NEW YORK (GenomeWeb) – Numerous challenges in the molecular diagnostics space are slowing down the implementation of precision medicine, a number of stakeholders in the field reported at last week's Arrowhead Personalized and Precision Medicine conference in San Francisco.
In presentations and a panel discussion, both molecular diagnostic developers and a payor representative discussed challenges with test development, reimbursement, and regulation. They cited the need for partnerships and novel regulatory structures.
Mirdula Iyer, general manager of precision medicine at Quest Diagnostics, said that the laboratory reports results from more than 70,000 tests in its precision medicine portfolio annually. But nonetheless, "we still find that there is limited compliance to guideline-based testing," she said, meaning that either patients are not getting testing who should or are not getting the right kind of testing.
Iyer cited internal data showing that between 25 percent and 45 percent of patients who could benefit from biomarker testing fail to be tested. And for those who are tested, reimbursement is initially denied in about one quarter of the cases.
For oncology-specific molecular diagnostic tests, around half of BCR-ABL tests for patients with chronic myeloid leukemia are not ordered based on guidelines, she said. Typically, the test is ordered one to two times per year instead of the recommended three to four times, she explained. Also, around 60 percent of patients with non-small cell lung cancer who receive molecular testingare treated before the oncologist receives their test results.
Even for companion diagnostic tests, insurance frequently denies payment. Claims are initially denied for 40 percent of HER2 tests, 17 percent of EGFR tests, and 25 percent of BRAF tests, she said.
Others reported challenges with getting pharmacogenomics tests reimbursed. Kristine Ashcroft, CEO of YouScript, a spin-off from pharmacogenetic testing firm Genelex, described obstacles it has encountered for panel-based PGx testing, for example.
Ashcraft noted that one challenge has been the shift from the 'one test for one marker for one drug' paradigm to the ability to do multimarker panel testing to assess patients' ability to metabolize a range of different drugs. That has posed problems with regulators and payors alike, she said.
Last year, Genelex collaborated with the University of Utah to demonstrate the efficacy of its panel-based PGx test, reporting in the Journal of Medical Economics that testing led to a 39 percent reduction in hospitalizations and a 79 percent reduction in emergency room visits. That observational study did not lead to improved reimbursement, Ashcraft previously told GenomeWeb.
Girish Putcha, director of laboratory science at Palmetto GBA's MolDx program, said that payors want to see clinical trial data. "Studies need to show utility," he said, "even if the study is small." Ideally, studies should be randomized, statistically powered, and have defined clinical utility endpoints that compare standard of care with and without testing. The challenge for payors though, he said, is that most studies are single-arm, non-randomized, and observational.
"There needs to be greater diligence up front to figure out what [testing] is needed and what a payor will pay" before companies develop tests, he said.
Ashcraft said she recognized the importance of conducting clinical trials, but noted that they are often difficult for diagnostic companies to conduct because they do not have the resources of large pharmaceutical companies.
She said that YouScript has recently run a small, randomized clinical trial of its PGx panel, which showed a more than 50 percent reduction in hospitalization rates and a more than 40 percent reduction in ER visits in patients who received the PGx panel compared to those who did not.
Moving forward, she said, it would be helpful to have a clearer idea of the type of data that payors wanted to see in order to reimburse. "We try to have conversations with payors," she said, but they often will not engage.
Jami Elliott, CEO of Quantigen Genomics, a genomic services company based in Indiana that offers PGx testing, agreed. Diagnostic companies "want to construct the right trial," he said, so it would be helpful to get guidance from payors.
Quest's Iyer said that her group has recently started to collaborate with pharmaceutical companies to sponsor testing for patients to ensure that the right patients get the right drugs. The sponsorship "removes the challenges of reimbursement," she said, and also helps improve consistency in testing, since it is performed at one or two laboratories. "It removes a lot of hurdles around companion diagnostic testing," she said.
Aside from providing clinical trial data showing the clinical utility of a test, another challenge for test providers has been that technology has moved so quickly that current guidelines and regulations are a bad fit for multi-marker molecular tests, panel participants said.
In addition, it is time consuming to bring diagnostic tests through FDA approval, and there is no guarantee that test developers will see a return on that investment.
Also, the FDA currently has "no structure" in place for submitting multi-marker pharmacogenetic tests for FDA clearance, Elliott said.
Putcha said that the industry as a whole should work together and "agree on principles" to figure out better processes for developing tests, bringing them through regulatory approval, and gaining reimbursement. For instance, laboratory developed tests and in vitro diagnostics should complement, not duplicate each other, he said. The FDA's "approach is evolving," he said, citing the agency's willingness to work with companies developing NGS-based companion diagnostic panels. Similar approaches could be taken with companies developing multi-marker pharmacogenomics tests, he added.