Coordinated patterns of genomic instability and immunogenicity predict 10-year breast cancer progression across subtypes and patient characteristics

Genomic instability is associated with breast cancer immune response. We sought to evaluate joint effects of genomic instability and immune response on recurrence using validated RNA-based assays and statistical models for assessing joint expression in a population-based dataset ( N = 2783). Prognostic value of the resulting coordinated signature, Repair, Immune, and Genomics for Oncologic Reclassification (RIGOR150), was assessed in estrogen receptor (ER)+ and ER− breast cancers ( N = 1899 with recurrence information) and validated in public datasets. RIGOR150 was strongly associated with recurrence risk in training (Hazard Ratio [HR] for high vs. low-risk = 5.10, 95% confidence interval [CI] = 3.19, 8.15) and test data (HR = 3.07, 95% CI = 1.73, 5.45) and was validated in external datasets. Among ER+ tumors, RIGOR150 high-risk was associated with early (0–5 years, HR = 5.75, 95% CI = 3.38, 9.77) and late recurrence (5–10 years, HR = 2.67, 95% CI = 1.26, 5.65). RIGOR150 added prognostic value independent of existing risk predictors (21-gene recurrence score and PAM50), and was associated with neoantigens, proliferation, and tumor-immune cell proximity. Overall, joint patterns of genomic instability and immunogenicity augment current breast cancer prognostic scores, providing early and late prognostication.