Impact of body composition in advanced hepatocellular carcinoma: A subanalysis of the SORAMIC trial

**Background:** Hepatocellular carcinoma (HCC) is the most common primary liver cancer and a leading cause of cancer-related death worldwide. For advanced-stage HCC, sorafenib has been the standard of care, with locoregional therapies such as selective internal radiation therapy (SIRT) used in addition to systemic therapy. The SORAMIC trial previously found no significant overall survival (OS) benefit for sorafenib plus SIRT versus sorafenib alone. Body composition parameters—skeletal muscle mass (SMM) and adipose tissue (AT)—have emerged as potential biomarkers in HCC, but data in the palliative setting are scarce, predominantly retrospective, and from small cohorts. This study aimed to assess the influence of baseline body composition on OS in both treatment arms of the SORAMIC trial.

**Methods:** This was an exploratory post hoc substudy of the prospective, randomized, phase II SORAMIC trial conducted at 38 sites in 12 European countries and Turkey. From 424 patients in the palliative arm, 369 were included (55 excluded for lack of baseline CT within 30 days). The cohort comprised 192 patients in the sorafenib/SIRT group and 177 in the sorafenib group; 84.8% were male, mean age 67±8.6 years (range 31–85). Body composition was measured semiautomatically on axial CT at L3 using ImageJ. Parameters included skeletal muscle area, SMI (normalized for height), total adipose tissue (TAT), visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), intramuscular adipose tissue (IMAT), and their indices. Low skeletal muscle mass (LSMM) was defined as SMI <52.4 cm²/m² for males and <38.5 cm²/m² for females (Prado cutoffs). High VAT and high SAT were defined as area >100 cm²; high VSR as >1.1. Univariable Cox regression was used; proportional hazards and linearity assumptions were checked.

**Key Results:** Median OS for the entire cohort was 9.9 months (10.8 months for SIRT/sorafenib, 9.2 months for sorafenib alone). LSMM was present in 206 patients (55.8%): 52.0% in the sorafenib group and 59.4% in the SIRT/sorafenib group. In the overall cohort, no body composition parameter showed a relevant association with OS. For SMI, HR was 1.005 (95% CI 0.992–1.019, p=0.446); for LSMM (low vs. high), HR 0.918 (95% CI 0.726–1.162, p=0.478); for high VAT, HR 1.011 (95% CI 0.793–1.290, p=0.929). In the sorafenib-only group, SMI showed HR 1.005 (95% CI 0.986–1.025, p=0.583); LSMM HR 0.997 (95% CI 0.705–1.409, p=0.985); high VAT HR 0.971 (95% CI 0.680–1.386, p=0.870). In the SIRT/sorafenib group, SMI HR was 1.006 (95% CI 0.987–1.026, p=0.513); LSMM HR 0.828 (95% CI 0.597–1.148, p=0.258); VAT HR 1.000 (95% CI 0.999–1.001, p=0.977). Kaplan-Meier curves showed overlapping survival for LSMM vs. non-LSMM, high vs. normal VAT, high vs. normal SAT, and high vs. normal VSR in all groups. Stratified analyses by ECOG status and BCLC stage also revealed no association. Patients with NAFLD/NASH had the highest values of AT, FM, and FFM.

**Clinical Implications:** This study does not support the use of CT-derived body composition parameters—including sarcopenia (LSMM), visceral adiposity, or fat distribution—for patient selection or prognostication in advanced HCC patients receiving palliative treatment with sorafenib with or without SIRT. The authors suggest that aggressive tumor characteristics and relatively short OS in this cohort may diminish any detectable influence of body composition. While sarcopenia screening remains important for functional assessment, these parameters should not guide treatment allocation in this setting. The study's strengths include its large sample size and prospective data collection; limitations include exclusion of patients without baseline CT and the use of fixed (non-BMI-stratified) cutoff values for LSMM.