Abstract P1-10-01: Utilization and outcomes of post-mastectomy radiotherapy in women with 1-3 positive axillary nodes following neoadjuvant chemotherapy: A multiply-imputed, propensity-adjusted National cancer database analysis

Abstract

Abstract Background: Use of post-mastectomy radiotherapy (PMRT) following neoadjuvant chemotherapy (NAC) varies widely. The combined analysis of NSABP B1724 studies suggests a higher risk of locoregional recurrence for women with persistently positive lymph nodes following NAC. The impact of PMRT on survival in the subset of patients with 1 to 3 positive nodes after NAC is not clear and the goal of the current study is to evaluate practice patterns and the impact of PMRT on survival in this subset of patients. Methods: The National Cancer Database was queried for women who underwent NAC followed by mastectomy and had 1-3 persistent axillary lymph nodes with or without adjuvant PMRT. A propensity score was generated to account for indication bias under two circumstances: one with the original dataset and one following a multiple imputation to complete the dataset to allow for a more accurate propensity score generation. Factors impacting utilization of PMRT were calculated with backwards-selection binary regression. Kaplan-Meier with log-rank test and Cox-regression analyses were used for survival. Results: Within the queried cohort of 14,895 women from 2006-2012, 70% of women received PMRT following NAC with 1-3 positive axillary lymph nodes. One lymph node was positive in 56%, 2 in 25.9%, and 3 in 18.1%. PMRT was utilized in 67.3% of women with 1 node positive, 71.5% with 2 nodes positive, and 76.2% with 3 nodes positive. Factors associated with the usage of PMRT were: residence on the east coast in a highly populated region with proximity to the treatment center, later year of diagnosis (71% of women in 2012), younger age, non-African American race, non-Hispanic descent, private insurance, higher clinical stage, triple negative/high grade histology, pathologic T4, greater number of axillary node positive, positive margins or LVSI, greater number of nodes examined, and use of adjuvant hormonal therapy (all p<.05). Median follow up was 40.1 months (range: 2.6-106.9 months). Prior to multiple imputation, a propensity score was generated which accounted for 28.7% of cases. PMRT did not have an impact on survival within the limited analysis. In a pooled propensity-adjusted Cox analysis following multiple imputation to complete the dataset, PMRT was found to have an OS HR of 0.878 (95%CI 0.806-0.957, p=.003). Five year overall survival rates with and without PMRT were 77.6% and 75.4% (p<.001). For 1-3 positive nodes respectively, 5-year overall survival rates with and without PMRT were: 79.7% vs. 79.1% (p=.041), 77.3% vs. 70.8% (p<.001), and 71.8% vs. 68.0% (p=.004). Conclusions: In a population-based analysis of women who underwent neoadjuvant chemotherapy prior to mastectomy, post-mastectomy radiotherapy appears to confer a survival advantage in women with 1-3 residual axillary lymph nodes. There continues to be a wide variation in practice nation-wide with a significant number of women with persistently positive 1-3 axillary lymph nodes not receiving PMRT. Citation Format: Horne ZD, Gebhardt BJ, Balasubramani GK, Beriwal S. Utilization and outcomes of post-mastectomy radiotherapy in women with 1-3 positive axillary nodes following neoadjuvant chemotherapy: A multiply-imputed, propensity-adjusted National cancer database analysis [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P1-10-01.