Predicting Survival In Head and Neck Cancer: The Protective Role of BMI and the Impact of Evolving Soft Tissue Volumes during Radiotherapy
Abstract
Purpose
The nutritional status of head and neck cancer (HNC) patients receiving radiotherapy often declines due to treatment-related side effects. This study characterizes the effect of baseline body mass index (BMI) on overall survival and analyzes how soft tissue volumes change during radiotherapy to improve clinical outcome predictions.
Methods
Clinical data (N=541) and daily treatment imaging (N=689) were analyzed for a cohort of HNC patients who received radiotherapy at our institution. Baseline BMI effects on survival were evaluated using Kaplan-Meier analysis. Multivariate Cox proportional hazards regression identified independent predictors of survival among clinical features, including age, staging, and performance status. Daily cone-beam CT scans were automatically contoured using TotalSegmentator, an autosegmentation tool, to track the average percent change in muscle, fat, and body volume throughout treatment.
Results
Underweight patients had a significantly lower median survival (4.2 years [1.2-NR]) and 5-year survival rate (46.9% [28.2-63.6]) compared to obese patients, who had a median survival of 10.9 years [8.7-10.9] and a 5-year survival rate of 81.2% [73.3-87.0]. Higher baseline BMI showed a protective effect, with obese (HR=0.46, [0.27-0.78]) and overweight (HR=0.52, [0.34-0.79]) cohorts demonstrating lower mortality risk. Significant independent predictors of poor survival included Age 65+ (HR=2.24, [1.54-3.26]), Stage III disease (HR=1.87, [1.14-3.08]), and poor performance status (ECOG 3) (HR=4.72, [1.63-13.66]). By treatment completion, average BMI declined by 9.9% (+/- 5.3 SD). While skeletal muscle and body volume tracked with this change, subcutaneous fat declined at a more drastic rate of 13.0% (+/- 15.2 SD).
Conclusion
Overweight and obese status at the start of radiotherapy serve as protective indicators for survival in HNC patients undergoing radiotherapy. Rapidly declining subcutaneous fat volumes during treatment may provide an earlier indication for outcome prediction than static clinical features. These findings support more personalized clinical decisions tailored to evolving soft tissue changes.