Hao Zhang1, Shan Tan2, Wengen Chen3, Seth Kligerman3, Grace Kim1, Warren D D'Souza1, Mohan Suntharalingam1, Wei Lu4. 1. Department of Radiation Oncology, University of Maryland School of Medicine, Baltimore, Maryland. 2. Department of Radiation Oncology, University of Maryland School of Medicine, Baltimore, Maryland; Department of Control Science and Engineering, Huazhong University of Science and Technology, Wuhan, China. 3. Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine, Baltimore, Maryland. 4. Department of Radiation Oncology, University of Maryland School of Medicine, Baltimore, Maryland. Electronic address: wlu@umm.edu.
Abstract
PURPOSE: To construct predictive models using comprehensive tumor features for the evaluation of tumor response to neoadjuvant chemoradiation therapy (CRT) in patients with esophageal cancer. METHODS AND MATERIALS: This study included 20 patients who underwent trimodality therapy (CRT+surgery) and underwent 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) both before and after CRT. Four groups of tumor features were examined: (1) conventional PET/CT response measures (eg, standardized uptake value [SUV]max, tumor diameter); (2) clinical parameters (eg, TNM stage, histology) and demographics; (3) spatial-temporal PET features, which characterize tumor SUV intensity distribution, spatial patterns, geometry, and associated changes resulting from CRT; and (4) all features combined. An optimal feature set was identified with recursive feature selection and cross-validations. Support vector machine (SVM) and logistic regression (LR) models were constructed for prediction of pathologic tumor response to CRT, cross-validations being used to avoid model overfitting. Prediction accuracy was assessed by area under the receiver operating characteristic curve (AUC), and precision was evaluated by confidence intervals (CIs) of AUC. RESULTS: When applied to the 4 groups of tumor features, the LR model achieved AUCs (95% CI) of 0.57 (0.10), 0.73 (0.07), 0.90 (0.06), and 0.90 (0.06). The SVM model achieved AUCs (95% CI) of 0.56 (0.07), 0.60 (0.06), 0.94 (0.02), and 1.00 (no misclassifications). With the use of spatial-temporal PET features combined with conventional PET/CT measures and clinical parameters, the SVM model achieved very high accuracy (AUC 1.00) and precision (no misclassifications)-results that were significantly better than when conventional PET/CT measures or clinical parameters and demographics alone were used. For groups with many tumor features (groups 3 and 4), the SVM model achieved significantly higher accuracy than did the LR model. CONCLUSIONS: The SVM model that used all features including spatial-temporal PET features accurately and precisely predicted pathologic tumor response to CRT in esophageal cancer.
PURPOSE: To construct predictive models using comprehensive tumor features for the evaluation of tumor response to neoadjuvant chemoradiation therapy (CRT) in patients with esophageal cancer. METHODS AND MATERIALS: This study included 20 patients who underwent trimodality therapy (CRT+surgery) and underwent 18F-fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) both before and after CRT. Four groups of tumor features were examined: (1) conventional PET/CT response measures (eg, standardized uptake value [SUV]max, tumor diameter); (2) clinical parameters (eg, TNM stage, histology) and demographics; (3) spatial-temporal PET features, which characterize tumor SUV intensity distribution, spatial patterns, geometry, and associated changes resulting from CRT; and (4) all features combined. An optimal feature set was identified with recursive feature selection and cross-validations. Support vector machine (SVM) and logistic regression (LR) models were constructed for prediction of pathologic tumor response to CRT, cross-validations being used to avoid model overfitting. Prediction accuracy was assessed by area under the receiver operating characteristic curve (AUC), and precision was evaluated by confidence intervals (CIs) of AUC. RESULTS: When applied to the 4 groups of tumor features, the LR model achieved AUCs (95% CI) of 0.57 (0.10), 0.73 (0.07), 0.90 (0.06), and 0.90 (0.06). The SVM model achieved AUCs (95% CI) of 0.56 (0.07), 0.60 (0.06), 0.94 (0.02), and 1.00 (no misclassifications). With the use of spatial-temporal PET features combined with conventional PET/CT measures and clinical parameters, the SVM model achieved very high accuracy (AUC 1.00) and precision (no misclassifications)-results that were significantly better than when conventional PET/CT measures or clinical parameters and demographics alone were used. For groups with many tumor features (groups 3 and 4), the SVM model achieved significantly higher accuracy than did the LR model. CONCLUSIONS: The SVM model that used all features including spatial-temporal PET features accurately and precisely predicted pathologic tumor response to CRT in esophageal cancer.
Authors: Marinke Westerterp; Henderik L van Westreenen; Johannes B Reitsma; Otto S Hoekstra; Jaap Stoker; Paul Fockens; Pieter L Jager; Berthe L F Van Eck-Smit; John T M Plukker; J Jan B van Lanschot; Gerrit W Sloof Journal: Radiology Date: 2005-09 Impact factor: 11.105
Authors: Alberto Ruol; Christian Rizzetto; Carlo Castoro; Matteo Cagol; Rita Alfieri; Gianpietro Zanchettin; Francesco Cavallin; Silvia Michieletto; Gianfranco Da Dalt; Vanna Chiarion Sileni; Luigi Corti; Silvia Mantoan; Giovanni Zaninotto; Ermanno Ancona Journal: Ann Surg Date: 2010-11 Impact factor: 12.969
Authors: Arta Monir Monjazeb; Greg Riedlinger; Mebea Aklilu; Kim R Geisinger; Girish Mishra; Scott Isom; Paige Clark; Edward A Levine; A William Blackstock Journal: J Clin Oncol Date: 2010-09-27 Impact factor: 44.544
Authors: Steven M. Larson; Yusuf Erdi; Timothy Akhurst; Madhu Mazumdar; Homer A. Macapinlac; Ronald D. Finn; Cecille Casilla; Melissa Fazzari; Neil Srivastava; Henry W.D. Yeung; John L. Humm; Jose Guillem; Robert Downey; Martin Karpeh; Alfred E. Cohen; Robert Ginsberg Journal: Clin Positron Imaging Date: 1999-05
Authors: Ben R Saleem; Roelof J Beukinga; Ronald Boellaard; Andor W J M Glaudemans; Michel M P J Reijnen; Clark J Zeebregts; Riemer H J A Slart Journal: Eur J Nucl Med Mol Imaging Date: 2016-12-24 Impact factor: 9.236