Decision analysis of PD-1 inhibitor combined with chemotherapy in neoadjuvant therapy of resectable locally advanced head and neck squamous cell carcinoma (LA HNSCC).

Abstract

6074 Background: Neoadjuvant therapy (NAT) with PD-1 inhibitors plus chemotherapy has been shown to have a high pathological response rate in LA HNSCC, but there is controversy over factors such as the number of treatment cycles and biomarkers. This trial explored multiple factors that may affect NAT and provided patients with an individualized treatment strategy. Methods: Untreated pts with AJCC 8 th edition stage III-IVB (HPV-positive oropharyngeal : stage II-III) LA HNSCC were selected between 2021 and 2024. After enrollment, pts received 2 cycles of PD-1 inhibitors combined with chemotherapy, and RT was used when CR was achieved on imaging. If PR, SD or PD was achieved, direct surgery or RT could be chosen, or 1-2 cycles of PD-1 inhibitors plus chemotherapy could be received again before curative treatment. The primary endpoint was 1y-DFS. The sample size was N=80, which provided 0.8 power based on Exact Test at One-Sided alpha level of 0.05. Results: A total of 82 pts were included. Median age was 59 yrs (23-76), and 81 (98.8%) were male. Location: oropharyngeal was 22 (26.8%) (HPV-positive was 61.9%), laryngeal was 10 (12.2%), hypopharynx was 48 (58.5%), nasal cavity and sinuses was 2 (2.4%). T3 +T4 was 55.3%, N2+N3 was 44.7%. CPS of pts who underwent testing was 89.0% (73/82), and 48.0% pts were CPS≥20. Pts received 2 cycle (45.1%), 3cycle (50%), 4 cycle (4.8%) NAT. Median follow-up time was 15.9 months, 1y-PFS was 95.9% and 1y-OS was 98.4%. The PFS of pts with pCR or (CR+PR) were both significantly higher than that of pts with non-pCR or (SD+PD)(p=0.049 and p=0.024). The ORR was 84.1%, and ORR of 2-cycle was lower than muti-cycle (81.1% vs 86.7%,p=0.544) . 49.1% pts achieved pCR of primary lesion. pCR of 2-cycle was little higher than muti-cycle (53.3% vs 44.4%, p=0.867). However, the T3-4 in the 2-cycle group was significantly lower than that in the multi-cycle group (43.2% vs 68.9%, p=0.037). In T3-4 pts, pCR in multiple cycles was higher than that in 2 cycles (56.3% vs 38.5%, p=0.339). In addition, pCR of CPS ≥ 20 was significantly higher than that of CPS < 20 (66.7% vs 34.6%, p=0.028). In T1-2 population, pCR of CPS≥20 vs <20 was 81.8% vs 28.6% (p=0.006), while no significant different in T3-4 population. Poorly differentiated patients with CPS ≥ 20 vs. moderately/highly differentiated pts was 85% and 30.6% (p<0.001). 1y - laryngeal function preservation rate was 98.3% (1/58). The overall TRAEs incidence rate was 72%, the most common Grade 3-4 TRAEs were myelosuppression (8.9%). Conclusions: Increasing the number of cycles may be beneficial for T3-4 pts. In addition, CPS≥20 pts can achieve higher pCR, especially in T1-2 pts. CPS is also associated with worse pathological differentiation type. Clinical studies (NCT06100497) are currently underway to further explore the efficacy and safety of PD-1 inhibitors combined with chemotherapy in poorly differentiated LA HNSCC.