P101. Surgical management of de novo stage IV breast cancer over the past decade: A propensity score-matched analysis

Surgical resection of de novo stage IV breast cancer (BC) remains controversial in light of inconclusive data on definitive survival benefit. We compared outcomes and trends in primary tumor surgery utilization among a large institutional cohort of de novo stage IV BC patients.


Methods:

This was a retrospective review of all stage IV BC patients treated at a comprehensive cancer center (2014-2022). Linear regression and mixed-effects models were used to identify temporal variability among patients undergoing primary breast surgery. 1:2 propensity score matching was performed to select no surgery (systemic therapy only) and surgery cohorts matched by age, race/ethnicity, receptor subtype, histology, and grade. Kaplan-Meier method, log-rank test, and multivariable Cox proportional hazards model were used for survival analysis.

 


Results:

In total, 1290 patients with de novo stage IV BC were treated at our institution over a 9-year period. Of these, 16.7% (215) underwent primary tumor resection after systemic therapy. Compared to the no surgery cohort, patients undergoing surgery were younger (median age 49 vs. 55) with higher nodal burden (N2-3: 48.8% vs. 28.2%), higher grade (52.1% vs. 31.4%) and greater rate of HER2+ disease (34.4% vs. 19.8%) (all p< 0.01). There were no significant temporal differences in the overall proportion of stage IV patients receiving surgery (p=0.46); however, surgery rates significantly decreased over time for clinical T2 (p=0.042) and N1 tumors (p=0.02) and increased for patients with inflammatory breast cancer (p=0.037). After matching, 199 and 398 patients were included in the surgery and no surgery cohorts, respectively. Surgical resection conferred a survival benefit on multivariable Cox regression hazards model both before (HR 0.33, 95% CI: 0.35-0.44, p< 0.001) and after matching (HR 0.42, 95% CI: 0.29-0.62, p< 0.001). Matched stage IV BC patients treated with surgery had significantly higher 5-year overall survival rate of 66.1% (95% CI: 59.3-73.7%) versus 32.2% (95% CI: 25.3-40.9%) for those receiving systemic therapy alone (p< 0.0001).

 


Conclusions: Despite temporal variability in primary tumor resection for de novo stage IV BC at our institution, surgery conferred a significant survival benefit compared to systemic therapy alone in our patient cohort. While selection bias cannot be ruled out, the large magnitude of difference observed between the two cohorts, even after propensity score matching, warrants re-exploration of surgical intervention with curative intent for stage IV BC patients.