Image-based analysis of skeletal muscle mass predicts cisplatin dose-limiting toxicity in patients with locally advanced head and neck cancer.

Abstract

METHODS

Patients with LA-HNC treated with cisplatin-based chemoradiotherapy (CRT) were included. SMM was measured using pre-treatment scans. Logistic regression analysis was performed to identify the predictive impact of low SMM for DLT.

PURPOSE

Evidence suggests that patients' skeletal muscle mass (SMM) can predict the patients at risk for cisplatin dose-limiting toxicities (DLT). Cisplatin is currently dosed on body surface area (BSA). The predictive value of SMM for cisplatin DLT in patients with locally advanced head and neck cancer (LA-HNC) is investigated.

RESULTS

In total, 343 patients were included of which 199 patients (58.0%) had low SMM and 154 patients (44.9%) experienced cisplatin DLT. In multivariate analysis, low SMM at diagnosis was the only predictive factor for DLT (HR 1.8, 95% CI 1.1-2.9).

CONCLUSIONS

Low SMM was associated with an increased risk of DLT. Trials are needed to investigate cisplatin dosing with consideration of SMM rather than solely BSA.