To investigate the efficacy of different first-line treatment methods in hormone receptor-positive, HER-2-negative (HR+/HER-2-) breast cancer patients with bone-only metastases (BOM).

Clinical data of 373 HR+/HER-2- breast cancer patients with BOM between October 2001 and November 2018 in the Senior Department of Oncology Medicine, Chinese PLA General Hospital were retrospectively analyzed. Patients were divided into two groups according to the different initial treatment, including chemotherapy (CT) group (initial CT group, n=165) and endocrine therapy (ET) group (initial ET group, n=208). Patients without disease progression at least 3 months after initial treatment were divided into different groups based on their maintenance therapy, including continuous CT (CT cohort), maintenance ET following initial CT (CT-ET cohort) and continuous ET (ET cohort). The clinicopathological characteristics and prognostic factors were analyzed between different treatment groups. The log-rank test was used for univariate analysis, and Cox regression model for multivariate analysis. Propensity score matching and standardized mean difference with inverse probability weighting were adopted to evaluate the balance between groups. Survival analysis was performed using the Kaplan-Meier method.

The median progression-free survival (PFS) of patients in the initial CT group and the initial ET group was 10 months (95%CI: 6.76-13.23) and 12 months (95%CI: 10.33-13.66), respectively, while the median overall survival (OS) was 61 months (95%CI: 45.42-76.57) and 52 months (95%CI: 41.28-62.71), respectively. No statistically significant difference was observed between the two groups (χ²=1.057, 1.044, both P>0.05). After propensity score matching, both initial CT and ET group had 106 cases, with median PFS of 12 months (95%CI: 8.70-15.29) and 14 months (95%CI: 11.48-16.51), indicating a significant difference (χ²=4.254, P=0.039); and the median OS was 68 months (95%CI: 49.20-86.79) and 64 months (95%CI: 51.21-76.79), indicating no significant difference (χ²=0.018, P=0.894). A total of 332 patients showed no disease progression for at least 3 months after initial treatment, including 58 in the CT cohort, 82 in the CT-ET cohort, and 192 in the ET cohort. The median PFS for the CT cohort, CT-ET cohort and ET cohort was 6 months (95%CI: 6.74-11.20), 19 months (95%CI: 21.60-30.43), and 13 months (95%CI: 15.45-19.62), indicating a significant difference (χ²=59.586, P<0.001). The median OS was 48 months (95%CI: 46.16-68.64), 72 months (95%CI: 70.78-93.54), and 54 months (95%CI: 61.40-80.91), indicating a significant difference (χ²=5.984, P=0.050). After inverse probability weighting, the median PFS was 7 months (95%CI: 5.00-12.00), 20 months (95%CI: 16.00-25.00), and 13 months (95%CI: 12.00-16.50) in the three corhort, with significant differences among groups (χ²=51.493, P<0.001). The median OS was 46 months (95%CI: 32.00 to not reached), 92 months (95%CI: 61.00-114.00), and 54 months (95%CI: 48.00-67.00), with no statistically significant difference (χ²=5.334, P=0.069). The Cox multivariate analysis results showed that different first-line treatment methods (CT-ET cohort vs CT cohort: HR=0.02, 95%CI: 0.01-0.06; ET cohort vs CT cohort: HR=0.57, 95%CI: 0.40-0.81) was an independent factor affecting PFS; the time from diagnosis to recurrence (≥24 months vs <24 months, HR=0.64, 95%CI: 0.46-0.89) was an independent influencing factor of OS.

In HR+/HER2- breast cancer patients with BOM, after propensity score matching, the median PFS of initial ET was better than that of initial CT among the patients in the front-line preferred endocrine group, and the median OS of the two groups of patients was similar. In patients without progression after initial first-line therapy, the CT-ET cohort yielded better outcomes than CT or ET cohort.