The overall survival benefit for PD-L1 CPS ≥ 5 tumors in CheckMate 649 is a game-changer. An oxaliplatin doublet plus chemotherapy should become a standard of care for these patients,” according to Elizabeth Smyth, MD, an oncology consultant at Cambridge University Hospital NHS Foundation Trust in the United Kingdom, the invited discussant of CheckMate 649 and ATTRACTION-4. She noted this represents the first “paradigm shift” in this malignancy in
20 years.

Such success with immune checkpoint inhibitors in gastric cancer is new. Aside from improved survival with nivolumab/chemotherapy in refractory disease, trials in earlier lines (or the maintenance setting) were not positive for nivolumab, pembrolizumab, or avelumab. “The take-home message from these previous trials was that the efficacy of anti–PD-1 monotherapy in gastric cancer is limited,” she said.

Elizabeth Smyth, MD

Impact of Posttrial Subsequent Therapy

As to why ATTRACTION-4 failed to show an overall survival benefit, Dr. Smyth speculated that subsequent therapy was the primary factor, as the studies were largely similar. “Historically, overall survival in gastric cancer trials has been confounded by high levels of second-, third- and fourth-line therapy in Asian patients. This likely resulted in the lack of overall survival benefit in ATTRACTION-4, as almost twice as many patients received second-line therapy as in CheckMate 649 and a quarter of the control group received posttrial immunotherapy,” she said. “An analysis censoring posttrial immunotherapy might be useful in understanding whether there’s a true survival benefit.”

The receipt of posttrial subsequent therapy may also explain why CheckMate 649 was “resoundingly positive” in patients with CPS ≥ 5 tumors, whereas KEYNOTE-062, with a similar CPS ≥ 10 population, was negative for pembrolizumab.¹ Twice as many patients in the control arm of KEYNOTE-062, compared with CheckMate 649, received immune blockade later, “and although the numbers are still low, this could chisel away at any overall survival benefit,” she suggested.

Selecting Patients for Treatment

Are all newly diagnosed patients with advanced gastric cancer candidates for immunotherapy, or are only those with high PD-L1 expression? In CheckMate 649, the overall survival benefit was more muted in the CPS ≥ 1 subgroup and in the population as a whole; however, it was still statistically significant—possibly because of PD-L1 enrichment overall. Among all patients with gastric cancer, in the general population, about 40% will have tumors with PD-L1 CPS ≥ 5 expression; this proportion was 60% in the CheckMate 649 population and 70% within the CPS ≥ 1 category.

“Therefore, these groups were enriched with highly immunogenic tumors to a degree not seen in unbiased nontrial populations,” Dr. Smyth noted. “So, we may not see these overall survival results outside of this trial…. We may need further analysis to generalize these results to less immunogenic tumors.” 

DISCLOSURE: KEYNOTE-062 was supported by Merck Sharp & Dohme. Dr. Smyth has served as a consultant or advisor to Astellas Pharma, AstraZeneca, Bristol Myers Squibb, Merck Serono, and Servier; holds pending patent no: 1716712.3; and has been reimbursed for travel, accommodations, or other expenses by Bristol Myers Squibb and Servier.

REFERENCE

1. Shitara K, Van Cutsem E, Bang YJ, et al: Efficacy and safety of pembrolizumab or pembrolizumab plus chemotherapy vs chemotherapy alone for patients with first-line, advanced gastric cancer: The KEYNOTE-062 phase III randomized clinical trial. JAMA Oncol 6:1571-1580, 2020.