Accelerated partial breast irradiation: An analysis of variables associated with late toxicity and long-term cosmetic outcome after high-dose-rate interstitial brachytherapy

Abstract

To perform a detailed analysis of variables associated with late tissue effects of high-dose-rate (HDR) interstitial brachytherapy accelerated partial breast irradiation (APBI) in a large cohort of patients with prolonged follow-up. Beginning in 1995, 75 women with Stage I/II breast cancer were enrolled in identical institutional trials evaluating APBI as monotherapy after lumpectomy. Patients eligible included those with T1-2, N0-1 (<or=3 nodes positive), M0 tumors of nonlobular histology with negative surgical margins, no extracapsular nodal extension, and negative results on postexcision mammogram. All patients underwent surgical excision and postoperative irradiation with HDR interstitial brachytherapy. The planning target volume was defined as the excision cavity plus a 2-cm margin. Treatment was delivered with a high-activity Ir-192 source at 3.4 Gy per fraction twice daily for 5 days to a total dose of 34 Gy. Dosimetric analyses were performed with three-dimensional postimplant dose and volume reconstructions. All patients were evaluated at 3-6-month intervals and assessed with a standardized cosmetic rating scale and according to Radiation Therapy Oncology Group late normal tissue toxicity scoring criteria. Clinical and therapy-related features were analyzed for their relationship to cosmetic outcome and toxicity rating. Clinical features analyzed included age, volume of resection, history of diabetes or hypertension, extent of axillary surgery, and systemic therapies. Therapy-related features analyzed included volume of tissue encompassed by the 100%, 150%, and 200% isodose lines (V100, V150, and V200, respectively), the dose homogeneity index (DHI), number of source dwell positions, and planar separation. The median follow-up of all patients was 73 months (range, 43-118 months). The cosmetic outcome at last follow-up was rated as excellent, good, and fair/poor in 67%, 24%, and 9% of patients, respectively. Suboptimal cosmetic outcome was significantly associated with the number of source dwell positions, V150, and V200 and inversely associated with DHI (0.77 vs. 0.73; p=0.05). Late skin toxicity was rated as Grade 0, 1, or 2 in 77%, 19%, and 4% of patients, respectively. The risk of Grade 1/2 skin toxicity was significantly associated with V150 and V200 and inversely associated with DHI (0.77 vs. 0.71; p=0.009). Late subcutaneous toxicity was rated as Grade 0, 1, 2, 3, or 4 in 55%, 15%, 12%, 5%, and 13% of patients, respectively. The risk of Grade 0/1 vs. Grade 2-4 subcutaneous toxicity was significantly associated only with a lower value of DHI (0.77 vs. 0.73; p=0.02). To further explore factors that might contribute to the risk of fat necrosis (symptomatic or asymptomatic), a separate analysis showed that only dose hotspots as reflected in V150 and V200 were significantly associated with elevated risk. The use of adriamycin-based chemotherapy after APBI was found to be associated with a significant increase in the incidence of higher-grade skin toxicity and a higher risk of fat necrosis and suboptimal cosmetic outcome. Patient age, volume of resection, extent of axillary surgery, a history of diabetes or hypertension, and the use of tamoxifen were not found to be significantly associated with cosmetic outcome or late normal tissue complications. Long-term cosmetic results and the risk of late skin and subcutaneous toxicity after APBI with interstitial HDR brachytherapy can be correlated with specific treatment-related variables. These data provide dosimetric parameters that might be used to minimize the risk of normal tissue injury after APBI interstitial brachytherapy.