Differential results of genetic risk scoring for multiple sclerosis in European and African American populations

Genetic risk scores (GRSs) for multiple sclerosis (MS) help identify high-risk individuals and stratify populations for clinical trials, but most are derived from European populations, raising questions about GRS accuracy in other ancestries.

We aimed to determine whether an MS GRS can stratify individuals of African and Latino/admixed ancestry and assess whether the JointPRS tool enhances GRS portability in African ancestry.

In this cross-sectional study using All of Us (2018-2022), we derived a GRS from 232 variants for 32,428 European, 32,428 African, and 32,428 Latino/admixed participants, each divided into quintiles. The outcome was MS ascertained through International Classification of Diseases (ICD)-9/10 and Systematized Nomenclature of Medicine (SNOMED) codes. JointPRS was used to improve GRS portability in the African-ancestry group.

MS prevalence was 1.0% in European, 0.56% in African, and 0.46% in Latino/admixed participants. The GRS stratified MS risk effectively in European (odds ratio (OR) = 2.30 (1.60-3.36); p-trend < 0.001) and Latino/admixed (OR = 2.53 (1.43-4.85); p-trend < 0.001) ancestry groups but did not significantly partition African participants (OR = 1.30 (0.88-1.95); p-trend = 0.17). After applying JointPRS, stratification in African ancestry improved (OR = 3.02 (1.00-8.95); p-trend = 0.007).

A GRS for MS stratified European and Latino/admixed individuals but not African ancestry. Incorporating African-specific data enhanced performance, underscoring the need for more ancestry-tailored GRS.