Abstract 3932: Values of single nucleotide polymorphisms identified from genome-wide association studies on risk prediction, risk reclassification, and population-based screening of breast-cancer

Abstract

Abstract Background: With an increasing number of single nucleotide polymorphisms (SNPs) being identified from genome-wide association studies (GWAS), it's important to examine whether these SNPs have values in public health. And it's also unclear how to select the targeted SNPs valuable for population-based screening from numerous GWAS-identified SNPs. Methods: Based on the fraction of the genetic risk explained by a SNP, the area under the receiver operating characteristic curve (AUC), integrated discrimination improvement, and net reclassification improvement were used to select targeted SNPs and to evaluate the values of genetic risk score (GRS) with targeted SNPs on risk prediction, risk reclassification, and effects of population-based screening among 2 million simulated Chinese women aged 35-69 years. Results: A total of 11 targeted SNPs from 23 GWAS-identified SNPs would be valuable for population-based screening of breast cancer among Chinese women. After incorporating GRS with targeted SNPs into traditional risk predication based on established risk factors (ERF) of breast cancer, the AUC significantly improved from 65.5% to 67.5% (P<0.001). Additional 25.5% (P<0.001) population would be correctly reclassified into cases or controls. If screening targeted at top 25% high-risk women, screening-detected cancer would significantly increase 4.2% (49.2% vs 45.0%, P<0.001) for screening strategy included GRS and ERF in risk assessment before screening compared with that only included ERF in risk assessment. Conclusion: GRS based on targeted GWAS-identified SNPs could significantly improve the risk prediction, risk reclassification, and the effects of population-based screening. Further real-world studies are needed to validate this method and these results. Citation Format: Yubei Huang, Fengju Song, Kexin Chen. Values of single nucleotide polymorphisms identified from genome-wide association studies on risk prediction, risk reclassification, and population-based screening of breast-cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3932.