Association of Survival with Locoregional Therapy (Surgery and Radiation) Following Systemic Treatment for Metastatic Inflammatory Breast Cancer

Abstract

Inflammatory breast cancer (IBC) is an aggressive and rare subtype that occurs in less than 5% of all breast cancers. Underusage of trimodality therapy (surgery, radiation and chemotherapy) for Stage III IBC has been described, despite improved overall survival (OS) benefit. De-novo metastatic IBC patients (Stage IV) have a dismal 5-year survival rate of 18% yet they account for 30% of all IBC cases. We evaluated the survival benefit of locoregional treatment (LRT) of surgery (Sg) and/or radiation (RT) following systemic treatment (ST) for metastatic IBC. 1,051 patients diagnosed with metastatic IBC from 2004-2015 and received ST with or without LRT were identified from the National Cancer Database (NCDB). Demographic, clinicopathologic and treatment variables were obtained. We excluded receipt of immunotherapy and patients likely upstaged during LRT and restricted a subset analysis to receipt of LRT following ST. Four treatment groups were defined: ST + adjuvant RT; ST + adjuvant Sg; ST + adjuvant Sg and RT; and ST only. Markov chain Monte Carlo (MCMC) estimation was used for imputation analysis of missing data. Differences in the distribution of time to the first treatment were assessed with non-parametric Kruskal-Wallis test. 3-year survival rates and median OS were assessed by Kaplan-Meier method. Other statistical computations were performed on SAS 9.3 system or GraphPad Prism software. Of 1,051 patients, 1,044 (99.3%) were female with average age of 57 years (interquartile range, IQR = 48-66). 83 (7.9%) had RT with ST; 181 (17.2%) had Sg with ST; 266 (25.3%) had Sg + adjuvant RT with ST; and 521 (49.6%) had only ST. The median OS was 23.4 months, with a 3-year survival rate of 33.9%. The 3-year survival rates for RT with ST, Sg with ST, Sg + adjuvant RT with ST and ST only were 36.6%, 39.0%, 52.6%, and 21.7%, respectively, P< 0.0001. On multivariate analysis (MVA), patients undergoing Sg with ST (adjusted HR 0.59, 95% CI 0.48-0.73, P< 0.0001) and Sg + adjuvant RT with ST (adjusted HR 0.43, 95% CI 0.36-0.52, P< 0.0001) had improved OS compared to only ST. This was consistent in the presence of hormone receptor status. In the multivariate cohort, 349 de novo patients had ST before LRT. 80% had definitive RT (40.5-66.4 Gy) and 92% had at least a simple total mastectomy. 3-year survival rates for ST + adjuvant RT, ST + adjuvant Sg, ST + adjuvant Sg and RT, and ST only groups were 34.9%, 45.0%, 49.3%, and 21.9%, respectively, P< 0.0001. On MVA, ST + adjuvant Sg (adjusted HR 0.54, 95% CI 0.43-0.69, P< 0.0001) and ST + adjuvant Sg and RT (adjusted HR 0.42, 95% CI 0.34-0.52, P < 0.0001) had improved OS compared to ST only. The significant association of overall survival benefit with trimodality therapy suggests a role for local therapy in the treatment of patients with de-novo Stage IV inflammatory breast cancer. Due to potential biases associated with a retrospective review, in order to make definitive conclusions about locoregional benefit, a randomized controlled trial is necessary.