High-throughput “omics” technologies that generate molecular profiles on tumor specimens are increasingly being incorporated into clinical trials, but some of these assays have not been well validated, leading many in the research community to question their fitness for use in patient-care decisions. Recently, the scientific community united in an effort to produce guidelines that would promote quality.
Tumor biomarker tests are key to accomplishing the goal of personalized medicine. However, much of the translational research regarding tumor biomarker tests has “ignored many of the principles of the scientific method,” according to Lisa M. McShane, PhD, of the Biometric Research Branch, Division of Cancer Treatment and Diagnosis, at the National Cancer Institute (NCI).
The recognition of this suboptimal situation led Dr. McShane and a group of other scientists to develop a “criteria checklist” for clinical trial use of these tests. The checklist should be applied when incorporating a tumor biomarker test into any clinical trial that prospectively evaluates its clinical utility, the group suggested. The criteria were recently published simultaneously in the journal Nature1 and, elaborated on in greater detail, in BMC Medicine.2
“It is hoped that this 30-point checklist will guide investigators toward the use of best practices in omics test development.… The ultimate goal is to develop a more efficient, reliable and transparent process to move omics assays from promising research results to clinically useful tests that improve patient care and outcome,” the authors wrote.1
What Are ‘Omics’?
The papers focus on molecular tests derived from high-throughput omics assays, generally used for predicting patient outcomes in clinical trials. The chief omics disciplines are genomics, transcriptomics, proteomics, metabolomics, and epigenomics. A distinguishing characteristic of the omics tests, which are the focus of the checklist, is that computational methods are applied to the data to build mathematical predictor models. This is in contrast to molecular tests based on such variables as genetic mutations, which can be used to screen patients for eligibility for a trial and to select patients for targeted treatments.
To date, relatively few omics tests have entered clinical use but they are increasingly incorporated into clinical trials, even though their performance, reproducibility, and robustness are often not proven, and they may be based on statistical and mathematical approaches that are flawed.
“It is essential to consider all these issues before launching into a clinical study using an omics test in a way that might influence the clinical management of patients,” Dr. McShane argued.
Group Concerns
Checklist coauthor William L. Bigbee, PhD, outgoing Chair of the National Institutes of Health Cancer Biomarkers Study Section and past Leader of the Cancer Biomarkers Shared Facility at the University of Pittsburgh Cancer Institute, provided some background in an interview with The ASCO Post.
“Omics-based tests are very powerful, emerging tools that are revolutionizing medicine, particularly in predicting and treating cancer. However, there are many variables and opportunities for error, including issues pertaining to study design, patient selection, biologic sample integrity, and data analysis and management. The NCI checklist is intended to provide clear expectations and guidelines for the development and implementation of omics-based tests and will hopefully eliminate unintentional errors,” he explained.
He said the scientific community has been witnessing the development and initial implementation of omics-based technologies, anticipating their impact on “personalized medicine. Their mutual concern has been a lack of rigorous clinical validation, and therefore premature use, of some of these tests. These thought leaders, who have broad expertise in the omics field, were brought together for a 2011 NCI Workshop to examine the issues.
The Workshop was followed by an Institute of Medicine (IOM) report that also called for an examination of the omics evaluation process.3 The level of interest in the issue led to further dialogue and the subsequent development of the published checklist to operationalize the principles set forth in the IOM report and the NCI workshop discussions.
“The criteria encompass all the interests, disciplines and technologies that need to weigh in on the development of omics-based predictors in a way that will have clinical relevance and application,” Dr. Bigbee said.
The authors scrutinized published literature pertaining to their specific areas of expertise and asked, “To what degree can we take these predictors from the research environment and have confidence that they could be used in different settings, specifically, clinical trials? Given their complexity, how can we assure that these predictors can go forward in ways that will have a clinical impact for patient selection?” he said.
The authors believe this document will be useful in three contexts: the basic or translational research setting, the clinical trial setting, and within the NCI itself.
“The NCI was very interested in having a reference document like this, together with the IOM report, as part of a more formal framework for the review process in evaluating applications for omics-based research and NCI-sponsored trials,” he said. “We also think it will be useful to journal reviewers and editors as a reference document as they think critically about the evaluation of manuscripts reporting omics-based studies. ”
What’s on the Checklist?
The checklist applies to any clinical trial that involves the investigational use of an omics test that will influence the clinical management of patients in the study. The criteria generally also apply to retrospective analysis of specimens in older studies. The researchers recommend that investigators consult the checklist during the research planning and early development phases of their trials.
The checklist contains specific recommendations in the following areas:
- Specimen collection, processing, storage, and so forth
- Technical issues related to omics-based assays (eg, reagents, specimens, instrumentation, scoring methods)
- Model development, specification, and preliminary performance evaluation
- Clinical trial design and conduct (eg, rigorous statistical design, informatics plan for the data, complete specification of the omics test)
- Ethical, legal, and regulatory issues (safety and privacy of patients, intellectual property issues)■
Disclosure: Dr. McShane and Bigbee reported no potential conflicts of interest.
References
1. McShane LM, Cavenagh MM, Lively TG, et al: Criteria for the use of omics-based predictors in clinical trials. Nature 502:317-320, 2013.
2. McShane LM, Cavenagh MM, Lively TG, et al: Criteria for the use of omics-based predictors in clinical trials: Explanation and elaboration. BMC Medicine 11:220, 2013.
3. Micheel CM, Nass SJ, Omenn GS (eds): Evolution of translational omics: Lessons learned and the path forward. Washington, DC, National Academies Press, 2012.
