Efficacy of pembrolizumab in advanced esophageal carcinoma: A systematic review and meta-analysis.

Abstract

438 Background: Pembrolizumab, a monoclonal immunoglobulin G4 antibody is a programmed cell death protein 1 inhibitor, approved for treatment of advanced (metastatic or recurrent) esophageal squamous cell carcinoma (ESCC) and esophageal adenocarcinoma (EAC). We performed a pooled analysis to assess the efficacy of pembrolizumab in esophageal carcinoma (EC) and to compare it with chemotherapy alone. Methods: We conducted a systemic literature search, using PubMed, EMBASE, and ClinicalTrials.gov, till July 30, 2024. Initial search yielded 412 articles. We excluded duplicates and irrelevant studies and included 4 clinical trials (CT) that reported pembrolizumab’s efficacy in EC. We estimated pooled objective response rate (ORR), partial response (PR), and hazard ratio (HR) for overall survival (OS) and progression-free survival (PFS), along with 95% confidence interval (CI) and p-value in RevMan v.5.4. Subgroup analysis, based on combined positive score (CPS) for programmed cell death ligand-1 (PDL-1) expression and tumor histology type was also performed. Results: We included four CTs in our study; their characteristics are shown in the table. The estimated pooled ORR and PR of pembrolizumab in all patients was 21% (CI: 0.10-0.41) (p < 0.001) and 19% (CI: 0.08-0.37) (p < 0.001), respectively. Pooled median PFS was 3.05 mo (CI: 0.92-5.17) (p = 0.004), and pooled median OS was 8.05 mo (CI: 5.21-10.88) (p < 0.001). Pembrolizumab use in EC showed a reduced risk of disease progression vs chemotherapy alone [HR: 0.85 (CI: 0.75-0.96) (p < 0.05)] and a 21% reduction in mortality [HR: 0.79 (CI: 0.71-0.88) (p < 0.001)]. Safety analysis revealed less grade ≥3 adverse events [OR: 0.56 (CI: 0.35-0.90), p < 0.05)] with pembrolizumab compared with chemotherapy alone. On subgroup analysis between ESSC and EAC, pembrolizumab exhibited statistically nonsignificant responses with ORR (OR: 1.29, CI: 0.89-1.88, p = 0.18), OS (pooled HR: 0.94, CI: 0.76-1.15, p = 0.53). Pembrolizumab in patients with CPS ≥10 vs CPS <10 also showed statistically non-significant responses with ORR [OR: 1.70, CI: 0.99-2.91, p = 0.06), OS (pooled HR: 0.63 vs 1.00) (p = 0.05) and PFS (pooled HR: 0.62 vs 0.83, p = 0.08). Conclusions: Pembrolizumab alone or combined with chemotherapy compared with chemotherapy alone exhibited favorable responses in terms of ORR, OS and PFS with a manageable toxicity profile, setting a benchmark for future studies. Characteristics of included studies. Trial ID, phase Histology Line Number of patients Regimen Median follow-up (mo) ORR (%) Median PFS (mo) Median OS (mo) KEYNOTE-590, III ESCC, EAC 1 st 749 Pembro + chemo vs chemo 58.8 45 vs 29.3 6.3 vs 5.8 12.3 vs 9.8 KEYNOTE-181, III ESCC, EAC 2 nd 628 Pembro vs chemo 7.1 13.1 vs 6.7 2.1 vs 3.4 9.3 vs 6.9 KEYNOTE-180, II ESCC, EAC 3 rd or later 121 Pembro 5.8 9.9 2.0 5.8 KEYNOTE-028, 1b ESCC, EAC 3 rd or later 21 Pembro 7 30 1.8 7 Pembro: pembrolizumab, mo: month.