First-line pembrolizumab plus chemotherapy versus chemotherapy in advanced esophageal cancer: Longer-term efficacy, safety, and quality-oflife results from the phase 3 KEYNOTE-590 study

Abstract

Background: At interim analysis of the phase 3, randomized, double-blind KEYNOTE-590 (NCT03189719) study, 1L pembrolizumab (pembro) + chemotherapy (chemo) vs chemo alone provided superior OS, PFS, and ORR with a manageable safety profile in patients (pts) with untreated, advanced/unresectable or metastatic adenocarcinoma or esophageal squamous cell carcinoma (ESCC) or Siewert type 1 esophagogastric junction adenocarcinoma (EGJ). We report efficacy, safety, and health-related quality of life (HRQoL) results with an additional 12 months (mo) of follow-up. Methods: 749 eligible pts were randomized 1:1 to pembro 200 mg or placebo Q3W for up to 2 yr + chemo. Randomization was stratified by geographic region, histology, and performance status. Treatment continued until progression, unacceptable toxicity, or withdrawal, or 2 yr. No crossover was permitted. Primary endpoints were OS in pts with ESCC PD-L1 combined positive score (CPS) ≥10 tumors, and OS and PFS (RECIST v1.1; by INV) in ESCC, PD-L1 CPS ≥10, and all pts. Secondary endpoints included ORR, DOR, safety, and HRQoL. HRQol was assessed in 711 treated pts with ≥1 HRQoL assessment (356 pembro + chemo; 355 chemo). Data cutoff was July 9, 2021. Results: At data cutoff, median follow-up (randomization to data cutoff) was 34.8 mo. Median OS was longer with pembro + chemo vs chemo in pts with ESCC CPS ≥10 (HR 0.59; 95% CI, 0.45-0.76), ESCC (HR 0.73; 95% CI, 0.61-0.88), CPS ≥10 (HR 0.64; 95% CI, 0.51-0.80), and all pts (HR 0.73, 95% CI, 0.63-0.86). In pts with adenocarcinoma OS HR was 0.73 (95% CI, 0.55-0.99). The 24-mo OS rate in all pts was 26.3% vs 16.1%. Median PFS was longer with pembro + chemo vs chemo in ESCC (HR 0.65; 95% CI, 0.54-0.78), CPS ≥10 (HR 0.51; 95% CI, 0.41-0.65), and all pts (HR 0.64; 95% CI, 0.55-0.75). The 24-mo PFS rate in all pts was 11.6% vs 3.3%. Confirmed ORR was 45.0% (25 CR [6.7%]) vs 29.3% (9 CR [2.4%]), with median DOR of 8.3 vs 6.0 mo. Approximately 20% vs 6% of pts had response duration ≥24 months. Grade 3-5 drug-related AE rates were 72% vs 68%. Discontinuation rates from drug-related AEs were 21% vs 12%. There was no significant difference in least square mean (LSM) change from baseline to wk 18 between arms in EORTC QLQ-C30 global health status/quality-of-life (LSM difference -0.10; 95% CI, -3.40-3.20). LSM change from baseline to wk 18 was better with pembro + chemo vs chemo for QLQ-OES 18 pain (-2.94; 95% CI, -5.86 to -0.02) and dysphagia (-5.54; 95% CI, -10.92 to -0.16). Conclusions: With an additional 12 months of follow-up, pembro + chemo continued to provide significant and clinically meaningful improvement in OS, PFS, and ORR vs chemo with a manageable safety profile, and stable quality-of-life for pts with untreated, advanced esophageal and EGJ cancer. These data continue to support 1L pembro + chemo as a new standard of care in these patients.