On cross-ancestry cancer polygenic risk scores

Abstract

Polygenic risk scores (PRS) can provide useful information for personalized risk stratification and disease risk assessment, especially when combined with non-genetic risk factors. However, their construction depends on the availability of summary statistics from genome-wide association studies (GWAS) independent from the target sample. For best compatibility, it was reported that GWAS and the target sample should match in terms of ancestries. Yet, GWAS, especially in the field of cancer, often lack diversity and are predominated by European ancestry. This bias is a limiting factor in PRS research. By using electronic health records and genetic data from the UK Biobank, we contrast the utility of breast and prostate cancer PRS derived from external European-ancestry-based GWAS across African, East Asian, European, and South Asian ancestry groups. We highlight differences in the PRS distributions of these groups that are amplified when PRS methods condense hundreds of thousands of variants into a single score. While European-GWAS-derived PRS were not directly transferrable across ancestries on an absolute scale, we establish their predictive potential when considering them separately within each group. For example, the top 10% of the breast cancer PRS distributions within each ancestry group each revealed significant enrichments of breast cancer cases compared to the bottom 90% (odds ratio of 2.81 [95%CI: 2.69,2.93] in European, 2.88 [1.85, 4.48] in African, 2.60 [1.25, 5.40] in East Asian, and 2.33 [1.55, 3.51] in South Asian individuals). Our findings highlight a compromise solution for PRS research to compensate for the lack of diversity in well-powered European GWAS efforts while recruitment of diverse participants in the field catches up.

Polygenic Risk Score Improves the Accuracy of a Clinical Risk Score for Coronary Artery Disease

“…The extent to which our results can be applied to larger non-European ancestries, in particular African ancestry, warrants further investigation. These results also highlight the urgency of developing novel cross-ancestry PRS methods 10,17,53,54,55 and using more diverse cohorts to construct PRSs. 17 In addition, as the case-control and cohort analyses are derived from the same study, more broad generalizability of the results requires further investigation.…”

Genotype imputation and polygenic score estimation in northwestern Russian population

“…We used the nested and full models as described in Martin et al [1] to evaluate the concordance of the polygenic score with actual phenotypes. The full linear model was given as: phenotype ~PGS + age + age 2 + sex + sex � age + sex � age 2 + PC (1)(2)(3)(4)(5)(6)(7)(8)(9)(10) and the nested model contained all covariates as full, excluding PGS. R 2 attributed to PGS was estimated as the difference between R 2 of the full and the nested models.…”

“…Over the last decade, genome-wide association studies (GWAS) have discovered a substantial number of associated variants for many complex traits. Yet, the non-uniform representation of populations in genetic studies considerably limits the applicability of GWAS-based resources for individual risk prediction [1][2][3][4]. For example, Martin et al [1] showed that polygenic scores (PGS) are far more accurate for European individuals than for non-Europeans.…”

An empirical comparison between polygenic risk scores and machine learning for case/control classification

“…A similar analysis can be performed using machine learning, as illustrated in this manuscript. PRS is a more substantial quantity than classification in machine learning because PRS predicts a particular person's tendency to have a specific disease or trait [6,7]. In contrast, machine learning classifies people into traits or categories [8].…”

“…This table shows which operations can be skipped for machine learning (Step number: 3.1,3.2,3.3,3.4). Secondly, it shows the alternative of PRS calculation steps for machine learning classification (Step number: 4,5,5 1,6). …”