Competing Mortality, Not Breast-Cancer Mortality, Explains the Survival Deficit of Breast Cancer After a Prior Non-Breast Cancer: A SEER Competing-Risks Analysis

Background and Objectives: Women who develop breast cancer after a previous cancer of another site have poorer overall survival, yet it is unsettled whether the breast cancer itself is more lethal or whether the deficit reflects competing causes of death in an older, previously treated population. We decomposed this survival deficit by cause of death, defining every endpoint identically across groups. Materials and Methods: Using the Surveillance, Epidemiology, and End Results (SEER) database (17 registries, 2010 to 2020), we studied women with invasive breast cancer as a single primary (n = 473,073), as a second or later primary after a non-breast index cancer (second primary breast cancer, SPBC; n = 26,832), and as a second breast primary (n = 40,252). Cause of death was assigned from the SEER cause-of-death-to-site recode, applied identically across cohorts. Cohorts were compared using cause-specific and Fine–Gray models weighted by an inverse-probability propensity score for age, race and ethnicity, subtype, grade, stage, nodal status, year, marital status, income, and rurality. Results: Breast-cancer mortality was not higher in SPBC than in single-primary disease, and was if anything lower (10-year cumulative incidence 12.0 versus 13.6 percent; weighted cause-specific hazard ratio 1.06). The deficit fell almost entirely to competing, non-breast mortality (35.3 versus 13.5 percent). In a five-way decomposition, cardiovascular disease was not the leading competing cause; non-cancer causes and the index cancer each accounted for more deaths. Once patient factors were balanced, cardiovascular and other non-cancer hazards were near unity (1.05 and 1.07), leaving further cancer death as the main excess. Conclusions: A breast cancer arising after a non-breast cancer is not intrinsically more lethal. Its survival deficit is competing mortality, driven mainly by further cancer death and, to a smaller and largely age-explained degree, by cardiovascular and other non-cancer causes.