As cancer care evolves into a field where genomic testing is used routinely to pinpoint diagnoses and determine treatments for specific subpopulations, a growing number of molecular tests are becoming available to oncologists. Invited speakers examined issues related to the role of these tests in precision medicine at a Senate briefing on Capitol Hill sponsored by the Friends of Cancer Research (FOCR) and the Deerfield Institute, which develops and analyzes data to enhance the understanding of innovation, emerging products, and trends within health care.
Concerns have been raised that some of the tests used, specifically laboratory-developed tests (LDTs), may not be as high in quality as FDA-approved tests sold as diagnostic kits and known as companion diagnostics in oncology. Unlike companion diagnostics, LDTs generally do not undergo a premarket review by the FDA. Oversight of LDTs is provided by the Clinical Laboratory Improvement Amendments (CLIA), but FOCR charges that CLIA oversight primarily “focuses on laboratory processes and personnel—not on premarket assessment of analytical and clinical validity to determine if they actually perform as claimed,” in the words of FOCR President and CEO Jeff Allen, PhD.
In October 2014, the FDA announced it would extend its oversight of diagnostic tests to LDTs because of their increasing complexity and the growing frequency of their use to guide treatment decisions. More than 11,000 LDTs are currently used in practice, FDA data state. To date, according to a published study of FDA-approved diagnostic tests and LDTs used in advanced non-small cell lung cancer (NSCLC) discussed in detail at the briefing, it is unclear how frequently LDTs are used compared to FDA-approved tests to guide clinical decision-making in treatment with targeted therapies (Personalized Medicine in Oncology 2016;5).
“These tests are incredibly important, but they need to work…and there’s a lot of variability,” said Ellen V. Sigal, PhD, Founder of the FOCR. The study found that testing in NSCLC patients was more commonly performed with LDTs than with FDA-regulated tests for EGFR mutations, and was evenly split between LDTs and FDA-regulated tests for ALK rearrangements. The published paper found that the high rate of overall testing for mutations is consistent with other studies, and indicates that molecular testing is now a routine part of the treatment of advanced NSCLC patients. “It is standard of care to test these patients for mutations,” said Jonathan Leff, MBA, Chairman of the Deerfield Institute.
The published study also makes the point that “steps should be taken to mitigate uncharacterized variability between tests used in the same clinical setting.” Leff noted that “by and large the physician does not know whether it’s an LDT or an FDA-approved test” when he or she orders the test. In fact, LDTs and FDA-regulated tests are often used in the same clinical setting, which the authors state may raise the concern “that an unknown degree of variability may exist between tests for the same intended use.
”There are pros and cons to the increased use of LDTs, as the study points out. They may offer rapid technical advances and even represent some advantages over FDA-approved companion diagnostics for a specific mutation, if any such tests exist. Some physicians are concerned about overregulation of LDTs.
The Lab’s View
“The quality of LDTs and FDA-approved tests is actually pretty similar,” emphasized A. John Iafrate, MD, PhD, Director of the Center for Integrated Diagnostics and Pathologist at Massachusetts General Hospital and Professor at Harvard Medical School, Boston. “We don’t need a large regulatory infrastructure” when it comes to LDTs, he explained, noting that his preference is to make some “common-sense changes” to CLIA and work closely with the College of American Pathologists (CAP) on these tweaks and updates to regulatory systems that already exist.
“Under CLIA and under CAP, there is regulation,” stressed Iafrate. He added that, in some cases, an LDT is the only choice in diagnosing an NSCLC patient subset, citing chromosomal rearrangements in the ROS1 gene as an example (ROS1 was approved as a target for crizotinib in March 2016). “I don’t mean to paint an Armageddon situation,” Iafrate told Oncology Times, but he predicted that, if there is extensive blanket regulation of LDTs, some laboratories will close. Also, he said, “Imposing a regulatory framework over all LDTs is going to be incredibly expensive…I don’t think that has gotten enough attention.” Iafrate noted that, under that new framework, diagnostic and treatment costs could rise.
CAP has issued risk-based recommendations aimed at strengthening oversight of LDTs. Under these recommendations, optimal oversight of LDTs would be achieved by strengthening CLIA accreditation standards on laboratories using low- and moderate-risk LDTs and requiring FDA review prior to offering the test clinically for all high-risk LDTs. Under these CAP recommendations, moderate-risk LDTs would include KRAS and HER2, while Oncotype Dx would be considered a high-risk test.
Challenges in Genomic Testing
“The linchpin for precision medicine is the diagnostic,” said Jeffrey Shuren, MD, JD, Director of the FDA’s Center for Devices and Radiological Health. “From our perspective, we’ve seen a variety of problems in genomic tests used in cancer.” He explained that some of these problems have been methodology issues, and added that what the FDA does not want is “imprecise medicine.” Shuren emphasized that a diagnostic test should demonstrate analytical validity, clinical validity, and—of key importance to insurance payors—clinical utility.
Shuren said the FDA is engaging with the clinical community to define what is really important when it comes to a genomic variant. He said many variants are very uncommon, and that even across laboratories there can be different results for the same patient. At the FDA, “we need a smart kind of oversight,” Shuren emphasized.
“The technology is going to continue to accelerate…the train has left the station,” said Michael Pellini, MD, President and CEO of Foundation Medicine. “This is not just a cancer issue,” he added. In such a fast-moving market, Pellini recommended thoughtful consideration of the tests that are really having an impact on clinical decision-making. He advised caution in next-generation sequencing, stating, “It turns out this is really difficult to apply to different tumor types and do so with accuracy.” He added it is important to take the development of diagnostic tests one step at a time.
Andrea Ferris, MBA, President and Chairman of LUNGevity, a nonprofit lung cancer patient advocacy organization, cautioned that “just because we can test for it doesn’t mean we should.” She also said it would be “irresponsible” for medical professionals to put the burden on the patient to ask what diagnostic test is being used—either an LDT or an FDA-approved test. Patients cannot and should not be expected to enter into this level of technical decision-making, she noted.
Leff pointed out that one issue in the current landscape of diagnostic testing that needs urgent attention is the fact that investment in the development of new molecular diagnostic tests lags way behind investment in the development of new drugs. More money needs to be put into development of diagnostics, according to Leff. In addition, he said, intellectual property issues are highly complicated for new diagnostics.
“Predictability is critical here,” Leff stressed of new regulation of LDTs. And, he concluded, any new regulatory framework of LDT oversight must not be overly restrictive, keeping the need for investment in innovation in mind.