 90 40.4 7.4 4.3 52.1 G3a 45 to 59 1.1 5.3 0.0 6.4 G3b 30 to 44 1.1 0.0 1.1 2.1 G4 15 to 29 0.0 0.0 0.0 0.0 G5 10). Presence of depressive symptoms (p six months of cART, from March 2017 to March 2018. The analyses focused on the HIVDQoL overview item measuring generic QoL (3 = ’excellent’ to −3 = ’extremely bad’) and the EQ‐5D visual analogue score (EQ‐VAS) measuring self‐rated health (100 = ’best imaginable health state’ to 0 = ‘worst’). Analyses included non‐parametric tests of difference and correlational analyses. Results: One hundred and thirty‐five patients were included (122 men; 13 women), mean age 43 (SD 12.25). One hundred and seven patients were on cART (NNSTI, N = 66; NNRTI, N = 23; PI, N = 15/r‐based regimen). Mode of transmission included: MSM (N = 76), heterosexual (N = 40) and IDU (N = 10). Mean CD4+ was 655/mm3 (SD 316) for those on cART and 429/mm3 (SD 259) for patients’ pre‐treatment. Mean self‐reported health (EQ‐VAS) was 79 (SD 14.57) for cART‐treated and 78 (SD 18.73) for those pre‐treatment. Generic QoL (HIVDQoL item (1) mean was 1.21 (SD 1.19) (>‘good’ QoL) for cART‐treated and 0.48 (SD 1.74) (midway between ‘neither good nor bad’ and ‘good’) for those pre‐treatment. EQ‐VAS health scores were found to differ by mode of transmission, with MSM reporting better health than IDU (p = 0.022) and those reporting heterosexual transmission (p = 0.043). However, there was no difference in QoL by mode of transmission. Treatment with cART was associated with better QoL than pre‐treatment status (p = 0.049), without differences in health ratings. QoL, but not health status, was significantly worse for patients with CD4 count of 200. Perceived health, but not QoL, differed with mode of HIV transmission. QoL is not simply a reflection of health status and it is important to measure both outcomes. P214 Syphilis on the rise in HIV‐positive MSM in Germany J Wesselmann, C Boesecke, J Wasmuth, J Rockstroh and C Schwarze‐Zander Department of Internal Medicine I, University Hospital Bonn, Bonn, Germany Background: STDs, such as syphilis, have been increasing in recent years among MSM, often HIV+ patients, due to more frequent condomless sex. Aim of our study was to evaluate incidence of syphilis infection, impact on immunological and laboratory markers and treatment response of serological markers in a German cohort. Methods: This retrospective study included 859 HIV+ patients screened for syphilis infection (TPPA, VDRL) November 2015 to May 2017 in the HIV outpatient clinic at Bonn University Hospital. The impact of syphilis and its treatment on renal function markers (serum creatinine, GFR), liver enzymes (gamma‐GT, ALT, AST), inflammatory parameters and blood count (CRP, Hb, LDH) and immune response (leucocytes, CD4 count, CD8 count, CD4/CD8 ratio) was investigated three to six months before, at time of syphilis diagnosis, and three to six months after treatment. Serological response to syphilis treatment (VDRL, TPPA) was investigated every three months after treatment. Results: In the study period 43/859 (5%) patients were newly diagnosed with syphilis. Of these 3/43 (7%) were re‐infected within the observation period. Compared to incidence of syphilis infection between 2000 and 2010 there was a 2.4‐fold increase in 2016. Past syphilis infection was detected in 28% (244/859). All patients with syphilis were male and 97% MSM, compared to the whole study population patients were younger (mean age 44 years vs. 49 years) and fewer had symptomatic HIV disease (77% CDC stage A vs. 57%). Only 37% developed symptoms of syphilis (47% exanthema, 20% chancres, 20% uveitis, 13% urethritis). At the three observed timepoints mean gamma‐GT increased from 49 U/L to 70 U/L (p = 0.001) and decreased to 53 U/L, respectively, CRP increased from 2.1 to 7.4 mg/dL and decreased after treatment to 1.6 mg/dL (p = 0.002) and the mean CD4 count dropped from 670/µL to 646/µL at time of syphilis diagnosis and increased significantly after treatment to 715/µL (mean, p = 0.022). The relative CD4 cell count did not change during the observation period. Following syphilis treatment VDRL titer showed a slow decrease. After three to six months only 50% showed a ≥ 4‐fold decrease, which increased after nine to twelve months to 86%. Conclusion: Syphilis co‐infection has dramatically increased in our HIV+ population, especially in younger, healthier MSM. Regular screening is extremely important in this group of HIV+ patients as more than half of syphilis cases miss symptoms of infection. Elevation of gamma‐GT and CRP and decrease of absolute CD4 cell count may be an indicator of syphilis infection. VDRL can show a slow decrease after treatment and requires monitoring. P215 An assessment of how effectively health systems monitor HIV‐associated comorbidities, using current global and European frameworks K Safreed‐Harmon1, J Pericàs2, M Kall3, U Davidovich4, J del Amo5, J Lazarus1 and J Anderson 6 1Barcelona Institute of Global Health (ISGlobal), University of Barcelona, Barcelona, Spain. 2Infectious Disease Service, Hospital Clínic, University of Barcelona, Barcelona, Spain. 3National Infection Service, Public Health England, London, UK. 4Public Health Service of Amsterdam, Amsterdam, Netherlands. 5National Center for Epidemiology, Institute of Health Carlos III, Madrid, Spain. 6Jonathan Mann Clinic, Homerton University Hospital NHS Foundation Trust, London, UK Background: Today we have the tools to deliver effective long‐term viral suppression of HIV. Data from continua of care show increasing proportions of PLWHIV progressing to viral suppression in countries at all income levels. Yet alongside this progress, there is a growing burden of non‐AIDS‐defining comorbidities and related health concerns for PLWHIV. Decisions about which elements of service coverage and which health outcomes countries monitor have important implications for how health systems focus their HIV responses. This study examines whether existing monitoring frameworks sufficiently enable European countries to observe and understand the comorbidities that impact on the health and well‐being of PLWHIV. Materials and methods: Drawing on recent literature, we identified 15 non‐AIDS‐defining comorbidity areas that contribute to poor health in PLWHIV globally: bacterial and viral infections (excluding bacterial STIs), bacterial STIs (chlamydia, gonorrhoea and syphilis), cardiovascular, digestive, drug toxicities, endocrine/metabolic, haematological, liver (including HBV and HCV), malignancies, malnutrition/wasting, neurological, parasitic infections (including malaria), renal, respiratory and psychiatric conditions. Three researchers independently assessed the extent to which each comorbidity area was monitored with regard to: (1) service access; and (2) disease burden in four monitoring frameworks: Global AIDS monitoring 2018 (UNAIDS); Modular framework handbook (The Global Fund); MER 2.0 indicator reference guide (PEPFAR); and the 2018 Dublin Declaration questionnaire (European Centre for Disease Prevention and Control). Researchers assigned grades of A when comorbidities were addressed comprehensively, B when comorbidities were addressed but not comprehensively and C when comorbidities were not addressed. Discrepancies were resolved through consultation. Results: Over half (8/15) of the comorbidities were not mentioned in any of the four monitoring frameworks (malignancies, parasitic infections, and digestive, endocrine/metabolic, haematological, neurological, renal and respiratory diseases/disorders). Across the four frameworks, there were more grades of A or B for access to services (11) than for comorbidity burden (four), and neither MER 2.0 nor the Dublin questionnaire included any indicators monitoring the comorbidity burden. The only item addressed comprehensively in Global AIDS Monitoring was comorbidity burden for drug toxicities. The only items addressed comprehensively in the Dublin questionnaire were access to services for bacterial STIs, liver diseases and psychiatric disorders. Conclusions: We found major HIV monitoring frameworks fail to comprehensively address most non‐AIDS‐defining comorbidities, particularly chronic conditions associated with ageing. As the continuum of HIV care is reconceptualised to reflect long‐term health and well‐being, monitoring frameworks must be revised to include non‐AIDS‐defining comorbidities. This will encourage prevention, diagnosis and treatment of comorbidities consequently improving long‐term health outcomes and quality of life. P216 Delayed but adequate serologic response to syphilis treatment in HIV‐positive adults M Ren 1, L Szadkowski2, D Tan3, S Walmsley4 1Medicine, University of Toronto, Toronto, Canada. 2Biostatistics Research Unit, University Health Network, Toronto, Canada. 3Infectious Diseases, St. Michael's Hospital, Toronto, Canada. 4Infectious Diseases, University Health Network, Toronto, Canada Background: Adequate response to syphilis treatment is a four‐fold decrease in serum RPR at 6 (early syphilis) and 12 months (latent syphilis). We characterized the timeline of serologic response to syphilis treatment in HIV‐positive adults. Methods: We conducted a chart review of 532 HIV‐positive adults with positive syphilis serology between 2000 and 2017. Inclusion criteria were: reactive pre‐treatment RPR titer; documentation of date and type of syphilis therapy; reversion to a non‐reactive RPR or at least six months or one year of follow‐up for early syphilis and late syphilis/neurosyphilis, respectively. The first eligible episode was included. Time to four‐fold decrease and non‐reactive RPR was calculated using Kaplan‐Meier estimates. Univariable proportional hazards models assessed associations between clinical covariates and time to four‐fold decrease and non‐reactive RPR. Results: Two hundred and thirty‐one patients did not have a reactive pre‐treatment RPR titer. One hundred and eighty‐nine male patients (87% MSM) met inclusion criteria. Median (IQR) age was 42 (35 to 48), median (IQR) CD4 count was 443 (273 to 609) and 56% of patients had a suppressed viral load. Seventy‐five percent were on ARVs and 57% had a baseline RPR titer ≥1:64. Twelve percent were primary syphilis, 28% secondary, 12% early latent, 28% late latent and 19% neurosyphilis. Median (IQR) follow‐up was 2.55 (1.53 to 6.14) years. Seventy‐two percent received IM doses of benzathine penicillin G (27% 1 dose, 45% 2 to 3 doses), 21% received intravenous penicillin G and 5% doxycycline. One hundred and eighty‐four (97.4%) had at least a four‐fold decrease in RPR and 96 (52.2%) reverted to a non‐reactive RPR. The median (95% CI) time to four‐fold decrease and non‐reactive RPR was 0.37 (0.33 to 0.45) years and 2.72 (2.30 to 4.53) years, respectively. Patients with baseline titer ≥1:64 were more likely to achieve a four‐fold decrease (HR 1.36, p = 0.05), but less likely to achieve a non‐reactive RPR (HR 0.41, p seven days) 9 (16.4%) 644 (9.3%) 0.16 Cardiac surgery in‐hospital 14 (25.5%) 3487 (47.7%) 40 IU/L, a significant proportion reduced this parameter (median change −23.5, −38.5 to −13.5, p 200 cells/mm3 in 296/323 (90%). At enrolment, past HBV infection was documented in 106 (32%) patients, 101 (31%) had been immunised and had documented anti‐HBS antibodies >100 IU/L [AH6] and 121 (37%) were susceptible or had unknown or incomplete HBV immunity. Patients up to date for “basic vaccines” (immunisations/immunity) were: HBV 206 (63%); HAV 189 (58%); varicella 184 (56%); measles 48 (15%); tetanus 8 (2%); and “complementary vaccines”: pneumococcus 0 (0%). By the first interim analysis (1 June 2018), immunisations/immunity rates had increased for all vaccine‐preventable diseases such as tetanus 141, 43% (difference 45.5%); pneumococcus 87, 26.5% (difference 26.5% [EN9]); measles 97, 29.6% (difference 15%); HAV 206, 63% (difference 5%); varicella 187, 57% (difference 0.9%); HBV 207, 63% (difference 0.3%) (Table 1). Conclusions: Despite regular visits, HIV patients in our expert care centre were poorly immunised / protected against vaccine‐preventable diseases. The implementation of a systematic evaluation, supported by an expert CDSS generating ready‐to‐use catchup plans, demonstrated that this did not reflect vaccine refusal but failure to identify missing immunisations during routine practice. Abstract P219 – Table 1. Up‐to‐date immunisation/immunity against vaccine preventable disease in HIV patients Up‐to date immunisation At enrolment n = 328 On June 1st 2018 n = 328 Difference n % (95% CIa) % (95% CIa) % (95% CIb) Tetanus 8 2.4% (1.1 to 4.7) 141 43.0% (37.6 to 48.5) 40.5% (35.2 to 46.1) Hepatitis A 189 57.6% (52.1 to 63.0) 206 62.8% (57.3 to 68.1) 5.2% (3.0 to 8.2) Hepatitis B 206 62.8% (57.3 to 68.1) 207 63.1% (57.6 to 68.3) 0.3% (0.0 to 1.7) Measles 48 14.6% (11.0 to 18.9) 97 29.6% (24.7 to 34.8) 14.9% (11.3 to 19.3) Varicella 184 56.1% (50.5 to 61.5) 187 57.0% (51.5 to 62.4) 0.9% (0.2 to 2.6) Pneumococcus 0 0.0% (0.0 to 1.1) 87 26.5% (21.8 to 31.7) 26.5% (21.8 to 31.7) a95% confidence intervals obtained with Clopper‐Pearson's method. bdifference in number of patients divided by 328 (95% confidence intervals obtained with Clopper‐Pearson's exact method). P220 Comparison of early serological response of early syphilis to treatment with a single‐dose benzathine penicillin G between HIV‐positive and HIV‐negative patients: a cohort study C Yang 1, W Liu2, L Chang2, C Wu2, Y Su2 and C Hung2 1Department of Internal Medicine, Far Eastern Memorial Hospital, New Taipei City, Taiwan. 2Department of Internal Medicine, National Taiwan University Hospital, Taipei, Taiwan Background: Serological response of early syphilis to treatment has been reportedly poorer in HIV‐positive patients compared with HIV‐negative patients; however, the interpretation of the published data is limited by the differences in study design, subjects with different stages of syphilis included, definition used for serological response, treatment administered and follow‐up frequency and duration. We aimed to compare the early serological response to benzathine penicillin G (BPG) during the monthly follow‐up for 3 consecutive months between HIV‐positive and HIV‐negative patients with early syphilis. Materials and methods: Since January 2015, adult patients aged 20 years or older who presented with early syphilis (primary, secondary and early latent syphilis) with baseline rapid plasma regain (RPR) titers of 4 or greater were included in this prospective observational study after the patients received a single dose of BPG for early syphilis according to the STD Treatment Guidelines 2015 of US CDC. RPR titers were determined at baseline and thereafter every 4 weeks for 12 weeks, followed by every 12 weeks. Serological response was defined as decline of RPR titer by four‐fold or greater at each time point compared with baseline. Serological failure was defined as an increase of RPR titer by four‐fold or greater during follow‐up after ever achieving a decline of the titer. Results: Between January 2015 and March 2018, 151 HIV‐positive and 48 HIV‐negative patients were included; all were men who have sex with men. Compared with HIV‐positive patients, HIV‐uninfected patients had more cases of secondary syphilis (52.5% vs. 32.9%, p = 0.027) and fewer early latent syphilis (40.0% vs. 58.9%, p = 0.048) and prior syphilis (20.0% vs. 67.8%, p 40 copies/mL. Conclusions: Hospitalisation among cART‐treated HIV‐positive patients declined in the recent years; INF, and in particular pneumonias, still represent the main cause of hospital admission in virally suppressed cART‐treated patients. With the ageing of HIV population, hospitalisations for non‐AIDS cancers doubled over 2010 to 2017. P223 Use of recreational drugs and sexually transmitted infection (STI) diagnosis among patients attending a STI/HIV reference clinic in Rome A Latini 1, M Colafigli1, M Frasca1, L Alei1, M Giuliani1, A Cristaudo1 and M Zaccarelli2 1STI/HIV Unit, San Gallicano Dermatologic Institute, IRCCS, Rome, Italy. 2Clinical Department, National Institute for Infectious Diseases Lazzaro Spallanzani IRCCS, Rome, Italy Background: To assess frequency and association with recent STI and viral hepatitis (VH) diagnosis with reported recreational drug (RD) use among clients of a reference centre in Rome. Methods: Patients attended the centre for HIV treatment or STI visit between January and May 2018. Patients, after signing informed consent, self‐compiled a questionnaire concerning RD and sexual behaviours. Data about drugs used during sex were collected: crystal meth, GHB and mephedrone (3 chems) and in addition with MDMA, cocaine, ketamine, erectile dysfunction agents (EDA), poppers and steroids (9 chems). Association with STI or VH diagnosis in the past six months, HIV status, demographic and behavioural data was assessed. Results: Overall, 401 patients were included (203 attending the centre for STI and 198 for HIV treatment), of them 235 (59%) were MSM, 113 (28%) hetero males (HM) and 53 (13%) females. The proportion of HIV patients was higher among MSM (76% vs. 10% among other groups). Although cannabinoids were the most widely used, cocaine, poppers, EDA, crack and mephedrone use are significantly more frequently reported by MSM, who reported more often poly‐drugs use. Among MSM, recent STI or VH diagnosis in the past six months were significantly associated with RD use, overall and for syphilis, hepatitis A and urethritis (Figure 1). Patients who reported any of 3 chems use also reported poly‐RD use (mean number of RD reported: 6.3 ± 3.0 vs. 0.8 ± 1.2 among patients who did not report use of any 3 chems). Condomless sex with non‐steady partners was significantly more frequently reported (p 3.25 were associated with higher risk of the outcome, while reduced risk was associated to baseline CD4 > 500 cells/mm3 and nadir CD4 > 200 cells/mm3. Conclusion: In our study population of HCV/HIV co‐infected subjects, CMV IgG positivity does not seem to be associated with a higher risk of liver and AIDS progression. The current use of DAA could allow us to evaluate the impact of HCV eradication in this specific population. Abstract P228 – Figure 1. KM analysis for the three endpoints: Fib‐4 index ≥3.25, end stage liver disease, AIDS events/death. References [1] Lichtner M, Cicconi P, Vita S, Cozzi‐Lepri A, Galli M, Lo Caputo S, et al. for the Icona Foundation Study. Cytomegalovirus coinfection is associated with an increased risk of severe non‐AIDS‐defining events in a large cohort of HIV‐infected patients. J Infect Dis. 2015;211:178‐86. [2] Hsue PY, Hunt PW, Sinclair E, Bredt B, Franklin A, Killian M, et al. Increased carotid intima‐media thickness in HIV patients is associated with increased cytomegalovirus‐specific T‐cell responses. AIDS 2006;20:2275‐83. [3] Kuniholm MH, Parrinello CM, Anastos K, Augenbraun M, Plankey M, Nowicki M, et al. Hepatitis C viremia is associated with cytomegalovirus IgG antibody levels in HIV‐infected women. PLoS One. 2013;8:e61973. P229 The importance of serological testing and risk factors of measles, mumps, rubella and VZV among HIV‐infected adults in Istanbul, Turkey during measles outbreak in Europe O Altuntas Aydin 1, S Senoglu1, H Kumbasar Karaosmanoglu1, Z Yesilbag1 and A Aydin2 1Infectious Diseases and Clinical Microbiology, University of Health Sciences, Bakirkoy Dr Sadİ Konuk Training and Research Hospital, Istanbul, Turkey. 2Neurosurgery, University of Health Sciences, FTR Training and Research Hospital, Istanbul, Turkey Background: Measles, mumps, rubella (MMR) and varicella zoster virus (VZV) infection can cause serious diseases and complications in HIV/AIDS patients. Due to successful vaccination programmes measles has become neglected in Turkey. Even though welcoming more than 3 million immigrants from Syria in the last seven years, measles outbreak has not been encountered in our country. However, recent outbreaks of measles are ongoing in Europe, especially in Italy, France, Romania and Turkey's neighbourhood Greece [1]. The objective of this study was to determine of MMR and VZV seronegativity and the risk factors associated being seronegative in HIV‐infected adults in Istanbul, Turkey. Materials and methods: All HIV‐infected patients in our cohort who had MMR and VZV serological tests performed between January 2016 and May 2018 were retrospectively identified. Sera were tested for MMR and VZV IgG using commercial immunoassays. Age, gender, CD4 cell counts, MMR and VZV IgG serological results were evaluated from the records. Statistical analyses were performed by using SPSS Statistics v.21.0. Categorical variables were compared using chi‐square test and Fisher's exact test. Results: MMR and VZV IgG serologies were available in 415 of 712 patients in active follow‐up (58.2%). Out of 415 naive patients 88.6% were men and median age was 35.6 years (range 18 to 67). Seronegativity was found in 3.9% for measles, 4.8% for rubella, 8.2% for mumps and 3.4% for VZV. Among 18.7% measles, 23.5% mumps, 35% rubella and 7.1% varicella IgG seronegative patients CD4 cell count was less than 200/mm3. Being born after 1983 is strongly associated with seronegativity against measles (p = 0.00) and a nadir CD4 cell count below 200/mm3 is independently associated with rubella seronegativity (p = 0.013). Conclusions: There is high need for MMR vaccination in HIV‐infected patients in Turkey born in 1983 or later. Measles continues to spread across Europe as the vaccination coverage in many European countries is suboptimal. Systematic measles antibody screening and vaccination should be performed in HIV‐infected individuals to prevent serious disease and complications in the era of vaccination. Reference: [1] European Centre for Disease Prevention and Control (ECDC). Monthly measles and rubella monitoring report, April 2018. Stockholm: ECDC; 2018. P230 Prescription patterns of comedication and potential for drug‐drug interactions with ART in HIV patients in a retrospective claims database in Germany: implications for adequate HIV treatment selection S Lopes 1, K O'Day2, J Van Stiphout3, Y Punekar1, M Radford4 and J Haas5 1Global Health Outcomes, Viiv Healthcare, London, UK. 2Global Health Economics, Xcenda LLC, Palm Harbor, FL, USA. 3Global HEOR and Market Access, Xcenda Switzerland GmbH, Bern, Switzerland. 4Health Outcomes Europe, Viiv Healthcare, London, UK. 5Real World Evidence, Xcenda GmbH, Hannover, Germany Background: ART has increased life expectancy of PLWHIV, with consequent increase in the prevalence of comorbidities, and polypharmacy. This in turn increases the risk of drug‐drug interactions (DDIs) and contraindications (CIs) with ART, which are amongst the main reasons for changing ART. Different first‐line recommended ART regimens have different DDI and CI profiles. The aim of this study was to estimate the rate of potential DDIs and CIs of real‐world prescribed non‐ART comedication, with first‐line recommended ART in PLWHIV in Germany. Materials and methods: A retrospective, cross‐sectional cohort study design was used to collect non‐ART comedication prescription data from a representative sample of German health insurance claims from the InGef research database, in 2016. Adult patients with an ICD‐10 diagnosis code for HIV and who were prescribed any ART during 2016 were included in the analysis. Patients were stratified by gender, age, comorbidities and time on ART. Prescribed comedications were used to estimate the rate of potential for DDIs and CIs for each recommended first‐line ART per patient for the year of 2016, based on criteria from an established DDI reference database (www.hiv‐druginteractions.org). Results: Two thousand six hundred and eighty patients were included in the study, 86% were male, with mean age of 45.6 years, 34% were aged 50 years or over, and 29% had been on ART for at least five years. A total of 8025 prescribed comedications were counted in 2016. Overall, approximately one out of seven patients could potentially have a DDI/CI when prescribed raltegravir + emtricitabine/tenofovir alafenamide (FTC/TAF), one in six patients could potentially have a DDI/CI when prescribed with dolutegravir (DTG) /abacavir/lamivudine (ABC/3TC) or DTG + FTC/TAF; and approximately one in five patients could potentially have a DDI/CI when prescribed bictegravir/FTC/TAF. Boosted regimens presented the highest potential for DDIs/CIs (Table 1). Abstract P230 – Table 1. Rate of potential drug‐drug interaction or contraindication with each first‐line recommended ART regimen per patient per year Regimens Overall population (n = 2680) Dolutegravir/abacavir/lamivudine 0.159 Elvitegravir/cobicistat/emtricitabine/tenofovir alafenamide 0.851 Bictegravir/emtricitabine/tenofovir alafenamide 0.218 Dolutegravir+emtricitabine/tenofovir alafenamide 0.174 Darunavir/cobicistat/emtricitabine/tenofovir alafenamide 1.014 Raltegravir+emtricitabine/tenofovir alafenamide 0.150 Rilpivirine/emtricitabine/tenofovir alafenamide 0.422 Conclusions: Results indicate that comedication with potential DDIs/CIs with first‐line recommended ART are frequently prescribed among real‐world HIV patients in Germany. The potential risk of DDIs and CIs varies greatly by ART regimen. As patients grow older, understanding each ART regimen's DDI profile, alongside the comorbidities and comedications of the HIV population, can help inform treatment decisions, based on potential risks and benefits of available recommended ART and may avoid future complications. P231 Syphilis and HIV: characteristics of the co‐infection in patients newly diagnosed with HIV I Abreu, P Palma, L Graça, R Filipe, R Ruas, E Branco, C Caldas, C Piñeiro, J Soares, M Tavares, R Serrão and A Sarmento Infectious Diseases, Centro Hospitalar São João, Porto, Portugal Background: The prevalence of syphilis continues to rise in Europe, particularly in groups with high risk sexual behaviour, like MSM and sex workers [1,2]. It is also increased in people living with HIV. Furthermore, in these patients, a case of untreated syphilis can have direct implications on the course of this disease, leading to increase in plasma HIV viral load and decrease of T CD4+ lymphocytes [3]. Our objective with this study was to identify every case of syphilis in newly diagnosed HIV patients followed in our outpatient clinic in Centro Hospitalar São João, while trying to establish possible risk factors for the acquisiton of syphilis. Materials and methods: Retrospective observational study. We reviewed all patients that were newly diagnosed with HIV in our patient clinic between January 2015 and December 2017. In our clinic, we use the Treponema pallidum particle agglutination assay (TPPA) for the initial screening of syphilis. If positive, we do the Venereal Disease Research Laboratory (VDRL) test. For discordant results, we confirm the diagnosis with the fluorescent treponemal antibody absorption test (FTA‐ABS). Results: We identified 217 patients: 168 (77.4%) were men; the median age was 39 years. The mode of HIV exposure was heterosexual in 108 cases (49.8%); MSM in 97 (44.7%); intravenous drug use in five cases (2.3%) and unknown in seven cases (3.2%). There were 59 cases (27.2%) of syphilis: 22 (37.3%) were past treated infections and 37 (62.7%) were unknown and untreated infections (54.1% late latent syphilis; 13.5% early latent syphilis; 18.9% primary syphilis, 0.08% neurosyphilis; 0.03% retinitis; and 0.03% secondary syphilis). When we compared patients with syphilis and patients without, we found a statistically significant association between being an MSM and being diagnosed with syphilis (p = 0.04). The percentage of cases diagnosed per year remained stable during the period of time we studied: 30% in 2015, 20.2% in 2016 and 32.8% in 2017. Conclusion: The overall prevalence of syphilis in our group of HIV patients was of 27.2%, with a frequency that has remained stable through the last three years. The association between this diagnosis and MSM reflects the need to develop better campaigns on high risk groups for the adoption of safer sexual practices and periodic screening of sexually transmitted diseases. Further studies in our clinic intend to see which of these patients got re‐infected with syphilis and if we can use the syphilis screening as a marker for risk of other sexually transmitted diseases. References [1] European Centre for Disease Prevention and Control (ECDC). Syphilis. In: ECDC. Annual epidemiological report for 2015. Stockholm: ECDC; 2017. [2] Mulhall BP, Wright ST, De La Mata N, Allen D, Brown K, Dickson B, et al. Risk factors associated with incident sexually transmitted infections in HIV‐positive patients in the Australian HIV observational database: a prospective cohort study. HIV Med. 2016;17:623‐30. [3] Kalichman SC, Pellowski J, Turner C. Prevalence of sexually transmitted co‐infections in people living with HIV/AIDS: systematic review with implications for using HIV treatments for prevention. Sex Transm Infect. 2011;87:183‐90. P233 How are HIV patients dying? F Duarte, J Laranjinha, R Correia de Abreu and I Neves Infectious Diseases, Hospital Pedro Hispano, Matosinhos, Portugal Background: The introduction of HAART remains a hallmark in increasing the average life expectancy and the quality of life in HIV‐infected patients. However, diagnostic delay still is a reality with a high impact on mortality. Objectives: To assess the causes of mortality in HIV‐infected patients with a follow‐up at a 330‐bed tertiary hospital in the last 20 years, after the introduction of HAART. Methods: Two hundred and three deaths were extracted from a hospital database of 1230 HIV‐infected patients followed at the infectious diseases department, between 1997 and 2017. Out of these 203 patients, 113 were studied based on the attending physician (by the updated registration probability) and 11 were excluded (cause of death not registered). Epidemiological, immunological, virological and therapeutic aspects were evaluated at the time of diagnosis and of death. Results: One hundred and two HIV‐infected patients (77 males) were evaluated. At diagnosis, with an average of 38 ± 13 years old, intravenous drug use being the major risk of transmission (62%), only one case of HIV type 2, 79% with the diagnosis of a concomitant AIDS‐defining illness and 62% were late presenters (CD4+ T lymphocyte count 300 mg/g in 34% of cases. Forty‐eight patients had a proctological consultation and an abnormality was detected during this examination in 37.5% of them. One prostate cancer was diagnosed. A gynecological consultation was performed in 27 women (62%) and a genital or breast complication was detected in 15% of them. A faecal occult blood test was performed in 13 patients, and positive for one. Regarding the antiretroviral treatment, a therapeutic modification was proposed in 20% of cases (complication, side effect, prevention of toxicity, etc). Conclusions: Management with an annual summary report is recommended for the follow‐up of people living with HIV in France. Its realisation during a day‐care hospitalisation and the use of a standardised CRF allows to detect number of comorbidities. We showed that even in a medically followed population, vaccination coverage is insufficient, cardiovascular risk under‐evaluated and proctologic screening poorly performed. P236 The coverage of influenza and pneumococcal vaccination among HIV‐infected patients in Denmark: a cross‐sectional survey L Larsen 1, M Nguyen2 and I Johansen1 1Department of Infectious Diseases, Odense University Hospital, University of Southern Denmark, Odense, Denmark. 2Department of Infectious Diseases, Odense University Hospital, Odense, Denmark Background: Annual influenza vaccination is recommended for all HIV‐infected persons and so is pneumococcal vaccination at least once [1]. This is due to higher incidence rates for both infections among HIV‐infected persons [2,3] compared to HIV‐negative controls even in the later ART era. To clarify the use of and attitude towards these recommended vaccines among HIV‐infected persons in a high income country a questionnaire survey was performed. Materials and methods: In a single HIV outpatient clinic at a tertiary hospital adult HIV‐infected persons were invited to participate in the survey during their regular visit in Spring 2017. The questionnaire consisted of four parts: (1) Demography, educational level and household annual income. (2) Influenza vaccination for season 2016/17 and previous/next year and reasons for not being vaccinated. (3) Pneumococcal vaccine uptake ever and reasons for not being vaccinated. (4) Source of information for the patient regarding vaccination. To identify predictors for the 2016/17 influenza vaccine we used Fisher's exact test and student t‐test. p 50 copies/mL 6.36 (4.64 to 8.73) 350 cells/μL. After a median follow‐up of 623 days (range 523 to 705), 23 (3.7%) seroreverted. In the case‐control study, peak anti‐HAV IgG titer and positive anti‐HAV response at six months were inversely associated with early seroreversion with adjusted odds ratio (aOR) of 0.13 (95% CI 0.01 to 0.41) and 0.02 (95% CI 0.001 to 0.19), respectively, while seroreversion was more likely to occur in patients with older age (aOR 1.50, 95% CI 1.13 to 2.64) or with HIV viraemia at 12 months (aOR 13.66, 95% CI 1.40 to 315.28) in multivariable analyses. Conclusions: During an outbreak setting, early seroconversion occurred in 3.7% of HIV‐positive patients after successful HAV vaccination after 20 months of follow‐up. Lower and delayed seroresponses to HAV vaccination, older age and failure to maintain HIV viral suppression at 12 months were associated with early seroreversion. P240 Current chronic hepatitis C treatment in HIV co‐infection in Portugal: a cohort of 2133 patients presented by GEPCOI (Portuguese Coinfection Study Group) A Miranda 1, J Mendez2, R Serrão3, F Vale4, M Manata5, S Pinto6, A Gomes7, M Prata8, P Pacheco9, R Pazos10, R Pereira11, A Martins12, I Germano13, S Rocha14, A Reis15 and R Sarmento e Castro2 1Serviço de Infecciologia e Medicina Tropical, Hospital de Egas Moniz, Centro Hospitalar de Lisboa Ocidental, Lisbon, Portugal. 2Serviço de Doenças Infecciosas, Centro Hospitalar do Porto, Porto, Portugal. 3Serviço de Doenças Infecciosas, Centro Hospitalar de São João, Porto, Portugal. 4Serviço de Doenças Infecciosas, Centro Hospitalar de Setúbal, Setúbal, Portugal. 5Serviço de Doenças Infecciosas, Hospital de Curry Cabral, Centro Hospitalar de Lisboa Central, Lisbon, Portugal. 6Serviço de Doenças Infecciosas, Centro Hospitalar de Gaia/Espinho, Gaia, Portugal. 7Serviço de Doenças Infecciosas, Hospital Garcia de Orta, Almada, Portugal. 8Serviço de Doenças Infecciosas, Centro Hospitalar e Universitário de Coimbra, Coimbra, Portugal. 9Serviço de Doenças Infecciosas, Hospital Fernando da Fonseca, Amadora, Portugal. 10Serviço de Medicina, Centro Hospitalar Universitário do Algarve, Hospital de Portimão, Portimão, Portugal. 11Serviço de Doenças Infecciosas, Centro Hospitalar Universitário do Algarve, Hospital de Faro, Faro, Portugal. 12Serviço de Doenças Infecciosas, Centro Hospitalar do Baixo Vouga, Aveiro, Portugal. 13Serviço de Medicina, Hospital de São José, Centro Hospitalar de Lisboa Central, Lisbon, Portugal. 14Serviço de Doenças Infecciosas, Unidade Local Saúde Alto Minho, Viana do Castelo, Portugal. 15Serviço de Doenças Infecciosas, Hospital dos Marmeleiros, Funchal, Portugal Background: Direct acting antiviral drugs (DAA) changed the paradigm of hepatitis C therapy, significantly improving treatment response rates, patient life expectancy and quality of life. In Portugal, DAA therapy has been sequentially reimbursed since 2015 and generalised use of interferon‐free regimens became current practice. HCV/HIV co‐infection is a priority to engage HCV treatment due to faster disease progression. Real‐world data regarding efficacy and safety of DAA treatment in HIV/HCV co‐infected patients remain scarce. Material and methods Multicentre, retrospective, observational study of a real‐life clinical cohort (15 Portuguese hospital centres) of HCV/HIV infected patients, HCV treated with DAA‐based regimens, who completed treatment and response evaluation (12/24 weeks after treatment). Demographic, epidemiological, clinical, virological and treatment response data were analysed. Hepatic fibrosis was assessed by non‐invasive methods. Statistical analysis was performed by Microsoft Excel and SPSS Statistics v.24.0. Results: A total of 2133 patients were included. Demographic characterisation revealed male predominance (83%), mean age of 46 years and 96% Portuguese origin. HCV was acquired by intravenous drug use in 91% and 69% of patients were naïve. Genotype prevalence was 67.8% G1, 1.2% G2, 15.3% G3 and 15.7% G4. Hepatic fibrosis evidenced METAVIR score ≤F2 in 59.2%, 18.7% F3 and 22.1% F4. Mean baseline TCD4 was 619 cells/mm3 and 96% of patients on antiretroviral therapy were virologically suppressed. Most prescribed DAA regimens were SOF/LDV (84%) and SOF+RBV (9.6%). Global sustained virological response (SVR) achieved was 95%. SVR by genotype was 96% G1, 76% G2, 91% G3 and 96% G4. SVR by fibrosis stage was 97% F1, 95% F2, 94% F3 and 92% F4. SVR by DAA regimen was SOF/LDV 96% (n = 1784), SOF+SMV 100% (n = 2), SOF+RBV 87% (n = 204), SOF+DCV 91% (n = 54), GZR/EBR 93% (n = 15), 3D 95% (n = 43), SOF/VEL 86% (n = 7) and SOF+PR 100% (n = 24). Linear regression analysis showed lower response rates in the treatment of genotypes 2/3 (p 9 kPa) by Fibroscan. Fifty‐eight (48.3%) IDUs were co‐infected with HIV (Table 1). Treatment with DAAs was initiated only in five co‐infected patients, four achieved sustained virological response, while one patient with genotype 3 was a non‐responder. Two educational HCV Masterclasses, organised within the HepEd work package, were attended by 150 health care practitioners (HCPs) and printed educational materials for patients, peers and social workers were distributed. Three peers were trained for awareness and linkage to testing and patients from key populations were addressed in nine peer support sessions (HepFriend). Abstract P244 – Table 1. Characteristics in HCV/HIV co‐infected patients from key populations with rapid HCV positive antibody test at screening Characteristics HIV‐positive patients (n = 58) CD4 cell count/µL median (IQR) 510 (327 to 659) HIV viral load (log10 copies/mL) median (IQR) 2.68 (1.27 to 4.57) cARTa n (%) 22 (37.9) HIV‐RNA undetectable n (%) 15/22b (68.1) METAVIR F score ≥9 kPa (Fibroscan evaluation) n (%) 13 (22.4) HCV viral load (log10 IU/mL) median (IQR) 6.27 (5.66 to 6.59) HCV‐RNA undetectable n (%) 6 (10.3) ALT level (IU/L) median (IQR) 66 (44 to 111) HBsAg positive n (%) 7 (12.0) acombined antiretroviral treatment. b22 IDUs on cART. Conclusions: HIV/HCV co‐infection among subjects from key populations was high, especially among IDUs from drug support centres. Socio‐economic and structural barriers limited the treatment with DAAs in this group. There is an urgent need to improve the HIV/HCV screening and linkage to care for patients from hard‐to‐reach populations, to actively involve HCPs in HIV/HCV management and to find the best modalities to overcome the barriers to treatment. P245 Increased total and LDL cholesterol plasma levels upon direct antiviral agents (DAAs) driven HCV eradication G Taliani 1, G Marchetti2, A Cingolani3, M Lichtner4, S Cicalini5, E Quiros Roldan6, M Ursitti7, E Girardi8, A Antinori5, M Puoti9, A d'Arminio Monforte2 and A Cozzi‐Lepri10, on behalf of the Icona and Hepalcona Foundation Study Group 1Department of Clinical Medicine, Sapienza University of Rome, Rome, Italy. 2Clinic of Infectious and Tropical Diseases, ASST Santi Paolo e Carlo, University of Milan, Milan, Italy. 3Institute of Clinical Infectious Diseases, Policlinic A. Gemelli, Catholic University of the Sacred Heart, Rome, Italy. 4Unit of Infectious Diseases, SM Goretti Hospital, Sapienza University of Rome, Latina, Italy. 5HIV/AIDS Unit, INMI L. Spallanzani IRCCS, Rome, Italy. 6Department of Clinical and Experimental Sciences, ASST Spedali Civili, University of Brescia, Brescia, Italy. 7Department of Infectious Diseases, S. Maria Nuova IRCCS Hospital, Reggio‐Emilia, Italy. 8Department of Epidemiology, INMI L. Spallanzani IRCCS, Rome, Italy. 9Division of Infectious Diseases, ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy. 10Institute for Global Health, University College London, London, UK. Background: HCV has complex interactions with human lipid metabolism leading to downregulation of cholesterol levels. Treatment with DAAs was proven to induce a sharp and significant increase in total and low‐density lipoprotein cholesterol (LDL) persisting after the end of treatment (EOT). It has been suggested that the increased risk of cardiovascular disease (CVD) associated with ritonavir‐boosted darunavir (DRV/r) use may be independent from dyslipidaemia [1]. The aim of this analysis was to examine cholesterol changes in HIV‐HCV co‐infected patients after HCV clearance and according to DRV/r, ATV/r or RAL exposure during DAA. Materials and methods: The analysis includes data of HIV/HCV co‐infected patients in the Icona and HepaIcona cohorts for whom pairs of biomarkers were available. The first pair includes the two most recent values in a window [‐12;0] months of the date of DAA initiation. The second pair uses the latest in the window [+4;+12] months of the date EOT. Mean values at each time point were calculated as well as the difference among pairs. Univariable paired t‐test were conducted to test whether the variations were significantly different from zero. An ANCOVA analysis was used to test whether there was an effect of DRV/r, ATV/r and RAL use. Results: We included 465 patients on ART, who achieved SVR; 22% on DRV/r, 20% on ATV/r and 24% on RAL. Patients’ characteristics: median age 52 (50 to 55) years; 26% female; median BMI 24 (21 to 26) kg/m2; median CD4 584 (360 to 828) cells/mm3; HCV genotype 1a (35%), 3a (18%), 4 (13%). Total and LDL‐cholesterol along with platelet count, which prior to DAA tended to be stable or decrease, significantly increased after HCV clearance whereas high‐density lipoprotein (HDL) cholesterol remained unchanged. These changes, which occur in a short time‐lapse, potentially contribute to an increase in CVD risk through shared or separate pathways (Figure 1). Moreover, in patients exposed to DRV/r a significant increase of total cholesterol was observed over T1–T2 compared to unexposed ones (Δ+10.7 mg/dL; p = 0.01); no evidence for a difference was found for ATV/r (Δ+2.1 mg/dL; p = 0.69) or RAL (Δ‐6.9 mg/dL; p = 0.08) treated patients. Abstract P245 – Figure 1. DAG for the model exploring the causal link between HCV‐RNA eradication and risk of CVD. Conclusions: A complex and rapid change of risk factors for CVD risk seems to occur in HIV‐HCV co‐infected patients after HCV eradication with DAA, including increase of total and LDL‐cholesterol and platelets. Exposure to DRV/r independently contributed to the increase of total cholesterol. It is increasingly important to fit the individuals’ risk profiles before HCV treatment. Reference: [1] Ryom L, Lundgren JD, El‐Sadr W, Reiss P, Kirk O, Law M, et al. Cardiovascular disease and use of contemporary protease inhibitors: the D:A:D international prospective multicohort study. Lancet HIV. 2018;5:e291‐300. P246 Molecular determining of HIV‐1 with the presence of hepatitis B virus and hepatitis C virus co‐infections M Sayan 1, M Ozguler2, F Sarigul Yildirim3, T Yidirmak4, A Gündüz5, B Dokuzoğuz6, M Çelen7, D İnan8, Y Heper9, G Munis Ersoz10, I Karaoglan11, N Ceran12, A Deveci13, S Ozturk14, S Sayin Kutlu15, H Ozkan Ozdemir16, A Akbulut17, S Yazici18, A Sener19, A Cagatay20 and S Unal21 1Clinical Laboratory and DESAM, Kocaeli University (Turkey) and Near East University (Northern Cyprus), Kocaeli and Nicosia, Turkey. 2Infectious Diseases, Elazıg Education and Research Hospital, Elazıg, Turkey. 3Infectious Diseases, Antalya Education and Research Hospital, Antalya, Turkey. 4Infectious Diseases, Okmeydani Education and Research Hospital, Istanbul, Turkey. 5Infectious Diseases, Sisli Etfal Education and Research Hospital, Istanbul, Turkey. 6Infectious Diseases, Ankara Numune Education and Research Hospital, Ankara, Turkey. 7Infectious Diseases, Dicle University, Faculty of Medicine, Diyarbakir, Turkey. 8Faculty of Medicine, Infectious Diseases, Akdeniz University, Antalya, Turkey. 9Infectious Diseases, Uludag University, Faculty of Medicine, Bursa, Turkey. 10Infectious Diseases, Mersin University, Faculty of Medicine, Mersin, Turkey. 11Infectious Diseases, Gaziantep University, Faculty of Medicine, Gaziantep, Turkey. 12Infectious Diseases, Haydarpasa Numune Education and Research Hospital, Istanbul, Turkey. 13Infectious Diseases, Ondokuz Mayıs University, Faculty of Medicine, Samsun, Turkey. 14Infectious Diseases, Fatih Sultan Mehmet Education and Research Hospital, Istanbul, Turkey. 15Infectious Diseases, Pamukkale University, Faculty of Medicine, Denizli, Turkey. 16Infectious Diseases, Bozyaka Education and Research Hospital, Izmir, Turkey. 17Infectious Diseases, Firat University, Faculty of Medicine, Elazig, Turkey. 18Infectious Diseases, Medeniyet University, Goztepe Education and Research Hospital, Istanbul, Turkey. 19Infectious Diseases, Onsekiz Mart University, Faculty of Medicine, Canakkale, Turkey. 20Infectious Diseases, Istanbul University, Faculty of Medicine, Istanbul, Turkey. 21Infectious Diseases, Hacettepe University, Faculty of Medicine, Ankara, Turkey Background: Because of the similar modes of transmission (include unsterile medical injection, blood transfusion, sexual intercourse and injecting drug use), co‐infection with viral hepatitis and HIV is increasingly seen as a major public health problem. In addition to that, co‐infection with HIV and HBV or HCV infection increases the urgency of starting ART [1]. On the other hand, increasing the demographic mass movements, may be changed the virus transmitted trends and may have a potential effect for HIV and co‐infection surveillance in the future. In this study, we aimed to determine the molecular characteristics of HIV‐1 in the presence of HBV and HCV co‐infections in Turkey. Material and methods The present study was conducted between March 2010 and March 2017. HIV‐1 RNA was detected and quantified by a commercial real‐time PCR assays. Subtyping and genotypic resistance analysis were performed by population sequencing of the viral protease and reverse transcriptase regions of HIV‐1 pol gene. Drug resistance mutations were defined according to the Surveillance Drug Resistance Mutation list as recommended by the World Health Organization [2]. Results: We detected totally 3896 HIV‐1 positive patients whose molecular laboratory tests were completed in Turkey. The viral hepatitis co‐infections were detected in 4.3% (n = 170) of all HIV‐1 infected patients in this study. HBV and HCV co‐infections were observed as 3.2% and 0.5% in HIV‐1 positive patients, respectively. Major HIV‐1 subtypes were detected as a group M, subtype B (67.5%). We observed 13.5% drug resistance mutation motifs in HIV‐1 genomes which included this study. NRTI, NNRT and PI resistance mutations have been investigated in a detailed manner and the mutation rates were determined 9.4%, 5.3% and 1.8%, respectively. Conclusions: HBV and HCV co‐infections can be seen more frequently in HIV‐positive patients because of similar transmission routes. However, the ART drug resistance mutation pattern are observed similar. The molecular characterisation of HIV‐1 genome for ART resistance is not different from non‐co‐infected. In the light of increasing demographic mass movements and their potential to affect infection transmission trends, patients with HIV‐1 and their viral hepatitis co‐infections should be recommended an carefully surveillance. References [1] A Working Group of the Office of AIDS Research Advisory Council (OARAC) [Internet]. Guidelines for the use of antiretroviral agents in adults and adolescents living with HIV [cited 2018 Jan 1]. Available from: http://https://aidsinfo.nih.gov/guidelines. [2] Bennett DE, Camacho RJ, Otelea D, Kuritzkes DR, Fleury H, Kiuchi M, et al. Drug resistance mutations for surveillance of transmitted HIV‐1 drug‐resistance: 2009 update. PLoS One. 2009;4:e4724. P247 Estimation of HCV infection prevalence in a Basic Health Area of Madrid (Spain) J Martínez‐Sanz1, P Pérez Elías2, M Herrero Delgado3, R Barea4, Y de la Fuente Cortés5, M Vivancos Gallego1, A Moreno Zamora1, S Ares Blanco3, L Polo Benito6, A Mesa7, C Labrador Manzanares8, P González Huerga9, C Chamorro Escobar5, A Cano Espín2, A Fernández Rivera2, C Santos Álvarez2, M Rodríguez10, B Romero10 and M Pérez‐Elías 1 1Infectious Diseases, University Hospital Ramón y Cajal, Madrid, Spain. 2Primary Care, C.S. García Noblejas, Madrid, Spain. 3Primary Care, C.S. Mar Báltico, Madrid, Spain. 4Primary Care, C.S. Panamá, Madrid, Spain. 5Primary Care, C.S. Aquitania, Madrid, Spain. 6Hospital General Universitario Gregorio Marañón, Madrid, Spain. 7Primary Care, C.S. Avenida de Aragón, Madrid, Spain. 8DUE, Hospital de la Princesa, Madrid, Spain. 9SUMMA, Madrid, Spain. 10Microbiology, University Hospital Ramón y Cajal, Madrid, Spain Background: In Spain, epidemiological data on HCV infection are scarce. Our objective is to estimate the prevalence of HCV infection in the Basic Health Area of the Ramón y Cajal Hospital (BHA‐RYC), and distinguish between new HCV diagnoses (NHCVD) and previously diagnosed infections, linked or not. Methods: Sub‐analysis of the DRIVE03 study, which was carried out in four health centres of the BHA‐RYC, where patients without HCV aged between 18 and 70 years were prospectively included. After completing an HCV risk exposure and indicator conditions questionnaire, screening was performed using rapid tests in those with a positive questionnaire and all those over 50 years of age. We assessed the number of tests performed, the NHCVD and the number of patients with previous diagnosis but not conscious or not linked to care. The NHCVD rates (NHCVD/test performed) were also estimated. Results: Seven thousand nine hundred and ninety‐one individuals were included, and a total of 4718 tests were performed (61.4% with a positive questionnaire, 38.6% for being >50 years), with a positive result in 56 (1.19%). The 65% were women and the median age was 51 years (IQR 38 to 58). Among total positive results, four (7.1%) were false positive, 15 (26.8%) were individuals with known HCV infection already treated and cured, 27 (48.2%) had at least one laboratory determination positive for HCV but they were not aware or linked to care, and 10 (17.9%) were NHCVD. Table 1 shows the prevalence by groups. Abstract P247 – Table 1. Estimation of prevalence and diagnostic rate of HCV by groups Total population (n = 7991) Anti HCV+ Global prevalence (n = 52) Previous diagnosis not treated prevalence (n = 27) New HCV diagnosis prevalence (n = 10) N (%) N Rate per 1000 (95% CI) N Rate per 1000 (95% CI) N Rate per 1000 (95% CI) Sex Male 2729 (34) 30 10.99 (7.06 to 14.93) 12 4.40 (1.91 to 6.89) 5 1.83 (0.23 to 3.44) Female 5262 (66) 22 4.18 (2.43 to 5.93) 15 2.85 (1.41 to 4.29) 5 0.95 (0.12 to 1.78) Age, years 18 to 30 1652 (21) 1 0.61 (0 to 1.79) 0 ‐ 0 ‐ 31 to 40 1814 (23) 4 2.21 (0.04 to 4.37) 2 1.10 (0 to 2.63) 1 0.55 (0 to 1.63) 41 to 50 1967 (25) 15 7.63 (3.77 to 11.5) 9 4.58 (1.59 to 7.56) 3 1.53 (0 to 3.25) 51 to 60 1729 (21) 29 16.8 (10.7 to 22.9) 16 9.25 (4.72 to 13.79) 6 3.47 (0.69 to 6.25) >60 829 (10) 3 3.62 (0 to 7.71) 0 ‐ 0 ‐ Country Spain 6032 (78) 46 7.63 (5.42 to 9.83) 24 3.98 (2.39 to 5.57) 8 1.33 (0.41 to 2.25) Eastern Europe 181 (2) 4 5.52 (0 to 16.35) 2 11.05 (0 to 26.36) 2 11.05 (0 to 26.36) Latin America 1228 (15) 1 0.81 (0 to 2.41) 1 0.81 (0 to 2.41) 0 ‐ Africa 86 (1) 1 11.63 (0 to 34.42) 0 ‐ 0 ‐ Education level Primary 2086 (26) 26 12.46 (7.67 to 17.26) 15 7.19 (3.55 to 10.83) 4 1.92 (0.04 to 3.80) Secondary 3243 (41) 21 6.48 (3.71 to 9.25) 10 3.08 (1.17 to 4.99) 4 1.23 (0.02 to 2.44) Higher 2662 (33) 5 1.88 (0.23 to 3.52) 2 0.75 (0 to 1.79) 2 0.75 (0 to 1.79) Total 7991 (100) 52 6.51 (4.74 to 8.28) 27 3.38 (2.10 to 4.65) 10 1.25 (0.48 to 2.03) Conclusion: In early era of direct‐acting antivirals for the treatment of HCV infection, the prevalence of both new HCV diagnosis and patients not aware or not linked to care remains high. The population with the highest rates were males, 51 to 60 years old, from Spain or sub‐Saharan Africa and with a low level of education. P248 Challenges in the micro‐elimination of HCV in HIV co‐infected individuals D Basoulis 1, M Papadopoulou1, E Cholongitas1, G Kalamitsis2, G Daikos1 and M Psichogiou1 11st Internal Medicine Department, Athens Laiko General Hospital, Athens, Greece2Hellenic Liver Patient Association ‘Prometheus’, Athens, Greece Background: In 2011 an epidemic outbreak of HIV infection amongst intravenous drug users (IDU) took place in Athens, Greece, increasing the burden of HIV‐HCV co‐infection and overcrowding the infectious diseases clinics [1]. Recent national policy changes have permitted the enrolment of HIV‐infected patients into HCV treatment programmes with direct acting antivirals (DAA) regardless of fibrosis status, allowing for micro‐elimination attempts to flourish [2,3]. Materials and methods: To facilitate this endeavour, we dedicated one of our clinic days solely to HCV treatment and made incremental steps to access the full range of health care services to individuals’ needs. Elastography was offered by a community organisation (Prometheus) at our practice. We systematically sought and arranged appointments and made sure that our patients adhere to treatment by frequently contacting them for encouragement and for rescheduling appointments if missed. Results: One hundred and ninety‐six patients with co‐infection were referred to our centre in a median time of 1.1 months (interquartile range [IQR] 0.00 to 3.4) after diagnosis. 94.4% (185/196) were IDUs. One hundred and sixty (81.6%) were male, predominantly of Greek descent (86.2%), with a mean age of 34.3 ± 8.27 years. Twenty‐seven (13.8%) attended only one visit. One hundred and sixty‐nine patients were successfully linked (>1 visit) and 159 (94.1%) started HAART. Thirty (18.9%) discontinued, and four (2.5%) were transferred to a different clinic. One hundred and twenty‐five (63.7%) of 196 patients were retained in our care with a median follow‐up of 48.4 months (IQR 21.9 to 59.6) and were all receiving HAART. In 108 (93.9%), HCV‐RNA was evaluated. Of these, nine (8.3%) had an undetectable viral load. Median HCV‐RNA load was 6.18 log10 IU/mL (IQR 5.71 to 6.61). Liver stiffness was measured in 59 patients (47.2%) with a median of 6.1 kPa (IQR 5.2 to 8.6, range 10.6). Ninety‐nine (91.3%) of 108 patients were tested for genotype. Our population was distributed as follows: 1a, 40.5%; 3, 38.2%; 4, 19.1%; mixed genotype 1 and 3, 2.2%. 89.9% (89/99) were treated for HCV infection. Five were treated with Peg‐interferon/ribavirin successfully, while 84/89 (94.4%) received DAA treatment. 47.6% (40/84) have already completed the antiviral scheme, two discontinued due to compliance issues and 42 are still undergoing treatment. Eleven have achieved successful SVR, while it is pending for the rest. Conclusions: Micro‐elimination is achievable only in patients retained to care. A large population of co‐infected IDUs is not retained to care and micro‐elimination seems challenging. Novel models of care are needed to link and retain key populations reducing barriers and to allow the therapeutic advances to deliver a public health benefit. References [1] Nikolopoulos G, Tsiara C, Paraskeva D, Sypsa V, Psichogiou M, Paraskevis D, et al. Rapid decline in HIV incidence among persons who inject drugs during a fast‐track combination prevention program after an HIV outbreak in Athens. J Infect Dis. 2017;215:1496‐505. [2] Marshall AD, Pawlotsky JM, Lazarus JV, Aghemo A, Dore GJ, Grebely J. The removal of DAA restrictions in Europe ‐ one step closer to eliminating HCV as a major public health threat. J Hepatol. 2018 Jun 28;https://doi.org/10.1016/j.jhep.2018.06.016. [Epub ahead of print]. [3] Lazarus JV, Wiktor S, Colombo M, Thursz M; EASL International Liver Foundation. Micro‐elimination ‐ a path to global elimination of hepatitis C. J Hepatol. 2017;67:665‐6. P249 Increase in 10‐year Framingham cardiovascular risk following HCV eradication with DAA‐based therapy in HIV/HCV co‐infected patients T Aldamiz‐Echevarría 1, F Tejerina1, L Perez1, C Diez1, P Miralles1, J Lopez1, F Parras1 and J Bellon2 1Infectious Diseases, Hospital Gregorio Marañón, Madrid, Spain. 2Bioestadística, Instituto de Investigación Biomédica Gregorio Marañón, Madrid, Spain Background: Clearance of HCV viral load can lead to a rise in cholesterol rate unmasking a hyperlipaemia condition previously blocked by the virus; but its real impact on cardiovascular risk profile is unknown. Our objective is to analyse a potential worsening in cardiovascular risk. Material and methods We performed an observational retrospective study nested in a prospective setting in all consecutive HIV‐HCV infected patients who were treated with direct antiviral agents (DAA) and achieved HCV clearance in Gregorio Marañón Hospital Madrid, from November 2014 until June 2017. All recruited patients had HIV and HCV infection, both confirmed with serology and PCR in our laboratory. HCV eradication was defined as getting sustained viral response (SVR) at Week 12 after treatment. Those patients who were on hypolipaemiant therapy before HCV treatment were excluded, even if it was removed the time on DAA therapy. Framingham score was calculated for all the included patients at baseline, and 12 weeks after treatment. Lipids and cardiovascular risk were analysed as continuous and ordinal variables. Multivariable logistic regression was used to test the association of variables with cardiovascular risk progression. Results: Two hundred and ninety‐seven HIV/HCV co‐infected, not previously hypercholesterolaemic, were treated with DAA for HCV infection in Gregorio Marañón Hospital from November 2014 till June 2017 achieving SVR12. Framingham score was performed before DAA and 12 weeks after HCV treatment in 190 of them. One hundred and fifty‐one (79.5%) were men. Median age was 51.2 years (SD 5.7). Seventy‐seven (47%) were stage C for CDC score. Except for four, all included patients were on ARV treatment at the moment DAA was started. Ninety (47.9%) required a change in its ARV regimen to make it compatible to HCV treatment. One hundred and seventy‐four (93%) were former or active smokers. Thirteen (6.8%) had diabetes mellitus. Sixty‐nine (37.3%) had cirrhosis and 10 (5.4%) decompensated cirrhosis. Eighty‐three (43.6) received a PI‐containing regimen for HCV treatment. Twenty‐five of them already had IP in their ARV therapy. Median cholesterol levels were 160 mg/dL (SD 34.6) and 181.9 mg/dL (SD 97), basal and Week 12 post‐treatment. Median Framingham score was 13.9 (SD 9.2) and 16.2 (SD 10.2). Forty‐four patients (23.15%) increased their Framingham score. Logistic regression showed no association with age, sex, logHCV viral load, F3/F4 fibrosis, body mass index, treatment with a PI‐containing regimen for HCV. Conclusions: Our study suggests that in HIV/HCV co‐infected patients, HCV eradication after DAA‐based therapy is associated with an increase in the 10‐year Framingham cardiovascular risk. P250 HBcAb positivity is an independent risk factor for HIV viral blips in HIV‐HBV co‐infected patients on antiretroviral therapy V Malagnino 1, C Cerva1, G Maffongelli2, E Teti2, L Foroghi Biland1, N Cesta1, M De Masi1, M Andreoni1 and L Sarmati1 1Clinic of Infectious Diseases, Policlinico Tor Vergata of Rome, Rome, Italy2Department of System Medicine, Tor Vergata University, Rome, Italy Background: Co‐infection with HBV and HIV is common; however, there are few data on the influence of HBV on HIV viral replication control and infection progression in the course of ART in HIV/HBV co‐infected patients. Materials and methods: A retrospective analysis of HIV‐positive patients, enrolled from 2007 to June 2018 at the Tor Vergata Infection Unit, was conducted grouping patients for HBV status and recording baseline viro‐immunological features, history of virological failure, efficacy of ART to achieving viral undetectability and HIV viral blip detection (intermittent episodes of detectable viraemia between 50 and 100 copies/mL). Results: Two hundred and thirty‐one patients were included, among them 10 (4.3%) were HBsAg+, 85 (36.8%) anti‐HBc+/anti‐HBs+/‐ and 136 (58.9%) HBV negative. At baseline, anti‐HBc‐positive patients were older (48 years, IQR 39 to 55, p = 0.0001), they had lower CD4+ cell count and CD4+ nadir (188 cells/mm3, IQR 78 to 334, p = 0.02 and 176 cells/mm3, IQR 52 to 284, p = 0.001, respectively). Compared to HBV‐negative patients, a significantly higher number of AIDS and non‐AIDS events were documented in the group of anti‐HBc‐positive subjects (41.1% vs. 19.1%, p = 0.002 and 56.5% vs. 28.7%, p = 350 cells/mm3 and achieving HIV viral suppression within six months of ART had a protective role from HIV viral blip appearance during ART (OR 0.17, 95% CI 0.05 to 0.57, p = 0.005 and OR 0.26, 95% CI 0.12 to 0.56, p = 0.001, respectively). Conclusion: Anti‐HBc‐positive status, regardless of anti‐HBs status, CD4‐nadir 40 years 1.67 (0.98 to 2.85) 0.057 0.84 (0.43 to 1.65) 0.62 Person who injected drugs 1.91 (0.89 to 4.13) 0.09 1.37 (0.52 to 3.58) 0.51 HIV‐RNA >10^6 copies/mL 2.48 (1.45 to 4.25) 0.001 1.52 (0.76 to 3.05) 0.23 CD4 + at baseline >350/mm3 0.35 (0.19 to 0.64) 0.001 1.85 (0.65 to 5.24) 0.24 CD4 + nadir >350/mm3 0.17 (0.08 to 0.37) 50 copies/mL. cIQR was not calculated because only 3 patients had quantifiable HIV load among anti‐HDV+ patients. dIQR was not calculated because only 1 HDV+ patient had a quantifiable HCV load. eone patient did not receive any treatment. Conclusions: HDV+ patients showed a different clinical profile respect to HDV‐ patients, being more frequently infected by HCV and exhibiting longer duration of HIV infection; additionally, HDV+ had more severe liver disease and higher necroinflammatory activity respect to HDV‐. Concerning virological interference, a suppressive effect of HDV on HCV replication was shown. Longer duration of ART and IVDU were independently associated with HDV positivity, likely reflecting an increased risk of exposure to HDV. In conclusion our data underline the need for screening and monitoring a population at high risk, as well as development of newer treatment option for HDV. P254 Use of urinary albumin as a marker of renal damage in patients with chronic hepatitis C treated with direct antiviral drugs B Granozzi 1, L Badia2, V Viotti2, V Guardigni2, P Viale2 and G Verucchi2 1Department of Clinical and Experimental Medicine, S. Orsola, Bologna, Italy2Department of Clinical and Experimental Medicine, Alma Mater Studiorum, University of Bologna, Bologna, Italy Background: Direct‐acting antiviral medications (DAAs) have revolutionised care for chronic hepatitis C virus infection. Anyway, data on kidney safety of these drugs are still scarce [1,2]. Aim of this study was to evaluate urinary albumin modifications and subclinical glomerular damage during and after treatment with DAAs. Material and methods We retrospectively evaluated patients treated with DAAs between February 2016 and November 2017 with baseline creatinine clearance >60 mL/min/1.73 m2 and who achieved SVR. Urinary albumin concentration and serum creatinine were measured at baseline and at the end of 12 weeks of follow‐up. Results: Eighty‐seven patients with HCV infection (55.2% HCV/HIV co‐infected) were included in this study. Hypertension was the highest represented comorbidity (29.9%), followed by diabetes (10.3%). There were no significant changes in serum creatinine concentration and eGFR during therapy and at the end of follow‐up. In 40 patients (46.0%) an increase in urinary albumin value was observed; the average increase was 36.9 mg/L (range 1 to 775). Univariate analysis revealed no significant associations between an increased urine albumin value in the follow‐up and the baseline characteristics of the population (BMI, HIV infection, cirrhosis, diabetes, hypertension, antiretroviral therapies with tenofovir). Quantitative analysis of urinary albumin concentration variation showed a statistically significant correlation, at the univariate analysis, with baseline urinary albumin values (p 200 are vaccinated with MMR. BHIVA also recommends vaccination against varicella zoster (VZV) if found to be VZVIg negative. Methods: All electronic patient records (EPR) and blood test results were checked on HIV‐positive patients attending their consultant over a 6‐month period in 2018 to evaluate screening for measles, VZV, hepatitis A and B. Results: Three hundred and twenty‐four HIV‐positive patients attended over a 6‐month period. One hundred and forty‐eight of 160 (92.5%) men and 152/164 (93%) women were screened for measles IgG and VZVIgG. Eight (5%) men and eight (5%) women were found to be measles IgG‐ve and eligible for vaccination. Sixteen of 16 (100%) GP letters had been completed requesting measles vaccination. Twenty‐four (7%) patients had not been screened yet and blood requests were added to their next clinic visit. Twenty‐two of 300 (7%) patients were found to be VZVIgG‐ve even with some having a documented history of childhood infection. Seventeen of 22 (77%) had a documented GP letter advising VZV vaccination. Four have not had GP letters sent yet and one has not given correct GP details so we are unable to contact. Twenty‐six MSM were found to be HepAIgG‐ve 24/26 (92%) had been offered and given hepatitis A vaccines. One patient defaulted care and one transferred care. Letters advising need for hepatitis A vaccination had been sent to their GP or next centre of care. Thirty‐two of 160 (20%) men and 45/164 (27%) women were found to be hepatitis B naive with HepBsAb 37 copies/mL, % 10 20 No. of comedications 2.3 ± 2.0 2.1 ± 2.0 athe sum is >100% because patients received more than one antiretroviral. INI = dolutegravir or raltegravir. Conclusions: Our study showed that, in real‐life setting, most of HIV‐infected patients had anti‐epileptic drug concentrations falling within the AGNP therapeutic targets. One important exception is represented by valproate, which was associated with a higher rate of subtherapeutic concentrations compared with other anti‐epileptic drugs. We hypothesise that the fear for potential drug‐drug interactions with some antiretrovirals might have resulted in the selection of inappropriate valproate doses. P266 Pharmacokinetic analysis for darunavir in HIV‐1 infected patients on the cobicistat‐boosted darunavir regimen in an Italian observational, multicentre, prospective study (the TMC114FD1HTX4003 ‐ ST.O.RE. study) E Focà 1, A Antinori2, D Ripamonti3, R Maserati4, S Rusconi5, G Rizzardini6, M Palma7, D Mancusi7, R Termini7 and A Uglietti7 1Tropical and Infectious Diseases, Spedali Civili, Brescia, Italy. 2HIV/AIDS Department, National Institute for Infectious Diseases Lazzaro Spallanzani IRCCS, Rome, Italy. 3Division of Infectious Diseases, ASST Papa Giovanni XXIII, Bergamo, Italy. 4Department of Infectious Diseases, Foundation IRCCS San Matteo Hospital, Pavia, Italy. 5Infectious Diseases Unit, DIBIC Luigi Sacco, Milan, Italy. 6Division of Infectious Diseases, Luigi Sacco Hospital, Milan, Italy. 7Medical Affairs, Janssen‐Cilag SpA, Cologno Monzese, Italy Background: The ST.O.RE. study is an observational, single arm, multicentre, prospective study aimed at collecting real‐life data regarding the effectiveness and safety of an ARV regimen based on cobicistat‐boosted darunavir (DRV/c) in HIV1‐positive patients. A secondary objective of the study was to describe steady‐state DRV Ctrough during observational period. Materials and methods: Darunavir Ctrough values were recorded in the electronic case report form (eCRF) by clinical centres that collect blood samples to perform Ctrough analysis in their routine practice. In 14 patients we obtained two repeated DRV Ctrough sampling. All patients included in this analysis were virologically suppressed when switched from a ritonavir‐boosted protease inhibitor to the DRV/c‐based regimen and at the time of sampling. Comparisons between groups were performed using Mann‐Whitney test. Results: We collected the DRV Ctrough values from blood samples of 56 patients. All of them were Caucasian; 39 were male; the mean (±SD) age was 47.6 (±9.5) years. Median (IQR) DRV Ctrough values were 2862.5 (2969.5) overall, 2634 (2322) ng/mL in males and 4221 (2881) ng/mL in females. Female DRV Ctrough values were statistically higher (p = 0.046). Fourteen patients had two repeated DRV Ctrough sampling. Patients characteristics are reported in Table 1. Data of Ctrough are shown in Figure 1. In first sampling all patients were virologically suppressed while in the second one, one patient had HIV‐RNA value of 81 copies/mL with Ctrough 1148 ng/mL. Among the 14 patients, six males and three females were taking concomitant medications (vitamin D, phosphate and antihypertensives). None of the recorded concomitant drugs is known to alter the pharmacokinetics of DRV/c so far. Considering 70 samplings (56 single and 14 repeated) no patients had values below the 55 ng/mL, the protein‐binding adjusted EC50 for wild‐type HIV; six reported values below the threshold of 550 ng/mL, the protein‐binding adjusted EC50 for PI‐resistant HIV: five males (100, 159, 296, 363 and 494 ng/mL) and one female (435 ng/mL). Only half of these 70 patients had a double sampling greater than 550 ng/mL. None showed adverse events at the time of samplings. Abstract P266 – Table 1. Patient characteristics with repeated DRV Ctrough Overall (N = 14) Male (N = 11) Female (N = 3) Age (years), mean (SD) 47.3 (8.3) 46.0 (9) 52.0 (2.6) BMI (kg/mg), mean (SD) 23.4 (7.32) 23.6 (3.5) 22.7 (6.2) Race (White), N (%) 14 (100) 11 (100) 3 (100) CD4 nadir (cell/mm3), mean (SD) 269.2 (231.4) 254.3 (262.8) 319.0 (75.9) CDC category C, N (%) 4 (28.6) 4 (36.4) 0 HCV positive, N (%) 4 (27.3) 3 (27.3) 1 (33.3) Ongoing and treated conditions, N (%) 10 (71.4) 7 (63.6) 3 (100) % of TDF/FTC as backbone therapy 57.2 54.5 66.6 % of ABC/3TC as backbone therapy 21.4 100 0 % dual therapies 21.4 18.2 33.3 Abstract P266 – Figure 1. DRV Ctrough values: repeated samplings. Conclusions: In clinical practice we observed a DRV threshold being effective in all patients as cobicistat allowed to obtain Ctrough values far above the protein‐binding adjusted EC50 for wild‐type HIV. The observed intra‐and interpatient variability can be due to the observational nature of this study where samplings might not have been collected at the real Ctrough. P267 Association of tenofovir level and discontinuation due to impaired renal function H Yagura 1, D Watanabe2, T Nakauchi1, K Tomishima1, Y Nishida2, M Yoshino3, K Yamazaki1, T Uehira2 and T Shirasaka2 1Pharmacy, Osaka National Hospital, Osaka, Japan. 2AIDS Medical Center, Osaka National Hospital, Osaka, Japan. 3Pharmacy, Utano Hospital, Kyoto, Japan Background: Tenofovir (TFV) preparations are nucleotide reverse transcriptase inhibitors with highly antiretroviral activity used as first‐line treatment in HIV‐1‐infected patients. Long term TFV treatment, however, is associated with a risk of renal impairment, especially cellular damage in renal tubules. Plasma trough concentrations of TFV are high in patients with renal impairment, but the mechanism by which high trough concentration of TFV following its long‐term administration affects renal function has not been clarified. Methods: A regimen including TFV disoproxil fumarate (TDF) was administered to 136 HIV‐1‐infected patients aged ≥18 years treated at the National Hospital Organization Osaka National Hospital between January 2007 and December 2011. Plasma trough concentrations of TFV were measured 20–28 hours after its administration within 3 months after starting treatment. The association between trough concentrations of TFV and discontinuation of its administration due to renal function‐related adverse events within 288 weeks after the start of therapy was measured. Results: The median age of the 136 patients (135 were male) was 39 years (range 18 to 68 years). TDF was discontinued due to renal function‐related adverse events in 34 patients (25%), and the median time to discontinuation was 967 days (range 183 to 1986 days). The median trough TFV concentration was significantly higher in discontinued than in continued patients (87.5 vs. 73.0 ng/mL; p = 0.0033). Construction of a receiver operating characteristic (ROC) curve showed that the cut‐off value of trough TFV concentration at TDF discontinuation was 98 ng/mL (area under the curve [AUC], 0.668; sensitivity, 0.471; specificity, 0.833). Trough TFV concentration was significantly correlated with TDF discontinuation due to renal function‐related adverse events (p = 0.0022). However, trough concentrations of TFV did not correlate with time to TDF discontinuation (p = 0.29). Conclusions: Discontinuation of long‐term TDF administration due to renal function‐related adverse events is associated with high trough TFV concentrations in Japanese patients infected with HIV‐1. This implied the importance of early after administration TFV concentration measurement for the risk assessment of renal dysfunction in long‐term administration. COMMUNITY INITIATIVES P268 Effectiveness of a pilot partner notification programme (PNP) for new HIV cases (NHIVC) in Málaga, Spain C Gómez‐Ayerbe 1, C González‐Doménech2, I Viciana3, M Villalobos1, J Santos1 and R Palacios1 1Infectious Diseases, Hospital Clínico Virgen de la Victoria, Málaga, Spain. 2Microbiología, Universidad de Granada/Red Investigación en SIDA, Granada, Spain. 3Microbiology, Hospital Clínico Virgen de la Victoria, Málaga, Spain Background: HIV epidemic is not stabilised and almost half of Spanish NHIVC are still late diagnosis ( 50 years undergoing stable ART from Italy, Australia and Hong Kong recruitment sites who completed 9 months of follow‐up. FI includes 37‐item health variables, each of them is scored from 0 to 1. FI >0.3 was used to identify most frail individuals. HI includes 12 variables from five IC domains: Locomotion, Cognition, Vitality, Sensory, Psychosocial. After obtaining HI scores, IC is calculated with a 0 to 1 range score; the lower is the score, the better is IC evaluation. Variables were collected through a fitness tracking wearable device (Garmin‐Vivofit 2) and through questionnaires provided via ecological momentary assessment (EMA) [1] using MySAwH App. HI was collected on a monthly basis, while the FI was assessed on the baseline and follow‐up visit. Statistical analyses were performed in R software. Results: One hundred and fifty‐three OALWH are included in this analysis. Median age is 57 years. Twenty‐two (14.38%) patients are women. Mean CD4 is 686.99 (324.14 SD) and 141 (92.76%) patients had undetectable HIV viral load. Median FI at baseline is 0.23 (0.2–0.32 IC) and at nine months follow‐up is 0.26 (0.2 to 0.31 IC) with no significant p value 0.37. Median IC at baseline is 0.33 (0.25 to 0.4 IC) and at nine months follow‐up is 0.3 (0.2 to 0.36) with significant p value 0.05. IC domains prevalence was calculated every month, but no domain shows a significant change after nine months. Conclusion: This study shows a continuous improvement in IC after nine months of follow‐up, but not decrease of FI. The presence of a health coach that provides information about HI variations and lifestyle promotion can stimulate patients to be personally empowered to change their health measured by IC. Reference: [1] ClinicalTrials.gov. My Smart Age With HIV: Smartphone Self‐assessment of Frailty (MySAwH) [Internet]. Available from: http://https://clinicaltrials.gov/ct2/show/NCT02663856. P273 Surveying Ontario nurses using the COM‐B framework shows a high level of readiness for nurse‐led PEP and PrEP M Clifford‐Rashotte 1, J Lee2, N Fawcett3, B Fowler4, J Reinhart5 and D Tan2 1Medicine, University of Toronto, Toronto, Canada2Division of Infectious Diseases, St. Michael's Hospital, Toronto, Canada3Sexual Health Clinics, Toronto Public Health, Toronto, Canada4Health Sexuality Program, Region of Peel Health Department, Peel, Canada. 5Sherbourne Health Centre, Toronto, Canada Background: Effective HIV prevention in Ontario requires more widespread implementation of post‐ and pre‐exposure prophylaxis (PEP and PrEP), including through nurse‐led models of care. To plan for further scale up of nurse‐led PEP and PrEP, we assessed nurses’ readiness to deliver these interventions, using a behavioral change framework. Materials and methods: We asked the managers of every sexual health clinic, HIV clinic and community health center in Ontario to distribute an online survey to nurses in their organizations between March and June 2018. Our primary objective was to determine the proportion of nurses who would support the implementation of nurse‐led PEP and PrEP in their workplace. We explored nurses’ readiness for these interventions using the COM‐B behavioral change framework [1], assessing “capabilities,” “opportunities” and “motivations” for providing PEP and PrEP. Data were analyzed using descriptive statistics, and a multivariable logistic regression model was constructed to identify variables associated with support for nurse‐led PEP and PrEP. Results: Of 470 surveys distributed, 165 had responses for the primary outcome and were included in the analysis. Respondents had a median of 16 (25th percentile=9, 75th percentile=25) years of nursing experience and most worked in sexual health clinics (65.5%). The largest proportion of respondents was from Central Ontario (29.5%), though all regions were represented. 72.7% of respondents supported implementation of nurse‐led PEP and PrEP. More experienced nurses were less likely to support nurse‐led PEP and PrEP (aOR 0.55 per decade nursing, 95% CI 0.37 to 0.82). Nurses’ self‐reported knowledge of topics related to PEP/PrEP and comfort performing relevant clinical tasks were high, with the exception of creatinine interpretation, which only 30.8% of respondents felt comfortable with. Most respondents had initiated conversations with patients about PEP (70.3%) or PrEP (62.2%), but few worked in institutions which provide them (22.4% and 13.3%, respectively). The most commonly cited barriers to implementation were a lack of physician support (38.8% for PEP, 42.9% for PrEP), followed by a lack of knowledge among nurses (38.8% for PEP, 37.4% for PrEP). Finally, the most popular modalities for receiving potential further education about PEP and PrEP were online modules (86.5%), followed by in‐person workshops (71.2%). Conclusions: Nurses at Ontario sexual health clinics, HIV clinics and community health centers exhibit a high level of support for nurse‐led PEP and PrEP, and are well positioned to provide these interventions. To increase their implementation, priorities should include increasing physician support and providing online and in‐person education for nurses, with an emphasis on renal monitoring. Reference: [1] Michie S, van Stralen MM, West R. The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implement Sci. 2011;6:42. P274 Abstract withdrawn MODELS OF CARE: COST EFFECTIVENESS P275 Exploring the correlation between price and affordability across 50 countries: is pricing of dolutegravir equitable? J Sim 1 and A Hill2 1School of Public Health, Imperial College London, London, UK. 2Department of Translational Medicine, University of Liverpool, Liverpool, UK Background: In the SINGLE trial, dolutegravir (DTG) showed fewer adverse events than efavirenz (EFV) as first‐line treatment, but no difference in virological suppression, quality of life or survival [1]. In switching studies (NEAT 022, SWORD, STRIIVING), DTG led to significantly higher rates of adverse events and no virological benefit [1]. Yet, in upper‐income countries such as UK, DTG currently costs £6068 per year compared to £108 for EFV. There are other low‐cost generic antiretrovirals which could be used as alternatives to DTG. Method: Lowest prices of DTG and EFV in over 50 countries were collected from national price databases, reimbursement authorities and WHO Global Price Reporting Mechanism database. Median prices calculated for each income level group. We analysed the correlation between (1) DTG prices or (2) ratio of DTG to EFV (DTG/EFV) and gross domestic product (GDP) per capita (2016) published by the World Bank. Prices were recorded in US$ per person‐year (PPY). Results: Annual prices of DTG ranged from $27 PPY in Georgia to $20,130 in the USA (Table 1). DTG prices in high‐income countries (HICs) and upper middle‐income countries (UMICs) excluded from VL arrangements are substantially higher compared to low‐ and lower middle‐income countries (LMICs): median price in HICs $9045 PPY versus $2682 (UMICs) and $60 (LMICs). The price in USA ($20,130) is almost twice of Denmark ($11,056), although their GDP per capita is comparable (USA: $57,638 vs. Denmark: $53,579). In UMICs, we did not observe a correlation between DTG price and GDP per capita: in Bulgaria, DTG costs $9656 PPY compared to $2682 in Argentina, even though her GDP/capita is much lower (GDP/capita for Bulgaria: $7469 vs. Argentina: $12,440). There was also no relationship between DTG/EFV price ratio and GDP per capita. However, DTG prices were found to be much higher than EFV in UMICs (median DTG/EFV ratio 6.9) compared HICs (ratio 4.9). In LMICs, DTG prices were close to EFV prices (ratio 1.8). Abstract P275 – Table 1. Highest and lowest prices of dolutegravir, by income groups Income group Country Price of DTG per person‐year (US$) GDP per capita in 2016 (US$) High‐income countries Highest price: USA $20,130 $57,638 High‐income countries Lowest price: Canada $5267 $42,349 Upper middle‐income countries Highest price: Bulgaria $9656 $7469 Upper middle‐income countries Lowest price: Iran $36 $5219 Low and lower middle‐income countries Highest price: India $538 $1710 Low and lower middle‐income countries Lowest price: Georgia $27 $3866 Conclusions: Prices of DTG vary widely between countries, from 360 days (N = 1022) (N = 132) (N = 102) (N = 143) (N = 51) (N = 122) (N = 114) (N = 358) Age at HIV diagnosis (years), mean ± SD [median] 40.0 ± 11.9 [40.9] 40.4 ± 11.7 [41.7] 40.2 ± 12.3 [41.4] 39.1 ± 13.7 [39.8] 41.1 ± 13.2 [41.5] 40.9 ± 11.9 [40.5] 40.2 ± 11.5 [40.8] 39.6 ± 11.0 [40.7] Range (years) (18.0 to 70.0) (18.2 to 66.6) (18.3 to 65.7) (18.0 to 65.3) (18.2 to 63.9) (18.0 to 62.6) (18.0 to 70.0) (18.0 to 66.1) Female, n (%) 451 (44.1%) 57 (43.2%) 51 (50.0%) 53 (37.1%) 22 (43.1%) 52 (42.6%) 49 (43.0%) 167 (46.6%) Race, n (%) White 248 (24.3%) 35 (26.5%) 30 (29.4%) 39 (27.3%) 7 (13.7%) 23 (18.9%) 33 (28.9%) 81 (22.6%) Black 582 (56.9%) 72 (54.5%) 47 (46.1%) 78 (54.5%) 35 (68.6%) 76 (62.3%) 54 (47.4%) 220 (61.5%) Other 192 (18.8%) 25 (18.9%) 25 (24.5%) 26 (18.2%) 9 (17.6%) 23 (18.9%) 27 (23.7%) 57 (15.9%) Patients who initiated a single‐tablet regimen, n (%) 592 (57.9%) 82 (62.1%) 60 (58.8%) 81 (56.6%) 26 (51.0%) 72 (59.0%) 65 (57.0%) 206 (57.5%) Patients with ≥1 diagnosis of an opportunistic infection between diagnosis and ART initiation, n (%) 69 (6.8%) 6 (4.5%) 7 (6.9%) 7 (4.9%) 1 (2.0%) 9 (7.4%) 11 (9.6%) 28 (7.8%) Patients with ≥1 emergency room between diagnosis and ART initiation, n (%) 438 (42.9%) 8 (6.1%) 11 (10.8%) 31 (21.7%) 12 (23.5%) 46 (37.7%) 61 (53.5%) 269 (75.1%) Patients with ≥1 inpatient visit between diagnosis and ART initiation, n (%) 362 (35.4%) 12 (9.1%) 26 (25.5%) 39 (27.3%) 18 (35.3%) 36 (29.5%) 43 (37.7%) 188 (52.5%) Conclusions: This study reveals that only 13% of Medicaid patients initiated ART within 14 days post‐diagnosis and that patients with early initiation of ART had better outcomes (fewer hospitalizations and OIs) between diagnosis and treatment initiation. Further research is warranted to assess the real‐world impact of rapid ART initiation on economic and clinical outcomes post‐initiation. P285 Quality of life and experience of patients with HIV and other chronic diseases with Spanish health system: insights from the IEXPAC project M Galindo 1, N Sanchez‐Vega2, M Cotarelo2, O Rincon3 and M Fuster4 1Internal Medicine Service, Hospital Clínico de Valencia, Valencia, Spain. 2Medical Department, Merck Sharp & Dohme de España, Madrid, Spain. 3Medical Department, Merck Sharp & Dohme de España, Sevilla, Spain. 4Research Department, Sociedad Española Interdisciplinaria de SIDA (SEISIDA), Madrid, Spain Improvements over the time in quality of care lead to a more positive experience for patients with chronic diseases. Careful measurement of patient's experience can provide meaningful data to further enhancements in quality of care, clinical effectiveness and patient's safety. The study describes the experience with health care system and health‐related quality of life (HRQoL) in Spanish patients with HIV (PHIV) infection and other chronic diseases. One thousand six hundred and eighteen patients participated in an observational cross‐sectional study. Surveys were handed to patients with four different chronic diseases with at least one comorbidity: PHIV, rheumatic diseases (RD), inflammatory bowel disease (IBD) or diabetes mellitus (DM). Patients filled anonymously the questionnaire at home and responded by pre‐paid mail. The experience with the health care system was measured through the validated IEXPAC scale (http://www.iemac.es/iexpac/). This scale contains 12 items with five responses from “always” to “never”, yields a score from 0 (worst) to 10 (best experience) and measures three dimensions: productive interactions, new relational model and patient self‐management. HRQoL and beliefs about medication were measured by EQ‐5D‐5L and BMQ questionnaires respectively. Two thousand four hundred and seventy‐four patients received the survey and 1618 were returned (65.4%): 467 corresponded to PHIV [mean age 51.5 ± 10.8 years, 27% women]. Responses to IEXPAC are displayed in Table 1. Mean IEXPAC score for PHIV was 6.6 ± 1.7. Patients declared a median of 8 visits to primary care or specialty clinics in the last year and 29% had visited an emergency room. In the last three years, 48% had been hospitalised. PHIV attended least frequently to primary care (76.20% vs. all 83.40%), declared a median of 3.89 visits to specialty clinics and reported higher % receiving once‐daily dosage (50.20% vs. all 31.80%). PHIV differ significantly in terms of: considering themselves well informed about their disease (82% vs. all 75%); their perception of need of medication, which is significantly the highest (22.20 ± 3.87); and being the least concerned about medication (13.32 ± 4.82). PHIV most often described no limitations in any of the five dimensions included in EQ‐5D‐5L; their scores on the visual analogue scale “Your health today” were the highest (73.3 ± 19.1) (all multiple comparison tests HIV infection vs other, p 0.98; average difference ≤0.2 log copies/mL) and retrospective longitudinal samples from patients on treatment. Aptima and RealTime showed similar high precision, meeting the 5σ‐criterion for the majority of samples across all labs and subtypes. Cobas was less precise, missing the 5σ‐criterion for the majority of samples. Aptima and RealTime results differed more noticeably from Cobas 6800 results, particularly for subtypes C and AE in clinical follow‐up samples (range 0.33 to 0.56 log). In precision analysis at 50 copies/mL and 200 copies/mL, mean, median and standard deviation turned out to be higher for Cobas 6800 as compared to Aptima and RealTime for all subtypes. Across all subtypes, coefficients of variation for absolute values at 50 copies/mL and 200 copies/mL ranged from 22% to 34% and 20% to 33% for Aptima, from 35% to 51% and 23% to 42% for Cobas and 26% to 37% and 21% to 35% for RealTime, respectively. Conclusions: In this analysis, results from Cobas appeared less reliable near the clinically relevant cutoff and should be interpreted with more caution in this context. Aptima and RealTime differed more noticeably from Cobas 6800 results but showed a high agreement between their results at this low viraemic level. As an important criterion, this may ease the interpretation of Aptima results in comparison to previous RealTime results. In this context, low viraemic Cobas 6800 results after RealTime monitoring may require a closer view. VIROLOGY AND IMMUNOLOGY: RESISTANCE P294 Week 48 resistance analyses of the once‐daily, single‐tablet regimen (STR) darunavir/cobicistat/emtricitabine/tenofovir alafenamide (D/C/F/TAF) in HIV‐1‐infected adults from the AMBER and EMERALD Phase III trials E Lathouwers 1, E Wong2, K Brown1, B Baugh3, A Ghys1, J Jezorwski3, E Van Landuyt1, M Opsomer1 and S De Meyer1 1Janssen, Pharmaceutica NV, Beerse, Belgium. 2Janssen, Scientific Affairs LLC, Titusville, NJ, USA. 3Janssen, Research & Development LLC, Raritan, NJ, USA The once‐daily STr D/C/F/TAF 800/150/200/10 mg is approved in the EU, under regulatory review in the US and is being investigated in two international, randomised Phase III trials. D/C/F/TAF had non‐inferior efficacy versus D/C+F/TDF in HIV‐1‐infected, ART‐naïve adults (AMBER; NCT02431247) and versus bPI+F/TDF in ART‐experienced, virologically suppressed adults on a bPI+F/TDF regimen (EMERALD; NCT02269917). In EMERALD, patients with prior antiretroviral experience (58% had received ≥5 antiretrovirals) and prior VF (15%), including patients with emtricitabine and tenofovir resistance‐associated mutations (RAMs), were eligible for study participation. Only if historical genotypes were available, patients with darunavir RAMs were excluded. Week 48 AMBER and EMERALD resistance analyses are presented. Samples for genotyping/phenotyping were selected in patients with protocol‐defined VF (PDVF) in AMBER (virological non‐response, virological rebound and/or viraemic at final timepoint) or EMERALD (rebound) and with VL ≥400 copies/mL at failure or later timepoints. In AMBER, resistance testing was performed using GenoSure®MG at screening and PhenoSense®GT post‐baseline. Deep sequencing was performed post‐hoc using Illumina MiSeq on samples from one patient. In EMERALD, GenoSure®MG was used for resistance testing post‐baseline. In addition, HIV‐1 proviral DNA from baseline samples (VL 50 copies/mL. One hundred patients failing INI‐based TT (RAL, EVG/cobi, DTG) and 100 failing PI‐based TT (DRV/r ATV/r (consecutive in the database)) were included, and all eligible patients failing DTG‐based 2DC (DTG+3TC, DTG+RPV) or PI‐based 2DC (PI + RAL, DTG or 3TC). Study period was 2015 to 2018 for genotypic testing. Results: Twenty‐three patients who had experienced their first virological failure were included in the DTG‐based 2DC group and 32 in the PI‐based 2DC group. Median age was 39 years for TT, 41 for 2DC; median number of previous regimens was 2 for TT, 3 for 2DC; median zenith VL was 79,239 copies/mL for TT, 156,966 for 2DC, and median VL at time of failure was 3019 copies/mL for TT, 4313 for 2DC. Three percent of PI TT patients, 7.0% of INI TT patients (0.0% of DTG‐based TT), 21.7% of DTG 2DC patients and 37.5% of PI 2DC patients presented emergent resistance after failure. Table 1 shows the characteristics of the subsequent regimen after failure (% of PI use, % of multi‐tablet regimens) depending on the initial regimen group. Abstract P295 – Table 1. Emergent resistance and characteristics of subsequent regimen after failure 2NRTIs+PI 2NRTIs+INI DTG‐based 2DC PI‐based 2DC N 100 100 23 32 Emergent resistance, % 3.0 7.0 21.7 37.5 PI use, % 3.0 7.0 39.1 65.6 MTR use, % 4.0 7.0 52.2 65.6 Conclusions: Higher rates of drug resistance selection were observed in patients failing a DTG‐ or PI‐containing 2DC comparing to INI‐ or PI‐containing TT. Patients on 2DC are switched after failure to more complex regimens, containing PIs and higher number of tablets than patients failing TT. P296 Frequent detection of drug resistance mutations by deep sequencing in patients with documented extensive resistance and long‐lasting viral suppression: the proviral DNA archive remains stable for decades C Hoffmann 1, A Thielen2, E Wolf3, M Bickel4, A Stoehr5, P Braun6, H Knechten6, S Esser7, C Wyen8, I Krznaric9, M Müller10, J Brust11, J Wasmuth12, H Horst13, S Holm14 and M Däumer2 1ICH Study Center, ICH Stadtmitte, Hamburg, Germany. 2Virology, Institute of Immunology and Genetics, Kaiserslautern, Germany. 3MUC Research, MVZ Karlsplatz, Munich, Germany. 4Infektiologikum, Frankfurt, Germany. 5IFI Institute, Hamburg, Germany. 6Praxiszentrum, Aachen, Germany. 7Dermatology, University of Essen, Essen, Germany. 8Praxis am Ebertplatz, Cologne, Germany. 9Medizinisches Infektiologiezentrum Berlin, Berlin, Germany. 10Gemeinschaftspraxis, Schwabstrasse, Stuttgart, Germany. 11Mannheimer Onkologie Praxis, Mannheim, Germany. 12Department of Medicine I, University of Bonn, Bonn, Germany. 13Department of Medicine II, University of Schleswig‐Holstein, Kiel, Germany. 14Praxis, Hannover, Germany Background: Deep sequencing (DS) assays may represent a reproducible approach to analyse HIV‐1 mutant spectra, even at variant frequencies well below those routinely detectable by population (Sanger) sequencing. DS data from viral reservoirs (i.e. peripheral blood mononuclear cells [PBMCs]) in patients with documented multiple viral drug resistance mutations (DRMs) and with long‐standing viral suppression are scarce. Material and methods This nationwide study included patients with a history of multi‐drug resistance (MDR) in 13 German HIV centres. MDR was defined to comprise at least one major DRM in at least three classes of NRTIs, NNRTIs and PIs or INSTIs. In patients with viral suppression ( 200 copies/mL after switch). Drug resistance was evaluated before (as cumulative plasma resistance) and after switch. Cumulative genotypic susceptibility score for companion drugs (cGSS) was also evaluated (HIVdb algorithm v.8.4). Results: Overall, 248 cART‐treated patients virologically suppressed from a median time of 1.9 (IQR 0.6 to 4.5) years and starting an INI‐based dual therapy in 2015 (IQR 2011 to 2016) were analysed. At switch, patients experienced a median number of 5 (IQR 2 to 9) previous regimens. One hundred and sixty‐six of 248 patients switched to raltegravir, administered mainly with darunavir (80, 48.2%) or etravirine (21, 58.8%). Eighty‐two of 248 patients switched to dolutegravir, administered mainly with lamivudine (46, 56.1%) or darunavir (22, 26.8%). Overall, 64.5% had ≥1 previous major resistance mutation (NRTI, 56.9%; NNRTI, 44.8%; PI, 27.4%; INI, 2.9%). cGSS revealed that 20.2% of patients harboured a virus with intermediate/full resistance to the companion drugs adminstered with INI. The overall probability of VF at 36 months after switch was 8.3% (median [IQR] viraemia at VF: 10,458 [1829 to 51,596] copies/mL). By stratifying for INI, the probability of VF with dolutegravir was 5.2%, while with raltegravir was 9.5% (p = 0.292). The duration of virological suppression before switch was strongly associated with probability of VF after switch ( 5 years, 0%; p 50 copies/mL measurement; blip) was observed in 4.8% (2/42) with M184V/I versus 5.5% (12/217) without M184V/I. Among participants with M184V/I, 95% (40/42) were suppressed at their last study visit at the time of the W48 Snapshot analysis: 37 had HIV‐1 RNA 400 copies/mL). Survival analysis was used to investigate predictors of VF. The GSS predicted by the latest and the cumulative genotype (CGSS), was calculated using the Stanford HIVdb (v.8.5) with respect to the two‐drug regimen started. Pre‐baseline viraemia copy‐years (VCY) were calculated using the trapezoidal rule on the VL log10 scale using all the available VL. Results: We included 2149 patients. Overall 68% were males, 31% MSM, 6% non‐Caucasians, with median age of 50 years (IQR 45 to 56), 15 years of HIV history (8 to 22). At baseline 71% of patients had HIV‐1 RNA 50 copies/mL was three years before baseline (1 to 6). Median GSS was 2 (1 to 2), with GSS 50 copies/mL and the second >200 copies/mL) undergo genotyping. P304 Impact of NRTI mutations on virological efficacy of antiretroviral regimens containing elvitegravir: an ARCA‐ECCO cohort study R Gagliardini1, S Modica1, D Redi1, E Giombini2, A Bezenchek3, D Di Carlo4, F Maggiolo5, F Lombardi6, A Borghetti6, A Callegaro7, M Gismondo8, M Colafigli9, G Sterrantino10, A Costantini11, S Ferrara12, S Rusconi13, S Di Giambenedetto6, M Zazzi1, A De Luca 1, B Rossetti14 and N Gianotti15 1Department of Medical Biotechnologies, University of Siena, Siena, Italy. 2National Institute for Infectious Diseases L. Spallanzani, Spallanzani, Rome, Italy. 3InformaPRO, Rome, Italy. 4Pediatric Clinical Research Center, University of Milan, Milan, Italy. 5Infectious Diseases Unit, Bergamo Hospital, Bergamo, Italy. 6Clinic of Infectious Diseases, Catholic University of Sacred Heart, Rome, Italy. 7Microbiology and Virology Unit, Bergamo Hospital, Bergamo, Italy. 8Microbiology Unit, Ospedale L. Sacco, Milan, Italy. 9STI/HIV Unit, IRCCS Rome, San Gallicano Dermatologic Institute, Rome, Italy. 10Division of Tropical and Infectious Diseases, Careggi Hospital, Florence, Italy. 11Clinical Immunology Unit, Università Politecnica delle Marche, Ancona, Italy. 12Clinic of Infectious Diseases, Azienda Ospedaliera‐Universitaria Ospedali Riuniti, Foggia, Italy. 13Infectious and Tropical Diseases Unit, DIBIC L. Sacco Hospital, Milano, Italy. 14Infectious Diseases Unit, AOU Senese, Siena, Italy. 15Infectious Diseases, IRCCS San Raffaele, Milano, Italy Background: Integrase inhibitor‐based regimens are recommended by current guidelines as first‐choice ARV therapy. ARV drug resistance mutations remain a major cause of treatment failure. The aim of study was to evaluate the effect of mutations on virological efficacy of elvitegravir‐containing ARV regimens in naïve and treatment‐experienced HIV‐1 patients in a real‐life setting. Material and methods From the ARCA and ECCO databases we selected naïve and treatment‐experienced HIV‐1 infected patients starting tenofovir disoproxil fumarate or alafenamide/emtricitabine/elvitegravir/cobicistat (from June 2012 to December 2017), with baseline PR/RT resistance genotype and at least one HIV‐1 RNA during follow‐up. NRTI resistance mutations were defined as the detection of at least one mutation according to Stanford algorithm. Primary endpoint was virological failure (VF, defined as an HIV‐RNA VL >1000 copies/mL or two consecutive values of >50 copies/mL after Week 24 for treatment‐experienced with baseline VL >50 copies/mL and at any time for treatment‐experienced with baseline values of 50 copies/mL (aHR 2.73, 0.95 to 7.81, p = 0.062) showed a trend with higher risk of VF. In the naïve group, 1/3 patients with any NRTI mutation had VF versus 1/43 without NRTI mutation (p 100 reads only) were analysed. Sequences were interpreted by HIV‐GRADE (http://www.hiv‐grade.de) for resistance mutations using 10%, 2% and 1% minority cut‐offs. A specific focus laid on differences in reported resistance‐associated mutations and resistance levels (e.g. additional drug class or further drugs same class). The proportion of subpopulations harbouring additional mutations with greater than 2000 copies/mL (= mutational load) were calculated, therapy data and follow‐up for those patients was monitored as far as available. Results: Four hundred and eighty‐three (74.9%) samples were identified as subtype B. No drug resistance‐associated mutations were reported by HIV‐GRADE for 44% with a 10% cut‐off, 29.5% and 19.7% with 2% and 1% respectively. This also correlates with an increase of resistance level in the interpretation, especially for NNRTIs. With a cut‐off of 10% in 148 samples (105 non‐B subtype) only PI relevant mutations were detected. We found mutations only relevant for NRTIs in 21 samples and for NNRTI in 100 samples. Additional mutations could be detected in 94 of the samples using a 2% minority cut‐off. This corresponds to an additional mutational load of >2000 copies/mL in 76 cases with a 2% minority cut‐off and additional 134 mutations at 1% cut‐off. Consequences on treatment concepts and regimen were found rarely in therapy data. Conclusions: A relative high proportion (56%) of investigated sequences showed resistance mutations at a minority cut‐off of 10% increasing substantially lowering the cut‐off range to 2% or 1% in number of mutations and also regarding resistance levels. Relevance of mutations in these low percentages is often discussed. The concept of “mutational load” tries to correlate the viral load with the proportion of mutation in the whole viral population. Despite the low percentage these viral quasispecies can be detected in relevant absolute quantities which increases the probability of detecting viable resistant virus. There is a clear need for clinical evaluation of the relevance of mutations in the low percentage range for resistance interpretation due to its broader use in clinical routine. P308 Quiescent profile of T cells from Colombian MSM with extremely high‐risk sexual behaviours and HIV‐1 specific CTL response A Ossa‐Giraldo 1, Y Blanquiceth2, K Contreras2, L Flórez2, J Hernández1 and W Zapata2 1Infettare Research Group, Universidad Cooperativa de Colombia, School of Medicine, Medellin, Colombia. 2Immunovirología Research Group, Universidad de Antioquia, School of Medicine, Medellin, Colombia Background: MSM still being a key population on HIV‐1 epidemiology [1]. MSM with high‐risk sexual behaviours are at great risk of exposure to infection [2]. Better intervention strategies are urgently needed, such as biomedical research to develop new options for prevention and treatment. The study of seronegative MSM with high‐risk behaviours represents an important opportunity to better understand HIV‐1 infection and immune response to improve the current intervention strategies [3,4]. Methods: Analysis of sociodemographic data and basal activation profile and functional response of T lymphocytes against stimuli with HIV‐1, in two groups of MSM from Medellín, Colombia, South America with different sexual behaviours. Results: We included 44 MSM with high and low risk of exposure (14 and 30, respectively). The high‐risk group presented a higher frequency of sexual partners in the three months prior to the inclusion of the study (Me = 31 vs. Me = 2; p 50 c/mL, 40 c/mL≤VL 50 occurred infrequently in all groups. These results suggest no difference in blip rates or any clinical consequences from VL elevations ≥200 copies/mL, as efficacy rates were high and equal between arms (95% each) and CVW numbers were low DTG+RPV and traditional 3DRs of therapy. P314 Association of new HIV diagnoses within long‐lived transmission clusters from Malaga area (Spain) I Viciana Ramos1, C González Domenech 2, G Sena Corrales1, C Gómez Ayerbe1, M Villalobos1, G Ojeda1, E Nuño1, E Clavijo1, R Palacios Muñoz1 and J Santos1 1UGC Infectious Diseases and Clinical Microbiology, Hospital Virgen de la Victoria, Malaga, Spain. 2Microbiology, University of Granada, Granada, Spain Background: The early HIV diagnoses and the decrease of the hidden infections rate are the goals of a proactive partner notification programme (PNP) among new HIV diagnoses (NHIVD), performing at Virgen de la Victoria Hospital (southern Spain) from September 2017. Our aim was the molecular epidemiological determination of the transmission clusters (TC) comprising these NHIVD, clarifying if they are expanding previous TCs in our area or, on the contrary, actively arising new TCs. Materials and methods: We considered all the HIV‐1 genotype resistance tests performed in NHIVD and participating in the PNP in our hospital, from January 2017 to June 2018. Drug resistance mutations were determined with Viroseq HIV® system and the partial sequence of HIV‐1 pol gene provided phylogenetically compared to a cohort of 451 naïve patients diagnosed between 2004 and 2015 and clustered in 86 TCs previously described. The belonging to a prior TC was based on a phylogenetic criterion as well as using the mean pairwise distance. Therefore, a NHIVD was phylogenetically considered within any cluster if the branch‐support measure (SH‐like aLRT test) was ≥90%. The alignment was done by ClustalX and the phylogenetic reconstruction inferred by maximum likelihood method (FastTree programme). Regarding pairwise distances, they were estimated directly from the sequence alignment (MEGA‐X software package) and the difference of 0.015 substitution/site with any sequence included in a particular TC, as maximum threshold of belonging to such TC. Results: During the study period, we had 59 NHIVD, 41 out of them with pol gene sequence available. Among them, 30 (73.2%) were phylogenetically grouped within some TC previously defined, with a mean pairwise‐distance intra‐cluster ≤0.015. Seven out of 30 sequences associated to any prior TC (23.3%) were clustered within the one comprising the CRF19_cpx variant, already described as an outbreak in our area. Seven out of the 41 sequences left (17%) were phylogenetically clustered to others but with SH‐like aLRT ≤90% and pairwise‐distance higher than 0.015. Finally, three of the NHIVD included in the contacts study (7.3%) were associated together and independently from the rest sequences, forming a new TC. Conclusions: Almost three out of four NHIVD included in this study expand previous TC, already defined in our area. The incidence of new TCs is relatively low. The molecular epidemiology of the partner notification programme points out to networks including already diagnosed and under follow‐up patients. P315 HIV‐1 diversity in the Moscow region, Russia: phylodynamics of the most common subtypes A Lebedev1, N Lebedeva2, F Moskaleychik1, A Pronin2, E Kazennova1 and M Bobkova 1 1N.F. Gamaleya National Research Center of Epidemiology and Microbiology, Ministry of Health, Moscow, Russian Federation. 2Moscow Regional AIDS Centre, Ministry of Health of the Moscow Region, Moscow, Russian Federation Background: The Moscow region is the most densely populated subject of the country with an HIV prevalence of about 0.76%. Being a highly developed transport hub the Moscow region is characterised by a high intensity of internal and external migration. Taken together, these factors can cause a wide range of HIV‐1 subtypes in the region, on the one hand, and contribute to the emergence of new recombinant forms of the virus, on the other hand. This is the first large‐scale study in the Moscow region that includes all groups at risk of contracting HIV. Materials and methods: We analysed demographic data and genotypes from 927 HIV‐infected individuals collected in Moscow region in 2011 to 2016. Phylogenetic analysis was performed with maximum‐likelihood and Bayesian coalescent‐based methods. The latter was used to estimated time of the most recent common ancestor and demographic history. Recombination analysis was performed using the jumping profile hidden Markov model algorithm. Results: Our analysis revealed a broad HIV‐1 diversity in Moscow region, including sub‐subtype A6 (85.4%), subtype B (7.4%) and CRF02_AG (1.2%). Forty‐seven (4.2%) unique recombination forms (URF_A6/B) were also found. Other HIV‐1 subtypes were detected as single cases. Among heterosexuals, IDUs and MTCTs, the A6 subtype was most prevalent (>85.0%, in all groups); in MSM group ‐ subtype B (74.4%). The sub‐subtype A6 sequences were combined into the epidemic cluster that arose approximately 1997. Within the subtype B five major epidemic clusters containing 95.0% of the total subtype B sequences were identified. Each cluster contained sequences associated with only one or two dominant transmission routes. The age of these clusters varied widely between 1978 and 1990. Reconstruction of the demographic history of sub‐subtype A6 and B identified at least two epidemic growth phases. Both subtypes displayed the initial phase of exponential growth (subtype A6 ‐ until 2010; subtype B – until 1995), followed by a decrease in the growth rate and the stationary phase approaching the present time. Conclusions: Our study demonstrated a high degree of HIV‐1 subtypes diversity in the Moscow region. We showed that HIV‐1 subtype B introduction in the Moscow region occurred much earlier than HIV‐1 A6 subtype, which agreed well with the known HIV‐1 epidemiological history in the region. Despite the different age of subtypes A6 and B populations, the rate of growth of epidemics caused by them stabilises, which may be the result of ARV therapy and/or preventive measures. P316 Effects of immune checkpoint inhibitors on HIV reservoirs and responses in patients with HIV‐associated cancers J Forni 1, A Dalla‐Pria1, W Ma2, M Liu2, N Immani2, J Evans1, T Newson‐Davis1, C Brock1, X Xu2 and M Bower1 1National Centre for HIV Malignancy, Chelsea and Westminster Hospital, London, UK. 2Centre for Immunology and Vaccinology, Imperial College London, London, UK Background: Immune checkpoint inhibitors including anti‐PD1 antibodies and anti‐CTLA4 antibodies are effective therapies for cancers. Latent reservoir cells infected with HIV express PD‐1 and are potential targets for these antibodies. A recent publication reported a decline in total HIV DNA in a patient treated with nivolumab for HIV‐associated lung cancer [1]. Methods: Peripheral blood samples were collected from HIV‐positive patients on antiretroviral therapy with undetectable plasma HIV RNA, who were receiving immune checkpoint inhibitor (ICI) therapy for cancer. Samples were collected before and after three cycles of ICI therapy (two months). HIV latent reservoirs were assayed by quantitative PCR for total HIV DNA and by Alu‐gag PCR for integrated proviral HIV DNA. Anti‐HIV T‐cell responses were evaluated by interferon gamma ELISPOT. Results: Three patients were enrolled (one melanoma, two NSCLC). The patient with melanoma was treated with a combination of ipilimumab (anti‐CTLA4) and nivolumab (anti‐PD1), whilst the other two patients were treated with pembrolizumab (anti‐PD1). The baseline CD4 median was 376/mm3 and all had undetectable plasma HIV viral loads. After three cycles of ICI therapy, no changes in CD4 counts or percentages were observed. Similarly, there were no significant changes in the HIV reservoirs as measured by total HIV DNA (per PBMC or per CD4 cell) and integrated HIV DNA (per PBMC or per CD4 cell). ELISPOT assays using patient PBMCs and pooled HIV‐derived peptides will be presented. Conclusions: Despite positive effects on HIV reservoirs reported elsewhere in one patient, we observed no significant effects on latent reservoirs in three patients. Reference: [1] Guihot A, Marcelin AG, Massiani MA, Samri A, Soulie C, Autran B, et al. Drastic decrease of the HIV reservoir in a patient treated with nivolumab for lung cancer. Ann Oncol. 2018;29:517‐18. P317 Western blot in treated people with HIV‐1 chronic infection: frequency of negative HIV‐1 pol genes F Rinaldi 1, S Rolla2, L Galli3, P Andrea3, A Bigoloni3, C Muccini4, A Mastrangelo4, S Nozza3, C Tavano2, M Clementi5, A Lazzarin3, A Bartoloni6 and A Castagna7 1Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy. 2Laboratory of Microbiology and Virology, IRCCS San Raffaele Scientific Institute, Milan, Italy. 3Infectious Diseases, IRCCS San Raffaele Scientific Institute, Milan, Italy. 4Infectious Diseases, Vita‐Salute San Raffaele University, Milan, Italy. 5Laboratory of Microbiology and Virology, Vita‐Salute San Raffaele University, Milan, Italy. 6Infectious and Tropical Diseases Unit, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy. 7Infectious Diseases, Vita‐Salute San Raffaele University, IRCCS San Raffaele Scientific Institute, Milan, Italy Background: The aim of this study was to evaluate immune responses against HIV‐1 gag, env and pol genes in adult people with HIV‐1 chronic infection, with long exposure to ART. Materials and methods: Retrospective analysis on all adult people with HIV‐1 chronic infection, followed at the Infectious Diseases Clinic of the San Raffaele Hospital, Milan, Italy, with a western blot (WB) test performed after at least 12 months of ART (if multiple WB tests per patient were available, the most recent was considered). Patients’ characteristics at WB determination described by median (quartiles) or frequency (%) and compared by Mann‐Whitney test or chi‐square test. A multivariate logistic regression, including age, gender, HCV co‐infection, years of ART, change in CD4+ at WB since ART start (CD4+ slope), years of HIV‐RNA 50 copies/mL, are observed in otherwise successfully treated patients. We evaluated the rate and the effect of TVL increase during the ST.O.RE. study, an Italian prospective, single‐arm, multicentre, non‐interventional study in patients switching from ritonavir‐boosted protease inhibitor to darunavir/cobicistat (DRV/c)‐based therapy, in order to collect data from clinical practice. Materials and methods: All available HIV‐1 RNA data, collected in patients on treatment, were included throughout 48 weeks of observation. TVL increase was defined as HIV‐1 RNA values ≥50 copies/mL both single or confirmed measurements occurring anytime during the 48 weeks whether or not leading to virology failure. We also analysed the CD4 cell count, CD8 cell count and CD4/CD8 ratio in patients showing TVL increase to observe potential change in immunological parameters. Continuous data were presented as median and interquartile range (IQR). Comparisons between groups were performed using Wilcoxon rank‐sum test and correlation analyses were performed by calculating Pearson coefficient. All tests were two sided and a p value 1000 copies/mL in eight patients (Group 1) while TVL between 50 and 1000 copies/mL (Group 2) was reported in 10. Reported reasons for TVL increase were virological blip (a single measurement between 50 and 1000 copies/mL) in nine patients, non‐adherence (VL >50 copies/mL and/or >1000 copies/mL in single or consecutive measurements during declared non‐adherence period) in seven, virological failure in one, drug‐drug interactions in one. We analysed comparison between TVL increase and immunological status when available. CD4 cell count and CD4/CD8 ratio decrease were statistically significant in patients with TVL >1000 copies/mL as compared to those with TVL 1000 copies/mL median (IQR) n = 8 Group 2 HIV RNA 1000 copies/mL) appears to affect the immunological status with possible increase of immune activation and inflammation. P319 Persistent outbreak of the HIV‐1 CRF19_cpx variant in treatment‐naïve MSM patients in Malaga area (Spain) I Viciana Ramos1, C González Domenech 2, M Mayorga3, J De La Torre Lima4, C Gómez Ayerbe1, M Castaño3, A Del Arco4, R Palacios Muñoz1 and J Santos1 1UGC Infectious Diseases and Clinical Microbiology, Hospital Virgen de la Victoria, Malaga, Spain. 2Microbiology, University of Granada, Granada, Spain. 3Infectious Diseases Unit, Hospital Carlos Haya, Malaga, Spain. 4Infectious Diseases Unit, Hospital Costa del Sol, Marbella, Spain Background: During the period 2011 to 2016, the HIV‐1 CRF19_cpx variant emerged as an outbreak in newly HIV diagnoses (NDVIH) in southern Spain. Our aim was to determine the current status of this outbreak, analysing the new cases of this variant in our area and their epidemiological relationship with the previous ones. Materials and methods: We considered all the HIV‐1 genotype resistance tests performed in NDVIH at Virgen de la Victoria Hospital, reference centre in southern Spain, from January 2017 to June 2018. Drug resistance mutations were determined with Viroseq HIV® system and the partial sequence of HIV‐1 pol gene provided submitted to REGA v.3.0 for subtyping. Sequences assigned as CRF19_cpx subtype were phylogenetically compared to the 254 reference sequences of the same variant retrieved from the LANL, as well as to the 55 ones comprising the already described CRF19_cpx variant outbreak. The alignment was done by ClustalX and the phylogenetic reconstruction inferred by maximum likelihood method (PhyML v.3.0 programme). The cluster reliability was supported on the value of SH‐like aLRT test. The resistance mutations were predicted using Stanford algorithm v.7.1.1. We also collected demographic, clinical and immunovirological data. Results: During the study period, 523 resistance studies were performed in NDVIH; 12 (2.3%) had sequences consigned in REGA as subtype CRF19_cpx. All the new cases conformed a very well‐defined transmission cluster (aLRT=92%) with the CRF19_cpx sequences from the previous outbreak, already comprising up to 67 patients. Eight of the new sequences were clustering within two subclusters previously defined: E and F, currently including 18 and three patients, respectively. We have not found the G190A mutation in any of the new sequences. The new cases of the CRF19_cpx were MSM, with an average age of 32.5 years (IQR 27.1 to 43.6) and Spaniards, except one Italian patient. Half of them were seroconverters (mean seroconversion time of 17.0 months, 8.3 to 81.3). The initial T‐lymphocyte CD4 count was 423 cells/μL (200 to 562) and initial viral load was 4.9 log copies/mL (4.6 to 5.2). Conclusions: All the new cases of the CRF19_cpx variant emerged in our area during 2017 and half this year are phylogenetically clustered with the previous outbreak, pointing out its active status. The NDVIH infected with this variant possess similar epidemiological, clinical and immunovirological characteristics to those already included in the outbreak. None of the new sequences of this subtype showed the G190A mutation. The active transmission of the CRF19_cpx variant in our area should warn us about the necessity of intense epidemiological surveillance programmes. P320 HIV‐1 subtype diversity and international travel in Romanian people who inject drugs R Jipa 1, S Paraschiv2, L Banica2, D Otelea2, E Manea1, I Nicolae2, A Abagiu1 and A Hristea1 1Clinical Department, National Institute for Infectious Diseases “Prof Dr Matei Bals”, Bucharest, Romania. 2Molecular Diagnostics Laboratory, National Institute for Infectious Diseases “Prof Dr Matei Bals”, Bucharest, Romania Background: Although subtype F1 is highly prevalent in Romania, increases in international travel have led to increased detection of other subtypes in newly diagnosed HIV patients [1]. Objective: Describe clinical and epidemiological patterns as correlated with recent history of international travel in newly diagnosed IVDUs with HIV and to assess how travel contributes to the spread of different subtypes. Methods: We retrospectively studied 270 IVDUs diagnosed with HIV infection between October 2012 and September 2017. Epidemiological, demographic and clinical data were collected through a questionnaire survey. We excluded patients with missing data regarding travel history and/or available HIV pol sequences. Subtype analysis was done using Rega HIV‐1 subtyping tool v2.0 and further phylogenetic analysis was performed with FastTree. Differences between groups were analysed using the Mann‐Whitney U test for continuous variables and the chi‐square test for dichotomous variables. Results: Sixty‐nine patients were excluded from the study. Out of 201 patients included, 92 (46%) had a recent history of international travel. The characteristics of patients from the two groups are shown in Table 1. Travellers who used intravenous drugs and had unprotected sex during travel were infected in a higher proportion with non‐F1 subtypes (73% vs. 61%, p = 0.162, OR [95% CI] 2.69 [0.7 to 10.31], and 38% vs. 36%, p = 0.798, OR [95% CI] 1.25 [0.45 to 3.41]). Travel to Spain was reported more frequently for IVDUs infected with F1 strains than for those infected with other subtypes (52% vs. 46%, p = 0.6, OR [95% CI] 0.74 [0.26 to 2.08]). More IVDUs with non‐F1 subtypes reported travel to Greece as compared to IVDUs with F1 subtype (p = 0.002, OR [95% CI] 6.71 [2.02 to 22.25]). Phylogenetic analysis indicated that IVDUs sequences are clustering together, regardless of their travelling status. The sequences of travelling IVDUs are intermixed in the phylogenetic tree with the sequences of non‐travelling IVDUs. For subtype B, the sequences from IVDUs who reported travelling abroad are more diverse and intermixed with reference sequences. Abstract P320 – Table 1. Epidemiological, clinical and laboratory characteristics in travellers versus non‐travellers International travellers N = 92 Non‐travellers N = 109 p OR [95% CI] Male, N (%) 78 (85) 91 (84) 0.84 0.91 [0.42 to 1.94] Age (years), median (IQR) 31 (26 to 34) 30 (26 to 35) 0.75 In prison, N (%) 63 (68) 56 (56) 0.01 2.05 [1.15 to 3.66] HIV stage: A, N (%) 48 (52) 78 (72) 0.09 0.42 [0.16 to 1.07] HIV stage: B, N (%) 11 (12) 7 (6) 0.07 2.53 [0.92 to 6.97] HIV stage: C, N (%) 33 (36) 24 (22) 0.01 1.97 [1.06 to 3.71] Chronic hepatitis B, N (%) 1 (1) 10 (9) 0.02 0.12 [0.01 to 1.01] Chronic hepatitis C, N (%) 91 (99) 105 (96) 0.62 2.6 [0.26 to 25.44] CD4 T‐cell count (cells/mm3), median (IQR) 404 (246 to 508) 402 (256 to 575) 0.91 Heroin abuse, N (%) 88 (96) 100 (92) 0.38 1.98 [0.58 to 6.65] New psychoactive substances abuse, N (%) 80 (87) 100 (92) 0.35 0.6 [0.24 to 1.49] HIV subtype F1, N (%) 66 (72) 82 (75) 0.63 0.83 [0.44 to 1.56] HIV subtype B, N (%) 5 (5) 2 (2) 0.25 3.07 [0.58 to 16.23] HIV subtype CRF14_BG, N (%) 14 (16) 14 (13) 0.68 1.21 [0.54 to 2.71] HIV subtype CRF14_F1, N (%) 5 (5) 3 (3) 0.47 2.03 [0.47 to 8.73] HIV other recombinant subtypes, N (%) 2 (2) 7 (6) 0.18 0.32 [0.06 to 1.59] Conclusions: Regardless of the history of international travel, non‐F1 subtypes were found in similar proportions. Among travellers, drug use and unprotected sex were reported especially in patients with non‐F1 subtypes. Non‐F1 subtypes were associated with travelling to Greece. Phylogenetic analysis also suggested international transmission events for CRF14_BG and B subtypes. Reference: [1] Niculescu I, Paraschiv S, Paraskevis D, Abagiu A, Batan I, Banica L, et al. Recent HIV‐1 outbreak among intravenous drug users in Romania: evidence for cocirculation of CRF14_BG and subtype F1 strains. AIDS Res Hum Retroviruses. 2015;31:488‐95.

0 comments Cited 14 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: not found
Article: not found

Genital HIV-1 Shedding With Dolutegravir (DTG) Plus Lamivudine (3TC) Dual Therapy.

Sara Gianella, Vincent C. Marconi, Baiba Berzins … (2018)

0 comments Cited 6 times – based on 0 reviews      Review now

Bookmark

All references

Author and article information

Contributors

Sandro Vento

Journal

Journal ID (nlm-ta): EClinicalMedicine

Journal ID (iso-abbrev): EClinicalMedicine

Title: EClinicalMedicine

Publisher: Elsevier

ISSN (Electronic): 2589-5370

Publication date PMC-release: 10 January 2019

Publication date Collection: December 2018

Publication date (Electronic): 10 January 2019

Volume: 6

Pages: 5-6

Affiliations

[a ]Diagnosis and Therapy of HIV Infection Unit, G.B. Rossi University Hospital, Verona, Italy

[b ]Department of Medicine, School of Medicine, Nazarbayev University, Astana, Kazakhstan

[c ]University Medical Center, Astana, Kazakhstan

Author notes

[* ]Corresponding author at: Dept. of Medicine, School of Medicine, Nazarbayev University, 5/1 Kerey and Zhanibek Khans Str., 010000 Astana, Kazakhstan. sandro.vento@ 123456nu.edu.kz

Article

Publisher ID: S2589-5370(19)30001-X

DOI: 10.1016/j.eclinm.2019.01.001

PMC ID: 6537522

PubMed ID: 31193618

SO-VID: 2300e393-4d6a-47d6-9e75-5bfe279e8338

License:

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

History

Date received : 11 December 2018

Date accepted : 4 January 2019

Comments

Comment on this article

scite_

Cited by 3

See all cited by

Most referenced authors 77

See all reference authors

- Version 1
- Version 1

Patient-centered Care and Treatment in HIV Infection

Read this article at

Abstract

Related collections

Decoding Infection and Transmission

Most cited references 7

Tackling Africa's chronic disease burden: from the local to the global

HIV Glasgow 2018, 28–31 October 2018, Glasgow, UK

Genital HIV-1 Shedding With Dolutegravir (DTG) Plus Lamivudine (3TC) Dual Therapy.

Author and article information

Contributors

Journal

Affiliations

Author notes

Article

History

Categories

Comments

Comment on this article

Similar content 124

Cited by 3

Most referenced authors 77