The Clinical Efficacy of Adjuvant Pembrolizumab Immunotherapy in Completely Resected Stage Iii Malignant Melanoma: A Systematic Review and Meta-Analysis of Randomized Controlled Trials and Primary Studies

Introduction: Patients with completely resected stage III cutaneous melanoma remain at substantial risk of locoregional and distant recurrence. Programmed cell death protein 1 (PD-1) blockade with pembrolizumab became a standard adjuvant option after the EORTC 1325/KEYNOTE-054 trial, but its long-term durability, effect modifiers, efficacy against prior active immunotherapies, survival impact and real-world tolerability have not been synthesised together. We aimed to quantify the clinical efficacy and safety of adjuvant pembrolizumab in completely resected stage III melanoma by integrating randomized and primary observational evidence. Methods: The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. PubMed, Europe PMC, the Cochrane Central Register of Controlled Trials (CENTRAL), ClinicalTrials.gov, Semantic Scholar and Google Scholar were searched for records published between November 2011 and 2 November 2026. Randomized controlled trials (RCTs), comparative studies, cohorts and case series reporting pembrolizumab-specific outcomes after complete resection of stage III melanoma were eligible; systematic reviews were excluded from synthesis. Risk of bias was assessed with RoB 2, ROBINS-I and the JBI checklist for case series. Hazard ratios (HR), risk ratios (RR) and logit-transformed proportions were pooled with DerSimonian–Laird random-effects models, supplemented by modified Hartung–Knapp–Sidik–Jonkman (HKSJ) confidence intervals (CI) and prediction intervals. Certainty of evidence was rated with GRADE. Results: Thirteen studies (29 reports; four RCTs, six comparative cohorts, two single-arm cohorts and one case series) including 2,651 pembrolizumab-treated and 1,220 control patients were included. Pembrolizumab reduced the risk of recurrence or death compared with control (HR 0.69, 95% CI 0.59–0.82; four studies; I² = 28.8%; HKSJ 95% CI 0.53–0.91), with HR 0.62 (0.53–0.73) versus placebo or observation and 0.78 (0.65–0.93) versus high-dose interferon or ipilimumab. The pooled HR in PD-L1-positive melanoma was 0.64 (0.55–0.75). In KEYNOTE-054 the effect remained stable from 1.25 to 6.9 years of follow-up (HR 0.57 to 0.63; 7-year RFS 50% versus 36%; distant metastasis-free survival HR 0.64), was consistent across AJCC-8 substages and BRAF status, and was greater after the onset of immune-related adverse events. Overall survival did not differ significantly in SWOG S1404 (HR 0.82, 96.3% CI 0.61–1.09) and the pooled estimate was heterogeneous and inconclusive. Pooled 12-month recurrence-free survival (RFS) in routine practice was 74.7% (72.4–77.0%; I² = 0%). Grade ≥3 treatment-related adverse events occurred in 17.1% (15.0–19.5%), discontinuation for toxicity in 14.0% (11.4–17.1%) and completion of the 1-year course in 52.0% (44.0–60.0%). Versus placebo, pembrolizumab increased grade ≥3 toxicity (RR 4.35, 2.61–7.26); versus high-dose interferon or ipilimumab, it reduced it (RR 0.36, 0.30–0.42). Health-related quality of life was preserved versus placebo and improved versus active comparators. No small-study effects were detected. Discussion: The 31–38% relative reduction in recurrence translates into approximately 133 additional patients per 1,000 remaining recurrence-free at 7 years. The benefit is durable and reproducible in routine practice, whereas the toxicity burden is moderate, predominantly endocrine and often permanent, and an overall survival advantage has not been demonstrated. Conclusion: High-certainty evidence shows that adjuvant pembrolizumab durably prolongs recurrence-free survival in completely resected stage III melanoma, with a more favourable toxicity and quality-of-life profile than prior immunotherapies. Mature overall survival data, biomarker-guided patient selection and integration with neoadjuvant strategies are required.