The Importance of 3-dimensional CT Treatment Planning to Optimize Outcomes From Accelerated Partial Breast Irradiation (APBI) With Brachytherapy in Early Stage Breast Cancer

While we await outcomes from randomized studies, APBI has seen increasing use as an alternative to whole breast irradiation (WBI) after lumpectomy for early stage breast cancer. Recently, there has been concern generated about excess toxicity associated with brachytherapy APBI. Multi-catheter (MCT) brachytherapy, as the initial APBI method developed, has the longest follow-up, and provides a means for looking at late outcomes. Between 1998 and 2008, 75 patients underwent MCT PBI. The patients had median age 70 years; 7% Tis, 81% T1, 12% T2; 5% node positive, 91% were ER positive, 79% got anti-endocrine therapy, and 9% got chemotherapy. Three dimensional conformal CT planning (3DCT) was done for 58, 52 had plans available for analysis. A 2 plane implant was done in 21%, 3 plane for 57%. Mean number of catheters was 17. Dose was prescribed to a PTV at 1-2cm from the lumpectomy cavity. A PTV was defined prospectively (pPTV) for most, 17% were done retrospectively (rPTV) for this analysis. The prescription dose (RxD) was 34 Gy/10 fractions BID for 87%, 39 Gy/10 fractions BID for the remaining. The mean planning target volume (PTV) and total breast volume were 142cc and 1075cc, respectively. On average, 92% of the PTV was covered by the RxD. Toxicity scored per Common Toxicity Criteria v.4, cosmesis by the Harvard Scale. Late toxicity/cosmesis was evaluated in those with >1year follow-up. With 7.8 years median follow-up, three (4%) had an in-breast recurrence. All three were in the ‘cautionary’ or ‘unsuitable’ category of the ASTRO APBI consensus statement. Grade 2 acute toxicity was seen in 33% (most common pain and dermatitis) with no grade 3 events or moist desquamation noted. Long term, radiation-related grade 2 toxicity was 31%, mostly fibrosis. There was one grade 3 event (nipple/areola deformity) and one grade 4 event (angiosarcoma in treated breast). Cosmesis was excellent or good in 78%, fair 19%, and 3% poor. The mean V100 for pPTV and rPTV patients was 197cc and 288cc, giving conformality index of 1.54 and 2.74 respectively (p < 0.005). V150 and V200 were also smaller with a prospectively defined PTV (p < 0.01). More fair-poor cosmesis occurred in those without CT planning (p = 0.02). The volume of breast receiving 100%, 150%, and 200% of the prescription dose was associated with worse toxicity and cosmetic outcome (p < 0.05). This series of MCT APBI demonstrates excellent local control and toxicity profile with long term follow-up. Careful 3DCT based planning for MCT APBI should be used to minimize toxicity and optimize cosmetic outcomes.