Diot Quentin, Kavanagh Brian, Vinogradskiy Yevgeniy, Garg Kavita, Gaspar Laurie, Miften Moyed
Department of Radiation Oncology, University of Colorado School of Medicine, Aurora, Colorado 80045.
Department of Radiology, University of Colorado School of Medicine, Aurora, Colorado 80045.
Med Phys. 2015 Nov;42(11):6477-87. doi: 10.1118/1.4932624.
To differentiate radiation-induced fibrosis from regional lung collapse outside of the high dose region in patients treated with stereotactic body radiation therapy (SBRT) for lung tumors.
Lung deformation maps were computed from pre-treatment and post-treatment computed tomography (CT) scans using a point-to-point translation method. Fifty anatomical landmarks inside the lung (vessel or airway branches) were matched on planning and follow-up scans for the computation process. Two methods using the deformation maps were developed to differentiate regional lung collapse from fibrosis: vector field and Jacobian methods. A total of 40 planning and follow-ups CT scans were analyzed for 20 lung SBRT patients.
Regional lung collapse was detected in 15 patients (75%) using the vector field method, in ten patients (50%) using the Jacobian method, and in 12 patients (60%) by radiologists. In terms of sensitivity and specificity the Jacobian method performed better. Only weak correlations were observed between the dose to the proximal airways and the occurrence of regional lung collapse.
The authors presented and evaluated two novel methods using anatomical lung deformations to investigate lung collapse and fibrosis caused by SBRT treatment. Differentiation of these distinct physiological mechanisms beyond what is usually labeled "fibrosis" is necessary for accurate modeling of lung SBRT-induced injuries. With the help of better models, it becomes possible to expand the therapeutic benefits of SBRT to a larger population of lung patients with large or centrally located tumors that were previously considered ineligible.
在接受立体定向体部放射治疗（SBRT）的肺癌患者中，区分高剂量区域外的放射性纤维化与局部肺不张。
使用点对点平移方法，根据治疗前和治疗后的计算机断层扫描（CT）图像计算肺变形图。在计划扫描和随访扫描中匹配肺内50个解剖标志点（血管或气道分支）用于计算过程。开发了两种利用变形图的方法来区分局部肺不张和纤维化：向量场法和雅可比方法。对20例接受肺SBRT治疗的患者的40次计划扫描和随访CT图像进行了分析。
使用向量场法在15例患者（75%）中检测到局部肺不张，使用雅可比方法在10例患者（50%）中检测到，放射科医生在其中12例患者（60%）中检测到。在敏感性和特异性方面，雅可比方法表现更好。仅观察到近端气道剂量与局部肺不张的发生之间存在弱相关性。
作者提出并评估了两种利用肺解剖变形来研究SBRT治疗引起的肺不张和纤维化的新方法。准确模拟SBRT引起的肺部损伤，需要区分这些不同于通常所说的“纤维化”的独特生理机制。借助更好的模型，有可能将SBRT的治疗益处扩展到更多患有先前被认为不符合条件的大肿瘤或中心型肿瘤的肺癌患者群体。
