Development of a Polygenic Risk Score for Metabolic Dysfunction-Associated Steatotic Liver Disease Prediction in UK Biobank

Abstract

Background: Metabolic dysfunction-associated steatotic liver disease (MASLD) is the leading cause of liver-related morbidity and mortality. Although the invasive liver biopsy remains the golden standard for MASLD diagnosis, Magnetic Resonance Imaging-derived Proton Density Fat Fraction (MRI-PDFF) is an accurate, non-invasive method for the assessment of treatment response. This study aimed at developing a Polygenic Risk Score (PRS) to improve MRI-PDFF prediction using UK Biobank data to assess an individual's genetic liability to MASLD.

Methods: We iteratively sequestered 10% of MRI-PDFF samples as a validation set and split the rest of each dataset into base and target partitions, containing GWAS summary statistics and raw genotype data, respectively. PRSice2 was deployed to derive PRS candidates. Based on the frequency of SNP appearances along the PRS candidates, we generated different SNP sets according to variable frequency cutoffs. By applying the PRSs to the validation set, we identified the optimal SNP set, which was then applied to a Greek nonalcoholic fatty liver disease (NAFLD) study.

Results: Data from 3553 UK Biobank participants yielded 49 different SNP sets. After calculating the PRS on the validation set for every SNP set, an optimal PRS with 75 SNPs was selected (incremental R² = 0.025, p-value = 0.00145). Interestingly, 43 SNPs were successfully mapped to MASLD-related known genes. The selected PRS could predict traits, like LDL cholesterol and diastolic blood pressure in the UK Biobank, as also disease outcome in the Greek NAFLD study.

Conclusions: Our findings provide strong evidence that PRS is a powerful prediction model for MASLD, while it can also be applied on populations of different ethnicity.

Keywords: UK Biobank; metabolic dysfunction-associated steatotic liver disease; polygenic risk score.

Figure 1

Flow chart of UKBB participants who met the inclusion/exclusion criteria for the study.

Figure 2

After completion of iterative and aggregation PRS derivation process, different sets of SNP contents were generated depending on frequency cutoffs. Validation of the 68 generated SNP sets resulted in 20 optimal SNP sets that were statistically significant and exhibited explanatory power for the MRI-PDFF validation dataset. X-axis: number of SNPs per SNP set; y-axis: incremental R², color-coding denotes the p-value. For the graph illustration, ggplot2 (v 3.5.1) [28] was used.

Figure 3

A final 75 SNPs containing PRS was selected based on high predictive value and low SNP content. The selected PRS was also evaluated for its predictive ability of NASH score in the Greek NAFLD study. The model that was generated when the 75 SNP set of the optimal PRS was applied yielded notable metrics, including p-value = 0.009 and incremental R² value = 0.003.

Figure 4

A final 75 SNPs containing PRS was selected based on high predictive value and low SNP content. The optimal PRS was applied to the whole set of UK Biobank samples and evaluated for its predicted ability of MRI-PDFF (p-value = 0.001 and incremental R² value = 0.025).

Figure 5

Phenotypic variance of UKBB explained for MASLD-related markers by the optimal 75 SNP-containing PRS. The graph illustration was performed using the software GraphPad Prism version 5.03.

Figure 6

Enrichment analysis of the gene set in order to investigate molecular functions in which genes associated with the SNP set of the optimal PRS are involved.