2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
In a 6 to 2 vote, the FDA’s Oncologic Drugs Advisory Committee voted against maintaining the accelerated approval of pembrolizumab for the treatment of patients with PD-L1–positive recurrent or advanced gastric or gastroesophageal junction adenocarcinoma who have received 2 or more lines of therapy.
In a 6 to 2 vote, the FDA’s Oncologic Drugs Advisory Committee (ODAC) voted against maintaining the accelerated approval of pembrolizumab (Keytruda) for the treatment of patients with PD-L1–positive (combined positive score [CPS] ≥1) recurrent or advanced gastric or gastroesophageal junction (GEJ) adenocarcinoma who have received 2 or more lines of therapy.
In explaining her decision, Susan Halabi, PhD, said, “While I do recognize that there is an unmet need, the reason I voted no is because I didn’t find the data compelling. As was presented, the response rate was low, and when we look at the risk/benefit ratio, the risk outweighs the benefit to the patient. Also, the fact that patients will have access through the expanded program persuaded me to vote no. Finally, because of the changing landscape, I wasn’t convinced that the monotherapy by itself is going to work.”
Mark Lewis, MD, added, “I voted no. The landscape of both therapeutics and even companion diagnostics looks so different now than when the accelerated approval was first granted. The timing of this meeting is providential coming after the seismic CheckMate-649 data and subsequent approval. I think we are going to see immunotherapy move earlier in treatment lines, and frankly the main reason I voted no, and it’s not to cast a deaf ear on the unmet needs of the patients, is that none of the pending studies as best I could tell would definitively answer the question of either monotherapy or the position in terms of lines of therapeutic sequencing of this agent."
The meeting roster comprised ODAC members Susan Halabi, PhD, Christopher H. Lieu, MD, and David E. Mitchell, temporary voting members, James Randolph Hillard, MD, Pamela L. Kunz, MD, Mark A. Lewis, MD, Diane Reidy-Lagunes, MD, and Colin D. Weekes, MD, PhD, FASCO.
Voting members, Susan Halabi, PhD, Christopher H. Lieu, MD, David E. Mitchell, Pamela L. Kunz, MD, Mark A. Lewis, MD, Colin D. Weekes, MD, PhD, FASCO, voted against the continued accelerated approval of pembrolizumab.
Non-voting members included acting designated federal officer of ODAC, Takyiah Stevenson, PharmD, acting industry representative to the committee, Albert L. Kraus, PhD, and FDA participants, Richard Pazdur, MD, Julia Beaver, MD, and Steven Lemery, MD, MHS.
The meeting represents the final day of a 3-day public review of 6 indications for checkpoint inhibitors granted under the agency’s accelerated approval process that later failed to reach thresholds for statistical significance for key end points in confirmatory clinical trials.
With regard to pembrolizumab, the PD-1 inhibitor demonstrated an objective response rate (ORR) of 13.3% (n = 143; 95% CI, 8.2%-20.0%) in patients with PD-L1–positive tumors (combined positive score ≥1) and microsatellite stable (MSS) tumor status or undetermined microsatellite instability or mismatch repair status enrolled in cohort 1 of the phase 2 KEYNOTE-059 trial.
The ORR in known MSS/mismatch repair proficient and CPS of 1 or greater (n = 100) was 11.0% (95% CI, 5.6%-18.8%).
The median duration of response among 19 responding patients ranged from 2.8+ to 19.4+ months. Responses were 6 months or longer in 11 (58%) patients and 12 months or longer in 5 (26%) patients.
Primary findings from cohort 1 of the trial served as the basis for the September 2017 accelerated approval of the agent for the treatment of patients with PD-L1–positive recurrent or advanced gastric/GEJ adenocarcinoma who have received 2 or more lines of chemotherapy, including fluoropyrimidine- and platinum-containing chemotherapy, and, if appropriate, HER2/neu-targeted therapy.
The continued approval of the agent was contingent on positive data from the phase 3, second-line KEYNOTE-061 and frontline KEYNOTE-062 trials. However, pembrolizumab missed each trial’s primary end point of overall survival (OS) and progression-free survival (PFS) in patients with a PD-L1 CPS of 1 or greater.
In KEYNOTE-061, patients with gastric/GEJ cancer were randomized to receive either 200 mg of pembrolizumab every 3 weeks (n = 296) or 80 mg/m2 of paclitaxel on days 1, 8, and 15 every 4 weeks (n = 296). The hazard ratio for OS and PFS was 0.82 (95% CI, 0.66-1.03) and 1.27 (95% CI, 1.03-1.57), respectively.
In KEYNOTE-062, patients with gastric/GEJ cancer and a PD-L1 CPS of at least 1 were randomized 1:1 to receive 200 mg of pembrolizumab every 3 weeks (n = 256), pembrolizumab plus 5-fluorouracil (5-FU)/cisplatin (n = 257), or 5-FU/cisplatin (n = 250). The hazard ratio for OS and PFS for pembrolizumab vs 5-FU/cisplatin was 0.91 (95% CI, 0.69-1.182) and 1.64 (95% CI, 1.36-1.98), respectively.
Moreover, the treatment landscape changed in April 2021, when the FDA approved the combination of nivolumab (Opdivo) with select types of chemotherapy for the frontline treatment of patients with advanced or metastatic gastric cancer, GEJ cancer, and esophageal adenocarcinoma.
Data from the phase 3 CheckMate-649 trial, which served as the basis for the approval, showed that nivolumab in combination with leucovorin, 5-FU, and oxaliplatin (FOLFOX) or capecitabine and oxaliplatin (CapeOX) resulted in a significant improvement in survival in treatment-naïve patients with advanced gastric cancer, GEJ cancer, and esophageal adenocarcinoma vs chemotherapy alone.
The median OS reported with nivolumab/chemotherapy was 13.8 months compared with 11.6 months with chemotherapy alone (HR, 0.80; 95% CI, 0.71-0.94; P = .0002).
Expressing concern over voting to maintain the accelerated approval, the FDA underscored the low response rate, immune-related adverse effects, and negative confirmatory trials associated with pembrolizumab.
Moreover, Richard Pazdur, MD, director of the FDA’s Oncology Center of Excellence stated that unintended consequences of maintaining the approval could result in the unwanted sequential use of nivolumab and pembrolizumab and preferential use of pembrolizumab in the third-line setting vs nivolumab in the frontline setting.
However, during the hearing, Ronan J. Kelly, MD, MBA, and Manish A. Shah, MD, maintained that checkpoint inhibition has shown a benefit in the third-line setting, and without pembrolizumab, many patients would be left to hospice. Moreover, Kelly maintained that despite the frontline approval of nivolumab/chemotherapy in the frontline setting, many patients have passed the window of opportunity for first-line therapy and are in need of additional treatment.
In response, the FDA stated that if the third-line indication is rescinded, the removal of the indication could be delayed and that patients could still receive pembrolizumab through expanded access.
Additionally, the accelerated approval of pembrolizumab for the treatment of adult and pediatric patients with unresectable or metastatic, microsatellite instability-high or mismatch repair deficient solid tumors that have progressed after prior treatment and who have no satisfactory alternative treatment options, as well as for patients with MSI-H or dMMR colorectal cancer following progression on a fluoropyrimidine, oxaliplatin, and irinotecan would remain intact, irrespective of the decision of the biologics license application in gastric cancer.
Because KEYNOTE-061 and KEYNOTE-062 failed to meet the post-marketing requirement for pembrolizumab, Merck highlighted 4 ongoing, randomized, phase 3 trials that have the potential to confirm the clinical benefit of the agent in gastric cancer within 1 to 3 years: KEYNOTE-859, KEYNOTE-811, KEYNOTE-585, and LEAP-015. However, none of the 4 trials will evaluable pembrolizumab alone.
Oncologic Drugs Advisory Committee (ODAC) Meeting. FDA. April 28, 2021. Accessed April 28, 2021. https://collaboration.fda.gov/ODAC04282021?disclaimer-consent=true&proto=true.