Predictive value of body composition changes in cancer-related fatigue in patients with lung cancer undergoing chemotherapy

Background: Cancer-related fatigue (CRF) is a common complication in lung cancer patients undergoing chemotherapy, and body composition alterations may be potential predictive factors. Objectives: To explore the dynamic changes of body composition during chemotherapy and analyze the predictive value of combined body composition indicators for severe CRF. Methods: A retrospective observational study was performed by extracting data from 300 lung cancer patients who received cisplatin-based chemotherapy at two tertiary hospitals between March 2022 and August 2024. CRF was assessed using the Revised Piper Fatigue Scale (PFS-R), and body composition (water content, skeletal muscle, fat, phase angle) was measured via InbodyS10. Repeated-measures ANOVA, multivariate Logistic regression, and receiver operating characteristic (ROC) curves were used for analysis. Results: CRF scores significantly increased over time, rising from mild at T0 (3.15±1.38) to moderate at T3 (6.08±1.52; F=42.36, P<0.001) (F=42.36, P<0.001), while all body composition indicators decreased significantly (all P<0.001). The combined model of body composition indicators showed higher predictive value for severe CRF at T3 (AUC=0.862, 95%CI:0.768–0.956) than T2 (AUC=0.756, P=0.021). Independent predictors of severe CRF at T3 included female gender (OR=1.789), TNM stage Ⅳ (OR=3.336), and decreased percentage changes in water content (OR=0.912), skeletal muscle (OR=0.883), and phase angle (OR=0.807) (all P<0.05). Conclusions: Combined body composition detection effectively predicts severe CRF in lung cancer patients undergoing chemotherapy, especially in the middle-late stage, providing a basis for personalized intervention.