Personalized medicine is a reality for many cancer patients, and getting closer for others, but the oncology community is struggling with a number of questions surrounding genomic testing, a key to providing that care.
As a result, oncology leaders have started to take a closer look at how to resolve these issues, and how to do it as quickly as possible.
Among the most basic questions are when to use the testing, what to do with the information that’s generated, and how to secure reimbursement for diagnostics that may not have fully proven their clinical utility.
The science of genomics – from profiling a tumor’s genetic make-up to sequencing a cancer patient’s entire genome – is growing exponentially, leading to a rush to commercialize diagnostics based on the discoveries and a push among big cancer centers to leverage the knowledge to help develop therapeutics and inform clinical trials. Meanwhile, the Food and Drug Administration hasn’t articulated a clear regulatory strategy around genomic testing.
Mark Fleury, Ph.D.
"We need an infrastructure in place that quickly translates verified advances in treatment into practice," said Mark Fleury, Ph.D., of the American Cancer Society Cancer Action Network (ACS CAN). Now, genomic advances are relying on what he calls "passive diffusion."
Currently, there’s a kind of free-for-all, agreed Dr. Richard L. Schilsky, medical director of the American Society of Clinical Oncology.
"People need tests to guide treatment decisions, but there are very few controls over how tests are offered and developed," said Dr. Schilsky. He also noted that there is not a huge evidence base yet on the usefulness of some of the testing.
The oncology community is trying to bring some civility to the "Wild West" of genomics, starting by establishing standards, Dr. Schilsky said.
Among the issues being discussed:
• What genetic variants should oncologists be testing for in all common cancers?
• What is level of evidence that supports testing for those particular variants?
• What are the recommended clinical actions to be taken once certain variants are discovered?
• What information should be reported to the oncologist?
• What information should be reported to the patient, and how?
In early April, ASCO convened a meeting of representatives from clinical oncology, pathology, the genetic sequencing community, and the regulatory community, among others, to discuss those issues and more, Dr. Schilsky said. "We didn’t come away with any consensus, but we put all the issues on the table," he said.
ACS CAN held a policy forum in April to delve into some of the same issues.
"There are three main things you need to make personalized medicine work," said Dr. Fleury. The information generated has to be accurate, it has to somehow translate to having meaning for the disease, and the diagnostic has to provide some sort of change in treatment for the better, said Dr. Fleury. Now, though, "there are breakdowns in all three of those steps."
Understandably, said Dr. Fleury, payers don’t want to reimburse for tests or services that don’t provide any measurable utility. And, he said, the patient won’t want the information if he or she "doesn’t understand the usefulness."
Dr. Tanguy L. Seiwert of the department of medicine at the University of Chicago who focuses on head and neck and lung cancer, said, "Physicians want to do this, and there’s a growing movement that [genomic testing] is beneficial." But he added that the uncertainty over payment had helped contribute to an uncertainty overall about personalized medicine.
There’s not a lot of good data on exactly how many genomic tests are available – either as a single mutation test or a panel of tests – and how often they are being used.
The FDA has approved 19 diagnostics that are meant to help clinicians determine whether certain targeted therapies would be useful. They are used in colorectal, lung and breast cancers, gastrointestinal stromal tumors, and melanoma.
But that’s only a small fraction of the available tests. In July 2013, the Agency for Healthcare Research and Quality identified 178 different genetic tests for 10 common cancer conditions. Sixty-six were new since it last surveyed the field in 2011; the largest number of tests were being used for breast cancer.
The commercial potential of genomic testing is growing, despite the many outstanding questions. Myriad Genetics said in its most recent quarterly report that there was "wider acceptance of our molecular diagnostic tests by the medical community and increased patient testing volumes." Still, Myriad’s oncology testing revenues grew relatively slowly, by about 12% from the previous year to $102 million for the second quarter. Most of Myriad’s diagnostics are predictive, like its BRACAnalysis test. But the company is also branching out into genomics that will help guide treatment decisions.