Distant relapse metastatic breast cancer (rMBC) incidence and survival are vital measures of breast cancer diagnosis and treatment progress over time.

We conducted a retrospective cohort study of stage I-III invasive breast cancer, 1990-2011, follow-up through 2016 [N = 8292, rMBC = 964 (12%)] at a community-based institution. Patient and tumor characteristics (treatment, distant recurrence, vital status) from BC registry data were evaluated. Survival analysis and Cox proportional hazards (HzR) with 95% confidence intervals (95% CI) were calculated using distant recurrence and distant disease-specific survival (DDSS) endpoints.

Both 5- and 10-year distant relapse (rMBC) declined over time from 1990-1998 to 2005-2011 [11% to 5%, 16% to 8% (p < 0.001)]. Proportionately, HER2 + BC distant relapse decreased 9% and triple negative (HR-/HER2-) increased 8% (p = 0.011). In the Cox model, lower stage [stage I: HzR = 0.08 (0.07, 0.10), stage II: 0.29 (0.25, 0.33)], more recent diagnosis years [1999-2004: HzR = 0.60 (0.51, 0.70), 2005-2011: HzR = 0.44 (0.38, 0.52)], HR+ [HzR = 0.62 (0.53, 0.72)], and age 40+ [HzR = 0.81 (0.67, 0.98)] had decreased rMBC risk. Compared to HR+/HER2- BC, triple-negative BC had increased rMBC risk [HzR = 2.02 (1.61, 2.53)] but HER2+ subtypes did not. HR-, age 70+, > 1, or visceral metastases and stage III disease were associated with worse DDSS. DDSS did not improve over time.

rMBC incidence declined over time with decreased HER2-positive distant recurrence, a shift to more triple-negative BC and consistently poor distant disease survival.