Plasma cathepsin D as an early indicator of alcohol-related liver disease

Abstract

Background & aims: People who drink alcohol excessively are at increased risk of developing metabolic dysfunction and alcohol-related liver disease (MetALD) or the more severe form alcohol-related liver disease (ALD). One of the most significant challenges concerns the early detection of MetALD/ALD. Previously, we have demonstrated that the lysosomal enzyme cathepsin D (CTSD) is an early marker for metabolic dysfunction-associated steatohepatitis (MASH). Here, we hypothesized that plasma CTSD can also serve as an early indicator of MetALD/ALD.

Methods: We included 303 persistent heavy drinkers classified as having MetALD or ALD (n = 152) and abstinent patients with a history of excessive drinking (n = 151). Plasma CTSD levels of patients with MetALD/ALD without decompensation were compared with 40 healthy controls. Subsequently, the relationship between plasma CTSD levels and hepatic histological scores was established. Receiver-operating characteristic curves were generated to assess the precision of plasma CTSD levels in detecting MetALD/ALD. Lastly, plasma CTSD levels were compared between abstainers and drinkers.

Results: Plasma CTSD levels were higher in patients with MetALD/ALD compared to healthy controls. While hepatic disease parameters (AST/ALT ratio, liver stiffness measurement) were higher at advanced histopathological stages (assessed by liver biopsy), plasma CTSD levels were already elevated at early histopathological stages. Furthermore, combining plasma CTSD levels with liver stiffness measurement and AST/ALT ratio yielded enhanced diagnostic precision (AUC 0.872) in detecting MetALD/ALD in contrast to the utilization of CTSD alone (AUC 0.804). Plasma CTSD levels remained elevated in abstainers.

Conclusion: Elevated levels of CTSD in the circulation can serve as an early indicator of MetALD/ALD.

Impact and implications: Alcohol-related liver disease is the leading cause of liver disease-related morbidity and mortality worldwide. However, the currently available non-invasive methods to diagnose MetALD/ALD are only able to detect advanced stages of MetALD/ALD. Here, we demonstrate that plasma levels of the lysosomal enzyme cathepsin D are already elevated at early stages of MetALD/ALD. Moreover, cathepsin D levels outperformed the currently available non-invasive methods to detect MetALD/ALD. Plasma levels of cathepsin D could therefore be a useful non-invasive marker for detection of MetALD/ALD.

Keywords: alcohol use disorder; alcohol-related liver disease; cathepsin D; persistent heavy drinking.

Graphical abstract

Fig. 1

Plasma CTSD levels and hepatic disease markers in MetALD/ALD patients and healthy controls. (A) AST/ALT ratio of healthy controls vs. patients with MetALD/ALD (healthy control, n = 31; MetALD/ALD, n = 148). (B) LSM of healthy controls vs. patients with MetALD/ALD (healthy control, n = 38; MetALD/ALD, n = 145). (C) CTSD in the plasma of healthy controls vs. patients with MetALD/ALD (healthy control, n = 38; MetALD/ALD, n = 152). ∗p <0.05, ∗∗∗∗p <0.0001 (Mann-Whitney test). All error bars are median with interquartile range. ALD, alcohol-related liver disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTSD, cathepsin D; LSM, liver stiffness measurement; MetALD, metabolic dysfunction and alcohol-related liver disease.

Fig. 2

Plasma CTSD levels and hepatic disease markers increased in patients with MetALD/ALD classified according to hepatic histological criteria. AST/ALT ratio, LSM and CTSD levels were analyzed in healthy controls (n = 40) and patients with MetALD/ALD (n = 152) and classified by liver biopsy-proven NAS steatosis score (0-3) (A-C), NAS inflamamtion score (including ballooning (0-2) and lobular inflammation (0-3)) (D-E), and fibrosis stage (0-4) (G-I) according to NAS. ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001, ∗∗∗∗p <0.0001 compared to healthy controls and ^#p <0.05, ^##p <0.01, ^###p <0.001, ^####p <0.0001 compared to score 0 MetALD/ALD patients (Kruskal-Wallis test). All error bars are median with interquartile range. ALD, alcohol-related liver disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTSD, cathepsin D; LSM, liver stiffness measurement; MetALD, metabolic dysfunction and alcohol-related liver disease; NAS, NAFLD activity score.

Fig. 3

Plasma CTSD levels in patients with MetALD/ALD stratified by AST/ALT ratio at cut-off level 2. ∗∗p <0.01, ∗∗∗∗p <0.0001 compared to healthy controls and ^#p <0.05 compared to MetALD/ALD patients with AST/ALT ratio ≤2 (indicated in graph as ‘AST/ALT ratio ≤2’) (Kruskal-Wallis test). All error bars are median with interquartile range. (healthy controls, n = 38; AST/ALT ratio ≤2, n = 116; AST/ALT ratio >2, n = 32). ALD, alcohol-related liver disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTSD, cathepsin D; MetALD, metabolic dysfunction and alcohol-related liver disease.

Fig. 4

Plasma CTSD levels correlate with hepatic fibrosis determinants in patients with MetALD/ALD. (A) Plasma CTSD levels vs. ELF (n = 152) and (B) plasma CTSD levels vs. LSM (n = 145). Spearman’s correlation was performed. p <0.05 was considered statistically significant. ALD, alcohol-related liver disease; CTSD, cathepsin D; ELF, enhance liver fibrosis; LSM, liver stiffness measurement; MetALD, metabolic dysfunction and alcohol-related liver disease.

Fig. 5

The diagnostic value of plasma CTSD, AST/ALT ratio and LSM. Receiver-operating characteristic curve analysis was performed to evaluate the AUC of plasma CTSD, AST/ALT ratio or LSM to predict MetALD/ALD. ALD, alcohol-related liver disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTSD, cathepsin D; LSM, liver stiffness measurement; MetALD, metabolic dysfunction and alcohol-related liver disease.

Fig. 6

The diagnostic value of adding CTSD to AST/ALT ratio and LSM. Receiver-operating characteristic curve analysis was performed to evaluate the AUC of using the combination of CTSD levels with AST/ALT ratio and LSM to predict MetALD/ALD. ALD, alcohol-related liver disease; ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTSD, cathepsin D; LSM, liver stiffness measurement; MetALD, metabolic dysfunction and alcohol-related liver disease.

Fig. 7

Plasma CTSD levels in drinkers and abstinent individuals. ^∗∗∗∗p <0.0001 compared to healthy controls, ^####p <0.0001 compared to drinkers (Kruskal-Wallis test). All error bars are median with interquartile range. (healthy controls, n = 38; drinking individuals, n = 152; abstinent individuals, n = 151).