Abstract P4-16-03: Patterns of failure after accelerated partial breast irradiation by consensus panel group: A pooled analysis of William Beaumont Hospital and the American Society of Breast Surgeons Trial data

• Published in: Cancer research (Chicago, Ill.) Vol. 72; no. 24_Supplement; pp. P4 - P4-16-03
• Main Authors: Wilkinson, JB, Beitsch, PD, Arthur, D, Shah, C, Haffty, BG, Wazer, D, Keisch, M, Shaitelman, SF, Lyden, M, Chen, PY, Vicini, FA
• Format: Journal Article
• Language: English
• Published: 15.12.2012
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/0008-5472.SABCS12-P4-16-03

Abstract Background: To determine six-year outcomes and patterns of failure following accelerated partial breast irradiation (APBI) within a pooled patient population from William Beaumont Hospital (WBH) and the American Society of Breast Surgeons (ASBrS) MammoSite® Registry Trial. Methods: 2,127 cases of early-stage breast cancer were treated using APBI (WBH: n=678; ASBrS: n=1,449). Three forms of APBI were used at WBH (interstitial, n=221; balloon-based, n=255; or 3D-CRT, n=206) while all Registry Trial patients received balloon-based brachytherapy. Patients with complete coding necessary for ASTRO Consensus Panel (CP) group assignment (n = 1,813) were divided into suitable (n = 661, 36.5%), cautionary (n = 850, 46.9%), and unsuitable (n = 302, 16.7%) categories. Tumor characteristics, clinical outcomes, and patterns of failure were analyzed according to CP group. Results: Median age was 65 years (32–94 years), median tumor size was 10.0mm (0–45mm), and median follow up was 59.4 months. The WBH cohort had more node-positive disease (6.9% vs. 2.6%, p < 0.01) and cautionary patients (49.5% vs. 41.8%, p = 0.06). Margin and estrogen receptor (ER) status were not different between WBH and ASBrS cohorts (p = 0.46, 0.82). Six-year rates of ipsilateral breast tumor recurrence (IBTR), regional nodal failure (RNF), and distant metastasis (DM) for the whole cohort were 3.2%, 0.7%, and 1.1%, respectively. Elsewhere failures (EF) were the predominant mode of in-breast recurrence for each CP group (suitable: 2.3%, cautionary: 2.5%, unsuitable: 4.9%, p = 0.16) as compared to true recurrences (TR) near the lumpectomy bed (suitable 0.9%, cautionary: 1.5%, unsuitable: 0.8%). No statistical difference in combined rates of ipsilateral recurrence (TR+EF) were observed between the three consensus panel groups (suitable: 3.2%, cautionary: 4.1%, unsuitable: 5.7%, p = 0.25). The six-year rate of distant metastasis was higher for cautionary and unsuitable groups (3.5–3.6% vs. 1.1% for suitable, p = 0.01), although cause-specific survival was the same for all patient categories (98.0–98.6%, all groups, p = 0.47). On multivariate analysis, no factor was associated with increased risk of true recurrence while ER negative status (OR: 4.13, p < 0.01) and a positive/close margin (OR: 2.70, p = 0.02) were associated with increased rates of elsewhere failure. Conclusions: Excellent overall outcomes following breast conserving surgery and APBI were seen at six years in our pooled analysis. A nonsignificant elevation in the rate of IBTR for cautionary and unsuitable consensus panel groups was due to new primary cancers elsewhere in the breast and not treatment failures. Despite a trend towards increased elsewhere failures in the unsuitable group, the current ASTRO CP guidelines were not able to adequately differentiate patients at an increased risk of IBTR or tumor bed failure within a large patient cohort treated with APBI. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P4-16-03.