Longitudinal Body Composition Identifies Hepatocellular Carcinoma With Cachexia Following Combined Immunotherapy and Target Therapy (CHANCE2213)

ABSTRACT Background Cancer cachexia can impact prognosis, cause resistance to anticancer treatments and affect the tolerability of treatments. This study aims to identify hepatocellular carcinoma (HCC) with cachexia by characterizing longitudinal body composition ( BC ) trajectories. Methods This longitudinal, multicentre cohort study included unresectable HCC patients treated with first‐line programmed death‐(ligand)1 inhibitors plus anti‐vascular endothelial growth factor antibody/tyrosine kinase inhibitors between 01/2018–12/2022. BC measurements including skeletal muscle mass (SMM) and total adipose tissue area (TATA) were evaluated by computed tomography at the third lumbar vertebra at baseline and follow‐up imaging. Unsupervised latent class growth mixed models were applied to distinguish potential longitudinal SMM and TATA trajectories for identifying cachexia. The primary study endpoint was overall survival (OS), with secondary endpoints including progression‐free survival (PFS), objective response rate (ORR) and safety. Multiple Cox proportional hazards models were used to calculate adjusted hazard ratios (HRs) for survival. Results A total of 411 patients with 2138 time‐point measurements were included. The median age was 56 years, and 50 (12.2%) patients were female. Two distinct trajectories were identified for SMM and TATA: sharp‐falling and stable. SMM sharply declined in 58 patients (14.1%) and TATA in 71 of 406 patients (17.5%) with significant worse OS (for SMM, 17.0 vs. 24.9 months; p < 0.001; HR = 0.59; for TATA, 15.3 vs. 25.1 months; p < 0.001; HR = 0.44). Patients were categorized into three phases based on trajectories: pre‐cachexia (SMM and TATA stable, n = 299, 73.6%), cachexia (SMM or TATA sharp‐falling, n = 86, 21.2%) and refractory cachexia (SMM and TATA sharp‐falling, n = 21, 5.2%). Patients with refractory cachexia exhibited the worst OS, PFS and ORR, followed by those with cachexia. The median OS was 11.5 months for refractory cachexia, 17.7 for cachexia and 26.0 for pre‐cachexia; median PFS was 6.0, 7.9 and 10.9 months, respectively, with ORR of 4.8%, 39.5% and 54.2%, respectively (all p s < 0.001). Multivariable Cox analysis identified refractory cachexia as an independent risk factor for both OS (HR = 3.31; p < 0.001) and PFS (HR = 2.94; p < 0.001), with cachexia also showing significant impacts. Grade 3–4 adverse events were higher in patients with refractory cachexia (23.8%) and cachexia (8.1%) compared with pre‐cachexia (6.0%; p = 0.010). Conclusions HCC patients with cachexia and refractory cachexia were identified by longitudinal BC trajectories. Falling trajectories of BC identified refractory cachexia patients with worst response, survival and poor tolerability from systemic therapy combinations. Trial Registration ClinicalTrials.gov identifier: NCT05278195