Toxicities and concerning patient survival data sparked a U.S. Food and Drug Administration advisory committee to recommend shutting the door on single-arm clinical studies of PI3K inhibitors.
Sarah Silbiger/Getty Images
Toxicities and concerning patient survival data sparked a U.S. Food and Drug Administration advisory committee to recommend shutting the door on single-arm clinical studies of PI3K inhibitors, a drug class that has been used to treat various forms of hematological cancers.
Instead, the FDA advisory committee recommended the use of randomized clinical trials in order to mitigate the safety concerns. On Thursday, the Oncologic Drug Advisory Committee overwhelmingly voted to call for future approvals of PI3K inhibitors to be supported by randomized trials in the future. The committee voted with 16 in favor of the resolution and one abstention.
Michele Nadeem-Baker, a chronic lymphocytic leukemia patient and an FDA Patient Representative, said the evidence presented during the meeting is compelling and shows that this class of drug needs to be supported by data that can only come through randomized studies.
Anthony Sung, a hematologist from Duke University and ODAC Committee member, was the one abstaining vote. Following the vote, he noted that the safety data presented by the FDA was problematic but said he had some concerns over the language of the resolution.
The vote came after a robust three-hour discussion of survival and toxicity concerns in the phosphatidylinositol-3-kinase inhibitors drug class. FDA hematologists laid out the history of approvals related to the four different drugs under discussion, Gilead Sciences’ Zydelig (idelalisib), which has been withdrawn from the market for certain kinds of cancer, Bayer’s copanlisib, Secura Bio’s develisib, and TG Therapeutics’ Ukoniq (umbralisib), as well as the history of safety issues patients experienced during clinical trials. The committee members honed in on the question of whether or not randomized data should be required to support a demonstration of substantial evidence of efficacy, and whether or not the drug class is safe for its intended use in the proposed patient population.
Richard Pazdur, director of the FDA’s Oncology Center of Excellence, said the examination of the safety data for PI3K drugs has revealed “unprecedented findings in oncology.” He expressed concern that in previous clinical studies, there has not been a careful evaluation of the efficacy dose-toxicity relationship.
PI3Ks regulate multiple cellular processes including cell growth, proliferation, differentiation, survival, migration and metabolism. However, the FDA raised concerns over a greater reliance on single-arm trials for this drug class. Throughout the development history of this drug class, several clinical studies have been halted due to toxicity issues, raising concerns about the drug’s impact on the overall survivability of patients who receive this class of drug for their type of hematological cancer.
Just last week, New York-based TG Therapeutics withdrew a pending Biologics License Application and supplemental New Drug Application for a combination therapy with its PI3K inhibitor, Ukoniq (umbralisib) due to an “imbalance” in overall survival seen in recent updated clinical data.
Concerns raised by the regulatory agency, which were previously highlighted by BioSpace, point to questions about the high doses of PI3K inhibitors that have received approval. The FDA team data suggests that drug developers have not done an effective job of exploring lower dosing levels in order to reduce toxicity while maintaining efficacy.
Nicole Gormley, the acting division director of hematology products at the FDA, noted that lower doses may be equally effective but there is not much data available. She said a randomized mid-stage trial could allow companies to collect the data exploring those lower doses. That way, a company could go into a registrational trial with more confidence about dose levels, she said.
Gormley noted today’s advisory committee meeting was a different type of ODAC meeting. She said it is forward-thinking regarding the PI3K class of drug and the known toxicities.
“This degree of safety findings that we’re seeing and multiple survival results… are all showing these consistent findings of overall survival data is concerning,” she said. Gormley added that it is imperative the FDA ensures that the medical products used for these cancer patients are safe and effective.
“All of these products have (effective) activity but also have a very concerning safety profile,” Gormley said.
In a presentation highlighting the regulatory agency’s concerns, Nicholas Richardson, an FDA hematologist and oncologist pointed out limitations of the use of single-arm trials in the class. In a single-arm trial it can be difficult to balance the efficacy and safety of the drug, he said.
“Randomized trials are the preferred approach to demonstrating casual effects and treatment,” he said. Richardson added that overall survival is an objective measure of clinical benefit that can only be assessed in a randomized trial.
Gormley noted that safety data collected in the single-arm studies of the PI3K drugs have shown different types of toxicities and safety issues. The delta- and gamma-isoform-specific inhibitors are linked to increased issues of pneumonia and immune-mediated toxicities, including hepatitis, pneumonitis, colitis and rash. Some of those rashes have become severe and led to infections, which have caused patients to succumb to their disease, Richardson said. Alpha-specific isoforms can lead to hyperglycemia and hypertension, Gormley said.
There were some concerns during the meeting that the recommended language in the ODAC resolution could inhibit the potential of future accelerated approvals in hematological cancers. Gormley dismissed that thought. She said the accelerated approval pathway does not equate to a single-arm trial. Gormley noted that randomized trials can be powered for early efficacy endpoints and later endpoints for progression-free survival that can lead to accelerated approval.
