Baseline skeletal muscle index as an independent prognostic factor in locally advanced head and neck squamous cell carcinoma: A post hoc analysis of the REACH trial.

6088 Background: The prognosis for locally advanced squamous cell carcinoma of the head and neck (LAHNSCC) is heterogeneous. A low skeletal muscle mass index (SMI) assessed by computed tomography reflects sarcopenia and is associated with a poor prognosis. However, evidence-based data on HNSCC remain scarce. Moreover, the prognostic value of SMI in patients treated with radiotherapy combined with different systemic treatments remains unclear. Methods: This is a post hoc analysis of the REACH trial (NCT02999087), conducted between 2017 and 2020, which treated LAHNSCC in two cohorts. Patients eligible (fit) for cisplatin were randomized between radiotherapy with cisplatin and radiotherapy with cetuximab-avelumab; those ineligible for cisplatin (unfit) were randomized between radiotherapy with cetuximab and radiotherapy with cetuximab-avelumab. Baseline SMI was assessed on a C3 vertebral slice using radiotherapy planning computed tomography and analyzed as a continuous variable. Associations of SMI with disease-free survival (DFS) and overall survival (OS) were assessed using Cox proportional hazards models, with multivariable adjustment for performance status, age, stage, p16 status, treatment and cisplatin fit/unfit status. Model performance was internally validated using bootstrapping. Results: Among the 694 patients included in REACH, SMI analysis was available for 623 patients (90%). Only the analyses performed for the 507 men are presented. Among them, 312 (62%) were eligible for cisplatin and 195 (38%) were ineligible. Median SMI was 47 (range 32-66) kg/m 2 , with no difference according to the fit/unfit status and treatment received. A higher SMI was observed in patients with ECOG 0 (vs 1, p<0.0001), younger patients (<65y, p=0.01) but not according to tumor stage (III vs IV). There was a linear relationship between SMI and DFS with a better DFS as SMI rises in both univariate (HR 0.81 every 5 SMI units, p<0.0001) and multivariate analyses (HR 0.88/5 units, p=0.013). Comparable results were observed for OS, with a significant linear relationship between SMI and OS (HR of death 0.76/5 units, p<0.0001, univariate analysis; HR 0.86/5 units, p=0.015, multivariate analysis). Other independent factors associated with improved OS were a p16+ oropharynx tumor, ECOG 0 (vs 1), Stage III tumor (vs IV) and eligibility for cisplatin. The prognostic impact of SMI was consistent across fit/unfit cohorts and treatment arms, with no significant interaction detected. Conclusions: Baseline SMI determined on C3 is a strong and independent prognostic factor for DFS and OS in patients with LAHNSCC treated with radiotherapy with cisplatin, radiotherapy with cetuximab, or radiotherapy with cetuximab and avelumab. The association between SMI and clinical outcomes is linear, arguing against the use of arbitrary prognostic thresholds for SMI.