Outcomes and Toxicities with Partial Breast Irradiation vs. Whole Breast Irradiation: A Meta-Analysis

Abstract

Whole breast irradiation (WBI) has traditionally been considered the primary adjuvant radiation therapy technique after breast conserving surgery. However, over the past two decades, partial breast irradiation (PBI) has emerged as an alternative to WBI, offering the potential for shorter courses of radiation and improved toxicity profiles. Several randomized trials using different PBI techniques have recently been published. We therefore performed a meta-analysis to compare outcomes between WBI and PBI as well as compare outcomes with various PBI techniques. For this analysis, seven randomized trials were included (National Institute of Oncology, GEC-ESTRO, University of Florence, Barcelona, RAPID, IMPORT-LOW, and NSABP B39). Older trials evaluating PBI were excluded as they were performed prior to the 3-dimensional treatment era and were not consistent with modern PBI techniques. Randomized trials evaluating intraoperative radiation therapy were excluded due to differences in technique as compared to traditional PBI techniques. PBI trials were divided into external beam (3D-CRT/IMRT; University of Florence, Barcelona, RAPID, IMPORT-LOW, and NSABP B39) and brachytherapy (National Institute of Oncology, GEC-ESTRO). As the majority of patients in NSABP B39 were treated with 3D-CRT (73%, n = 1536), this was considered an external beam PBI trial in this analysis. Outcomes evaluated included ipsilateral breast tumor recurrence (IBTR) at 5 years as well as acute (RAPID, University of Florence and Barcelona trials) and late toxicities (RAPID, University of Florence, Barcelona; GEC-ESTRO, IMPORT LOW trials provided Grade 2+ toxicities). A Bayesian logistic regression model evaluated the risk of above outcomes for evidence of deviation by WBI and PBI techniques. Statistical estimation reports risk of rate by posterior median and 95% highest posterior density (HPD) interval. A total of 9,758 patients were included with 4,840 receiving WBI and 4,918 PBI. At 5 years, no difference in IBTR was noted between PBI (1.8%, 95% CI 0.68%-3.2%) and WBI (1.7%, 95% CI 0.92%-2.4%). By PBI technique, the 5-year IBTR rate for 3DCRT/IMRT was 1.7%, and 2.2% for brachytherapy. With respect to acute toxicities, grade 2/3 rates were 7.1%/0% with PBI as compared to 40.0%/3.6% with WBI. For late toxicities, grade 2/3 rates were 0%/0% with PBI as compared to 1.0%/0% with WBI and for grade 2+ were 1.2% with PBI versus 3.6% with WBI. IBTR rates were similar between PBI and WBI with no differences noted by PBI technique. Acute and late toxicities were lower with PBI as compared to WBI. Additional data from randomized trials is needed to compare toxicity profiles between various PBI techniques.