Developing An Integrated Prognostic Score Using Germline Risk Variants And Rna Expression From Primary Breast Cancers For Prognostication Of Survival Across Subtypes.

Abstract Introduction: RNA expression offers prognostication in early-stage, hormone-receptor positive breast cancer. We hypothesized that germline risk variants may influence patients after breast cancer onset, and that integration of common germline risk variants with RNA expression may broaden prognostication across subtypes. Methods: We designed two prognostic scores for women of European ancestry using genome-wide association study (GWAS) summary statistics on cancer risk from the Breast Cancer Association Consortium (133,384 cases and 113,789 controls) and individual-level genotype/clinical information from 694 women in The Cancer Genome Atlas (TCGA). PrediXcan, a transcriptome-wide association tool family, was used to generate gene-level weights of association to breast cancer risk that were then multiplied by gene expression levels. A predicted transcriptomic score (PTRS) used predicted expression levels based on germline genotypes, and an observed transcriptomic score (OTRS) used normalized RNA expression. Cox proportional hazards tested score performance on overall survival (OS) and progression-free survival (PFS). Results: Estimation performance is shown in Table 1. A 1388-gene PTRS with range from -0.5 to 4 offered initial prognostic ability in OS across subtypes (HR: 0.69 per 1-point increase, 95% CI: 0.48-1.00, p-value 0.05). This trend persisted with clinical covariates in a multivariate regression (HR: 0.61, 95% CI: 0.36-1.00, p=0.06). No associations were seen between PTRS/PFS or OTRS/OS. OTRS/PFS trended toward statistical significance with clinical factors included (age, AJCC staging). PTRS for OS did not show subtype-specific performance. Conclusion: Patients with breast cancer do not have broad access to prognostic data across subtypes. Integration of germline and somatic data types offers potential to improve current limitations in prognostic testing for patients. Table 1.Performance of predicted transcriptomic score (PTRS) and observed transcriptomic score (OTRS)Cox regression hazard ratio95% CIp-valuePolygenic score - OS0.960.64-1.400.84PTRS-OS0.690.48-1.000.05OTRS-OS0.990.97-1.000.66Polygenic score - PFS1.100.77-1.600.55PTRS - PFS1.100.80-1.600.49OTRS - PFS0.980.96-1.000.10 Citation Format: Padma Sheila Rajagopal, Yanyu Liang, Alvaro Barbeira, Owen Melia, Jiamao Zheng, Yonglan Zheng, Toshio Yoshimatsu, Dezheng Huo, Guimin Gao, Olufunmilayo F. Olopade, Hae K. Im. Developing an integrated prognostic score using germline risk variants and RNA expression from primary breast cancers for prognostication of survival across subtypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 215.