Background Conventional cytotoxic regimens for recurrent or metastatic head and neck squamous cell carcinoma (R/M HNSCC) offer limited durable benefit and are associated with significant toxicity. Immune checkpoint inhibitors targeting the PD-1/PD-L1 pathway have emerged as alternative treatments; however, variability in response, survival, and toxicity across studies necessitates comprehensive synthesis. This systematic review and meta-analysis evaluated the efficacy and safety of PD-1 inhibitors in adults with R/M HNSCC. Methods A systematic search of PubMed (MEDLINE), Embase, Scopus, and Cochrane databases was conducted. Study selection followed the PICOS framework and PRISMA guidelines, with screening performed using Rayyan. Data synthesis was conducted in R (version 4.1.1) using the meta and metafor packages. The protocol was registered in PROSPERO (CRD420251169189). Results Fourteen studies (n = 1,231) were included: nine in non-NPC HNSCC (primary analysis) and five NPC-only studies (sensitivity subgroup). In non-NPC HNSCC, pooled ORR was 22% (95% CI: 8%–48%, I 2 = 87%) for combination therapy and 14% (95% CI: 3%–46%, I 2 = 29%) for monotherapy. For combination regimens, pooled best responses were: CR 5.6%, PR 21.7%, SD 28.2%, and PD 34.9%. One-year PFS was 29.3% (95% CI: 13.5%–52.4%, I 2 = 88%, k = 3), and 1-year OS was 59.0% (95% CI: 42.3%–73.8%, I 2 = 82%, k = 8). Grade 3–5 treatment-related adverse events occurred in 16.3% of monotherapy patients and 48.2% of combination therapy patients. In NPC-only studies (k = 5), pooled ORR was 61.1% (95% CI: 13.4%–94.1%, I 2 = 89%). Estimates are descriptive and largely derived from single-arm studies with substantial heterogeneity. Conclusion PD-1 inhibitor combinations may be associated with higher response rates and 1-year survival compared with monotherapy in non-NPC HNSCC, at the cost of increased toxicity. These findings are hypothesis-generating and do not establish causal efficacy. Well-designed randomized trials stratified by disease subtype and treatment context are needed, alongside greater focus on quality-of-life outcomes.
Alabbasi et al. (Fri,) studied this question.