Background: Asymptomatic metastatic breast cancer (mBC) is often detected using tumor marker or imaging tests in Japan. At the ASCO 2019, we reported on mBC detection, the distribution of symptomatic and asymptomatic disease by subtype (abstract e12568). We aimed to determine whether there are differences in post-recurrence survival (PRS), and treatment between asymptomatic and symptomatic MBCs, and identify factors associated with PRS. Methods: We performed a multicenter, retrospective analysis of patients with mBC treated in our hospitals from 2008 to 2018. Patients were divided into asymptomatic and symptomatic MBCs to compare their prognosis by breast cancer (BC) subtypes: luminal (hormone receptor positive/human epidermal growth factor 2 [HER2] negative), HER2 (any hormone receptor/HER2 positive), and triple-negative (TN) (hormone receptor negative/HER2 negative). Results: Of 204 patients with mBC (114 asymptomatic, 90 symptomatic), PRS was longer in asymptomatic mBC than in symptomatic mBC (median survival: 55 months vs. 29 months; p< 0.001) and tended to have longer overall survival (OS) (110 months vs 72 months, respectively; p = 0.09). In multivariate analysis, TN, recurrence-free survival (RFS), multiple metastasis sites, and symptomatic disease were independently predictive of PRS (Table). In luminal and HER2, PRS trended higher in the asymptomatic group than in the symptomatic group (luminal: 54 months vs. 41 months; p = 0.08, HER2: 71 months vs. 27 months; p = 0.09), but without significant difference in OS (luminal: 112 months vs. 124 months; p = 0.91, HER2: 113 months vs. 72 months; p = 0.40). In the luminal group, 13 patients (11%) were treated with cyclin-dependent kinase 4/6 (CDK4/6) inhibitors. The median PRS was 80 months for luminal patients with three factors — longer than 4 years of RFS, a single metastasis site, and asymptomatic disease. The duration of endocrine therapy did not differ between groups; however, the patients with luminal BC in the asymptomatic group had longer chemotherapy than those in the symptomatic group. In TN, PRS was very short (asymptomatic, 28 months; symptomatic, 10.5 months; p = 0.01). Conclusions: We demonstrated that asymptomatic MBC and symptomatic MBC differed in terms of subtypes, prognosis, and duration of chemotherapy in the luminal group. Therefore, unique treatment strategy for asymptomatic or symptomatic MBC should be developed.

HR = hazard ratio; CI = confidence interval; TN = triple-negative; RFS = recurrence-free survival.