- Record: found
- Abstract: found
- Article: found
Seasonal malaria vaccination: protocol of a phase 3 trial of seasonal vaccination with the RTS,S/AS01 E vaccine, seasonal malaria chemoprevention and the combination of vaccination and chemoprevention
research-article
Author(s):
Daniel Chandramohan
1
,
,
Alassane Dicko
2 ,
Issaka Zongo
3 ,
Issaka Sagara
2 ,
Matthew Cairns
4 ,
Irene Kuepfer
1 ,
Modibo Diarra
2 ,
Amadou Tapily
2 ,
Djibrilla Issiaka
2 ,
Koualy Sanogo
2 ,
Almahamoudou Mahamar
2 ,
Frederic Sompougdou
3 ,
Serge Yerbanga
3 ,
Ismaila Thera
2 ,
Paul Milligan
4 ,
Halidou Tinto
3 ,
Opokua Ofori-Anyinam
5 ,
Jean-Bosco Ouedraogo
3 ,
B Greenwood
1
Publication date (Electronic):
15 September 2020
Journal:
BMJ Open
Publisher:
BMJ Publishing Group
Read this article at
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
Introduction
Seasonal malaria chemoprevention (SMC), with sulphadoxine–pyrimethamine plus amodiaquine (SP+AQ) is effective but does not provide complete protection against clinical malaria. The RTS,S/AS01 E malaria vaccine provides a high level of protection shortly after vaccination, but this wanes rapidly. Such a vaccine could be an alternative or additive to SMC. This trial aims to determine whether seasonal vaccination with RTS,S/AS01 E vaccine could be an alternative to SMC and whether a combination of the two interventions would provide added benefits.
Methods and analysis
This is an individually randomised, double-blind, placebo-controlled trial. 5920 children aged 5–17 months were enrolled in April 2017 in Mali and Burkina Faso. Children in group 1 received three priming doses of RTS,S/AS01 E vaccine before the start of the 2017 malaria transmission season and a booster dose at the beginning of two subsequent transmission seasons. In addition, they received SMC SP+AQ placebo on four occasions each year. Children in group 2 received three doses of rabies vaccine in year 1 and hepatitis A vaccine in years 2 and 3 together with four cycles of SMC SP+AQ each year. Children in group 3 received RTS,S/AS01 E vaccine and four courses of SMC SP+AQ. Incidence of clinical malaria is determined by case detection at health facilities. Weekly active surveillance for malaria is undertaken in a randomly selected subset of children. The prevalence of malaria is measured in surveys at the end of each transmission season. The primary endpoint is the incidence of clinical malaria confirmed by a positive blood film with a minimum parasite density of 5000 /µL. Primary analysis will be by modified intention to treat defined as children who have received the first dose of the malaria or control vaccine.
Related collections
Most cited references12
- Record: found
- Abstract: found
- Article: not found
Efficacy of RTS,S/AS02 malaria vaccine against Plasmodium falciparum infection in semi-immune adult men in The Gambia: a randomised trial.
T. Doherty, P Gothard, M Delchambre … (2001)
RTS,S/AS02 is a pre-erythrocytic malaria vaccine based on the circumsporozoite surface protein of Plasmodium falciparum fused to HBsAg, incorporating a new adjuvant (AS02). We did a randomised trial of the efficacy of RTS,S/AS02 against natural P. falciparum infection in semi-immune adult men in The Gambia. 306 men aged 18-45 years were randomly assigned three doses of either RTS,S/AS02 or rabies vaccine (control). Volunteers were given sulfadoxine/pyrimethamine 2 weeks before dose 3, and kept under surveillance throughout the malaria transmission season. Blood smears were collected once a week and whenever a volunteer developed symptoms compatible with malaria. The primary endpoint was time to first infection with P. falciparum. Analysis was per protocol. 250 men (131 in the RTS,S/AS02 group and 119 in the control group) received three doses of vaccine and were followed up for 15 weeks. RTS,S/AS02 was safe and well tolerated. P. falciparum infections occurred significantly earlier in the control group than the RTS,S/AS02 group (Wilcoxon's test p=0.018). Vaccine efficacy, adjusted for confounders, was 34% (95% CI 8.0-53, p=0.014). Protection seemed to wane: estimated efficacy during the first 9 weeks of follow-up was 71% (46-85), but decreased to 0% (-52 to 34) in the last 6 weeks. Vaccination induced strong antibody responses to circumsporozoite protein and strong T-cell responses. Protection was not limited to the NF54 parasite genotype from which the vaccine was derived. 158 men received a fourth dose the next year and were followed up for 9 weeks; during this time, vaccine efficacy was 47% (4-71, p=0.037). RTS,S/AS02 is safe, immunogenic, and is the first pre-erythrocytic vaccine to show significant protection against natural P. falciparum infection.
- Record: found
- Abstract: found
- Article: found
Intermittent Preventive Treatment of Malaria Provides Substantial Protection against Malaria in Children Already Protected by an Insecticide-Treated Bednet in Mali: A Randomised, Double-Blind, Placebo-Controlled Trial
Introduction An estimated 863 million people live in sub-Saharan Africa of whom 16.2% are under 5 y of age [1]. About 300 million people live in areas where malaria transmission is highly seasonal. Malaria remains a major cause of morbidity and mortality and is estimated to cause 881,000 deaths globally per year and sub-Saharan Africa is disproportionately affected, suffering 91% of global malaria deaths with 88% occurring in children under 5 y of age [2]. Thus, in the absence of a vaccine, simple and effective control strategies are urgently needed to reduce the malaria burden in sub-Saharan Africa. Vector control, using insecticide-treated bednets (ITNs), insecticide-treated curtains, or indoor residual spraying (IRS), can reduce mortality and morbidity from malaria substantially [3], but in high transmission settings, these interventions provide only partial protection and additional control measures are needed. Intermittent preventive treatment (IPT) is a new approach in the prevention of malaria in infants and older children. Several randomised controlled trials have demonstrated that IPT of malaria in infants (IPTi) with sulphadoxine pyrimethamine (SP) given during routine vaccinations at approximately 2, 3, and 9 mo of age, reduces the incidence of clinical malaria by 22% to 59% [4], and this strategy has been shown to be safe and cost effective. However, in many regions of Africa, the main burden of malaria falls not on infants but on older children [5]. In parts of Africa, such as much of the Sahel and sub-Sahel, where malaria transmission is very seasonal, the incidence of severe malaria currently peaks at 2 to 3 y of age. As the overall incidence of malaria decreases in Africa in response to enhanced control efforts, an effect already being seen in some countries, it can be anticipated that the mean age of cases of malaria will increase further. For these reasons, trials have been undertaken in areas of seasonal malaria transmission to determine whether IPT in children (IPTc) could be used as an effective malaria control tool in older children. In Mali, a 69% reduction in the incidence of clinical malaria was seen in children 0–5 y old when two doses of SP were given 8 wk apart during the malaria transmission season [6]. In Senegal, SP plus a single dose of artesunate (AS), administered on three occasions at monthly intervals during the peak malaria season, reduced the incidence of clinical malaria by 86% [7]. A subsequent trial of different drug regimens showed that IPT with SP and amodiaquine (AQ) was even more effective than SP+AS, providing approximately 95% protection [8]. A further study, conducted in an area of Ghana with more prolonged transmission, found that AS+AQ monthly was more effective than AS+AQ or SP alone given every 2 mo, suggesting that for drugs such as SP and AQ, monthly administration is needed to achieve effective IPTc [9]. Bednet coverage among young children was low at each of the sites where these trials were conducted and use of ITNs was very uncommon. Use of ITNs is now a favoured approach to the control of malaria in most parts of Africa and major efforts are being made to scale up their use. With international support, ITN coverage is increasing in many malaria endemic countries in sub-Saharan Africa [10] and it is expected that almost universal coverage with ITNs in high risk groups, as called for in the Global Malaria Action Plan [11], will be achieved in many malaria endemic countries. Thus, following on the initial encouraging results obtained with IPTc, an issue that needs to be addressed urgently is whether IPTc can provide significant added benefit to the protection against malaria provided by ITNs to warrant its use as a malaria control tool in areas with seasonal transmission of malaria and a high use of ITNs. It was initially planned to address this question simultaneously in each of the three countries Mali, Burkina Faso, and Ghana, using a similar design and methods. However, the site in Ghana had to be abandoned because of delays in obtaining regulatory approval for the use of SP+AQ, the drug combination chosen for the study on the basis of the results of previous trials and knowledge of the sensitivity of Plasmodium falciparum to these drugs in the proposed study areas. Very similar protocols were used for the studies conducted in Burkina Faso and Mali. Methods The protocol of the trial (Text S1), protocol amendment (Text S5), and CONSORT checklist (Text S2) are available as supporting information. Objectives The primary objective of the study was to determine the degree to which IPTc given during the malaria transmission season reduces the incidence of clinical malaria in children who sleep under a long-lasting insecticide-treated net (LLIN). Secondary objectives were determination of the impact of this strategy on severe malaria, all cause hospital admissions, anaemia, nutrition (wasting, stunting, and being underweight), malaria infection, and molecular markers of resistance to SP and AQ. Study Sites The study was conducted in two rural villages, Djoliba and Siby, and the small town of Ouelessebougou situated in the district of Kati in the savannah region of Mali. Djoliba and Siby are located 40 and 30 km south west of the capital city Bamako, respectively, and Ouelessebougou is located 80 km south of Bamako. In Djoliba and Siby, community health centres are staffed with a physician and nurses. In Ouelessebougou, the community health centre is staffed by an assistant physician and nurses, but located less than 100 m from a district health centre staffed by four physicians and six nurses. A research team composed of physicians and medical residents was established in each of the three sites to follow up and provide health care to the study participants. Malaria transmission in the study area is highly seasonal and 80%–90% of malaria cases occur between August and November. The entomological inoculation rate (EIR) was 9.4 and 6.6 infective bites per person per season, respectively, in Siby and in Ouelessebougou, two localities far from any river and 37.3 infective bites per person per season in Djoliba located on the bank of the Niger River (Text S3). The coverage of ITNs at baseline was 33.4% (312/935) in Siby, 84.7% (563/665) in Djoliba, and 89.8% (2,207/2,458) in Ouelessebougou. Study Design and Participants The study was designed as an individually randomised, placebo-controlled trial of IPTc with SP+AQ in children who received a LLIN. Children aged 3–59 mo were enumerated and given a census identification number including a house number to facilitate their identification at screening, enrolment, and follow-up. Recruitment was started in Djoliba followed by Siby. In these communities all available children in the target age group who were not selected for the baseline survey of drug resistance were screened and enrolled if they met the inclusion criteria. In the larger community of Ouelessebougou, children were screened for enrolment on a first-come first-served basis until the required sample size was met. Children were eligible to join the study if they were aged 3–59 mo at the time of enrolment and permanent residents of the study area with no intention of leaving during the study period. Exclusion criteria were the presence of a severe, chronic illness, such as severe malnutrition or AIDS, and a history of a significant adverse reaction to SP or AQ. Cases of an acute illness, such as malaria, were not excluded. Such cases were treated appropriately and the child randomised and retained in the trial. Ethics The study protocol was reviewed and approved by the Ethical Committee of the Faculty of Medicine, Pharmacy and Dentistry, University of Bamako, Mali and by the Ethics Committee of the London School of Hygiene and Tropical Medicine. Community consent was obtained at meetings with leaders, heads of families, and other community members of each locality prior to the start of the study. Individual, written, informed consent was obtained from a parent or guardian of each child prior to screening and enrolment. A Data and Safety Monitoring Board (DSMB) was established and monitored the trial with the support of a local medical safety monitor. Current good clinical practices (cGCP) monitoring of the trial was performed by PharmaClin (http://www.pharmaclin.com). Interventions Every child who was screened was provided with a LLIN (Permanet, Vestergaard Frandsen) that was marked with the child's identification number regardless of whether or not the child was enrolled. Instructions were given to the parent or guardian on how to use the net and the importance of using the net regularly was emphasized. Monitoring of utilisation of ITNs by study participants was made in 150 randomly selected children each week and in all study children during the cross-sectional survey conducted at the end of the malaria transmission season. Eligible children were treated with a course of SP+AQ or matching placebos on three occasions at monthly intervals during the malaria transmission season, starting in August 2008. SP and AQ were manufactured by Kinapharma Limited and quality control checks on the drugs for solubility and content were performed at the London School of Tropical Medicine and Hygiene, prior to their use in the trial. Tablets met internal standards for drug solubility and content. Doses of SP and AQ were based on weight with children stratified into one of the three weight categories (5–9 kg, 10–18 kg, and ≥19 kg). SP was given at a dose of 175/8.75 mg to children 5–9 kg, 350/17.5 mg to children 10–18 kg, and 550/26.25 mg to those who weighed ≥19 kg. The corresponding doses for AQ were 70 mg, 140 mg, and 220 mg, respectively. AQ was given over 3 d. Drugs were prepackaged to facilitate administration and put in envelopes with colour codes, one for each weight group. Within each weight stratum, children were individually randomised using a computer-generated random number sequence and blocks of varying length. Treatment allocations were provided within sealed, opaque envelopes. Drugs were given under direct observation at a research clinic by study staff. Children were observed for 30 min after drug administration. If vomiting occurred during this 30-min period, drugs were readministered. If vomiting occurred on a second occasion, this was noted but the drugs were not given again. Such children were not excluded from the trial and they were eligible to receive drugs on the subsequent 2 d and during subsequent monthly IPT rounds. If a child missed the day set for treatment, a home visit was made to enquire why the child had not been brought for treatment and the reason was recorded. If the family wished to continue with treatment but was unable to attend on the specified day, then treatment was reoffered within an interval of 7 d of the designated date. Children with an acute malaria episode were treated with artemether-lumefantrine (AL) and did not receive IPT with SP+AQ if the treatment for acute malaria was received within 7 d of the scheduled date of IPT. Such children were eligible for treatment in future treatment rounds Outcomes The primary endpoint of the study was the incidence of clinical malaria; this was defined as the presence of fever (axillary temperature ≥37.5°C) or a history of fever in the past 24 h and the presence of P. falciparum asexual parasitaemia at a density greater or equal to 5,000 parasites per microlitre. Secondary endpoints were: (i) the incidence of clinical malaria defined as the presence of fever or a history of fever in the past 24 h and the presence of P. falciparum asexual parasitaemia at any density; (ii) incidence of severe malaria defined according to the WHO criteria [12]; (iii) malaria infection defined as the presence of asexual parasitaemia; (iv) mild, moderate, or severe anaemia defined as an haemoglobin (Hb) concentration <11 g/dl, <8 g/dl, and <5 g/dl, respectively; (v) hospital admission defined as a stay of at least 24 h in hospital for treatment; (vi) anthropometric indicators including wasting, stunting, and being underweight as defined by WHO [13]; and (vii) safety and tolerability measured by the occurrence of nonserious and serious adverse events. Passive surveillance for clinical malaria started at the time of the administration of the first dose of IPTc in August 2008 and continued until the end of the malaria transmission season in November/December 2008, 6–7 wk after the last round of IPTc. Parents were encouraged to bring their child to a study health centre, where medical staff were available 24 h a day and 7 d a week, if the child became unwell. A finger prick blood sample was be obtained from all study children with fever (an axillary temperature of 37.5°C or higher) or a history of fever within the previous 24 h for preparation of a blood film, measurement of Hb concentration, and for a rapid diagnostic test (RDT) OPTIMAL_IT (Diamed AG) for malaria. Children who had a positive RDT for malaria were treated immediately with AL. Severe cases were admitted to the health centre or referred to the paediatric ward of the Gabriel Touré Hospital in Bamako. Causes of death were assessed within a month of death using a modified version of the INDEPTH post mortem questionnaire (http://www.indepth-network.org/index.php?option=com_content&task=view&id=95&Itemid=183). Use of a LLIN was assessed by asking if a child had slept under an LLIN the previous night and the presence of the net was checked by field staff. During these home visits, the axillary temperature of each child was taken and a blood film obtained regardless of whether or not the child had fever. A RDT was performed if a child had measured fever or a history of fever within the previous 24 h and if this was positive, treatment with AL was given according to national guidelines. At the end of the malaria transmission season, a cross-sectional survey was undertaken at which every child was examined, their height and weight recorded, and a finger prick blood sample obtained for determination of Hb concentration, preparation of blood films, and collection of a filter paper sample for subsequent molecular studies. Safety and tolerability of SP and AQ were monitored passively during the study period in all the children and actively in a subset at the time of the administration of IPT (days 0, 1, and 2) and 1 d after the last dose of treatment (day 3) at each round. Assessment of Molecular Markers of Drug Resistance Monitoring of the frequency of molecular markers of resistance to sulphadoxine, pyrimethamine, and AQ was performed in two cross-sectional surveys, the first at baseline in August 2008 and the second during the survey undertaken at the end of malaria transmission season. The baseline survey was conducted in 256 children randomly selected from the screening list. These children were not enrolled in the placebo control trial. Participants enrolled in the placebo control trial were surveyed about 6 wk after the third course of IPTc, at the end of malaria transmission season, to assess whether administration of IPT with SP+AQ had lead to an increase in molecular markers of resistance to these drugs. Thick and thin blood smears and blood blotted onto filter papers were collected during both surveys for molecular analysis as described below. Laboratory Methods Thick blood films were air dried, stained with Giemsa, and examined for malaria parasites by two well-trained technicians. 100 high power fields were counted before a film was declared negative. Parasite density was determined by counting the number of parasites present per white blood cell (WBC) on a thick smear and assuming a WBC count of 8,000 per µl. In the case of a discrepancy (positive/negative or a difference in parasite density greater than 30%), a third reading was done. The median parasite density of two or three readings was used. An external quality control of slide reading performed by the Malaria Diagnosis Centre of Excellence (MDCoE) of the Walter Reed/Kenya Medical Research Institute, in Kisumu, Kenya, showed an overall concordance of more than 90% on parasite detection and 100% on species identification (Text S4). Hb concentrations were measured using a haemoglobin analyzer (Hemocue HB 301) on blood obtained by finger prick. Filter paper samples from children with a mono-infection of P. falciparum on blood smears were analysed by nested PCR for mutations at codons 51, 59, and 108 of the dhfr gene, 437 and 540 of the dhps gene, 76 of mutations in the P. falciparum chloroquine transporter gene (pfcrt), and 86 of the P. falciparum multidrug resistance gene one (pfmdr1) according to published methods [14]–[16]. Cases of mixed infection (wild type and mutant) were categorized as mutant. Sample Size Calculation of sample size was based on the assumptions that the clinical attack rate measured by passive surveillance would be 1.0–2.0 attacks per child per year in unprotected children aged 3–59 mo living in the study areas and that sleeping under an LLIN would reduce this attack rate by half to 0.5 to 1.0 clinical episode per child per year. Assuming that children experienced an average of 0.5 clinical episodes per child per year of sufficient severity to present to a health facility, to detect a 20% reduction in this incidence (i.e., from 0.5 to 0.4 attacks per child per year) in children who receive IPTc, the smallest reduction that would be likely to make IPTc a worthwhile investment, and allowing for a 20% loss to follow-up, we estimated that approximately 2,000 children (1,000 in each arm) were required for a study with 90% power at the two-sided 5% level of significance [17]. After the site in Ghana was dropped, the sample size was increased to 1,500 participants per arm, after an amendment was made to the protocol (Text S5), which would have 80% power to detect a two-thirds reduction in the incidence of severe malaria, assuming an incidence of 2% in children in the control arm. The study was not powered to detect a smaller reduction in the incidence of severe malaria but the analysis plan included provision for combination of the results of this trial with those of a parallel study conducted in Burkina Faso to provide sufficient size to allow detection of a smaller impact of IPTc on this end point. Data Management and Analysis Data were collected on standardized forms, double-entered, and verified using MS Access and then exported to Stata (StataCorp) for additional cleaning and analysis. A data analysis plan was written and submitted to the DSMB prior to analysis. The final, cleaned database was locked and a copy sent to the DSMB. An intention-to-treat analysis was performed. Incidence rates of clinical malaria, severe malaria, and hospital admissions were calculated by dividing the number of episodes by the total child days at risk. Children were not considered at risk for 21 d after each type of a malaria episode and these days were not included in the calculation of the child days at risk. The incidence rates in the two treatment groups were compared using Cox regression to estimate the incidence rate ratio, with adjustment for age, gender, and locality, and using a robust standard error to allow for the lack of independence among repeated episodes in the same child. The protective effect (PE) of IPTc was computed as 1 minus the incidence rate ratio. Time to first episode of clinical malaria in the two arms was examined using Kaplan-Meier plots and compared using log rank test. Anthropometric data at enrollment and at the end of season cross-sectional survey were converted into weight-for-age, height-for-age, and weight-for-height z-scores using WHO's anthropometric software (www.who.int/childgrowth/software/en). Underweight, stunting, and wasting were defined as z-scores of <−2 for the relevant indicator [13]. Changes in weight and height between the two groups were compared using Student's t test. Frequencies of single mutations as well as the triple mutant (dhfr 51+59+108) and quadruple mutant (triple mutant + dhps 437) genotypes were determined and compared between treatment arms and between the beginning and end of the study. Proportions of children with binary outcomes were compared between the two groups using Pearson's Chi square test or generalized linear models adjusted for age, gender, and locality. Results Trial Profile and Baseline Data The trial profile is summarised in Figure 1. A total of 3,065 children were screened of whom 3,017 (1,509 in the IPTc arm and 1,508 in the placebo arm) (98%) were enrolled. Reasons for exclusion are shown in Figure 1. The proportion of children who completed the follow-up to day 42 after the last round of IPTc was similar in the control and in the intervention arms (98.5% and 98.1%, respectively). The reasons for withdrawal were withdrawal of consent (n = 29), migration to another location (n = 15), a history of allergy to study drugs (n = 4 with two cases confirmed), and death (n = 3). There were no significant differences between intervention and control groups with regard to their age and gender distribution, nor in the prevalence of fever, wasting, or stunting at the time of enrolment (Table 1). 10.1371/journal.pmed.1000407.g001 Figure 1 Trial profile. 10.1371/journal.pmed.1000407.t001 Table 1 Baseline characteristics of enrolled children at the time of administration of the first dose of IPTc. Characteristics IPTc Placebo Percent (n/N) Percent (n/N) Age (mo) 3–11 18.2 (274/1,509) 18.5 (278/1,508) 12–23 22.5 (339/1,509) 20.5 (309/1,508) 24–35 20.5 (310/1,509) 22.0 (332/1,508) 36–47 20.0 (302/1,509) 19.4 (293/1,508) 48–59 18.8 (284/1,509) 19.6 (296/1,508) Gender Male 47.7 (720/1,509) 50.1 (755/1,508) Female 52.3 (789/1,509) 49.9 (753/1,508) Weight (kg) 5–9 34.8 (525/1,509) 34.7 (523/1,508) 10–8 63.1 (952/1,509) 63.2 (953/1,508) ≥19 2.1 (32/1,509) 2.1 (32/1,508) Nutritional factors Underweight 16.1 (238/1,480) 15.1 (223/1,477) Wasting 11.0 (163/1,480) 12.5 (185/1,477) Stunting 22.7 (336/1,480) 23.8 (352/1,477) Fever 7.2 (105/1,460) 7.6 (111/1,464) LLIN Usage Usage of LLINs was assessed for 590 children in the control group and for 591 children in the intervention group during weekly home visits, undertaken without prior warning, during the course of the intervention period. Usage of an LLIN was high in each of the three study localities and similar between the two groups (99.7% in the control group versus 99.3% in the intervention arm; p = 0.45). The Impact of IPTc on Malaria Among children with fever or history of fever who had an RDT positive result, 8.8% (112/1,277) turned out to have negative parasitaemia after microscopical diagnosis of malaria. The impact of IPTc on episodes of malaria detected through passive surveillance is presented in Table 2. The incidence of episodes of uncomplicated malaria (fever or a history of fever in the last 24 h and asexual parasitaemia ≥5,000/µl) was much lower among children in the IPTc arm than among those in the control arm (0.34 episodes per child/year versus 1.9 episodes per child/year). The PE against malaria adjusted for age, gender, and location was 82% (95% confidence interval [CI] 78%–85%) (p<0.001). An analysis of time to the first episode of clinical malaria, defined as above, also indicated a strong protective effect of IPTc (p<0.001) (Figure 2). The incidence of malaria defined as fever or a history of fever in the last 24 h and positive asexual parasitaemia of any density was also much lower in children in the IPTc arm compared to those in the control arm (0.41 episodes per child/year versus 2.4 episodes per child/year), giving a protective efficacy of 83% (95% CI 80%–86%) (p<0.001). Only 17 cases of severe malaria occurred during the follow-up period, 15 in the control group, and two in the intervention group (Table 2), giving a protective efficacy of 87% (95% CI 42%–99%) (p = 0.001). The two cases of severe malaria in the intervention arm, one of whom died, occurred more than 3 wk after the third course of IPT. 10.1371/journal.pmed.1000407.g002 Figure 2 Time to first episode of clinical malaria defined as fever (temperature ≥37.5°C) or history of fever in the last 24 h and parasitaemia ≥5,000/µl in the intervention and control arms. Kaplan-Meier survival estimates with pointwise 95% confidence bands. 10.1371/journal.pmed.1000407.t002 Table 2 Impact of IPTc on episodes of clinical malaria in children in Mali. Outcomes IPTc Placebo Unadjusted IRRs (95% CI) p-Value Adjustedc IRRs (95% CI) PE (95% CI) p-Value n Episodes Years at Riska Incidence Rate (95% CI)b n Episodes Years at Risk Incidence Rate (95% CI)b Fever or history of fever and any asexual parasitaemia 149 362.15 0.41 (0.35–0.48) 832 345.64 2.40 (2.25–2.58) 0.17 (0.14–0.20) <0.001 0.17 (0.14–0.20) 83 (80–86) <0.001 Fever or history of fever and parasitaemia ≥5,000 126 369.41 0.34 (0.29–0.41) 672 354.14 1.90 (1.76–2.05) 0.18 (0.15–0.22) <0.001 0.18 (0.15–0.22) 82 (78–85) <0.001 Severe malaria 2 399.10 0.005 (0.0006–0.0181) 15 400.87 0.037 (0.0209–0.0617) 0.13 (0.01–0.58) 0.001 — 87 (42– 99) 0.001 a Children were not considered at risk for 21 d after each type of a malaria episode. b Incidence rate/child/year. Note the incidence relate refers to only the 3-mo surveillance period and is not an annual rate. c Adjusted for age, gender, and location. 95% CI constructed using a robust standard error. IRR, incidence rate ratio. Incidence rates and the PE of IPTc against clinical malaria by locality and age category are presented in Table 3. Although the incidence of clinical malaria varied substantially between the three study localities, the PE of IPTc was similar in all three areas regardless of the definition of clinical malaria used. PE was higher in the lower age groups (3–11 mo and 12–23 mo) compared to the older age groups (≥24 mo) when the definition of clinical malaria that incorporated the presence of parasitaemia ≥5,000/µl or any parasitaemia was used (test for effect modification p≤0.001 and p = 0.003, respectively). 10.1371/journal.pmed.1000407.t003 Table 3 Effect of area of residence and age on the protective efficacy of IPTc against clinical episodes of malaria. Outcomes According to Area of Residence and Age Category IPTc Placebo Unadjusted RR (95% CI) p-Value Adjusted RR (95% CI) PE (95% CI) p-Value Episodes (Years at Risk) Incidence Ratea Episodes (Years at Risk) Incidence Ratea Clinical malaria defined as fever or history of fever in the last 24 h and asexual parasitaemia ≥5,000/µl Locality Djoliba 11 (73.77) 0.15 (0.08–0.27) 74 (73.49) 1.00 (0.80–1.26) 0.13 (0.10–0.18) <0.001 0.15 (0.08–0.28) 85 (72–92) <0.001 Siby 70 (93.32) 0.75 (0.59–0.94) 308 (90.57) 3.40 (3.04–3.80) 0.13 (0.07–0.24) <0.001 0.22 (0.17–0.29) 78 (71–83) <0.001 Ouelessebougou 45 (202.55) 0.22 (0.17–0.30) 292 (190.20) 1.53 (1.37–1.72) 0.21 (0.16–0.26) <0.001 0.14 (0.10–0.20) 86 (80–90) <0.001 Age (mo) 3–11 6 (68.64) 0.09 (0.04–0.19) 52 (66.60) 0.78 (0.59–1.02) 0.11 (0.05–0.26) <0.001 0.13 (0.05–0.29) 87 (71–95) <0.001 12–23 12 (83.00) 0.14 (0.08–0.25) 134 (72.40) 1.85 (1.56–2.19) 0.08 (0.04–0.14) <0.001 0.07 (0.04–0.13) 93 (87–96) <0.001 24–35 38 (76.63) 0.50 (0.36–0.68) 173 (77.96) 2.22 (1.91–2.58) 0.22 (0.15–0.33) <0.001 0.23 (0.15–0.34) 77 (66–85) <0.001 36–47 36 (72.51) 0.50 (0.36–0.69) 153 (67.84) 2.26 (1.92–2.64) 0.22 (0.15–0.31) <0.001 0.21 (0.15–0.31) 79 (69–85) <0.001 48–59 34 (66.91) 0.51 (0.36–0.71) 156 (66.43) 2.34 (2.00–2.74) 0.21 (0.14–0.32) <0.001 0.22 (0.15–0.32) 78 (68–85) <0.001 Clinical malaria defined fever or history of fever in the last 24 h and asexual parasitaemia regardless of the density Locality Djoliba 12 (72.05) 0.17 (0.09–0.29) 90 (73.40) 1.22 (1.0–1.50) 0.13 (0.10–0.18) <0.001 0.14 (0.10–0.18) 86 (82 –90) <0.001 Siby 83 (90.86) 0.91 (0.74–1.13) 372 (86.41) 4.30 (3.88–4.76) 0.13 (0.07–0.24) <0.001 0.14 (0.07–0.26) 86 (74 –93) <0.001 Ouelessebougou 54 (199.25) 0.27 (0.21–0.35) 370 (185.82) 1.99 (1.80–2.20) 0.21 (0.16–0.26) <0.001 0.21 (0.16–0.26) 79 (74–84) <0.001 Age (mo) 3–11 10 (68.15) 0.15 (0.08–0.27) 72 (65.96) 1.09 (0.86–1.38) 0.13 (0.07–0.26) <0.001 0.14 (0.07–0.27) 86 (73–93) <0.001 12–23 15 (81.60) 0.18 (0.11–0.30) 153 (71.07) 2.15 (1.83–2.52) 0.08 (0.05–0.14) <0.001 0.08 (0.05–0.14) 92 (86–95) <0.001 24–35 47 (75.18) 0.62 (0.47–0.83) 206 (76.20) 2.70 (2.36–3.10) 0.23 (0.17–0.31) <0.001 0.23 (0.16–0.33) 77 (67–84) <0.001 36–47 39 (70.23) 0.55 (0.40–0.76) 200 (65.44) 3.06 (2.66–3.51) 0.18 (0.13–0.25) <0.001 0.18 (0.13–0.25) 82 (75–87) <0.001 48–59 38 (65.04) 0.58 (0.42–0.80) 194 (63.98) 3.03 (2.63–3.49) 0.19 (0.13–0.27) <0.001 0.19 (0.13–0.28) 81 (72–87) <0.001 a Incidence rate expressed as number of episodes/child/year. Note that this is based on the 3-mo surveillance period and does not correspond to an annual rate. The percentage of children with malaria infection detected at weekly active surveillance visits was 13.2% (74/563) in the control group compared to 1.9% (11/575) in the intervention group, giving a protective efficacy of 85%, (95% CI 73%–92%) (p<0.001). At the end of the transmission season, 13.2% (188/1,423) of children in the control group were parasitaemic compared to 7.2% (101/1,405) in the intervention group, giving a protective efficacy of 46% (95% CI 31%–68%) (p<0.001). The Impact of IPTc on Anaemia At the end of the malaria transmission season, the proportion of the children with anaemia (Hb <11 g/dl), was significantly higher in the control group compared to the intervention group (61.1% [875/1,433] versus 53.9% [766/1,422]) (PE = 12%; 95% CI 3%–20%) (p<0.001). The relative difference was larger for moderate anaemia (Hb <8 g/dl) with a prevalence of 3.5% (50/1,433) versus 1.9% (27/1,422) in the control and intervention groups, respectively (PE = 47%; 95% CI 15%–67%) (p = 0.007). No cases of severe anaemia (Hb <5 g/dl) were observed in either treatment group at the time of the postintervention survey. However, during the follow-up period, a total of eight cases of severe anaemia occurred, two in the intervention arm and six in the control arm. The two participants in the intervention group who developed severe anaemia had not received a complete course of IPT at the time that they developed their severe anaemia. The Impact of IPTc on Nutritional Indicators The impact of IPTc on nutritional indicators is presented in Table 4. The proportions of children with wasting, stunting, and being underweight at the end of the malaria transmission season were similar between the control and intervention arms However, weight gain during the intervention period was 97 g (95% CI 37 g–157 g) more among children in the intervention arm compared to that recorded among children in the control arm. Changes in height were similar between the two arms with an increase of 2.3 cm (95% CI 2.2 cm–2.5 cm) in children in the intervention arm compared to an increase of 2.4 cm (95% CI 2.2 cm–2.5 cm) in children in the control arm. 10.1371/journal.pmed.1000407.t004 Table 4 Effect of IPTc on nutritional indicators in children at the end of the malaria transmission season. Nutritional Indicators Placebo IPTc Adjusted Analysis Percent n Percent n OR (95% CI)a p-Value Wasting 5.6 1,364 4.3 1,360 0.75 (0.53–1.07) 0.12 Stunting 25.2 1,365 24.6 1,361 0.96 (0.81–1.15) 0.69 Underweight 12.8 1,365 10.9 1,361 0.84 (0.66–1.06) 0.15 a Adjusted for age, sex, and locality. The Impact of IPTc on Molecular Markers of Antimalarial Drug Resistance The frequencies of molecular markers associated with resistance to SP and AQ in the two groups at baseline and postintervention are presented in Table 5. The frequencies of individual and multiple dhfr and dhps mutations in the placebo group were similar in pre- and postintervention periods. The frequencies of all individual dhfr and dhps and of the triple dhfr (51, 59, 108) and quadruple dhfr (51, 59, 108) + dhps 437 mutations were higher in the intervention than in the control group at the end of the surveillance period and, for the dhfr 59, dhps 437, triple and quadruple mutations, differences between groups were statistically significant. Frequencies of the pfcrt 76 and pfmdr1 86 did not change significantly over time and were similar postintervention in the intervention and control groups. 10.1371/journal.pmed.1000407.t005 Table 5 Frequencies of molecular markers of resistance to SP and AQ at baseline and at the end of the intervention period in intervention and control arms. Molecular Markers Baseline Postintervention Baseline Versus Overall Postintervention p-Value IPTc Placebo n Percent Mutant n Percent Mutant n Percent Mutant p-Value DHFR 51 48 62.5 78 75.6 148 66.2 0.144 0.35 DHFR 59 48 60.4 78 76.9 148 59.5 0.009 0.50 DHFR 108 41 78.0 76 78.9 139 71.2 0.217 0.58 DHPS 437 47 38.3 83 67.5 165 43.6 <0.001 0.09 DHPS 540 45 0 82 7.3 165 3.6 0.205 0.82 Triple DHFR mutations 41 58.5 76 69.7 139 54.0 0.024 0.90 Quadruple mutants (triple DHFR + DHPS 437) 41 22.0 75 53.3 139 28.1 < 0.001 0.07 Pfcrt-76 46 80.4 79 84.8 156 75.0 0.085 0.25 Pfmdr1-86 46 45.6 76 36.8 156 34.6 0.739. 0.19 n = number of participants with parasitaemia at blood smear tested. The Impact of IPTc on Hospital Admissions and Death Hospital admissions and deaths that occurred during the study period are listed in Table 6. 19 hospital admissions of at least 24 h were recorded; nine of these were recorded in children in the control arm and ten in children in the intervention arm. The incidence rates of hospital admissions per child/year were 0.0225 episodes in the control group versus 0.0251 in the intervention arm (p = 0.81). There were five deaths, two in the control arm and three in the intervention arm. Two of the five deaths were due to malaria (one in each group). Both occurred during hospitalisation while the remaining three deaths occurred at home. On the basis of the results of a verbal autopsy, these deaths were thought to be due to poisoning by traditional medicines, meningitis and anaemia, and secondary bleeding following a circumcision, respectively. 10.1371/journal.pmed.1000407.t006 Table 6 Hospital admissions and deaths by treatment arms. Numbering Treatment Arm Date Cause Outcome Hospital admission 1 Placebo 11/8/2008 Severe malaria Recovered 2 Placebo 10/25/2008 Severe anaemia Recovered 3 Placebo 9/27/2008 Severe malaria Recovered 4 Placebo 11/9/2008 Severe malaria Recovered 5 Placebo 12/7/2008 Severe malaria Recovered 6 Placebo 9/18/2008 Severe malaria Recovered 7 Placebo 9/16/2008 Severe malaria Death 8 Placebo 9/18/2008 Severe malaria Recovered 9 Placebo 11/16/2008 Severe malaria Recovered 10 IPTc 11/23/2008 Gastro-enteritis Recovered 11 IPTc 11/5/2008 Severe malaria Death 12 IPTc 12/2/2008 Respiratory infection Recovered 13 IPTc 10/11/2008 Gastro-enteritis Recovered 14 IPTc 8/13/2008 Severe anaemia Recovered 15 IPTc 9/21/2008 Asthma Recovered 16 IPTc 12/3/2008 Severe malaria Recovered 17 IPTc 11/11/2008 Respiratory infection Recovered 18 IPTc 11/4/2008 Febrile convulsions Recovered 19 IPTc 9/19/2008 Respiratory infection Recovered Deaths out of hospital a 1 Placebo 11/08/08 Intoxication to traditional medicines — 2 IPTc 09/11/08 Meningitis — 3 IPTc 09/17/08 Anaemia secondary to circumcision — a Does not include death occurred following hospital admissions listed above. Safety and Tolerability There was no serious adverse event related to the study drugs. The frequencies of adverse events following the administration of IPTc with SP+AQ or placebo, using active surveillance are summarized in Table 7. The frequencies of adverse events were similar between the control and intervention arms. However, there was a tendency toward a higher frequency of vomiting and of loss of appetite in the intervention arm compared to the control arm (4.0% versus 1.9%, p = 0.06 for vomiting and 1.9% versus 0.8%, p = 0.08 for loss of appetite). Proportions of children with skin rash and itching on at least at one occasion were similar between the two arms. Four participants in the intervention arm were withdrawn from the study because of reactions to study drug versus none in the control arm. Two of these children had a documented skin rash at physical examination (one after the first dose of IPT and the other after the second dose of IPT) and these were assessed as being related to study drugs. Both were moderate in intensity, did not involve bullous eruptions, and resolved within 2 d. The parent of the third participant reported itching. Physical examination was normal but the child was withdrawn from the study on precautionary grounds. The fourth participant had an acute respiratory infection at the time of administration of the first dose of IPT. No adverse event was recorded at the time of routine surveillance but the parents requested withdrawal of their child from the study at the time of the second round of IPTc. 10.1371/journal.pmed.1000407.t007 Table 7 Proportions of children with adverse events on at least one occasion during three rounds of IPTc treatment using the active surveillance. Adverse Events IPTc Placebo ORs (95% CI) p-Value Percent (n/N) Percent (n/N) Fever 10.1 (69/686) 9.9 (66/669) 1.02 (0.72–1.46) 0.91 Vomiting 4.0 (19/475) 1.9 (9/473) 2.1 (0.96–4.80) 0.06 Drowsiness 0.1 (1/686) 0 (0/669) — — Itching 1.0 (7/686) 0.6 (4/667) 1.7 (0.50–5.86) 0.39 Diarrhoea 6.7 (46/686) 4.6 (31/669) 1.48 (0.92–2.36) 0.10 Skin rash 0.3 (2/686) 0.8 (5/668) 0.39 (0.7–2.0) 0.26 Coughing 8.2 (56/686) 6.0 (40/631) 1.40 (0.92–2.13) 0.12 Loss of appetite 1.9 (13/686) 0.8 (5/668) 2.56 (0.90–7.22) 0.08 Jaundice 0 (0/686) 0.1 (1/667) — — Discussion This study has shown that three doses of IPTc with SP+Q given at monthly intervals during the peak transmission season reduced the incidence of uncomplicated and severe malaria by 80% in children 3–59 mo of age who slept under an ITN in three localities in Mali despite the difference in ITN use at baseline. This level of protective efficacy is similar to that reported in a previous trial conducted in an area of Senegal with a coverage of ITNs of less than 1% [7], suggesting that the relative efficacy of IPTc is not reduced by the use of an ITN at the time of the intervention. Two studies have shown that in pregnant women, IPT adds little benefit to the protection afforded by an ITN, at least in multigravidae [18],[19]. This finding is not the case for IPTc in children, as the strategy remained highly efficacious even when deployed in a community with a high usage of ITNs. Despite the large difference in background incidence of malaria in the three sites, suggesting high variability in transmission intensity, the protective efficacy of IPTc against clinical malaria was high and similar between the three sites. This suggests that similar efficacies of IPTc against clinical malaria can be expected in areas with different transmission intensities and baseline ITN coverage. Surprisingly, Siby and Ouelessebougou, which had a low EIR (less than ten infective bites per person/season), had a higher malaria attack rate than Djoliba, which had a higher EIR (37 infective bites per person/season). High malaria infection and attack rates have been reported previously in the context of a low EIR (3.5 infective bites per person/season) in Mali [20], and similar malaria incidence rates were found in children aged 0–5 y in two areas despite a more than 10-fold difference in EIR [21]. However, these apparently anomalous results could have also been due to imprecision in the determination of the EIR, which can vary markedly with time and space or to a difference in the efficiency of transmission. Early detection and treatment of malaria cases is known to reduce hospital admission and deaths due to malaria [22],[23]. Early detection and prompt treatment was available in our carefully controlled study and the protective effect of IPTc on severe malaria or death might be more marked than we observed if IPTc was deployed in a community that did not have such ready access to health care. Parasite prevalence, as assessed by weekly surveys during the intervention period was reduced by 85% in children who received IPTc, but this difference dropped to 46% at the end of the intervention period suggesting that the prophylactic effect of the last dose of SP+AQ had begun to decline 6 wk after administration, as has been found in studies of IPTi [24]. We observed a 47% reduction in the proportion of children with moderately severe anaemia (Hb <8 g/dl) as a result of administration of IPTc. This impact on anaemia is consistent with the reduction of 45% in incidence of anaemia observed when AS+AQ was given at monthly intervals over 6 mo in Ghana, although in the Ghanaian study there was no difference in the proportion of children with anaemia at the end of the 6-mo intervention [9]. We did not detect any difference between the intervention and control arms in wasting, stunting, or under weight. This finding is consistent with a previous study in Senegal [25], which did not find evidence of an impact of IPTc on wasting, stunting, or being under weight at the end of the transmission season but only on triceps and subscapular skinfold, indicators that were not assessed in our study. However, in line with the Senegalese study, we found an increase in weight gain in the IPTc arm compared to the control arm during the course of the intervention period. More marked effects on nutritional measurements were found during a parallel study conducted in Burkina Faso [26], perhaps because the force of infection was higher in the Burkina Faso than in the Mali study areas and malaria, thus, a more important contributor to impairment of weight gain in the Burkina Faso than in the Mali study areas. SP+AQ was chosen as the drug combination for use in the trial on the basis of the results of previous studies that had shown this to be an effective combination for IPTc. This drug combination was generally well tolerated and no serious adverse event attributable to the study drugs was reported. The proportions of children with mild-to-moderate adverse events using active surveillance were not significantly different between the two arms, although there was a trend towards a higher frequency of vomiting and loss of appetite in the intervention group. In the parallel study in Burkina using the same drugs, a higher frequency of vomiting was found in the intervention arm [26]. However, even in the placebo group the frequency of vomiting was higher than in this study, suggesting a difference in the way in which minor side effects were solicited in the two study areas. Cisse et al. [7] reported a modest increase in vomiting in children who took SP+AS compared to those who took placebo in Senegal, while Kweku et al. [9] found no difference in incidence of these adverse events between IPTc intervention and control arms when using SP or AQ. Four withdrawals in the intervention arm were reported to be due to reactions to study drugs. In two cases, the presence of a skin rash was confirmed, another child had itching, and the final withdrawal followed the occurrence of an acute respiratory tract infection at the time of administration of the first round of IPTc. It is possible that this event was considered by the parents as a reaction to the study drugs. The safety of SP and AQ has been a concern in relation to their use for IPTc [27]–[30]. However, there is a growing body of evidence from studies in the last few years [4],[6],[7],[9],[26],[30] that these drugs are safe when used for IPT in pregnant women, infants, or children, and no safety concerns have arisen following the use of SP+AQ for IPTc on a large scale in Senegal. The efficacy of IPTc against clinical malaria has now been demonstrated in a number of studies, including the current trial and a parallel one conducted in Burkina Faso [26]. Is the evidence now strong enough to support the introduction of IPTc into countries with seasonal malaria transmission? Evidence from studies of IPT in infants [31],[32] suggests that prophylaxis is the key protective mechanism of IPT and that long-acting drugs are needed for effective IPTc. Currently, the SP+AQ combination meets this requirement in West Africa where both of these drugs are still reasonably effective as has been shown to be the case in the study area (Text S6). Studies conducted in Senegal and in Ghana [8],[30] have compared different drug combination and regimens and shown that currently SP+AQ at monthly intervals is the best combination. However, the continuing efficacy of SP cannot be guaranteed and alternative regimens for IPTc will be required in the future, which might include the long-acting drug piperaquine. Unlike the case of IPT in pregnant women and infants, IPT in children has no established delivery system, raising concerns as to whether it could be implemented as a control measure. However, studies conducted in Ghana and The Gambia have shown that high coverage with IPTc can be obtained using community health workers [30],[33], and this appears the most promising way of delivering this intervention. Another concern over the widespread deployment of IPTc is that this will enhance the spread of drug resistance. Therefore, we studied the presence of molecular markers associated with resistance before and after the intervention in children in the intervention or control group. The dhfr 59 and dhps 437 mutations associated with pyrimethamine and sulphadoxine resistance, respectively, were found significantly more frequently at the end of the malaria transmission season in parasites obtained from children in the intervention group than in those obtained from children in the control group, and this led to higher frequencies of the triple dhfr mutants and the quadruple mutant (triple dhfr + dhps 437) associated with significant resistance to SP in children who had received IPTc. This increase in the frequency of these mutations is consistent with a previous report in Senegal [7]. As in Senegal, the number of children in the intervention group carrying a resistant parasite was less than in children in the control group because of the substantial reduction in the overall prevalence of parasitaemia. Although IPTc may have contributed to the increase in frequency of some of resistant markers in this and other studies, the true impact on the resistance of SP and AQ remains to be established. Despite a prevalence of quadruple mutants of about 37%, the SP+AQ combination was highly effective in clearing parasitaemia from children resident in the study area with asymptomatic parasitaemia (Text S6). As is the case with any successful malaria intervention, administration of IPTc to children during several, successive malaria transmission seasons could interfere with the development of naturally acquired immunity, raising concerns that there would be an increased period of risk (rebound malaria) during the period immediately after the intervention was stopped if exposure levels remained high. The risk of malaria for children in this trial in the year after the intervention was stopped has been studied and the results are currently being analysed. However, several years of administration would be needed to define the degree to which acquisition of natural immunity would be impaired. It is very unlikely that this would outbalance the substantial gains made during the period when the drug was given. Our study has several strengths. First, the double-blind, randomised controlled design prevented a number of biases in the selection assignment of the participants to the two arms as well as in assessing the outcomes. A second strength is that this is the largest IPTc efficacy trial done so far, providing a more precise estimation of the outcomes measured. Third, the trial was conducted in three localities with different malaria incidence rates, allowing the efficacy of this strategy under different levels of malaria transmission to be assessed. The design would have been stronger if a factorial design had been used to assess the individual and combined impact of IPTc and ITN, but such a trial would be unethical as the efficacy of ITN is already established [3] and use of ITNs is policy in Mali. Other potential limitations of the study include the duration of evaluation, which focused only on about 15 wk of follow-up during the malaria transmission season. However, it is well established that the in the Sahel region of Mali, 85%–90% of clinical malaria cases occur during the period of August to November, and efficacy of this strategy remained high in a previous, smaller study when efficacy was computed over 12 mo period [6],[34]. In summary, IPTc given during the malaria transmission season, provided substantial additional protection against clinical malaria, infection with malaria, and anaemia to that provided by ITNs. IPTc with SP+AQ was safe and well tolerated. As the international community moves towards the target of malaria elimination, new malaria control tools will be needed [11]. IPT in children targeting the transmission season appears to be one of the strongest available tools to achieve this goal. Our findings support the need for an early review of whether IPTc can now be recommended as a component of malaria control in areas with seasonal malaria transmission. Supporting Information Text S1 Study protocol: A trial of the combined impact of IPT and ITNs on morbidity from malaria in African children. (0.19 MB PDF) Click here for additional data file. Text S2 CONSORT checklist. (0.22 MB DOC) Click here for additional data file. Text S3 Entomological investigations. (0.45 MB PDF) Click here for additional data file. Text S4 External quality assurance of malaria microscopic diagnosis. (0.10 MB PDF) Click here for additional data file. Text S5 Protocol amendment. (0.11 MB PDF) Click here for additional data file. Text S6 In vivo efficacy of the SP+AQ combination used for IPTc in the study area. (0.51 MB PDF) Click here for additional data file.
- Record: found
- Abstract: found
- Article: found
Intermittent Preventive Treatment of Malaria Provides Substantial Protection against Malaria in Children Already Protected by an Insecticide-Treated Bednet in Burkina Faso: A Randomised, Double-Blind, Placebo-Controlled Trial
Introduction Significant efforts have been made in recent years to improve malaria control. However, malaria still remains a major public health problem in sub-Saharan Africa, responsible for about 800,000 deaths annually [1], and existing malaria control strategies provide only partial protection. The pressing need for new malaria control tools has led to evaluation of the strategy of intermittent preventive treatment (IPT) of malaria. IPT involves administration of antimalarial drugs at defined time intervals to individuals regardless of whether they are known to be infected with malaria to prevent morbidity and mortality from the infection [2]. IPT was initially recommended for pregnant women involving the administration of at least two doses of sulphadoxine pyrimethamine (SP) during antenatal visits after the first trimester of pregnancy. Recently the strategy was extended to infants (IPTi) with the administration of three doses of an antimalarial drug during the expanded programme of immunization (EPI) visits [3]. An Institute of Medicine report [4] indicated that IPTi is associated with a 30% (95% confidence interval [CI] 20%–39%) reduction in the incidence of clinical malaria. However in areas of seasonal malaria transmission, such as the Sahel and the sub-Sahelian regions of Africa, the main burden of malaria is in children under 5 y of age [5]. The strategy of IPT of malaria in children (IPTc) was designed for regions where malaria transmission is seasonal [6]. IPTc involves the administration of two to three doses of antimalarial drug during the high malaria transmission season. IPTc with artesunate (AS) plus amodiaquine (AQ) given to children aged 3–59 mo old on three occasions at monthly intervals during the malaria transmission season led to an 86% reduction in the incidence of clinical episodes of malaria in Senegal [6]. In an earlier study conducted in Mali, IPTc with sulphadoxine pyrimethamine (SP), administered on two occasions during the malaria transmission season with a 2-mo interval between treatments, reduced the incidence of malaria in children aged 6 mo to 10 y by 65% [7]. Similar findings were reported in Ghana in an area with perennial malaria transmission but a major seasonal peak, where six monthly rounds of administration of AS+AQ led to a 69% decrease in the incidence of malaria in children [8]. IPTc was safe and well tolerated in each of these studies [6–8], and no evidence of a rebound in the incidence of malaria was observed in the year after the intervention was stopped [6]. Insecticide-treated bednets (ITNs) provide at least 50% protection against morbidity from malaria [9] and are currently the cornerstone of malaria control in many countries in sub-Saharan Africa, although coverage with ITNs is still low in some countries. For example, in Burkina Faso it was estimated that between 2003 and 2006, fewer than 20% of households owned an ITN and less than 10% of children aged below 5 y of age slept under an ITN [1]. However, strenuous efforts are being made to increase coverage in endemic areas and in Burkina Faso, the National Malaria Control Programme (NMCP) has started procedures to purchase about 6 million long-lasting insecticide-treated bednets (LLINs). The successful trials of IPTc described above were conducted in areas with relatively low coverage of ITNs. Thus, it is not known whether IPTc will be as effective in children who sleep under an ITN as has been found in communities where ITN usage is low. To determine this difference, we have conducted a randomised, placebo-controlled trial of IPTc with SP + AQ in children who slept under an ITN in an area of seasonal malaria transmission in Burkina Faso. A parallel study has been conducted in Mali employing a very similar protocol [10], and it was planned to involve a third site from Ghana. However, this site could not participate because of delays in getting regulatory approval for the use of SP + AQ for IPTc from the Ghana Food and Drug Board. SP + AQ combination was chosen because this drug combination is cheap and remains highly efficacious for malaria treatment in Burkina Faso [11,12]. SP was considered as a second-line drug for the treatment of malaria before the introduction of artermisinin-based combination therapy, but the drug was seldom used, as was AQ. A previous study in Senegal showed that the SP + AQ combination was very effective for reducing the incidence of clinical malaria when used for IPT in children [13]. However, the main concern with this drug combination is the increased risk of vomiting associated with AQ. Materials and Methods The protocol of the trial (Text S1), protocol amendment (Text S2), and CONSORT checklist (Text S3) are available as supporting information. Study Design An individually randomised, double-blind, placebo-controlled trial was carried out during the 2008 malaria transmission season to evaluate the efficacy of IPTc in children who slept under an LLIN. All children enrolled in the trial (aged 3–59 mo) were given an LLIN at the start of the study and their family instructed in the use of the net. Children were then randomised to receive three courses of IPTc with SP plus AQ or placebos given at monthly intervals during the peak malaria transmission season. The combination of SP plus AQ was chosen for the trial following a pilot study conducted in the study villages in 2006, which showed that this combination was effective at treating uncomplicated falciparum malaria (Text S4). The incidence of malaria was monitored throughout the malaria season and a cross-sectional survey was performed at its end. Ethical Approval Ethical clearance for the trial was obtained from the Health Ethics Committee of Burkina Faso and from the London School of Hygiene and Tropical Medicine ethics committee. The study's objectives and methods were explained to each study community prior to the start of the trial and written informed consent was obtained from care givers of children before enrolment in the study. The trial was monitored by an independent Data Safety and Monitoring Board (DSMB). Study Area The study was carried out in four villages (Laye, Niou, Sao, and Toeghin) in Boussé health district, Kourweogo province, approximately 40 km northwest of Ouagadougou. The climate in the study area is characteristic of the Sudanese savannah with a dry season from November to June and a rainy season from July to October. The main malaria vectors are Anopheles gambiae s.s., An. Arabiensis, and An. funestus [14]. Malaria transmission is high but seasonal. In 2002, the entomological inoculation rate (EIR) was estimated to be 173 infective bites per person per year [15] with a peak in September. Plasmodium falciparum is responsible for more than 95% of malaria infections in the study area. Study Endpoints The primary endpoint of the study was the incidence of clinical malaria defined as the presence of fever (axillary temperature ≥37.5°C) or a history of fever in the past 24 h, the absence of any other obvious cause of fever, and the presence of at least 5,000 asexual parasites of P. falciparum per microlitre. This threshold has previously been shown to be of value in differentiating symptomatic malaria from other causes of fever with coincidental parasitaemia [16]. The secondary endpoints were: (1) the incidence of clinical malaria defined as the presence of fever or history of fever, the absence of any other obvious cause of fever, and the presence of P. falciparum asexual parasites at any density; (2) the incidence of severe malaria defined according to WHO criteria [17]; (3) the prevalence of anaemia at the end of malaria transmission season; (4) the prevalence of parasitaemia at the end of the malaria transmission season; (5) the prevalence of wasting, stunting, and being underweight at the end of malaria transmission season; (6) the incidence of all-cause hospital admissions during the surveillance period. Hospital admission was defined as any event that involved a child staying at a hospital or health centre for at least 24 h for medical care. Anaemia was defined as a haemoglobin (Hb) concentration 18 g/dl). The performance of the Hemocue was checked weekly with samples of known Hb concentration. DNA was extracted from filter papers collected before and after the intervention to look for the presence of genetic markers of resistance to SP and AQ. Nested PCR was performed to detect the presence of mutations at codons 51, 59, and 108 of the dhfr gene and at codons 437 and 540 of the dhps gene as described previously [21]. Restriction fragment length polymorphism was used to determine the presence or absence of mutations at codon 76 of the P. falciparum chloroquine transporter gene (pfcrt-76) and at codon 86 of the P. falciparum multidrug resistance gene one (pfmdr1-86) [22]. Data Handling and Statistical Analysis Double data entry by two independent data entry clerks was performed. Data were entered using Microsoft Access. Consistency checks and audit trail programs were developed and the database was cleaned and validated before analysis. The trial analysis plan was approved by the DSMB. Data analysis was performed using STATA version 10 (www.stata.com). Analyses were performed on the basis of intention to treat (ITT). Episodes of uncomplicated or severe malaria that occurred after the first dose of IPTc and within 42 d of the third course of IPTc administration were included in the analysis. The starting dates of episodes of malaria meeting the primary or secondary definitions above were identified. Child-days at risk were calculated and used as the denominator for the estimation of the incidence of malaria. Any child who experienced an episode of malaria and was treated was not considered at risk for the next 21 d, which were deducted from the child's time at risk. Children who migrated, died, were lost to follow-up, or were withdrawn from the study contributed to the denominator up to the date of the event or up to the date when they were seen for the last time by the research team. The crude incidence rate ratio (IRR) for the effect of IPTc on the incidence of malaria was estimated using Cox regression models with robust standard errors to account for repeated episodes within the same child. In secondary analyses, age, sex, and village were included in the regression model as covariates to obtain adjusted IRRs. Protective efficacy (PE) was derived from the IRR as follows: PE = (1 − IRR) ×100. Kaplan-Meier survival curves were plotted to compare the times to the first episode of malaria between children who received IPTc and those allocated to the placebo arm. Incidence rates of all-cause hospital admissions were estimated as the number of all hospital admissions divided by the number of child-days at risk computed as described above. Malaria parasitaemia was defined as the presence of malaria parasites irrespective of the developmental stage or species. The presence or absence of malaria infection was coded as a binary variable. Proportions of children with malaria parasites during weekly visits and at the cross-sectional survey carried out 6 wk after the last dose of IPTc were estimated for placebo and intervention arms. The presence of anaemia or moderately severe anaemia 6 wk after the last dose of IPTc was also coded as a binary variable. The risk ratios (RRs) for malaria infection, anaemia, and moderately severe anaemia among IPTc children compared with controls were estimated using generalized linear models. Age, sex, and village were included as covariates to obtain adjusted RRs. The PE of IPTc was calculated as follows: PE = (1 − RR) ×100. Anthropometric data collected 6 wk after the last course of IPTc were transferred from STATA version 10 to WHO's anthropometric software [18] to obtain z-scores for weight-for-age (WAZ, underweight), height-for-age (HAZ, stunting), and weight-for-height (WHZ, wasting). Wasting, stunting, and being underweight were defined according to the WHO child growth standards [18] and coded as binary variables based on z-scores <−2. Proportions of wasted, stunted, and underweight children were calculated and compared between intervention and control arms by fitting a generalized linear model to estimate crude and adjusted RRs. Results Baseline Characteristics 3,052 children were screened and 3,014 were enrolled in the trial (Figure 1); 1,509 were allocated to the intervention group (SP + AQ) and 1,505 to the control group. The mean age of study children at enrolment was 30.4 mo. The age and sex distributions of children were similar in the intervention and control groups (Table 1), as was their weight distribution. No important differences in the proportions of wasted, stunted, or underweight children at enrolment were observed between the groups. The prevalence of fever at enrolment was also comparable between groups and the proportions of children who used an ITN before the intervention were also similar (less than 0.5%). Similar proportions of children were allocated to the control and intervention groups in each of the four study villages. 10.1371/journal.pmed.1000408.g001 Figure 1 Trial profile. 10.1371/journal.pmed.1000408.t001 Table 1 Characteristics of the study children at the time of administration of the first dose of treatment and their use of ITNs before intervention. Characteristics Intervention Control Percent (n) Percent (n) n = 1,509 n = 1,505 Age (mo) 3–11 15.3 (231) 16.9 (255) 12–23 23.1 (348) 21.6 (325) 24–35 20.3 (307) 21.0 (316) 36–47 22.6 (341) 20.1 (302) 48–59 18.7 (262) 20.4 (307) Mean age 30.48 (SD = 15.76) 30.45 (SD = 16.32) Sex Male 52.3 (789) 50.5 (760) Female 47.7 (720) 49.5 (745) Weight (kg) 5–9 39.3 (593) 39.0 (587) 10–18 60.4 (911) 60.8 (915) ≥19 0.3 (5) 0.2 (3) Mean weight 11.05 (SD = 3.01) 11.03 (SD = 3.07) Nutritional factors n = 1,496 n = 1,488 Wasting 11.9 (178) 11.4 (170) Stunting 38.9 (582) 38.3 (570) Underweight 24.6 (368) 25.3 (376) Fever 20.0 (302) 19.9 (299) Use of bednets n = 1,428 n = 1,421 Any net 0.6 (18) 0.8 (11) Treated net 0.2 (3) 0.5 (7) SD, standard deviation. Treatment and Follow-up Similar proportions of children in the intervention and control groups received at least the first dose of the first treatment course (77% versus 78%). However, a higher proportion of children in the intervention than in the control group received at least the first dose of the second and the third courses of treatment (85% versus 78% and 86% versus 78%, respectively). The main reason for not receiving treatment was an acute illness. The numbers of children who received a full course of treatment (three doses) at each round were 1,055 (70%), 1,102 (73%), and 1,104 (73%) in the control group and 1,084 (72%), 1,213 (80%), and 1,272 (84%) in the intervention group for the first, second, and third rounds of treatment, respectively. Use of LLINs was monitored in a subsample of children through weekly home visits. Similar proportions of children in the control and in the intervention group slept under an LLIN (92.7% versus 92.8%). Effect of IPTc on the Incidence of Clinical Malaria 169 (8.2%) children who had a positive RDT test were parasite negative by microscopy. Incidence of malaria defined as fever or history of fever with asexual parasitaemia ≥5,000/µl (primary endpoint) was estimated at 1.92 (95% CI 1.73–2.14) per child during the study period in Toeghin; 2.12 (95% CI 1.92–2.36) in Niou; 1.62 (95% CI 1.44–1.81) in Laye; and 1.65 (95% CI 1.47–1.84) in Sao. Overall the incidence of malaria was 1.3 (95% CI 1.11–1.53) in children aged 3–11 mo; 2.50 (95% CI 2.26–2.76) in children aged 12–23 mo; 2.56 (95% CI 2.31–2.83) in children aged 24–35 mo; 1.36 (95% CI 1.19–1.56) in children aged 36–47 mo; and 1.27 (95% CI 1.10–1.47) in children aged 48–59 mo. The incidence of clinical malaria was highest in children aged 12–35 mo (Table 2). 982 episodes of clinical malaria with asexual parasitaemia of 5,000/µl or more were recorded in children in the control group compared with 332 episodes in children in the intervention group. 523 children had one episode, 201 had two episodes, and 19 had three episodes in the control group. In the intervention group, 246 children had one episode, 43 had two episodes, and no child experienced three episodes. The unadjusted IRR was 0.30 (95% CI 0.26–0.34) (p<0.001), indicating a PE of 70%. Adjustment for age, sex, and village did not alter this estimate. There was strong evidence that the IRR varied with age (p<0.0001) with the effect of IPTc being strongest in the youngest children (3–23 mo). IPTc with SP + AQ was effective in reducing clinical malaria in the four study villages (Table 2). There was weak evidence to suggest that the IRR varied with village (p = 0.06); the smallest protective effect was observed in the village of Sao and in the remaining villages, the protective effect was similar. Very few children were reported as not having used a net the previous night and so whether the effect of IPTc on clinical malaria varied with ITN use could not be examined. 10.1371/journal.pmed.1000408.t002 Table 2 Effect of IPTc on the incidence of malaria, defined as fever or history of fever with 5,000 or more asexual forms of P. falciparum per µl, by age group and locality. Age and Locality Intervention (SP + AQ) Control Unadjusted IRR (95% CI) p-Value Adjusted IRR (95% CI)a PE (1 − RR) (95%CI) p-Value Episodes (Child Years) Incidence Rate (95%CI) Episodes (Child Years) Incidence Rate (95% CI) Age (mo) 3–23 107 (147.4) 0.73 (0.60–0.87) 439 (127.5) 3.44 (3.13–3.78) 0.21 (0.16–0.26) <0.001 0.21 (0.16–0.26) 79 (74–84) <0.001 24–59 225 (231.7) 0.97 (0.85–1.11) 543 (213.7) 2.54 (2.33–2.76) 0.38 (0.32–0.44) <0.001 0.38 (0.32–0.44) 62 (56–68) <0.001 Overall 332 (380.3) 0.87 (0.78–0.97) 982 (341.3) 2.88 (2.70–3.06) 0.30 (0.26–0.34) <0.001 0.30 (0.26–0.34) 70 (66–74) <0.001 Locality Toeghin 84 (96.4) 0.87 (0.70–1.08) 267 (85.6) 3.12 (2.77–3.52) 0.27 (0.21–0.36) <0.001 0.27 (0.21–0.35) 73 (65–79) <0.001 Niou 82 (87.0) 0.94 (0.76–1.17) 268 (77.4) 3.46 (3.07–3.90) 0.27 (0.21–0.35) <0.001 0.26 (0.20–0.33) 74 (67 –80) <0.001 Laye 73 (99.0) 0.74 (0.59–0.93) 231 (88.8) 2.60 (2.29–2.96) 0.28 (0.21–0.37) <0.001 0.27 (0.21–0.36) 73(64–79) <0.001 Sao 93 (97.8) 0.95 (0.78–1.16) 216 (89.5) 2.41 (2.11–2.75) 0.39 (0.30–0.50) <0.001 0.39 (0.30–0.50) 61(50 –70) <0.001 a IRRs were adjusted for age sex and village using random effect Cox Regression model. PE was obtained using as follows. Note the incidence rates relate only to the three mo surveillance period are not an annual rate. Administration of IPTc delayed the time until children experienced their first clinical attack of malaria defined as the presence of fever or a history of fever together with P. falciparum asexual parasitaemia at a density of 5,000 parasites/µl or more (p<0.0001) (Figure 2). 10.1371/journal.pmed.1000408.g002 Figure 2 Time to first episode of clinical malaria defined as fever (≥37.5°C) or history of fever in the last 24 h and parasitaemia ≥5,000/µl in the intervention and control arms. Kaplan-Meier survival estimates with pointwise 95% confidence bands. Analysis using the secondary endpoint definition of clinical malaria (fever or history of fever and the presence of asexual parasitaemia of any density) showed a similar reduction (IRR = 0.29; 95% CI 0.26–0.32) (p<0.001) to the reduction observed when clinical malaria was defined as fever or history of fever with the presence of at least 5,000 asexual parasites/µl. Severe malaria was observed in 13 children in the control group and in four children in the intervention group (PE = 69%; 95% CI 6%–90%) (p = 0.04) (Table 3). The incidence of all-cause hospital admissions (20 cases in the intervention group compared with 37 in the control group) was 46% lower in the IPTc arm (95% CI 7%–69%) (p = 0.03) (Table 3). 10.1371/journal.pmed.1000408.t003 Table 3 Effects of IPTc on clinical malaria, defined as fever or history of fever with asexual P. falciparum parasitaemia of any density, severe malaria, and all-cause hospital admission. Secondary Endpoint Intervention (SP + AQ) Control Unadjusted IRR (95% CI) p-Value Adjusted IRR (95% CI)a PE (95% CI) p-Value n episodes Days at Risk Incidence Rate (95% CI) n episodes Days at Risk Incidence Rate (95% CI) Clinical malaria 416 373.6 1.11 (1.00–1.22) 1,232 325.5 3.78 (3.58–4.00) 0.29 (0.26–0.32) <0.001 0.29 (0.26–0.32) 71 (68–74) <0.001 Severe malariab 4 406.0 0.01 (0.004–0.026) 13 402.8 0.032 (0.019–0.056) 0.31 (0.10–0.94) 0.038 0.31 (0.10–0.94) 69 (6–90) 0.039 All-cause hospital admission 20 405.4 0.05 (0.03–0.08) 37 401.7 0.09 (0.07–0.13) 0.54 (0.31–0.92) 0.024 0.54 (0.31–0.93) 46 (7–69) 0.026 Deaths 3 406.5 7.4 † (2.4–22.9 7 403.4 17.4† (8.3–36.4)c 0.43 (0.11–1.64) 0.22 0.43 (0.11–1.64) 0.57 (−0.64 to 89) 0.22 a IRRs adjusted for age, sex, and village. Note the incidence rates relate only to the 3-mo surveillance period and are not an annual rate. b Severe malaria was defined according to the WHO definition [17] “presence asexual forms of P. falciparum and any other of danger signs of severe malaria in the absence of any other cause of illness.” c Death rates per 1,000 child year. Effect of IPTc on the Prevalence of Malaria Infection The prevalence of malaria parasitaemia among children visited at home during the intervention period was lower in the intervention than in the control group (18.6% versus 45.8%) (PE = 59%, 95% CI 50%–67%) (p<0.001) (Table 4). A lower prevalence of gametocytes was observed in children who received the intervention than in children who did not (2.4% versus 10.3%; RR = 0.23; 95% CI 0.13–0.40) (p<0.001). 10.1371/journal.pmed.1000408.t004 Table 4 Effect of IPTc on malaria infection, anaemia, and anthropometric indicators at the end of the malaria transmission season. Secondary Endpoint Intervention (SP+AQ) Control Unadjusted Analysis Adjusted Analysis Percent (n) n Percent (n) n RR (95% CI) p-Value RR (95% CI) p-Value Weekly survey of malaria infection Proportion with parasitaemia 18.6 (133) 715 45.8 (317) 692 0.40 (0.34–0.48) <0.001 0.41 (0.33–0.50) <0.001 End of transmission survey Proportion with parasitaemia 11.4 (164) 1,436 41.5 (594) 1,430 0.27 (0.23–0.32) <0.001 0.27 (0.23–0.32) <0.001 Proportion with anaemia (Hb<11 g/dl) 44.2 (638) 1,444 65.5 (944) 1,441 0.67 (0.63–0.72) <0.001 0.67 (0.61–0.75) <0.001 Proportion with moderate anaemia (Hb<8 g/dl) 2.7 (39) 1,444 6.2 (89) 1,441 0.44 (0.30–0.63) <0.001 0.44 (0.30–0.64) <0.001 Proportion with wasting 8.8 (122) 1,391 11.2 (156) 1,389 0.78 (0.62–0.98) 0.031 0.79 (0.65 –1.00) 0.049 Proportion with stunting 37.8 (526) 1,391 39.0 (542) 1,389 0.97 (0.88–1.06) 0.52 0.96 (0.85–1.08) 0.50 Proportion with underweight 20.8 (289) 1,391 24.7 (343) 1,389 0.84 (0.73–0.97) 0.015 0.84 (0.72–0.99) 0.034 RRs adjusted for age, sex and village using a Poisson regression generalized linear model (GLM). Wasting was defined as <−2 z-score weight for age. Stunting was define as <−2 z-score of height for age. Underweight was defined as <−2 z-score of weight for height. The overall prevalence of malaria parasitaemia was 26.4% during the survey conducted at the end of the malaria transmission season (6 wk after the last course of IPTc treatment); 94.2% (714) and 1.2% (9) of these infections were single infections with P. falciparum and P. malariae, respectively, the remainder being mixed infections. The prevalence of malaria infection increased with age (p<0.001) and varied between villages (unpublished data). 6 wk after the last course of IPTc treatment, 11.4% (164) of children in the IPTc arm had malaria parasitaemia compared to 41.5% (594) in the control group (PE = 73%; 95% CI 68%–77%) (p = 0.001). There was strong evidence that the effect of IPTc on malaria infection varied with age (p = 0.003) with the effect being strongest in children aged 3–23 mo. The proportion of children who carried gametocytes postintervention as assessed by microscopy was also lower in children from the intervention arm than in the control arm (2.0% versus 9.3%) indicating a 79% (95% CI 68%–86%) reduction in the risk of gametocytes carriage in children who had received IPTc (p<0.001). Effect of IPTc on Anaemia The prevalence of anaemia (Hb<11 g/dl) overall was high but decreased with age, from 70% (772) in children aged 3–23 mo to 45% (809) in children aged 24–59 mo (p<0.0001) but there was no evidence that it varied between village (p = 0.42). Hb concentration was higher in the intervention group than in the control group at the end of the malaria transmission season (mean Hb 11.01 g/dl [95% CI 10.94–11.08 g/dl] versus a mean of 10.35 g/dl [95% CI 10.27–0.42 g/dl]) (p<0.001). Anaemia (Hb<11 g/dl) was less common in children in the IPTc arm than in the control arm (RR = 0.67; PE = 33%, 95% CI 25–39) (p<0.001). 89 (6.2%) children in the control group had moderately severe anaemia (Hb<8 g/dl) versus 39 (2.7%) children in the intervention group (PE = 56%; 95% CI 36%–70%) (p<0.001). There were too few cases of severe anaemia (Hb<5 g/dl) to perform any meaningful comparative analysis between the intervention and control groups (zero and four cases, respectively). Effect of IPTc on Anthropometric Indicators At the end of the malaria transmission season, IPTc was associated with a 21% reduction (95% CI 0%–35%) (p = 0.05) in the risk of wasting (Table 4). Children in the intervention group were also less likely to be underweight (PE = 16%; 95% CI 1–28) (p = 0.03). The prevalence of stunting was similar in the two groups (p = 0.5). Comparison of the mean weight gain in the two treatment arms using data from children who were present both at the baseline and postintervention survey indicated greater weight gain in the intervention than in the control group (0.72 kg versus 0.57 kg) (p<0.0001). The effect of IPTc on clinical malaria varied in stunted and nonstunted children (p = 0.0001) with PEs of 62% (95% CI 54%–69%) and 78% (95% CI 73%–81%), respectively. No evidence of an effect modification was observed for underweight (p = 0.18) or wasting (p = 0.07). Effect of IPTc on Antimalarial Drug Resistance The prevalence of genetic markers of resistance to SP and AQ was assessed before the intervention in children 3–59 mo old who were not enrolled in the trial but who lived in the same communities. The proportions of children carrying parasites with triple dhfr mutations at codons 51, 59, and 108 associated with resistance to pyrimethamine or the triple dhfr plus single dhps mutation (mutation at codon 437), associated with resistance to sulphadoxine, were estimated at 32.6% and 25%, respectively (Table 5). No child carried an infection with triple dhfr and dhps mutations at codons 540 and 437. There was no evidence of a difference in the proportions of children who carried mutant parasites between the intervention and control groups during the postintervention survey. However, there was an overall increase between baseline and postintervention surveys in the proportions of children with triple dhfr mutations only (p<0.001) and triple dhfr plus a single dhps (codon 437) mutation (p = 0.001). The proportions of children harbouring parasites with the pfcrt-76 or pfmdr1-86 mutations were similar in the control and intervention arms and did not increase postintervention as compared to baseline. 10.1371/journal.pmed.1000408.t005 Table 5 Percentage of children carrying parasites harbouring genetic mutation associated with resistance to SP and AQ at baseline and 6 wk postintervention in intervention and control arms. Genetic Mutation Baseline, n = 132, Percent (n) Postintervention IPTc, n = 114, Percent (n) Placebo, n = 122, Percent (n) Overall postintervention, n = 236, Percent (n) Baseline versus postintervention, p-Value Triple dhfr mutation (51-59-108) 32.6 (43) 50 (57) 53.3 (65) 51.7 (122) <0.001 Triple dhfr single dhps (437) 25.0 (33) 40.4 (46) 44.3 (54) 42.4 (100) 0.001 Triple dhfr/double dhps (437–540) 0 (0) 0.9 (1) 0.0 (0) 0.4 (1) 0.45 Pfcrt-76 63.9 (83) 60.5 (69) 61.5 (75) 61.0 (144) 0.72 Pfcrt-76/pfmdr1-86 20.5 (27) 19.3 (22) 19.7 (24) 19.5 (46) 0.82 A small in vivo study conducted in asymptomatic children resident in the study area in the year after the intervention study confirmed the efficacy of the SP + AQ combination in clearing asymptomatic P. falciparum infections (Text S4). Adverse Events Ten deaths were observed during the intervention period, seven in children in the control group and three in children in the intervention group. Verbal autopsies suggested that four deaths (three deaths in the control group) were associated with malaria, three with acute diarrhoea and malnutrition, two with diarrhoea only, and that one death was due to pneumonia. No drug-related serious adverse events were observed during the follow-up period. The risk of itching, skin rash, diarrhoea, drowsiness, or loss of appetite did not differ between children who received SP plus AQ and those who received placebo (Table 6). The proportion of children who vomited at least once during the three courses of treatment was higher in children who received SP + AQ, 368/1,339 (27.6%), than in children who received placebos of SP plus AQ, 122/1,257 (9.7%). Amongst children who received SP plus AQ, those aged 3–11 mo were at highest risk of vomiting and the risk of vomiting decreased with age (Table 7). The proportions of children who vomited during the first, second, and third round of treatment were 17% (197), 15% (184), and 13% (160), respectively, in children who received SP + AQ. The RR for vomiting remained similar between the first (RR = 2.9), the second (RR = 3.0), and the third (RR = 3.3) course of treatment in the intervention group when compared to the control group. 10.1371/journal.pmed.1000408.t006 Table 6 Percentage of children with adverse events on at least one occasion during the three rounds of treatment in intervention and control arms. Adverse Event IPTc Placebo RRs (95% CI) p-Value Percent (n/n) Percent (n/n) Fever 13.4 (179/1,339) 14.9 (187/1,257) 0.90 (0.74–1.11) 0.33 Vomiting 27.6 (368/1,339) 9.7 (122/1,257) 2.83 (2.31–3.47) <0.001 Drowsiness 0.1 (1/1,339) 0.1 (1/1,257) 0.94 (0.06–15.0) 0.96 Itching 2.2 (30/1,339) 2.4 (30/1,257) 0.94 0.56–1.56) 0.80 Diarrhoea 7.2 (96/1,339) 7.0 (88/1,257) 1.02 (0.77–1.37) 0.86 Skin rash 1.5 (20/1,339) 1.6 (20/1,257) 0.94 (0.50–1.74) 0.84 Coughing 5.2 (70/1,339) 5.9 (74/1,257) 0.92 (0.66–1.27) 0.60 Loss of appetite 1.4 (5/1,339) 1.3 (3/1,257) 1.51 (0.36–6.34) 0.57 10.1371/journal.pmed.1000408.t007 Table 7 Risk of vomiting by age in children who received SP + AQ. Age group (mo) Percent (n/m) RR (95% CI) p-Value 3–11 54.5 (114/209) 9.9 (5.6–17.2) <0.001 12–24 38.7 (117/302) 7.0 (4.0–12.2) <0.001 24–35 28.6 (75/262) 5.2 (2.92–9.2) <0.001 36–47 14.5 (48/309) 2.8 (1.5–5.1) 0.001 48–59 5.5 (14/253) 1 — Discussion To the best of our knowledge, this is the first study to test if IPTc provides protection against malaria in children who are already protected by an ITN. Our results show conclusively that IPTc does provide substantial protection against clinical malaria episodes, severe malaria, and all-cause hospital admissions in children using an ITN. The primary role of ITNs is to prevent mosquito bites, thus reducing the risk of malaria infection, whereas IPTc clears existing infections and has a prophylactic effect preventing new blood stage infections. The high PE of IPTc in children using ITNs may partly reflect protection from infections acquired because of exposure to mosquitoes at night outside sleeping hours. Coverage of ITNs was low among older children and adults in the study area; the absolute reduction in malaria incidence due to IPTc may have been less if coverage of ITNs in the population as a whole had been higher. A similar additional effect of combining chemoprevention with ITNs was observed in a community randomised study of ITNs and chemoprophylaxis with Maloprim (dapsone-pyrimethamine) given every 2 wk in Sierra Leone [23]. In this study, ITNs and chemoprophylaxis alone were associated with 49% and 42% protection against malaria, respectively, whereas the combined effect of the two interventions was 72% (95% CI 67%–76%). Consistent findings have also been reported from The Gambia, where seasonal chemoprophylaxis with Maloprim protected children from malaria attacks in villages where ITNs were being used [24]. The efficacy of IPTc against clinical attacks of malaria was a little less in our study than that seen in Senegal [6] but very similar to that obtained in Mali [7] and Ghana [8] in populations with a low use of ITNs. Varying efficacy of IPTc between trials may be explained by a number of factors including transmission intensity and nutritional status. Our results showed that IPTc was more beneficial to children who were not stunted. A previous study by Danquah and collaborators [25] had reported that IPT of malaria in infants was less effective in malnourished than in non-malnourished infants. In our study, the impact of IPTc was more marked in children less than 24 mo, who have yet to achieve significant acquired immunity to malaria, than in older children, a similar pattern to that observed in Senegal [6]. A total of 3,756 courses of IPT treatment were administered in the SP + AQ arm, and 650 cases of malaria were averted, thus 5.8 IPT administrations were needed to prevent one episode of malaria. The reduction in the prevalence of anaemia seen in our study is consistent with the results of the study conducted in Hohoe, Ghana [8] where a 45% reduction in anaemia was observed in children who received IPTc with AS plus AQ monthly, and a 30% reduction was seen in children who received bimonthly IPTc with SP or AS plus AQ. However, IPTc with AS plus SP did not result in a detectable reduction in anaemia in Senegal [6]. It is likely that the fraction of anaemia attributable to malaria differs between settings and this fraction is likely to be lower in areas with lower transmission intensity such as Senegal. Modest reductions were observed in the proportion of wasted and underweight children after IPTc administration, but there was no effect on stunting. The lack of an effect on stunting is perhaps not surprising as this is generally held to be a measure of chronic undernutrition, which is less likely to change substantially over the period of follow-up reported in this paper than indicators of acute undernutrition. IPTc increased the mean weight gain during the rainy season as previously reported in a study that also demonstrated an increase in subcutaneous fat reserves in Senegalese children who received IPTc with SP plus AS [26]. The study was not powered to detect an impact on cases of severe malaria, hospital admissions, or deaths. However, encouraging results were found with a reduction in deaths, overall hospital admissions, and severe malaria in children who received IPTc with differences in cases of severe malaria and hospital admissions being statistically significant. Thus, it seems likely that if IPTc was widely deployed in areas with a similar pattern of malaria transmission to that of the study area it would have a significant impact on severe morbidity and mortality from malaria. This supposition is supported by the results of an earlier study conducted in The Gambia [27], which showed that chemoprophylaxis with Maloprim given fortnightly during the rainy season reduced overall mortality in children aged less than 5 y by about 40%. We monitored adverse events over 4 d during each course of treatment and detected no serious adverse events related to the study drugs, which were reasonably well tolerated and safe. However, we observed a higher risk of vomiting in the intervention group, as has been reported in previous IPTc studies that used drug combinations containing AQ [8,13,28]. Unlike the Hohoe study, the frequency of vomiting did not decrease with subsequent courses of IPTc. Our data showed a considerably higher risk of vomiting in younger children. Vomiting induced by AQ was not a significant deterrent to the use of SP plus AQ because compliance with the full course of treatment was higher in children who received IPTc than in those who received placebo. Nevertheless, the acceptability of SP plus AQ for IPTc would be improved if this problem could be overcome. There are at least two possible ways that this might be accomplished. A recent study undertaken in Senegal showed that the frequency of vomiting following the use of AQ containing combinations for IPTc is dose dependent [29]. Thus, one possibility would be to adjust the AQ content of tablets used for IPTc, allowing the optimum spread of dose per kg body weight for each weight group and ensuring that as few as possible children are overdosed. A second approach would be to produce a formulation of AQ that was more palatable than standard tablets. A number of concerns have been raised about the adoption of IPTc as a malaria control strategy. These include the possibility that IPTc will encourage the spread of resistance to the drugs used for IPTc. We observed an increase in the proportion of children harbouring genetic mutations associated with resistance to SP at the end of the malaria transmission season compared with the preintervention period but not in the prevalence of genetic mutations associated with resistance to chloroquine. However, in contrast to the findings of a study conducted in Senegal [6], we found no difference in the prevalence of resistance markers between intervention and control groups. This difference between studies might be due to the higher level of malaria transmission in the study area compared with Senegal, resulting in a higher rate of exchange of parasite strains between the two treatment groups and the rest of the population. An overall increase in the prevalence of drug resistance markers over the course of the malaria transmission season has been seen previously in areas of seasonal malaria transmission in the absence of any chemopreventive strategy [30], probably reflecting the selection pressure of an overall increase in the use of antimalarials for the treatment of febrile illnesses at this time of the year. However, we cannot exclude the possibility that the use of IPTc made some contribution to the overall increase in the prevalence of dhfr resistance markers seen at the end of the transmission season and sustained implementation of IPTc would inevitably increase drug pressure, a risk that would need to be balanced against the marked benefit that can result from this intervention. This risk can be reduced by using a drug combination rather than a single drug for IPTc and by using a different drug combination for IPTc than the one used for first-line treatment of symptomatic malaria. Another concern is that IPTc might interfere with the development of naturally acquired immunity to malaria, leading to an increase in the incidence of malaria in children after they move out of the age range in which IPTc is given. Administration of IPTc for 1 y did not lead to an increase in the incidence of clinical attacks of malaria in the following year in children in Senegal [6] or in older children in Ghana [8]. Nevertheless, we are investigating this possibility in the current study and the surveillance procedure described in this paper was reestablished during the 2009 malaria transmission season to look for any evidence of “rebound” malaria. Even if no increase in risk is found, this does not exclude the possibility that IPTc could significantly impair the development of natural immunity if administered to children for several consecutive years and this would have to be monitored carefully if IPTc is implemented as a malaria control strategy. Another major concern is whether IPTc could be delivered on a large scale. As this was an efficacy trial, medications were given under the control of project staff and this study did not address the issue of the implementability of IPTc. This issue has been addressed in a previous study undertaken in Ghana [31] and a study that has investigated two possible modes of delivery in The Gambia, use of immunisation trekking teams or community volunteers is described in an accompanying paper [32]. A study of the feasibility and safety of large scale implementation of IPTc with SP + AQ is currently underway in Senegal. A notable feature of this study was the high incidence of malaria in children in the control group who slept under an LLIN: 982 of 1,505 children experienced a clinical attack during the 3-mo observation period that corresponded to the peak malaria transmission season. A number of possible reasons for this result have been considered. Home visits indicated that more than 90% of the children had used their LLIN the previous night so that failure to use a bednet is unlikely to have been a major factor. It is possible that some children experienced mosquito bites before they retired to bed but this is less likely to be a problem than in older children or adults. LLINs (PermaNet) were obtained from a well-established manufacturer (Vestergaard). Checks on the deltamethrin content of ITNs obtained at the end of the malaria transmission season showed protective levels of deltamethrin in most samples and this was confirmed by bioassays (Text S7). However, some kdr-mediated resistance to pyrethroids was found (Text S7). Although the entomological inoculation rate (EIR) (34 infective bites per person per year) was significantly less than had been recorded in the same area 6 y previously, perhaps in part due to the use of LLINs, this was sufficient to sustain a high clinical attack rate in young children. LLINs were provided only to children in the study so it is likely that, counting these children, their mothers, and their older siblings, only about a third of the population were protected by an LLIN. To obtain maximum benefit from ITNs in high transmission settings such as this, universal coverage with ITNs is probably required. Conclusion IPT of malaria with SP plus AQ was safe and provided substantial additional protection against severe and uncomplicated malaria to children who slept under an LLIN. There is now strong evidence to support the integration of IPTc into malaria control strategies in areas of seasonal malaria transmission. Further research is needed to identify alternative drugs as the long-term use of SP for IPTc is uncertain. Supporting Information Text S1 Study protocol. (0.19 MB PDF) Click here for additional data file. Text S2 Protocol amendment. (0.11 MB PDF) Click here for additional data file. Text S3 CONSORT checklist. (0.09 MB PDF) Click here for additional data file. Text S4 In vivo efficacy of the SP + AQ combination. Results of a pilot study of in vivo efficacy of SP plus AQ in children with uncomplicated malaria in the study area. (0.51 MB PDF) Click here for additional data file. Text S5 In vivo efficacy of the SP + AQ combination. Results of an in vivo study of the efficacy of SP plus AQ in clearing malaria infection in children with asymptomatic malaria. The study was conducted 1 y after the intervention. (0.44 MB PDF) Click here for additional data file. Text S6 External quality assurance of malaria microscopic diagnosis. Report of external quality control of malaria microscopic diagnosis. (0.10 MB PDF) Click here for additional data file. Text S7 Entomological investigations. Reports on malaria transmission intensity estimation, bioassays, vector resistance to insecticide, and concentration of insecticide on LLINs. (0.46 MB PDF) Click here for additional data file.
Author and article information
Journal
Journal ID (nlm-ta): BMJ Open
Journal ID (iso-abbrev): BMJ Open
Journal ID (hwp): bmjopen
Journal ID (publisher-id): bmjopen
Title:
BMJ Open
Publisher:
BMJ Publishing Group
(BMA House, Tavistock Square, London, WC1H 9JR
)
ISSN
(Electronic):
2044-6055
Publication date Collection: 2020
Publication date
(Electronic):
15
September
2020
Volume: 10
Issue: 9
Electronic Location Identifier: e035433
Affiliations
[1
]departmentDepartment of Disease Control , London School of Hygiene and Tropical Medicine , London, UK
Author notes
Article
Publisher ID: bmjopen-2019-035433DOI: 10.1136/bmjopen-2019-035433
PMC ID: 7493088
SO-VID: f2d5e780-c59c-4a6d-80ba-17acde340318
Copyright © © Author(s) (or their employer(s)) 2020. Re-use permitted under CC BY-NC. No commercial
re-use. See rights and permissions. Published by BMJ.
License:
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
History
Date
received
: 19
November
2019
Date
revision received
: 27
May
2020
Date
accepted
: 31
July
2020
Funding
Funded by: UK Joint Global Health Clinical Trials;
Award ID: MR/P006876/1
Funded by: PATH MVI;
Award ID: 18269
Categories
Subject:
Epidemiology
Subject:
1506
Subject:
1692
Series title:
Protocol
Custom metadata
special-feature unlocked
ScienceOpen disciplines: Medicine