Discovery of pre-therapy 2-deoxy-2- 18 F-fluoro-D-glucose positron emission tomography-based radiomics classifiers of survival outcome in non-small-cell lung cancer patients

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

Purpose

The aim of this multi-center study was to discover and validate radiomics classifiers as image-derived biomarkers for risk stratification of non-small-cell lung cancer (NSCLC).

Patients and methods

Pre-therapy PET scans from a total of 358 Stage I–III NSCLC patients scheduled for radiotherapy/chemo-radiotherapy acquired between October 2008 and December 2013 were included in this seven-institution study. A semi-automatic threshold method was used to segment the primary tumors. Radiomics predictive classifiers were derived from a training set of 133 scans using TexLAB v2. Least absolute shrinkage and selection operator (LASSO) regression analysis was used for data dimension reduction and radiomics feature vector (FV) discovery. Multivariable analysis was performed to establish the relationship between FV, stage and overall survival (OS). Performance of the optimal FV was tested in an independent validation set of 204 patients, and a further independent set of 21 (TESTI) patients.

Results

Of 358 patients, 249 died within the follow-up period [median 22 (range 0–85) months]. From each primary tumor, 665 three-dimensional radiomics features from each of seven gray levels were extracted. The most predictive feature vector discovered (FVX) was independent of known prognostic factors, such as stage and tumor volume, and of interest to multi-center studies, invariant to the type of PET/CT manufacturer. Using the median cut-off, FVX predicted a 14-month survival difference in the validation cohort ( N = 204, p = 0.00465; HR = 1.61, 95% CI 1.16–2.24). In the TESTI cohort, a smaller cohort that presented with unusually poor survival of stage I cancers, FVX correctly indicated a lack of survival difference ( N = 21, p = 0.501). In contrast to the radiomics classifier, clinically routine PET variables including SUV _max, SUV _mean and SUV _peak lacked any prognostic information.

Conclusion

PET-based radiomics classifiers derived from routine pre-treatment imaging possess intrinsic prognostic information for risk stratification of NSCLC patients to radiotherapy/chemo-radiotherapy.

Electronic supplementary material

The online version of this article (10.1007/s00259-018-4139-4) contains supplementary material, which is available to authorized users.

Related collections

Most cited references 22

Record: found
Abstract: found
Article: not found

Epidemiology of lung cancer: Diagnosis and management of lung cancer, 3rd ed: American College of Chest Physicians evidence-based clinical practice guidelines.

Anthony J Alberg, Malcolm Brock, Jean Ford … (2013)

Ever since a lung cancer epidemic emerged in the mid-1900 s, the epidemiology of lung cancer has been intensively investigated to characterize its causes and patterns of occurrence. This report summarizes the key findings of this research. A detailed literature search provided the basis for a narrative review, identifying and summarizing key reports on population patterns and factors that affect lung cancer risk. Established environmental risk factors for lung cancer include smoking cigarettes and other tobacco products and exposure to secondhand tobacco smoke, occupational lung carcinogens, radiation, and indoor and outdoor air pollution. Cigarette smoking is the predominant cause of lung cancer and the leading worldwide cause of cancer death. Smoking prevalence in developing nations has increased, starting new lung cancer epidemics in these nations. A positive family history and acquired lung disease are examples of host factors that are clinically useful risk indicators. Risk prediction models based on lung cancer risk factors have been developed, but further refinement is needed to provide clinically useful risk stratification. Promising biomarkers of lung cancer risk and early detection have been identified, but none are ready for broad clinical application. Almost all lung cancer deaths are caused by cigarette smoking, underscoring the need for ongoing efforts at tobacco control throughout the world. Further research is needed into the reasons underlying lung cancer disparities, the causes of lung cancer in never smokers, the potential role of HIV in lung carcinogenesis, and the development of biomarkers.

0 comments Cited 240 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

Decoding global gene expression programs in liver cancer by noninvasive imaging.

Eran Segal, Claude Sirlin, Clara Ooi … (2007)

Paralleling the diversity of genetic and protein activities, pathologic human tissues also exhibit diverse radiographic features. Here we show that dynamic imaging traits in non-invasive computed tomography (CT) systematically correlate with the global gene expression programs of primary human liver cancer. Combinations of twenty-eight imaging traits can reconstruct 78% of the global gene expression profiles, revealing cell proliferation, liver synthetic function, and patient prognosis. Thus, genomic activity of human liver cancers can be decoded by noninvasive imaging, thereby enabling noninvasive, serial and frequent molecular profiling for personalized medicine.

0 comments Cited 204 times – based on 0 reviews      Review now

Bookmark

Record: found
Abstract: found
Article: not found

Hypoxic radiosensitization: adored and ignored.

Jens Overgaard (2007)

Since observations from the beginning of the last century, it has become well established that solid tumors may contain oxygen-deficient hypoxic areas and that cells in such areas may cause tumors to become radioresistant. Identifying hypoxic cells in human tumors has improved by the help of new imaging and physiologic techniques, and a substantial amount of data indicates the presence of hypoxia in many types of human tumors, although with a considerable heterogeneity among individual tumors. Controlled clinical trials during the last 40 years have indicated that this source of radiation resistance can be eliminated or modified by normobaric or hyperbaric oxygen or by the use of nitroimidazoles as hypoxic radiation sensitizers. More recently, attention has been given to hypoxic cytotoxins, a group of drugs that selectively or preferably destroys cells in a hypoxic environment. An updated systematic review identified 10,108 patients in 86 randomized trials designed to modify tumor hypoxia in patients treated with curative attempted primary radiation therapy alone. Overall modification of tumor hypoxia significantly improved the effect of radiotherapy, with an odds ratio of 0.77 (95% CI, 0.71 to 0.86) for the outcome of locoregional control and with an associated significant overall survival benefit (odds ratio = 0.87; 95% CI, 0.80 to 0.95). No significant influence was found on the incidence of distant metastases or on the risk of radiation-related complications. Ample data exist to support a high level of evidence for the benefit of hypoxic modification. However, hypoxic modification still has no impact on general clinical practice.

0 comments Cited 156 times – based on 0 reviews      Review now

Bookmark

All references

Author and article information

Contributors

Eric O. Aboagye: eric.aboagye@imperial.ac.uk

Journal

Journal ID (nlm-ta): Eur J Nucl Med Mol Imaging

Journal ID (iso-abbrev): Eur. J. Nucl. Med. Mol. Imaging

Title: European Journal of Nuclear Medicine and Molecular Imaging

Publisher: Springer Berlin Heidelberg (Berlin/Heidelberg )

ISSN (Print): 1619-7070

ISSN (Electronic): 1619-7089

Publication date (Electronic): 1 September 2018

Publication date PMC-release: 1 September 2018

Publication date (Print): 2019

Volume: 46

Issue: 2

Pages: 455-466

Affiliations

[1 ]ISNI 0000 0001 0705 4923, GRID grid.413629.b, Imperial College London Cancer Imaging Centre, Department of Surgery & Cancer, , Hammersmith Hospital, ; Du Cane Road, London, W12 0NN UK

[2 ]ISNI 0000 0001 0705 4923, GRID grid.413629.b, Imperial College Healthcare NHS Trust, Departments of Clinical Oncology, Radiology and Nuclear Medicine, , Hammersmith Hospital, ; Du Cane Road, London, W12 0HS UK

[3 ]ISNI 0000 0001 2191 5195, GRID grid.413820.c, Charing Cross Hospital, ; Fulham Palace Road, London, W6 8RF UK

[4 ]GRID grid.443984.6, Department of Nuclear Medicine, Level 1, Bexley Wing, , St James’s University Hospital, ; Beckett Street, Leeds, LS9 7TF UK

[5 ]ISNI 0000 0004 1936 8403, GRID grid.9909.9, Leeds Institute of Cancer and Pathology, School of Medicine, , University of Leeds, ; Leeds, UK

[6 ]ISNI 0000 0001 2322 6764, GRID grid.13097.3c, Department of Cancer Imaging, School of Biomedical Engineering and Imaging Sciences, King’s College London, St. Thomas’ Hospital, ; Westminster Bridge Rd, London, SE1 7EH UK

[7 ]ISNI 0000 0004 0417 0461, GRID grid.424926.f, Department of Nuclear Medicine, , The Royal Marsden Hospital, ; Downs Rd, Sutton, London, SM2 5PT UK

[8 ]ISNI 0000 0004 0641 4263, GRID grid.415598.4, Department of Nuclear Medicine, Queen’s Medical Centre, , Nottingham University Hospital, ; Derby Rd, Nottingham, NG7 2UH UK

[9 ]ISNI 0000 0004 0400 1238, GRID grid.416188.2, Department of Clinical Oncology, , Mount Vernon Hospital, ; Rickmansworth Road, Northwood, HA6 2RN UK

Article

Publisher ID: 4139

DOI: 10.1007/s00259-018-4139-4

PMC ID: 6333728

PubMed ID: 30173391

SO-VID: 18bba695-8338-4a60-8a9a-7ba6ed28601b

License:

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

History

Date received : 14 March 2018

Date accepted : 16 August 2018

Funding

Funded by: NIHR Biomedical Research Council

Award ID: MR/N020782/1 and C2536/A16584

Award Recipient : Eric O. Aboagye

Custom metadata

ScienceOpen disciplines: Radiology & Imaging

Keywords: radiomics,nsclc,survival,pet,risk stratification

Data availability:

ScienceOpen disciplines: Radiology & Imaging

Keywords: radiomics, nsclc, survival, pet, risk stratification

Comments

Comment on this article

scite_

Cited by 34

See all cited by

Most referenced authors 691

See all reference authors