Multivariate analysis of pulmonary fibrosis after electron beam irradiation for postmastectomy chest wall and regional lymphatics: evidence for non-dosimetric factors

Background and purpose: To evaluate the factors associated with pulmonary f ibrosis after postmastectomy electron beam irradiation of chest wall and re gional lymphatics in patients with breast cancer. Materials and methods: From July 1987 through July 1994, 109 women with sta ge II and III breast cancer receiving modified radical mastectomies were ma naged by postoperative electron beam irradiation. Doses of 46 to 50.4 Gy we re delivered to the chest wall covered with bolus, internal mammary nodes, supraclavicular nodes and axillary lymph nodes via 12 or 15 MeV single port al electron beam. Seventeen patients received additional 10-16 Gy surgical scar boost via 9 MeV electron beam. Comparison of pre-treatment and post-tr eatment chest X-ray films were used to monitor the development of pulmonary fibrosis. Results: Only Grade 1 radiation-induced late pulmonary toxicity was noted i n 33 patients (29%). Twenty-six patients (24%) developed pulmonary fibrosis under unbolused chest wall. Lung fibrosis under bolused chest wall was not ed in 11 patients (10%). Statistical difference (P < 0.01) was noted betwee n the incidence of fibrosis in these two sites. In multivariate analysis of lung fibrosis under unbolus-covered chest wall, the independent prognostic factors, are low body mass index (BMI) (P < 0.01), tamoxifen taking (P = 0 .03), and no treatment interruption (P = 0.03). No independent factor was a ssociated with lung fibrosis under bolus-covered chest wall in multivariate analysis. Conclusions: In the analysis of pulmonary fibrosis induced by unbolused ele ctron beam, BMI rather than body weight and body height is a strong prognos tic factor. Tamoxifen and short overall time can predispose the development of lung fibrosis. (C) 2000 Elsevier Science Ireland Ltd. All rights reserv ed.