NEW YORK (GenomeWeb) – The US Food and Drug Administration on Friday approved the first drug with a complementary diagnostic — a type of test that is not essential for determining who should receive the drug, but can give doctors an idea of how well their patients might respond to treatment.
Bristol-Myers Squibb's anti-PD-1 drug Opdivo (nivolumab), which the FDA earlier approved for previously treated squamous non-small cell lung cancer patients, is now available for all advanced NSCLC patients who have progressed despite platinum-based chemotherapy. The FDA approved Opdivo's expanded indication with breakthrough therapy designation three months ahead of its PDUFA date of Jan. 2, 2016.
Simultaneously, the FDA said it had also approved PD-L1 IHC 28-8 pharmDx, developed by Agilent Technologies subsidiary Dako, to identify patients who might benefit most from treatment. However, the agency didn't say the test was essential to giving the drug safely and effectively.
"Biomarker testing is not required for Opdivo," BMS said in a statement, but added that Dako's test was used in the pivotal study for Opdivo and can provide additional information to doctors on how best to treat patients. In a separate statement, Agilent highlighted that the PD-L1 test was the only FDA-approved assay to assess the survival benefit associated with Opdivo.
The drug label does describe the use of Dako's test in the "clinical studies" section, but in the "indications" section there is no mention of testing to assess PD-L1 expression. Moreover, the term "complementary diagnostic" doesn't show up in FDA's announcement of the new Opdivo indication or in the drug label.
A companion test is required for the safe and effective use of a drug, according to an FDA guidance on the topic. Although a complementary test isn't required in the same way, it can still be used to guide treatment strategies and identify patients likely to derive the most benefit from a drug.
At a recent meeting in Washington, DC, FDA officials said they were internally working on formally defining the term "complementary diagnostic," but noted that such tests would require premarket approval just like companion tests.
Companion versus complementary
The data that led to the approval of Opdivo, and to the approval of Merck's Keytruda (pembrolizumab) a week earlier, provides some insights into when the agency might require a companion diagnostic with a drug and when it would consider a complementary test useful in informing therapeutic strategies.
In the 600-patient pivotal study for Opdivo, patients lived longer on the immunotherapy compared to docetaxel, regardless of PD-L1 expression (12.2 months versus 9.4 months). Moreover, 19 percent of Opdivo-treated patients experienced complete or partial tumor shrinkage that lasted 17 months on average, compared to 12 percent of patients on docetaxel, whose response lasted six months on average.
"The clinical trials for Opdivo (nivolumab) showed an overall survival benefit across the study, which entered patients regardless of whether their tumors had PD-L1 expression and this was demonstrated in a randomized trial for traditional approval," FDA spokesperson Sarah Peddicord told GenomeWeb. "However, BMS also did exploratory studies to better understand the relationship between PD-L1 expression and tumor response for NSCLC [patients] following treatment with Opdivo."
Using Dako's IHC test, researchers stratified patients into groups based on whether they were PD-L1 negative (expression in less than 1 percent of tumor cells) or PD-L1 positive. The PD-L1 positive groups were defined as those that had PD-L1 expression in 1 percent to less than 5 percent of tumor cells; greater than 5 percent to less than 10 percent of tumor cells; and in 10 percent or more of tumor cells.
Opdivo-treated patients in the highest PD-L1 expression group had median overall survival of 19.4 months compared to 8 months in the docetaxel arm. Patients with lower levels of PD-L1 expression also lived longer on Opdivo compared to docetaxel, but the highest expressing group fared best. However, when PD-L1 was not expressed in the tumor, there was no difference in median overall survival between the treatment arms.
As such, in approving a complementary diagnostic, FDA attempted to balance the survival advantage seen in the overall population with the increased level of benefit seen in PD-L1 expressers. "Given that Opdivo extended survival in the entire population, the FDA did not want to restrict the indication to only the patients with high PD-L1 expression," Peddicord said.
In comparison, FDA approved Keytruda simultaneously with Dako's PD-L1 IHC 22C3 pharmDx Kit as a companion test, because Merck garnered accelerated approval for the drug using data in the high PD-L1-expressing group. Keytruda's approval was based on data from a subset of 61 patients who had PD-L1 expression in 50 percent or more tumor cells as determined by Dako's IHC test. In this subset, 41 percent of patients saw their tumors shrink, but these were all partial responses.
"This is how their initial study was designed and it's important to note that this was for accelerated approval in a small population where, in order to be granted accelerated approval, the drug needed to show an improvement in available therapy," Peddicord said.
Test to fit the drug
Although both the complementary and companion tests are developed by Dako on an IHC platform, there are key differences between the assays. The most important is how differently these tests define PD-L1-positive tumors, which in turn characterizes the best responder populations for these two drugs.
In the Keytruda development program, the PD-L1 test was analytically validated to gauge a PD-L1 expression tumor proportion score (TPS), and FDA granted accelerated approval based on data from a subset of patients with the highest score — meaning they had PD-L1 expression in half or more tumor cells. "The specimen should be considered PD-L1 positive if TPS greater than or equal to 50 percent of the viable tumor cells exhibit membrane staining at any intensity," Dako's PD-L1 IHC 22C3 pharmDx companion test labeling notes.
For Opdivo, the highest PD-L1-expressing group includes patients with expression in 10 percent or more tumor cells. The complementary PD-L1 IHC 28-8 pharmDx test's indication states that PD-L1 expression is defined as "the percentage of tumor cells exhibiting positive membrane staining at any intensity," and that a detection of PD-L1 expression "may be associated with enhanced survival from" the drug.
The FDA and drug firms advancing immunotherapies (Merck, AstraZeneca, Roche/Genentech, and BMS) have said they will compare and publish the analytical features of IHC tests they're using to stratify patients in trials, so healthcare providers can understand the differences between them.
Nuanced approach
Opdivo and Keytruda's approvals reflect growing acceptance on the part of industry that diagnostics don't necessarily restrict the market of a drug but help define it. And as research continues to reveal new ways of identifying those that respond well to immunotherapies, the FDA is also taking a more nuanced regulatory approach.
"There is still a lot to learn about the PD-1/PD-L1 pathway and its effects in lung cancer, as well as other tumor types," Richard Pazdur, director of the Office of Hematology and Oncology Products in the FDA's drug center said in a statement.
Based on the "indications" section of the drug label, Opdivo could be an option for any advanced NSCLC patients. Doctors would have to peruse the "clinical studies" section of the label to learn about how to define the best responder population. "The FDA provided information on survival based on tumor PD-L1 expression levels in product labeling so that health care providers and patients can be better informed on what their outcome may be based on tumor expression levels of PD-L1," Peddicord said. "BMS continues to conduct studies to explore these findings further."
Keytruda's label is more prescriptive about the intent-to-treat population, indicating it for patients whose tumors are positive for PD-L1 expression. However, the drug's indication doesn't specify the cut-off for determining PD-L1-positive tumors. That information is detailed in the "clinical studies" section of drug labeling and in the indication for the PD-L1 IHC 22C3 pharmDx test.
Still, doctors may not hold fast to the labeling language, since studies have shown that some patients with expression in less than 50 percent of tumor cells respond to the drug. It is likely, researchers wrote regarding a Keytruda study published in the New England Journal of Medicine earlier this year that "tumor PD-L1 expression alone does not accurately assess the dynamic immune microenvironment." As such, Merck is continuing to investigate Keytruda.
"As a condition of the approval, the sponsor has ongoing randomized studies in lung cancer to confirm that the patients with the highest PD-L1 expression will have long-term benefit, including testing whether Keytruda can delay disease progression or prolong survival in these patients," Peddicord said. "They are also testing whether patients with tumors with lower expression also have delayed disease progression and longer survival when taking Keytruda."