Evaluating Polygenic Score Transferability for Lipid Traits in Underrepresented Populations: Evidence from Samoan...

Area of Science:

• Genetics and Genomics
• Cardiovascular Disease Research
• Population Health

Background:

• Dyslipidemia is a major risk factor for cardiovascular disease (CVD), the leading cause of death in Samoa.
• Polygenic scores (PGS) are emerging tools for CVD risk prediction, but their utility in underrepresented populations like Pacific Islanders is largely unknown.
• Pacific Islanders are significantly underrepresented in genetic studies, including genome-wide association studies (GWAS).

Purpose of the Study:

• To evaluate the transferability and performance of multi-ancestry polygenic scores (PGS) for lipid traits in a Samoan population.
• To establish the first systematic benchmarks for lipid PGS in Samoans.
• To identify critical factors for successful PGS application in underrepresented ethnic groups.

Main Methods:

• Utilized data from 4,342 Samoan adults across five cohorts (1990-2010).
• Harmonized multi-ancestry PGS for LDL cholesterol (LDL-C), HDL cholesterol (HDL-C), triglycerides (TG), and total cholesterol (TC) with genome-wide imputed genotypes using a Samoan-specific reference panel.
• Assessed PGS performance using incremental R² from linear mixed models with bootstrapped confidence intervals.

Main Results:

• Lipid PGS demonstrated meaningful transferability in Samoans, with HDL-C showing the highest performance (incremental R² 5.0-15.0%).
• Performance varied by lipid trait: TC (5.0-10.7%), LDL-C (5.7-8.6%), and TG (3.5-7.0%).
• Effective LDL-C prediction required genome-wide variant matching (PRS-CS score), while limited matching yielded near-zero performance.

Conclusions:

• Lipid PGS can be meaningfully transferred to Samoan populations, providing valuable insights for CVD risk prediction.
• Ensuring comprehensive genome-wide variant coverage through methods like PRS-CS is critical for successful PGS transferability.
• Variant harmonization and population-specific reference panels are essential preconditions for deploying PGS in underrepresented groups.