T. Chanenchuk1, S.M. Thomas2,3, T. Wang1,2, K.J. Modell Parrish1,2, A. Chiba1,2, A. Botty Van Den Bruele1,2, L.H. Rosenberger1,2, J.K. Plichta1,2 1Duke University Medical Center, Department Of Surgery, Durham, NC, USA 2Duke University Medical Center, Duke Cancer Institute, Durham, NC, USA 3Duke University Medical Center, Department Of Biostatistics And Bioinformatics, Durham, NC, USA
Introduction: Given improvements in systemic therapy for de novo metastatic breast cancer (dnMBC), a novel staging system stratified patients into prognostic subgroups (IVA/B/C/D), based on tumor characteristics (size, grade, receptor), and extent of metastases. The overall survival (OS) in these subgroups varies from >6 years (IVA) to <1 year (IVD), therefore, it is possible that different treatment approaches may be considered based on stage subgroup. We aim to explore the potential association of surgery of the primary tumor with OS among the subgroups.
Methods: Patients with dnMBC (2010-2020) who received at least 1 systemic therapy were selected from the NCDB and stratified by stage group (IVA-D). Surgery patients with less than 30 days of follow-up after surgery were excluded, as were non-surgery patients with less than 11 months of follow-up after diagnosis (timepoint at which 90% of surgery patients had surgery). OS was estimated from date of surgery and from the date of diagnosis plus 11 months for non-surgery patients, to account for potential lead time bias. The Kaplan-Meier method was used to estimate OS, and log-rank tests were used to test for differences. Cox Proportional Hazards models were used to estimate the association of surgery with OS, after adjustment.
Results: The study included 44,178 patients: IVA 3,163 (7.2%), IVB 24,291 (55.0%), IVC 12,162 (27.5%), IVD 4,562 (10.3%). Median follow-up was 54.5 mo. Surgery receipt varied by subgroup: IVA 34.3%, IVB 23.8%, IVC 22.3%, IVD 32.6% (p<0.001). After adjustment, patients were more likely to undergo surgery if they were younger, had fewer comorbidities, had private insurance, and were treated at a non-academic facility (all p<0.05). Compared to patients with IVA disease, those with less favorable disease (IVB and IVC) were less likely to undergo surgery (both p<0.001), although those with IVD disease (worst prognosis) had similar odds of undergoing surgery to IVA (p=0.83).
Median OS varied by subgroup: IVA 87.5mo, IVB 50.4mo, IVC 31.1mo, IVD 14.9mo. Among those with IVA disease, surgery patients had a better unadjusted 5-yr OS (78.8% vs 49.7%, log rank p<0.001). Although the absolute survival rates were lower, patients with less favorable disease (IVB-D) who underwent surgery had a significantly higher OS than those who did not (Figure). After adjustment, surgery receipt remained associated with improved OS (HR 0.59, 95% CI 0.56-0.61).
Conclusion: Among those with dnMBC, a significant number of patients underwent surgery, even among those with less favorable prognosis. Although the absolute survival benefit associated with surgery may be lower with more advanced disease, resection of the primary tumor may be considered for select dnMBC patients.