Body Mass Index Predicts the Incidence of Radiation Pneumonitis in Breast Cancer Patients

ABSTRACTIn patients receiving breast radiotherapy, the risk of radiation pneumonitis has been associated with the volume of irradiated lung, and concomitant methotrexate, paclitaxel, and tamoxifen therapy. Many of the studies of radiation pneumonitis are based on estimates of pulmonary risk using central lung distance that is calculated using two-dimensional techniques. With the treatment of internal mammary nodes and three-dimensional treatment planning for breast cancer becoming increasingly more common, there is a need to further consider the impact of dose-volume metrics in assessing radiation pneumonitis risk. We herein present a case control study assessing the impact of clinical and dose-volume metrics on the development of radiation pneumonitis in patients receiving sequential chemotherapy and local-regional radiotherapy. MATERIALS AND METHODSAfter obtaining institutional review board consent, we retrospectively analyzed the records of 200 patients with node-positive breast cancer treated with computed tomography-based treatment planning at the University of Michigan and Duke University between October 1997 and April 2002. Those who were scored as having clinical radiation pneumonitis were selected for further analysis. A randomly selected group of unaffected patients was chosen as controls in a 3:1 ratio. Patients who received previous radiotherapy, received bilateral breast radiotherapy, were treated for chest wall recurrences, or received high-dose chemotherapy were excluded. All patients had irradiation to the primary site (breast/chest wall) and to the supraclavic-ular and internal mammary nodes. All patients received sequential chemotherapy, and 31 patients received tamoxifen. All patients were treated with three-dimensional conformal radiotherapy, with dose calculated to all structures including both lungs by University of Michigan Plan (Umplan) or Plan of the University of North Carolina (PLUNC). Grades 1–3 radiation pneumonitis (National Cancer Institute Common Toxicity Criteria) were considered events. Traditional dose-volume metrics such as the mean lung dose, V20, integral lung dose, and maximum lung dose were computed. A step-wise regression analysis was used to determine the correlation between the incidence of radiation pneumonitis and the clinical and dose-volume factors. RESULTSFourteen out of 200 patients (7%) developed radiation pneumonitis. All cases of radiation pneumonitis resolved with no clinically significant sequelae. A stepwise regression analysis showed body mass index to be the strongest predictor of radiation pneumonitis (P = 0.003). Of the 14 patients with radiation pneumonitis, nine (64%) had body mass index > 27, compared with nine of 42 (21%) of the controls (P = 0.0065). The presence of pulmonary comorbidities was found to be of borderline significance (P = 0.06) on univariate analysis and was selected as the second predictor on multivariate analysis (P < 0.05). A trend was suggested between maximum dose and radiation pneumonitis on univariate analysis (P = 0.08), and it was used as the third predictor on multivariate analysis (P = 0.09). None of the remaining dosimetric or clinical variables (including the use of a taxane, the use of tamoxifen, smoking history, and laterality) were significantly correlated with radiation pneumonitis. CONCLUSIONIn a group of patients treated with radiotherapy to the breast/chest wall and regional nodes, including the internal mammary nodes, the risk of radiation pneumonitis was 7%. Although body mass index was highly correlated with radiation pneumonitis, pulmonary comorbidities and maximum dose were also of interest. Physicians should take these risk factors into account when treating these patients.