ObjectiveTo explore the efficacy of serplulimab plus chemotherapy in esophageal squamous cell carcinoma (ESCC) patients with liver metastases.MethodsA post hoc exploratory analysis of ASTRUM‐007 study was performed, focusing on the association between the liver metastases status and the clinical outcomes. A systematic literature search of electronic databases was conducted to identify eligible randomized controlled trials for the meta‐analysis. Study‐level pooled analyses of hazard ratios (HRs) for PFS according to liver metastases were performed.ResultsThe post hoc analysis of ASTRUM‐007 showed that although patients with liver metastases had a worse prognosis comparing with the non‐liver metastases patients in both treatment arms (serplulimab plus chemotherapy arm: median PFS, 5.7 vs. 6.6 months, HR 1.57 [95% CI, 1.15–2.13]; median OS, 13.7 vs. 15.3 months, HR 1.48 [95% CI, 1.09–1.98]; placebo plus chemotherapy arm: median PFS, 4.3 vs. 5.5 months, HR 1.58 [95% CI, 1.01–2.39]; median OS, 10.3 vs. 11.2 months, HR 1.32 [95% CI, 0.84–2.00]), OS and PFS benefits derived from serplulimab plus chemotherapy versus placebo plus chemotherapy in this study were observed in both patients with liver metastases (HR of PFS: 0.60; 95% CI, 0.37–0.97; HR of OS: 0.68; 95% CI, 0.43–1.11) and the non‐liver metastases patients (HR of PFS: 0.62; 95% CI, 0.49–0.80; HR of OS: 0.69; 95% CI, 0.55–0.87) with similar magnitude. Three randomized controlled trials were included in the meta‐analysis. Pooled HRs demonstrated that the addition of anti‐PD‐1 antibodies significantly improved PFS compared to chemotherapy alone regardless of liver metastases status.ConclusionsThis study reveals that the presence of liver metastases is a poor prognostic factor but does not affect the improvements in both PFS and OS brought by adding PD‐1 blockade to chemotherapy in ESCC patients. Predictive biomarkers for survival in these patients warrant further investigation.