Muscle matters: automated CT-based body composition analysis predicts survival in patients with head and neck cancer treated with immunotherapy

Purpose: The objective of this study was to evaluate the prognostic significance of CT-based body composition markers in patients with head and neck squamous cell cancer (HNSCC) treated with immunotherapy. Material and methods: Forty-five HNSCC patients (24.4% female, median age: 66 years) treated with Nivolumab or Pembrolizumab were retrospectively assessed. Automated body composition analysis was performed on thoracic CT scans. The analysis included skeletal muscle (SM), bone (B), visceral adipose tissue (VAT), and volumes of various adipose tissue compartments. Overall survival (OS) was estimated using the Kaplan-Meier method, and hazard ratios (HR) were calculated using the Cox proportional hazards model. In the multivariate analysis, the strongest body composition parameter was entered into the model. Results: The median OS was 8.13 months (95% CI: 4.8-21.9). Univariate analysis identified baseline high SM/B ratio, high (SM+VAT)/B ratio, albumin levels >3.4g/dl, low Eastern Cooperative Oncology Group (ECOG) performance status, body mass index ≥18.5 kg/m2, and male sex as significant prognostic factors for longer OS. In multivariate analysis, SM/B ratio (HR: 0.25, 95% CI: 0.1-0.64, p=0.004) and albumin ≤3.4 g/dl (HR: 0.31, 95% CI: 0.12-0.76, p=0.01) remained independent. Patients with both high SM/B and albumin survived the longest (median not reached) compared to either high SM/B or high albumin (9.6 months) vs. low SM/B and albumin (2.7 months). A decrease in SM/B of ≥ 8% after three months was associated with a lower median OS of 6.7 vs. 26.2 months, p=0.032. Conclusions: Automated CT-based body composition analysis, particularly the thoracic SM/B ratio and serum albumin level, provide valuable prognostic information on OS for HNSCC patients receiving immunotherapy and may guide clinical decision-making in this patient population.