Systematic review and meta-analysis of neoadjuvant anti–PD-(L)1 inhibitors in head and neck squamous cell carcinoma.

Abstract

e18069 Background: Neoadjuvant PD-(L)1 inhibitors have emerged as a promising therapeutic option in multiple cancer types, including head and neck squamous cell carcinoma (HNSCC). This study aimed to assess the efficacy and safety outcomes of neoadjuvant PD-(L)1 inhibitor monotherapy in HNSCC through a systematic review and meta-analysis of clinical trials. Methods: PubMed and Embase were searched for studies published from inception to 26 July 2024. We used MeSH terms and keywords for HNSCC and neoadjuvant immunotherapy. No filters were applied for the search. Data were extracted following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. Only prospective trials were included. Pooled analysis was done using the meta-package by Schwarzer et al. Proportions and risk ratios (RR) with 95% confidence intervals (CI) were computed. Results: Total 7 neoadjuvant PD-(L)1 trials encompassing 282 patients were included. Number of neoadjuvant anti-PD-(L)1 cycles, which included pembrolizumab, nivolumab, and durvalumab, administered ranged 1-4 cycles. 2 studies mandated administration of adjuvant PD-(L)1 therapy for 6 cycles. The estimated pooled rate of pathologic complete response (pCR), major pathologic response (mPR), and partial pathologic response (pPR) were 5% (95% CI: 2%-12%), 8% (95% CI: 5%-13%), and 19% (95% CI: 12%-30%) respectively. The objective response rate (ORR), as assessed by pre-operative imaging per RECIST version 1.1 criteria, was 6% (95% CI: 2%-16%). Grade >3 treatment-related adverse event rate was 16% (95% CI: 4%-43%) and all-grade treatment-related adverse event rate was 59% (95% CI: 18%-90%). The 1-year disease-free survival and overall survival rates were 78% (95% CI: 63%-88%) and 89% (95% CI: 75%-96%), respectively. There were no reported cases of delayed surgery among included trials. Conclusions: Neoadjuvant anti-PD-(L)1 monotherapy, while safe and does not delay surgery, has modest pathological response rates. Furthermore, long-term follow up data as well as robust evidence for the correlation between pathological responses and event-free and overall survival are sparse, limiting further conclusions that inform the treatment’s impact on survival outcomes. Further studies are warranted to improve upon single agent neoadjuvant anti-PD-(L)1 inhibitor therapy in HNSCC and long term outcomes are awaited.