Association of facility volume with pathologic complete response and overall survival in patients with non-metastatic breast cancer.

Abstract

586 Background: High facility volume has been previously shown to be associated with a higher likelihood of pCR among patients with breast cancer following neoadjuvant chemotherapy. However, factors underlying the association of high facility volume with pCR and the impact of facility volume on survival remain unclear. We performed an observational cohort study to investigate the association of overall survival (OS) with facility volume and identify variables associated with high facility volume. Methods: The National Cancer Database (NCDB) was queried for women diagnosed between 2010 and 2017 with non-metastatic breast cancer treated with neoadjuvant chemotherapy followed by surgery. Facility volume was stratified by tertiles of low, intermediate, and high. Logistic and Cox multivariable analyses (MVA) were performed to evaluate factors associated with facility volume and OS, respectively. Propensity score matching was used to reduce selection bias. Results: A total of 105,804 women met our inclusion criteria (low facility volume: n=6,172; intermediate facility volume: n=20,199; high facility volume: n=79,433). The median follow up was 49.2 months (IQR 32.7-71.3). On logistic MVA, both intermediate (adjusted odds ratio [aOR] 1.24, 95% confidence interval [CI] 1.14-1.35, p<0.001) and high (aOR 1.53, 95% CI 1.42-1.65, p<0.001) facility volumes were associated with pCR. On Cox MVA, high (adjusted hazards ratio [aHR] 0.89, 95% CI 0.83-0.96, p=0.001), but not intermediate (aHR 0.97, 95% CI 0.90-1.04, p=0.40), facility volume was associated with improved OS. Similar findings were noted in 18,671 matched pairs between high versus low or intermediate facility volumes (HR 0.90, 95% CI 0.85-0.94, p<0.001). On logistic MVA, patients were more likely to receive treatments at high volume facilities if they were academic (adjusted odds ratio [aOR] 5.47, 95% CI 5.24-5.71, p<0.001), African American (aOR 1.25, 95% CI 1.20-1.30, p<0.001) with private insurance (aOR 1.13, 95% CI 1.04-1.22, p=0.005), and higher tumor grades (grade 2: aOR 1.10, 95% CI 1.03-1.18, p=0.008; grade 3: aOR 1.12, 95% CI 1.05-1.20, p=0.001). Patients were less likely to undergo treatments at high volume facilities if they reside in urban (aOR 0.63, 95% CI 0.60-0.66, p<0.001) or rural (aOR 0.75, 95% CI 0.67-0.85, p<0.001) areas with limited education (aOR 0.90, 95% CI 0.87-0.93, p<0.001), had government-led insurance (aOR 0.87, 95% CI 0.80-0.94, p=0.001), and had time interval of <180 or >240 days between diagnosis and surgery (aOR 0.92, 95% CI 0.90-0.95, p<0.001). Conclusions: To our knowledge, this is the largest study using a nationwide oncology database to report the association of high facility volume with pCR and improved OS, with various patient demographics associated with high facility volume. Our findings suggest an integral role of facility volume on pCR and survival for breast cancer.