Serum Myostatin Predicts the Risk of Hepatocellular Carcinoma in Patients with Alcoholic Cirrhosis: A Multicenter Study

Abstract

Background and Aim: Previous studies reported that serum myostatin is associated with sarcopenia. We aimed to elucidate the association between serum myostatin levels and hepatocellular carcinoma (HCC) development in patients with alcoholic liver cirrhosis (ALC). Methods: This retrospective, multicenter study assessed 1077 Asian ALC patients enrolled from 2007 to 2017. The primary endpoint was the development of HCC within 5 years. Cox proportional hazards model analyses were used to assess the association of serum myostatin levels and HCC development. The time-dependent areas under the receiver operating characteristic curve (AUROC) of serum myostatin for 5-year HCC development were calculated. Serum myostatin levels were measured using an enzyme-linked immunosorbent assay with samples collected on the index date. Results: During a median follow-up of 2.5 years, 5-year cumulative HCC incidence rates were 6.7% in the total population. The median level of serum myostatin was 3.3 ng/mL (interquartile, 2.1-5.2 ng/mL). The AUROC of serum myostatin for 5-year HCC development was 0.78 (95% confidence interval [CI], 0.76-0.81). In Cox proportional hazards model analyses, age, gender, platelet counts, and serum myostatin levels were independent risk factors for HCC development (adjusted hazard ratios [HRs] of age, male gender, platelet counts, and serum myostatin: 1.03, 2.79, 0.996, 1.18, respectively; all p < 0.05). Patients with high myostatin levels had a significantly higher risk of 5-year HCC development than those with low myostatin levels (HR 7.53, p < 0.001). Conclusion: Higher serum myostatin levels were significantly associated with a higher risk of developing HCC in ALC patients, which could identify high-risk patients who need stringent surveillance.

Keywords: cirrhosis; hepatocellular carcinoma; myostatin.

Figure 1

The predictive performance of serum myostatin levels and the cumulative probability of HCC in alcoholic liver cirrhosis (ALC) patients. (A) Time-dependent areas under receiver operating characteristic curves (AUROCs) of serum myostatin to predict HCC development within five years. (B) Risk stratification for HCC development in ALC patients according to serum myostatin levels (p < 0.001 by log-rank test). Abbreviations: ALC, alcoholic liver cirrhosis; AUROC, area under the receiver operating characteristic curve; HCC, hepatocellular carcinoma.

Figure 2

Kaplan–Meier estimates of HCC development in ALC patients according to the Toronto HCC risk index (THRI). (A) Risk stratification for HCC development in ALC patients according to the THRI (p < 0.001 by log-rank test). (B) Risk stratification for HCC development according to serum myostatin levels in the low risk group stratified by the THRI; (C) in the intermediate risk group; (D) in the high-risk group (all, p < 0.05 by log-rank test). Abbreviations: ALC, alcoholic liver cirrhosis; THRI, Toronto HCC risk index; HCC, hepatocellular carcinoma.

Figure 3

Kaplan–Meier estimates of HCC development in ALC patients in each group who had similar residual liver function according to Child–Pugh class or model for end-stage liver disease (MELD) scores. Risk stratification for HCC development according to serum myostatin levels among patients who were in Child–Pugh class A (A), class B (B), and class C (C) (all, p < 0.001 by log-rank test). Risk stratification for HCC development according to serum myostatin levels among patients who had low MELD scores (≤13, median value) (D) and high MELD scores (>13) (E) (all, p < 0.001 by log-rank test). Abbreviations: ALC, alcoholic liver cirrhosis; HCC, hepatocellular carcinoma; MELD, model for end-stage liver disease.

Figure 4

Kaplan–Meier estimates of HCC development in ALC patients in each group who had similar residual liver function according to the albumin–bilirubin (ALBI) grade or fibrosis-4 (FIB-4) scores. Risk stratification for HCC development according to serum myostatin levels among patients who were in ALBI grade 1 (A), grade 2 (B), and grade 3 (C) (all p < 0.01 by log-rank test). Risk stratification for HCC development according to serum myostatin levels among patients who had low FIB-4 scores (≤6.46, median value) (D) and high FIB-4 scores (>6.46) (E) (all p < 0.001 by log-rank test). Abbreviations: ALBI, albumin–bilirubin grade; ALC, alcoholic liver cirrhosis; FIB-4, fibrosis-4; HCC, hepatocellular carcinoma.

Figure 4

Kaplan–Meier estimates of HCC development in ALC patients in each group who had similar residual liver function according to the albumin–bilirubin (ALBI) grade or fibrosis-4 (FIB-4) scores. Risk stratification for HCC development according to serum myostatin levels among patients who were in ALBI grade 1 (A), grade 2 (B), and grade 3 (C) (all p < 0.01 by log-rank test). Risk stratification for HCC development according to serum myostatin levels among patients who had low FIB-4 scores (≤6.46, median value) (D) and high FIB-4 scores (>6.46) (E) (all p < 0.001 by log-rank test). Abbreviations: ALBI, albumin–bilirubin grade; ALC, alcoholic liver cirrhosis; FIB-4, fibrosis-4; HCC, hepatocellular carcinoma.

Figure 5

Kaplan–Meier estimates of hepatic decompensation event (A) and liver-related death (B) in ALC patients according to serum myostatin levels (all p < 0.001 by log-rank test). Abbreviations: ALC, alcoholic liver cirrhosis.