Trends and outcomes of systemic therapy in hormone receptor-positive early-stage male breast cancer: A propensity score analysis.

557 Background: Male breast cancer (MBC) is a rare disease. Research on trends and survival outcomes in systemic treatment for MBC is limited. This study aims to describe trends and survival outcomes of different systemic treatments in hormone receptor-positive (HR+) early MBC and to assess the factors related to treatment selection. Methods: We identified stage I-III HR+ invasive MBC patients who have undergone surgery and systemic therapy diagnosed during 2004-2014 from the National Cancer Database (NCDB). Treatment groups include: endocrine therapy (ET), chemotherapy (ET), combination therapy (ET+CT), and None. Using Cochran-Armitage trend tests to describe the temporal trends of different treatments. Performing propensity score models to evaluate overall survival. Conducting a logistic regression to analyze factors associated with ET+CT. Results: Among 6217 patients included (median follow-up 50.1months), 1290 (20.7%) patients received no systemic therapy (None), 705 (11.3%) patients only received CT, 2358 (37.9%) patients only received ET, and 1864 (30.0%) patients received ET+CT. From 2004 to 2013, there was a significant increase in the use of ET ( P(trend) < .001), which was accompanied by a decrease in CT and no treatment (all P(trend) < .001). And the use of ET+CT showed no significant change. After using overlap weighting (OW) to control confounders, multivariable cox regression analysis showed that ET+CT was associated with improved survival compared with ET (hazard ratio [HR], 0.46; P< .001). Sensitivity analyses (including the inverse probability of treatment weighting (IPTW) (HR,0.49; P < .001), IPTW with stabilized weight (HR,0.49; P< .001), propensity score matching (PSM) (HR,0.51; P < .001), PS regression adjustment (HR,0.47; P < .001), and PS stratification (HR 0.44; P < .001)) and exploratory subgroup analyses (except for well-differentiated, stage I, and breast conservation + radiotherapy subgroups) indicated similar outcome. Factors associated with receiving ET+CT include younger age, private insurance, Charlson Comorbidity Index was 0, poorer differentiation, higher stage, mastectomy and radiotherapy, estrogen receptor-positive and progesterone receptor-negative, and HER2-positive. Conclusions: ET+CT was associated with improved survival compared with ET in HR+ early MBC patients, except for some low-risk subgroups.