In an analysis reported in the Journal of Clinical Oncology, Zhao et al found that the addition of an immune checkpoint inhibitor to chemotherapy in trials of first-line treatment of advanced gastric or esophageal adenocarcinoma was not associated with benefit in low PD-L1–expressing subgroups not analyzed in the trials.
As stated by the investigators, “The U.S. Food and Drug Administration has granted regulatory approval for the use of [the immune checkpoint inhibitor] nivolumab in the first-line treatment of advanced gastric or esophageal adenocarcinoma, regardless of PD-L1 expression. However, the efficacy of immune checkpoint inhibitors in low PD-L1–expressing tumors remains unclear.”
Study Details
The study involved reconstruction of unreported Kaplan-Meier plots of PD-L1 combined positive score (CPS) subgroups from randomized phase III trials comparing the addition of immune checkpoint inhibitors to conventional chemotherapy in the first-line treatment of gastric or esophageal adenocarcinoma. Analysis was performed using KMSubtraction, a method that retrieves patients from unreported subgroups by matching patients on Kaplan-Meier curves of the overall cohort to patients on Kaplan-Meier curves of a known subgroup with follow-up time.
Trials included in the analysis were CheckMate 649 (studying nivolumab plus chemotherapy), KEYNOTE-062 (studying pembrolizumab plus chemotherapy and pembrolizumab as a single agent) and KEYNOTE-590 (studying pembrolizumab plus chemotherapy).
Key Findings
Two PD-L1 subgroups were identified with data unreported in the primary manuscripts of two trials: hazard ratios and Kaplan-Meier plots for overall and progression-free survival were not reported for patients with a CPS of 1 to 4 in CheckMate 649 or for a CPS of 1 to 9 in KEYNOTE-062.
No significant difference between nivolumab plus chemotherapy vs chemotherapy in overall survival (hazard ratio [HR] = 0.950, 95% confidence interval [CI] = 0.747–1.209, P =.678) or progression-free survival (HR = 0.958, 95% CI = 0.743–1.236, P = .743) was observed in the CheckMate PD-L1 CPS 1 to 4 subgroup.
In the KEYNOTE-062 PD-L1 CPS 1 to 9 subgroup, no significant difference in overall survival was observed between pembrolizumab plus chemotherapy (HR = 0.836, 95% CI = 0.658–1.061, P = .141) or pembrolizumab alone (HR = 1.027, 95% CI = 0.811–1.300, P = .827) vs chemotherapy. No significant difference in progression-free survival for pembrolizumab plus chemotherapy vs chemotherapy was observed (HR = 0.924, 95% CI = 0.730–1.169, P = .510), and pembrolizumab alone was associated with significantly poorer progression-free survival vs chemotherapy (HR = 2.092, 95% CI = 1.661–2.635, P < .001).
The investigators concluded, “Using KMSubtraction, data of PD-L1 subgroups previously unreported by primary manuscripts of pivotal clinical trials were retrieved. These data suggest the lack of benefit in the addition of immune checkpoint inhibitors to chemotherapy in low PD-L1–expressing gastric or esophageal adenocarcinoma tumors.”
Raghav Sundar, MBBS, PhD, of the Department of Haematology-Oncology, National University Cancer Institute, Singapore, is the corresponding author for the Journal of Clinical Oncology article.
Disclosure: The study was supported by the Singapore National Medical Research Council. For full disclosures of the study authors, visit ascopubs.org.
