Background: Patients who complete definitive therapy for primary, stage I-III breast cancer may develop recurrent disease. However, little is known about the outcomes patients experience after recurrence when treated outside of a clinical trial because population-based datasets usually do not capture recurrence status. We describe the overall survival of patients after developing recurrent breast cancer, identify factors independently associated with improved survival, and compare survival for patients with recurrent versus de novo stage IV metastatic disease.

Methods: The cancer registries from two Kaiser Permanente (KP) sites participating in the Cancer Research Network, KP Colorado and KP Northwest, provided data on adult women diagnosed 2000-2011 with primary breast cancer and followed through death, disenrollment or study end (12/31/2012). Among patients with stage I-III disease who completed definitive therapy, recurrence was captured via manual chart abstraction. Survival time was calculated from the date of recurrence or the date of de novo stage IV disease. Multivariable modeling identified factors independently associated with restricted mean survival time (RMST) through 7 years, controlling for age, race, income, co-morbidity, year of recurrence, time from primary diagnosis to recurrence, type of recurrence (local vs. regional/distant), use of chemotherapy or radiation at recurrence, and characteristics of the primary cancer that pre-dated recurrence (i.e., primary stage, grade, hormone-receptor status, and use of chemotherapy or radiation therapy for the primary diagnosis). We compared overall survival after developing recurrent versus de novo stage IV disease after matching for age, race, income, co-morbidity and year.

Results: From 7,216 breast cancer diagnoses we identified 506 cases of recurrent disease and 219 cases of de novo stage IV disease (7% and 3%, respectively). Most recurrences were regional or distant (81%). From the time of recurrence, median survival was 21 months and 2-year survival was 47%. Factors significantly associated with inferior RMST included regional/distant vs. local recurrence (-35.6 months; P<.01), primary stage III vs. stage I disease (-13.6 months; P<.01), and chemotherapy for the primary diagnosis (-9.6 months; P=0.02), but not race, income, grade, or primary cancer hormone-receptor status. Patients for whom the interval from diagnosis to recurrence was >4 years vs. <1 year had a longer RMST (+18.9 months; P<.0-1). Receipt of chemotherapy at the time of recurrence was associated with inferior RMST; the magnitude of this association was higher among patients with local (-18.5 months) versus regional/distant disease (-3.2 months). Women with regional/distant recurrence had significantly worse RMST than those with de novo stage IV disease (-10.3 months; P<.01).

Conclusions: Recurrent breast cancer is at least two-fold more common than de novo stage IV disease. Among patients who develop recurrence, characteristics of the primary cancer and its treatment are associated with survival after recurrence. Survival differences between patients with recurrent and de novo stage IV disease suggest that prognostic estimates and treatment paradigms should be tailored.

Citation Format: Hassett MJ, Cronin AM, Carroll NM, Uno H, Hornbrook MC, Ritzwoller D. The incidence of and survival after breast cancer recurrence. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P1-07-07.