Lack of Consensus Guideline Driven Approach for Utilization of Accelerated Partial Breast Irradiation in the ASTRO “Cautionary” Group: A National Cancer Database Analysis

Abstract

Development of accelerated partial breast radiotherapy (APBI) consensus guidelines arose in response to an increasing utilization of APBI, despite a lack of phase III data at the time. In 2009, ASTRO published their initial guidelines distinguishing three cohorts: suitable, cautionary, and unsuitable. These definitions differed, however, in comparison to those of other guidelines. Therefore, given the conflicting definition of “cautionary” among these guidelines, our intent was to evaluate utilization trends and factors influencing APBI use in patients defined as cautionary by the 2009 ASTRO guidelines. We retrospectively analyzed breast cancer patients who underwent breast conserving surgery and adjuvant radiation in the National Cancer Database from 2004-2013 and met the initial ASTRO criteria for “cautionary” use of APBI (i.e. Age >50, negative margins, pTis-pT2 (<3cm), pN0i-/N0i+, invasive lobular or ductal, ER+/-, no neoadjuvant chemo). We compared utilization of APBI to whole breast irradiation for this cohort, and completed a logistic regression analysis to identify independent predictors of APBI utilization (Odds ratio [OR]). From 2004-2013, the “cautionary” group included 163,344 women, of which 10.6% received APBI. On multivariate analysis, factors most associated with increased utilization of APBI in the “cautionary” group were living >50 miles from center (OR:2.2), south region (OR:1.7) and urban setting (OR:1.5), all p<0.0001. Factors associated with decreased odds were being at academic center, Her2-Neu positive, use of non-neoadjuvant chemo, and being pN0i+ (all p<0.002). Of note, the ASTRO “cautionary” factors of age 50 - 59, ER status, tumor size, and histology did not impact odds of receiving APBI. From 2004-2013, the 2009 ASTRO “cautionary” factors had little impact on APBI utilization in comparison to other more strongly associated factors, including treatment location (south region and academic center), patient convenience (distance to center) and signs of aggressive malignancies, including pN0i+, Her2Neu status and receipt of non-neoadjuvant chemotherapy. This demonstrates the lack of a guideline driven approach in the utilization of APBI, and necessiates further evaluation of the limitations present in adhering to guidelines.