Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) Impacts Long-Term Outcomes After Curative-Intent Surgery for Hepatocellular Carcinoma

Abstract

Background: Curative surgery for hepatocellular carcinoma (HCC) includes liver resection (LR) and orthotopic liver transplantation (OLT). Due to the obesity epidemic, metabolic dysfunction-associated steatotic liver disease (MASLD) is a frequent HCC aetiology that often coincides with increased alcohol consumption, termed MetALD, or even alcohol-associated liver disease (ALD).

Methods: Patients undergoing LR or OLT for HCC at Charité-Universitätsmedizin Berlin (2010-2020) were included in this retrospective cohort study investigating disease aetiology, time to recurrence (TTR), overall survival (OS) and CT-based body composition.

Results: Out of 579 patients with HCC, 417 underwent LR and 162 OLT. Tumour aetiologies were viral n = 191 (33.0%), MASLD n = 158 (27.3%), MetALD n = 51 (8.8%), ALD n = 68 (11.7%) and other/cryptogenic n = 111 (19.2%). Patients with MASLD and MetALD had more intramuscular (p < 0.001, p = 0.015) and visceral fat (both p < 0.001) than patients with non-metabolic dysfunction aetiologies. Patients with MASLD-HCC had comparable TTR (median 26 months, [95% CI: 23-31] vs. 30 months [95% CI: 4-57], p = 0.425) but shorter OS than patients with other HCC aetiologies (63 months [95% CI: 42-84] vs. 80 months [95% CI: 60-100], hazard ratio: 1.53 [95% CI: 1.050-2.229], p = 0.026) after LR. Multivariate analysis confirmed MASLD aetiology, portal vein thrombosis and MELD score ≥ 10 as independent prognostic factors for OS in LR (adjusted p = 0.021,p < 0.001,p = 0.003), even after excluding in-hospital mortality (adjusted p = 0.016,p = 0.002,p = 0.002). Causes of death were similar in MASLD and non-MASLD aetiology.

Conclusions: Patients with HCC undergoing LR and meeting the new MASLD criteria have significantly shorter OS. This study provides empirical prognostic evidence for the novel MASLD/MetALD classification in a large European cohort of patients undergoing curative-intent HCC therapy.

FIGURE 1

Definition and incidence of HCC aetiologies. (A) Flowchart of the definition of HCC aetiology, adapted from the 2023 Delphi consensus statement [11]. Retrospective quantification of alcohol consumption was considered inaccurate, and therefore patients with relevant alcohol consumption were differentiated by their incidence of steatosis and cardiovascular risk factors. (B) Tumour aetiologies of the overall patient cohort. (C) Tumour aetiologies of the cryptogenic/other group. Panel 1A was created with Biorender.com.

FIGURE 2

Predominant HCC aetiology and patient characteristics. (A) Liver disease severity, assessed by Child‐Pugh Score and MELD in the LR and OLT groups, divided by tumour aetiology. MELD was compared between aetiologies with Kruskal‐Wallis/Deniz Uluknn‘s multiple comparisons test. (B) Patient and laboratory parameters. Kruskal–Wallis test with significances given for Deniz Uluknn's multiple comparisons test of the LR/body composition cohort (n = 309). (C) Fibrosis and steatosis stage by HCC aetiology in the overall cohort. A total of n = 484 patients had a documented fibrosis stage and n = 322 had a documented steatosis stage of the non‐tumorous liver tissue. Friedman test with post hoc correction. Only significant comparisons are indicated with *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001. AFP, α‐fetoprotein; ALD, alcohol‐related liver disease; ALT, alanine transaminase; AST, aspartate aminotransferase; BMI, body mass index; CRP, C‐reactive protein; LiMAx, liver maximum capacity test; LR, liver resection; MASLD, metabolic dysfunction‐associated steatotic liver disease; MetALD, metabolic and alcohol‐related liver disease; OLT, orthotopic liver transplantation; INR, international normalised ratio.

FIGURE 3

Distribution of cardiovascular risk factors and long‐term outcomes in patients with and without MASLD aetiology. Frequency (%) of cardiovascular risk factors among patients undergoing liver resection (MASLD n = 107, MetALD n = 51, ALD n = 28, viral n = 128, other/cryptogenic n = 103) (A) and transplantation (MASLD n = 51 ALD n = 40, viral n = 63, other/cryptogenic n = 8) (B). Pearson's Chi‐square p < 0.05 was considered significant. Time to recurrence (C, D) and overall survival (E, F) of patients with HCC, undergoing liver resection (C, E) or orthotopic liver transplantation (D, F). Log‐rank p < 0.05 was considered significant. (G) Causes of death, stratified by type of operation, with a total of n = 177 deaths and n = 52 deaths recorded in the LR and OLT cohorts, respectively. ALD, alcohol‐related liver disease; HCC, hepatocellular carcinoma; MASLD, metabolic dysfunction‐associated steatotic liver disease; MetALD, metabolic and alcohol‐related liver disease; LR, liver resection; OLT, orthotopic liver transplantation.

FIGURE 4

Association of HCC aetiology with body composition pathologies. CT‐based body composition assessment in patients with HCC, with lean muscle mass in red, fatty‐infiltrated muscle in violet, visceral fat in dark green, subcutaneous fat in light green. Representative images of typical body composition found in patients with viral aetiology: (A) Physiological/normal body composition (BMI: 21 kg/m²), (B) Physiological body composition (BMI: 26 kg/m²), (C) Visceral and subcutaneous obesity (increased light and dark green areas) with preserved muscle mass (BMI: 24 kg/m²). Representative images of patients with MetALD and MASLD: (D) Sarcopenia, sarcopenic obesity and visceral obesity with preserved muscle quality in a patient with MASLD (BMI: 25 kg/m²), (E) Isolated visceral obesity, muscle mass and quality preserved in a patient with MetALD (BMI: 26 kg/m²), (F) Myosteatosis (increased violet muscle area) and visceral obesity, with preserved overall muscle quantity (BMI: 28 kg/m²). (G) Skeletal muscle index (SMI), subcutaneous fat index (SFI), visceral fat index (VFI) and skeletal muscle (SM)‐attenuation in Hounsfield units (HU), with lower values representing a higher amount of intramuscular fat. Kruskall–Wallis test with significances given for Deniz Uluknn's multiple comparisons test of the LR/body composition cohort (n = 309). Only significant p‐values are indicated, with *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001. (H) Correlation of HCC aetiology/underlying liver disease and body composition. Spearman's r (left) and p‐values (right) are given (n = 309). ALD, alcohol‐related liver disease; BMI, body mass index; CT, computed tomography; HCC, hepatocellular carcinoma; HU, Hounsfield units; MASLD, metabolic dysfunction‐associated steatotic liver disease; MetALD, metabolic and alcohol‐related liver disease; SFI, subcutaneous fat index; SMI, skeletal muscle index; VFI, visceral fat index.