Timothée Zaragori1,2, Julien Oster2, Véronique Roch1, Gabriela Hossu2,3, Mohammad B Chawki1, Rachel Grignon1, Celso Pouget4,5, Guillaume Gauchotte4,5, Fabien Rech6,7, Marie Blonski7,8, Luc Taillandier7,8, Laëtitia Imbert1,2, Antoine Verger9,2. 1. Department of Nuclear Medicine and Nancyclotep Imaging Platform, CHRU-Nancy, Université de Lorraine, Nancy, France. 2. IADI UMR 1254, INSERM, Université de Lorraine, Nancy, France. 3. CIC 1433 Innovation Technologique, INSERM, CHRU-Nancy, Université de Lorraine, Nancy, France. 4. Department of Pathology, CHRU-Nancy, Université de Lorraine, Nancy, France. 5. INSERM U1256, Université de Lorraine, Nancy, France. 6. Department of Neurosurgery, CHRU-Nancy, Université de Lorraine, Nancy, France. 7. Centre de Recherche en Automatique de Nancy CRAN UMR 7039, CNRS, Université de Lorraine, Nancy, France; and. 8. Department of Neuro-Oncology, CHRU-Nancy, Université de Lorraine, Nancy, France. 9. Department of Nuclear Medicine and Nancyclotep Imaging Platform, CHRU-Nancy, Université de Lorraine, Nancy, France; a.verger@chru-nancy.fr.
Abstract
The assessment of gliomas by 18F-FDOPA PET imaging as an adjunct to MRI showed high performance by combining static and dynamic features to noninvasively predict the isocitrate dehydrogenase (IDH) mutations and the 1p/19q codeletion, which the World Health Organization classified as significant parameters in 2016. The current study evaluated whether other 18F-FDOPA PET radiomics features further improve performance and the contributions of each of these features to performance. Methods: Our study included 72 retrospectively selected, newly diagnosed glioma patients with 18F-FDOPA PET dynamic acquisitions. A set of 114 features, including conventional static features and dynamic features, as well as other radiomics features, were extracted and machine-learning models trained to predict IDH mutations and the 1p/19q codeletion. Models were based on a machine-learning algorithm built from stable, relevant, and uncorrelated features selected by hierarchic clustering followed by a bootstrapped feature selection process. Models were assessed by comparing area under the curve using a nested cross-validation approach. Feature importance was assessed using Shapley additive explanations values. Results: The best models were able to predict IDH mutations (logistic regression with L2 regularization) and the 1p/19q codeletion (support vector machine with radial basis function kernel) with an area under the curve of 0.831 (95% CI, 0.790-0.873) and 0.724 (95% CI, 0.669-0.782), respectively. For the prediction of IDH mutations, dynamic features were the most important features in the model (time to peak, 35.5%). In contrast, other radiomics features were the most useful for predicting the 1p/19q codeletion (up to 14.5% of importance for the small-zone low-gray-level emphasis). Conclusion: 18F-FDOPA PET is an effective tool for the noninvasive prediction of glioma molecular parameters using a full set of amino-acid PET radiomics features. The contribution of each feature set shows the importance of systematically integrating dynamic acquisition for prediction of the IDH mutations as well as developing the use of radiomics features in routine practice for prediction of the 1p/19q codeletion.
The assessment of gliomas by 18F-FDOPA PET imaging as an adjunct to MRI showed high performance by combining static and dynamic features to noninvasively predict the isocitrate dehydrogenase (IDH) mutations and the 1p/19q codeletion, which the World Health Organization classified as significant parameters in 2016. The current study evaluated whether other 18F-FDOPA PET radiomics features further improve performance and the contributions of each of these features to performance. Methods: Our study included 72 retrospectively selected, newly diagnosed glioma patients with 18F-FDOPA PET dynamic acquisitions. A set of 114 features, including conventional static features and dynamic features, as well as other radiomics features, were extracted and machine-learning models trained to predict IDH mutations and the 1p/19q codeletion. Models were based on a machine-learning algorithm built from stable, relevant, and uncorrelated features selected by hierarchic clustering followed by a bootstrapped feature selection process. Models were assessed by comparing area under the curve using a nested cross-validation approach. Feature importance was assessed using Shapley additive explanations values. Results: The best models were able to predict IDH mutations (logistic regression with L2 regularization) and the 1p/19q codeletion (support vector machine with radial basis function kernel) with an area under the curve of 0.831 (95% CI, 0.790-0.873) and 0.724 (95% CI, 0.669-0.782), respectively. For the prediction of IDH mutations, dynamic features were the most important features in the model (time to peak, 35.5%). In contrast, other radiomics features were the most useful for predicting the 1p/19q codeletion (up to 14.5% of importance for the small-zone low-gray-level emphasis). Conclusion: 18F-FDOPA PET is an effective tool for the noninvasive prediction of glioma molecular parameters using a full set of amino-acid PET radiomics features. The contribution of each feature set shows the importance of systematically integrating dynamic acquisition for prediction of the IDH mutations as well as developing the use of radiomics features in routine practice for prediction of the 1p/19q codeletion.
Authors: Nathalie L Albert; Michael Weller; Bogdana Suchorska; Norbert Galldiks; Riccardo Soffietti; Michelle M Kim; Christian la Fougère; Whitney Pope; Ian Law; Javier Arbizu; Marc C Chamberlain; Michael Vogelbaum; Ben M Ellingson; Joerg C Tonn Journal: Neuro Oncol Date: 2016-04-21 Impact factor: 12.300
Authors: Jeanette E Eckel-Passow; Daniel H Lachance; Annette M Molinaro; Kyle M Walsh; Paul A Decker; Hugues Sicotte; Melike Pekmezci; Terri Rice; Matt L Kosel; Ivan V Smirnov; Gobinda Sarkar; Alissa A Caron; Thomas M Kollmeyer; Corinne E Praska; Anisha R Chada; Chandralekha Halder; Helen M Hansen; Lucie S McCoy; Paige M Bracci; Roxanne Marshall; Shichun Zheng; Gerald F Reis; Alexander R Pico; Brian P O'Neill; Jan C Buckner; Caterina Giannini; Jason T Huse; Arie Perry; Tarik Tihan; Mitchell S Berger; Susan M Chang; Michael D Prados; Joseph Wiemels; John K Wiencke; Margaret R Wrensch; Robert B Jenkins Journal: N Engl J Med Date: 2015-06-10 Impact factor: 176.079
Authors: Philipp Lohmann; Christoph Lerche; Elena K Bauer; Jan Steger; Gabriele Stoffels; Tobias Blau; Veronika Dunkl; Martin Kocher; Shivakumar Viswanathan; Christian P Filss; Carina Stegmayr; Maximillian I Ruge; Bernd Neumaier; Nadim J Shah; Gereon R Fink; Karl-Josef Langen; Norbert Galldiks Journal: Sci Rep Date: 2018-09-06 Impact factor: 4.379
Authors: Cristina Barca; Claudia Foray; Bastian Zinnhardt; Alexandra Winkeler; Ulrich Herrlinger; Oliver M Grauer; Andreas H Jacobs Journal: Cancers (Basel) Date: 2022-06-27 Impact factor: 6.575
Authors: Zhicong Li; Lena Kaiser; Adrien Holzgreve; Viktoria C Ruf; Bogdana Suchorska; Vera Wenter; Stefanie Quach; Jochen Herms; Peter Bartenstein; Jörg-Christian Tonn; Marcus Unterrainer; Nathalie L Albert Journal: Eur J Nucl Med Mol Imaging Date: 2021-09-07 Impact factor: 9.236