Noah S Kalman1, Geoffrey D Hugo2, Rebecca N Mahon3, Xiaoyan Deng4, Nitai D Mukhopadhyay4, Elisabeth Weiss3. 1. Department of Radiation Oncology, Virginia Commonwealth University, Richmond, USA. Electronic address: dr.n.kalman@gmail.com. 2. Department of Radiation Oncology, Virginia Commonwealth University, Richmond, USA; Department of Radiation Oncology, Washington University in St. Louis, St. Louis, USA. 3. Department of Radiation Oncology, Virginia Commonwealth University, Richmond, USA. 4. Department of Biostatistics, Virginia Commonwealth University, Richmond, USA.
Abstract
BACKGROUND: Radiographic radiation induced lung injury (RILI) is frequently observed after stereotactic body radiotherapy (SBRT). Models of radiographic change can identify patient risk factors that predict clinical toxicity. We examined the association between radiographic lung changes and lung tissue dose-density response over time with clinical risk factors for RILI, such as diabetes. METHODS: 424 baseline and follow up CT scans at 3, 6, and 12 months post-treatment were analyzed in 116 patients (27 with diabetes) undergoing thoracic SBRT. Volumes of dense/hazy regions and lung parenchyma dose-density response curves were evaluated with respect to follow up time, diabetes, and other factors. RESULTS: Dense and hazy tissue regions were larger in the diabetic population, with the effect most pronounced at 3 months. Similarly, dose-density response curves showed greater density change versus dose in the diabetic group (all p < 0.05). Diabetes, time, the interaction of diabetes and time, smoking status, African American race, baseline lung density, and tumor location were significantly associated with radiographic changes on mixed effect analyses. PTV size, pulmonary function, and medication exposure did not significantly impact RILI. Clinical grade 1-2 pneumonitis was more prevalent in diabetic patients (p = 0.02). However, radiographic change did not correlate with clinical pneumonitis. CONCLUSIONS: The presence of diabetes and other clinical factors is associated with increased volume and density of radiographic RILI after lung SBRT, most prominently early after treatment. This is the first report demonstrating the increased severity of RILI after SBRT in diabetic patients. Increased caution treating diabetic patients may be warranted.
BACKGROUND: Radiographic radiation induced lung injury (RILI) is frequently observed after stereotactic body radiotherapy (SBRT). Models of radiographic change can identify patient risk factors that predict clinical toxicity. We examined the association between radiographic lung changes and lung tissue dose-density response over time with clinical risk factors for RILI, such as diabetes. METHODS: 424 baseline and follow up CT scans at 3, 6, and 12 months post-treatment were analyzed in 116 patients (27 with diabetes) undergoing thoracic SBRT. Volumes of dense/hazy regions and lung parenchyma dose-density response curves were evaluated with respect to follow up time, diabetes, and other factors. RESULTS: Dense and hazy tissue regions were larger in the diabetic population, with the effect most pronounced at 3 months. Similarly, dose-density response curves showed greater density change versus dose in the diabetic group (all p < 0.05). Diabetes, time, the interaction of diabetes and time, smoking status, African American race, baseline lung density, and tumor location were significantly associated with radiographic changes on mixed effect analyses. PTV size, pulmonary function, and medication exposure did not significantly impact RILI. Clinical grade 1-2 pneumonitis was more prevalent in diabeticpatients (p = 0.02). However, radiographic change did not correlate with clinical pneumonitis. CONCLUSIONS: The presence of diabetes and other clinical factors is associated with increased volume and density of radiographic RILI after lung SBRT, most prominently early after treatment. This is the first report demonstrating the increased severity of RILI after SBRT in diabeticpatients. Increased caution treating diabeticpatients may be warranted.
Authors: Varun Kumar; Raman Agrawal; Aparamita Pandey; Stefan Kopf; Manuel Hoeffgen; Serap Kaymak; Obul Reddy Bandapalli; Vera Gorbunova; Andrei Seluanov; Marcus A Mall; Stephan Herzig; Peter P Nawroth Journal: EMBO J Date: 2020-04-27 Impact factor: 11.598