Cross-ancestry discovery of genetic risk variants for lean metabolic dysfunction-associated steatotic liver disease.

Abstract

Background: The presence of metabolic dysfunction-associated steatotic liver disease (MASLD) in lean patients challenges the traditional association between fatty liver and obesity. While genome-wide association studies (GWAS) have identified key genetic variants linked to hepatic steatosis, most have focused on Western obese populations and were based on the former non-alcoholic fatty liver disease (NAFLD) criteria. Genetic studies targeting lean MASLD, particularly among the Taiwan Han Chinese population, remain limited. This study addresses this knowledge gap by investigating the genetic architecture of lean MASLD in a Taiwan Han Chinese cohort.

Methods: We analyzed data from the Taiwan Biobank, integrating genomic data with abdominal ultrasound findings to define lean MASLD cases with BMI less than 23. A total of 1,720 lean MASLD individuals and 5,331 lean controls were included. A metabolic risk score ranging from 1 to 5 was calculated for each participant, reflecting the number of metabolic criteria met for MASLD classification. GWAS was conducted using logistic regression, adjusting for age, sex, and top 10 principal components. Conditional analyses were used to identify independent signals. Significant variants were replicated using data in the All of Us Research Program (1,017 lean MASLD and 1,553 lean healthy).

Results: Lean MASLD patients exhibited significantly worse metabolic profiles than lean controls. GWAS identified 47 genome-wide significant variants, predominantly in PNPLA3 and SAMM50, with rs9625962 in PNPLA3 emerging as the most significant and independent variant. These associations were replicated in the All of Us lean MASLD cohort, affirming their cross-population relevance. Notably, the C allele of rs9625962 increased the risk of MASLD in both lean and non-lean individuals. In the lean MASLD group, fibrosis risk scores (FIB-4 and NFS) positively correlated with the metabolic risk scores.

Conclusion: This study identified significant associations between genetic variants in PNPLA3 and SAMM50 and lean MASLD in the Taiwan Han Chinese population. The variant rs9625962 exhibited pleiotropic effects, contributing to MASLD regardless of BMI category, triglyceride levels, and liver function profiles. The replication of all significant variants in the All of Us dataset underscores the robustness and generalizability of our findings. These results highlight the importance of genetic factors in lean MASLD pathogenesis and support further investigation into gene-environment interactions.