Five speakers covered diverse aspects of the development of methodologies and tools that, as the statement of task put it, are related to demonstrating the evidentiary requirements for clinical validity and clinical utility that meet the needs of all stakeholders. Discussions included guidelines for test development, the role of comparative-effectiveness research (CER) in demonstrating clinical utility, statistical techniques, cost-utility analyses, and innovation mechanisms in small companies. Common themes included the need for clearly defined standards for analyses, the importance of context in determining clinical utility, and the importance of access to welldocumented biospecimens.

DEVELOPING OMICS TESTS

Debra Leonard, professor and vice chair in the Department of Pathology and Laboratory Medicine and director of the Clinical Laboratories at Weill Cornell Medical Center, summarized the findings of a recent Institute of Medicine (IOM) report titled Evolution of Translational Omics: Lessons Learned and the Path Forward (2012b). The report was written by an IOM committee in response to the development of gene-expression array tests at Duke University that were said to predict sensitivity to chemotherapeutic agents. Papers written about the tests suggested that they represented a major advance and would better direct cancer therapy. Clinical trials were initiated in 2007, with the tests being used to select which chemotherapeutic agent patients would receive.

A paper by Baggerly and Coombes (2009), however, pointed to numerous errors and inconsistencies in the data and stated that the results could not be reproduced. Following a 2010 letter from more than 30 bioinformaticians and statisticians to the National Cancer Institute (NCI) urging the suspension of the clinical trials and an investigation of the test and computational models by the NCI, the clinical trials were stopped. The NCI then asked the IOM to review the situation and provide guidance for the field.

The IOM committee was charged with recommending an evaluation process to determine when omics-based tests are fit for use in a clinical trial. It also was asked to apply these criteria to omics-based tests used in the three cancer clinical trials conducted by the Duke investigators and to recommend ways to ensure adherence to the developed framework.

A Recommended Framework

An omics test is defined as being composed of or derived from multiple molecular measurements and interpreted by a fully specified computational model to produce a clinically actionable result (IOM, 2012b). The test can assess genomics, transcriptomics, proteomics, epigenetics, and



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement