Makers of phosphatidylinositol 3-kinase (PI3K) inhibitors in blood cancers should be required to complete randomized phase 3 trials with active comparators before they submit applications for approval, an FDA advisory committee says.

The nearly unanimous vote (16 for and one abstaining) of the Oncologic Drugs Advisory Committee is in response to what committee members called a trend that’s “unprecedented in oncology” of high levels of toxicity and low rates of overall survival across one drug class.

Most of the four FDA-approved PI3K inhibitors used the agency’s accelerated approval pathway, which only requires a single-arm trial. Subsequent randomized phase 3 trials of the PI3K inhibitor class showed troubling toxicities, inadequate dose optimization and trial design considerations regarding the limitations of single-arm trials, as well as low overall survival rates.

Members were in agreement that randomized trials made sense for this potentially dangerous class of drugs that, due to single-arm trials, hasn’t been explored as deeply as it needed to be.

“If you had to do a development plan over,” said Oncology Center of Excellence Director Richard Pazdur, “most people would agree that there should have been random studies done earlier.”