FACTORS INFLUENCING COSMETIC RESULTS AFTER CONSERVATION THERAPY FOR BREAST-CANCER

Purpose: Host, tumor, and treatment-related factors influencing cosmet ic outcome are analyzed for patients receiving breast conservation tre atment. Methods and Materials: Four-hundred and fifty-eight patients w ith evaluable records for cosmesis evaluation, a subset of 701 patient s treated for invasive breast cancer with conservation technique betwe en 1969 and 1990, were prospectively analyzed. In 243 patients, cosmet ic evaluation was not adequately recorded. Cosmesis evaluation was car ried out from 3.7 months to 22.3 years, median of 4,4 years. By pathol ogic stage, tumors were 62% T1N0, 14% T1N1, 15% T2N0, and 9% T2N1. The majority of patients were treated with 4-6 MV photons. Cosmetic evalu ation was rated by both patient and physician every 4-6 months. A logi stic regression analysis was completed using a stepwise logistic regre ssion. P-values of 0.05 or less were considered significant. Excellent cosmetic scores were used in all statistical analyses unless otherwis e specified. Results: At most recent follow-up, 87% of patients and 81 % of physicians scored their cosmetic outcome as excellent or good. Ei ghty-two percent of physician and patient evaluations agreed with exce llent-good vs. fair-poor rating categories. Analysis demonstrated a lo wer proportion of excellent cosmetic scores when related to patient ag e > 60 years (p = 0.001), postmenopausal status (p = 0.02), black race (p = 0.0034), and T2 tumor size (p = 0.05). Surgical factors of impor tance were: volume of resection > 100 cm(3) (p = 0.0001), scar orienta tion compliance with the National Surgical Adjuvant Breast Project (NS ABP) guidelines (p = 0.0034), and > 20 cm(2) skin resected (p = 0.0452 ). Extent of axillary surgery did not significantly affect breast cosm esis. Radiation factors affecting cosmesis included treatment volume ( tangential breast fields only vs. three or more fields) (p = 0.034), w hole breast dose in excess of 50 Gy (p = 0.0243), and total dose to tu mor site > 65 Gy (p = 0.06), as well as optimum dose distribution with compensating filters (p = 0.002). Daily fraction size of 1.8 Gy vs. 2 .0 Gy, boost vs, no boost, type of boost (brachytherapy vs, electrons) , total radiation dose, and use of bolus were not significant factors. Use of concomitant chemotherapy with irradiation impaired excellent c osmetic outcome (p = 0.02). Use of sequential chemotherapy or adjuvant tamoxifen did not appear to diminish excellent cosmetic outcomes (p = 0.31). Logistic regression for excellent cosmetic outcome analysis wa s completed for age, tumor size, menopausal status, race, type of surg ery, volume of breast tissue resected, scar orientations, whole breast radiation dose, total radiation dose, number of radiation fields trea ted, and use of adjuvant chemotherapy. Significant independent factors for excellent cosmetic outcome were: volume of tissue resected (p = 0 .0001), type of surgery (p = 0.0001), breast radiation dose (p = 0.005 ), race (p = 0.002), and age (p = 0.007).