Radiomics as prognostic factor in brain metastases treated with Gamma Knife radiosurgery. Huang Chih-Ying,Lee Cheng-Chia,Yang Huai-Che,Lin Chung-Jung,Wu Hsiu-Mei,Chung Wen-Yuh,Shiau Cheng-Ying,Guo Wan-Yuo,Pan David Hung-Chi,Peng Syu-Jyun Journal of neuro-oncology PURPOSE:Gamma Knife radiosurgery (GKRS) is a non-invasive procedure for the treatment of brain metastases. This study sought to determine whether radiomic features of brain metastases derived from pre-GKRS magnetic resonance imaging (MRI) could be used in conjunction with clinical variables to predict the effectiveness of GKRS in achieving local tumor control. METHODS:We retrospectively analyzed 161 patients with non-small cell lung cancer (576 brain metastases) who underwent GKRS for brain metastases. The database included clinical data and pre-GKRS MRI. Brain metastases were demarcated by experienced neurosurgeons, and radiomic features of each brain metastasis were extracted. Consensus clustering was used for feature selection. Cox proportional hazards models and cause-specific proportional hazards models were used to correlate clinical variables and radiomic features with local control of brain metastases after GKRS. RESULTS:Multivariate Cox proportional hazards model revealed that higher zone percentage (hazard ratio, HR 0.712; P = .022) was independently associated with superior local tumor control. Similarly, multivariate cause-specific proportional hazards model revealed that higher zone percentage (HR 0.699; P = .014) was independently associated with superior local tumor control. CONCLUSIONS:The zone percentage of brain metastases, a radiomic feature derived from pre-GKRS contrast-enhanced T1-weighted MRIs, was found to be an independent prognostic factor of local tumor control following GKRS in patients with non-small cell lung cancer and brain metastases. Radiomic features indicate the biological basis and characteristics of tumors and could potentially be used as surrogate biomarkers for predicting tumor prognosis following GKRS. 10.1007/s11060-019-03343-4

CT-Based Radiomics Model for Predicting Brain Metastasis in Category T1 Lung Adenocarcinoma. Chen Aiping,Lu Lin,Pu Xuehui,Yu Tongfu,Yang Hao,Schwartz Lawrence H,Zhao Binsheng AJR. American journal of roentgenology The purpose of this study is to develop and evaluate an unenhanced CT-based radiomics model to predict brain metastasis (BM) in patients with category T1 lung adenocarcinoma. A total of 89 eligible patients with category T1 lung adenocarcinoma were enrolled and classified as patients with BM ( = 35) or patients without BM ( = 54). A total of 1160 quantitative radiomic features were extracted from unenhanced CT images of each patient. Three prediction models (the clinical model, the radiomics model, and a hybrid [clinical plus radiomics] model) were established. The ROC AUC value and 10-fold cross-validation were used to evaluate the prediction performance of the models. In terms of predictive performance, the mean AUC value was 0.759 (95% CI, 0.643-0.867; sensitivity, 82.9%; specificity, 57.4%) for the clinical model, 0.847 (95% CI, 0.739-0.915; sensitivity, 80.0%; specificity, 81.5%) for the radiomics model, and 0.871 (95% CI, 0.767-0.933; sensitivity = 82.9%, specificity = 83.3%) for the hybrid model. The hybrid and radiomics models ( = 0.0072 and 0.0492, respectively) performed significantly better than the clinical model. No significant difference was found between the radiomics model and the hybrid model ( = 0.1022). A CT-based radiomics model presented good predictive performance and great potential for predicting BM in patients with category T1 lung adenocarcinoma. As a promising adjuvant tool, it can be helpful for guiding BM screening and thus benefiting personalized surveillance for patients with lung cancer. 10.2214/AJR.18.20591