FDA Emphasizes Risks of Pembrolizumab in Multiple Myeloma After Reviewing Halted Trials

Less than 2 months after it ordered a stop to 3 trials of pembrolizumab (Keytruda) in multiple myeloma due to safety concerns, the FDA has released a more detailed analysis of what went wrong in these trials.

Merck, the immuno-oncology drug’s manufacturer, announced in July that the FDA had placed 2 phase 3 studies on full hold and a phase 1 study on partial hold, citing a pattern of patient deaths. The FDA’s safety alert, issued recently to healthcare providers, oncology researchers, and the public, analyzes data from the 2 phase 3 trials that were halted completely.

KEYNOTE-183 had been studying pembrolizumab in combination with pomalidomide and low-dose dexamethasone in patients with relapsed/refractory multiple myeloma. KEYNOTE-185 was assessing pembrolizumab in combination with lenalidomide and low-dose dexamethasone in those with newly diagnosed and treatment-naïve multiple myeloma.

In KEYNOTE-183 documented 29 deaths in the investigational arm receiving pembrolizumab and 21 deaths in the control arm. The FDA determined that the relative risk of death was 61% higher for participants in the pembrolizumab arm. This increased hazard was not accompanied by improved outcomes, as the objective response rate was 34% in the investigational arm but 40% in the control group.

Patients receiving pembrolizumab were also more likely to experience a higher rate of grade 3 to 5 toxicities than their counterparts in the control arm (83% vs 65%, respectively), as well as serious adverse events (63% vs 46%). The most common causes of death unrelated to cancer progression in the pembrolizumab arm included several heart-related conditions, neutropenic sepsis, and multiple organ dysfunction.

The KEYNOTE-185 trial appeared to pose even more risk to participants. There were 19 deaths in the investigational arm compared with 9 in the control group, indicating a risk of death that was over 2-fold for the patients receiving pembrolizumab. Higher incidences of grade 3 to 5 toxicities (72% vs 50%) and serious adverse events (54% vs 39%) were observed in the pembrolizumab arm than in the control arm.

Non-cancer causes of death in the investigational arm of KEYNOTE-185 included the cardiac events seen in KEYNOTE-183, but also intestinal ischemia, suicide, and sudden death. The objective response rate in the pembrolizumab arm was 64%, not significantly different from the 62% rate seen in the control arm.

Both the data analysis and a press statement from the FDA emphasized that these safety risks only apply to multiple myeloma, which is not an approved indication of pembrolizumab, and that patients taking the drug for approved indications should continue their regimen without worry. These approved indications include melanoma, lung cancer, head and neck cancer, classical Hodgkin lymphoma, and others.

“Today’s alert underscores the importance of why new therapies are thoroughly studied to ensure the benefits of taking them outweigh the risks to patients, and we will continue to aggressively monitor clinical trials to ensure patients are protected when safety concerns arise,” said Janet Woodcock, MD, director of the FDA’s Center for Drug Evaluation and Research, in the announcement.