The FDA’s Oncologic Drugs Advisory Committee (ODAC) voted 6-2 April 29 to recommend withdrawing accelerated approval for Merck & Co. Inc.’s Keytruda (pembrolizumab) as a third-line treatment for a subgroup of patients with recurrent locally advanced or metastatic gastric or gastroesophageal junction cancer.

The vote was based on FDA assurances that, if it withdrew the approval, it would work with Merck to delay the withdrawal or set up an access program to ensure the estimated 1,000 patients who are beyond first-line treatment could still get Keytruda. Following the vote, Steven Lemery, acting director of the FDA’s Division of Oncology 3, reiterated that promise, saying the agency would think carefully about the approach to take going forward to ensure it does the best it can for the benefit of those patients.

Although data from two confirmatory trials didn’t verify Keytruda’s benefit as a monotherapy in late-stage gastric cancer, the biggest issue for the FDA and ODAC members was the changing treatment landscape for gastric cancer with the approval earlier this month of Bristol Myers Squibb Co.’s Opdivo (nivolumab), another checkpoint inhibitor immunotherapy (IO), as first-line treatment in combination with chemotherapy. Opdivo is expected to quickly become standard of care.

When the FDA has an accelerated approval with a confirmatory trial that failed to demonstrate efficacy, agency staff has to ask, “‘Would we grant this indication at this time?’” said Richard Pazdur, director of the FDA’s Oncology Center of Excellence. In the case of Keytruda’s third-line gastric cancer indication, “the definitive answer is no,” he added.

To get accelerated approval today for the late-stage indication, Merck would have to show that Keytruda worked in patients whose disease progressed after treatment with Opdivo. “That’s the essence of why we’re bringing this to the committee,” Pazdur said.

Keytruda’s third-line approval as a monotherapy for gastric cancer is part of an FDA evaluation of accelerated approvals granted to anti-PD-1/PD-L1 antibodies over the past six years. The evaluation showed 10 accelerated approvals that either hadn’t been confirmed or the results of confirmatory trials were not considered clinically meaningful, said Julia Beaver, chief of medical oncology at the FDA.

After discussions with the FDA, sponsors of four of those approved supplemental biologic license applications (sBLAs) voluntarily withdrew those indications. One of those withdrawals involved another accelerated approval for Keytruda – as third-line treatment for metastatic small-cell lung cancer.

Pazdur said the agency is working with those sponsors to delay the withdrawals or create access programs to ensure current patients can continue to get therapies that are working for them.

The patient need in late-stage gastric cancer was driven home in the public hearing session of the ODAC meeting when a person tearfully reminded the panel that patients like him are more than numbers on a sheet of paper that are analyzed to assess a drug’s effectiveness. “My grandchildren thank you, and my daughters thank you,” he said, referring to the 18 months he’s had with his family since first starting treatment with Keytruda after his cancer progressed. “It doesn’t make any sense to take a treatment off the table that works for some of us.”

Currently, Keytruda is one of two third-line options in gastric cancer, as most patients at that stage can’t take any more chemo. In the trial used for accelerated approval, Keytruda demonstrated about a 13% response rate, with two total responses and the duration of the response lasting up to five years for some patients, said Pooja Bhagia, clinical development lead for GI cancers, oncology clinical research at Merck.

The other option is Taiho Pharmaceutical Co. Ltd.’s Lonsurf (trifluridine/tipiracil), which had a 4% response rate, with a median response of six months, said Peter Enzinger, a professor at Harvard Medical School and Dana-Farber Cancer Institute who consulted for Merck.