Comparative Efficacy of KEYNOTE-B10 Regimen as First-Line (1L) Therapy in Recurrent/Metastatic (R/M) Head and Neck Squamous Cell Carcinoma (HNSCC): A Network Meta-Analysis (NMA) with Aggregate-Level Matching.

In KEYNOTE-B10, a phase IV, single-arm trial, the combination of pembrolizumab+carboplatin+paclitaxel (referred to as the B10 regimen) demonstrated antitumor activity with a manageable safety profile as 1L therapy for R/M HNSCC. Due to the single-arm design of KEYNOTE-B10, an NMA was conducted using methodologies to incorporate data from a disconnected trial into a connected network of randomized controlled trials (RCT) to estimate the comparative efficacy of the B10 regimen versus other interventions in 1L R/M HNSCC. A systematic literature review (SLR) was conducted on June 30, 2023 to identify RCTs evaluating relevant interventions in the target population. The objective response rate (ORR) from the interim analysis of KEYNOTE-B10 was extracted and compared with all trials in the network. Aggregate-level matching (ALM) was used to integrate the KEYNOTE-B10 trial into the connected network by matching it to the RCT within the network with the closest trial design and patient population, thus creating a new pseudo-RCT. The best matching patient population was determined as that with the smallest sum of absolute standardized differences across the following characteristics: age, sex, race, ECOG performance status, metastatic disease, tumor location, human papillomavirus infection, and tumor stage at diagnosis. The SLR identified 25 trials, of which 6 matched the eligibility criteria of KEYNOTE-B10 and were included in the ALM/NMA. KEYNOTE-048 had the smallest overall difference in patient characteristics/highest number of balanced variables compared to KEYNOTE-B10, with the latter being incorporated as a new treatment arm in the former. Results from the fixed-effects NMA for the B10 regimen are summarized in Table 1. For ORR, the B10 regimen was more efficacious compared to the platinum+5FU, platinum+taxane, platinum, 5-fluorouracil, and methotrexate regimens. There was no statistically significant difference in ORR between the B10 regimen and pembrolizumab+platinum+5-fluorouracil, cetuximab+platinum+5-fluorouracil, and cetuximab+platinum+taxane. This NMA including interim results from KEYNOTE-B10 demonstrates improved or comparable ORR outcomes versus alternative 1L R/M HNSCC interventions. Once available, analyses should be completed utilizing the final analysis of KEYNOTE-B10 as well as survival outcomes. [ABSTRACT FROM AUTHOR]