Developing and validating polygenic risk scores for colorectal cancer risk prediction in East Asians

Abstract

AbstractImportanceSeveral polygenic risk scores (PRSs) have been developed to predict the risk of colorectal cancer (CRC); however, virtually all these PRSs were constructed in European descendants.ObjectiveTo develop and validate PRSs using data from large genome-wide association studies (GWAS) conducted in East Asians.Design, Setting, and ParticipantsPRSs were developed using GWAS data from 22,702 cases and 212,486 controls and validated in two case-control studies (1,454 Korean and 1,736 Chinese). PRSs were derived using three approaches: (1) leading risk variants in GWAS-identified CRC loci; (2) risk variants independently associated with CRC in East Asians that were identified via fine-mapping of known GWAS-identified risk loci; and (3) genome-wide risk prediction algorisms using LDpred2 and PRS-CS.Main Outcomes and MeasuresLogistic regression models were used to examine associations of PRSs with CRC risk by estimating odds ratios (ORs) with 95% confidence intervals (CIs) and area under the receiver operating characteristic curve (AUC).ResultsIn the validation sets, PRS115-EAS, a PRS with 115 GWAS-reported leading risk variants derived from East Asian data performed significantly better in discriminating CRC cases from non-cases than PRS115-EUR, a PRS derived using GWAS data from European descendants. In the Korea validation set, ORs per SD increase of PRS115-EAS were 1.63 (95%CI = 1.46 - 1.82; AUC = 0.63), compared with OR of 1.44 (95%CI = 1.29 - 1.60, AUC = 0.60) for PRS115-EUR. PRS115-EAS/EUR derived using results from meta-analyses of East Asian and European-ancestry data slightly improved the AUC to 0.64 (95% CI = 0.61 - 0.67). Similar, but less strong, associations were found in the China validation set. Individuals in the top 5% of PRS115-EAS/EUR were at a 2.52-folded (95% CI = 2.27 - 2.81) elevated risk of CRC as compared with the average risk group and have a 12% or higher risk of developing CRC by age 85 years.Conclusions and RelevanceOur results indicate that PRSs derived using GWAS-identified CRC risk variants are promising in predicting CRC risk in East Asians. Our study highlights the importance of using population-specific data to build CRC risk prediction models.Key PointsQuestionHow well do polygenetic scores (PRS) developed using data from genome-wide association studies (GWAS) predict colorectal cancer (CRC) risk in East Asians?FindingsUsing GWAS data from more than 230,000 cases and controls in East Asian, 11 PRSs were developed with variants ranging from 115 to 747,643. PRS including 115 variants derived from meta-analyses of East Asian and European-ancestry GWAS data showed the highest performance in discriminating CRC cases and controls in validation sets including 1,490 cases and 1,700 controls.MeaningThese findings support the utility of PRS in identifying high-risk individuals for CRC prevention and highlight the importance of using population-specific data to build CRC risk prediction models.