Prognostic value of computed tomography-based volumetric body composition analysis in patients with head and neck cancer: Feasibility study

Background: This study aimed to evaluate whether computed tomography (CT)-based volumetric body composition analysis has prognostic value in head and neck cancer (HNC) patients. Methods: This single-center retrospective study included 79 patients with HNC treated with definitive radiotherapy from March 2009 to December 2018. The patients were assessed for (a) weight-based variables and (b) pretreatment and posttreatment CT-based body composition variables. Overall survival (OS) and recurrence-free survival (RFS) analyses were conducted using Cox proportional hazards analyses. Results: Depletion of cervical skeletal muscle volume on presentation was associated with poor OS (hazard ratio [HR] = 3.1; 95% CI = 1.2-7.8; P =.016). Low fat proportions before and after treatment were associated with poor OS (HR = 2.5-3.5; 95% CI = 1.3-9.3; P =.013-.026). In multivariate Cox analysis, increased posttreatment fat attenuation demonstrated the greatest prognostic value for both OS (HR = 4.7; 95% CI = 2.2-10.3; P <.001) and RFS (HR = 4.3; 95% CI = 2.0-9.5; P <.001). Conclusions: CT-based body composition analysis has the potential for risk assessment in patients with HNC.