Text

Background and objective: The prognostic value of body composition on the outcome of various cancer types is well described. However, its impact on recurrence and survival in patients with hepatocellular carcinoma (HCC) undergoing liver transplantation is still unclear. Therefore, this study aimed to evaluate the prognostic relevance of AI-based body composition analysis in HCC patients undergoing liver transplantation.

Method: AI-based body composition was assessed in all patients who underwent liver transplantation between 2007 and 2025 due to HCC and had a computed tomography scan of the abdomen available within three months prior to transplantation. Pre-, intra- and postoperative data were retrospectively collected. Patients were stratified according to the lowest 30^th percentile of the Sarcopenia Index (SI; most sarcopenic) and the highest 30th percentile of the Myosteatosis Fat Index (MFI) and Visceral Fat Index (VFI) (highest myosteatosis and visceral fat distribution).

Result: A total of 126 patients were included. Sarcopenia and myosteatosis were identified as significant independent prognostic parameters for overall survival after transplantation (p=0.048 and p=0.002, respectively), while BMI class and VFI showed no significant influence on survival (p=0.826 and p=0.599).

Summary: The results of our study show that sarcopenia and myosteatosis have a significant impact on the outcome of patients with HCC undergoing liver transplantation, while plain abdominal fat does not appear to affect overall survival. Optimal patient selection can be improved by fully automated, AI-based assessment of these factors prior to transplantation.