Michael Zhang1,2, Edward Wang3, Derek Yecies1,2, Lydia T Tam4, Michelle Han5, Sebastian Toescu6, Jason N Wright7, Emre Altinmakas8, Eric Chen9, Alireza Radmanesh10, Jordan Nemelka11, Ozgur Oztekin12, Matthias W Wagner13, Robert M Lober14, Birgit Ertl-Wagner13, Chang Y Ho9, Kshitij Mankad15, Nicholas A Vitanza16, Samuel H Cheshier11, Tom S Jacques17, Paul G Fisher18, Kristian Aquilina6, Mourad Said19, Alok Jaju20, Stefan Pfister21, Michael D Taylor22, Gerald A Grant23, Sarah Mattonen3, Vijay Ramaswamy24,25, Kristen W Yeom2. 1. Department of Neurosurgery, Stanford Hospital and Clinics, Stanford, California, USA. 2. Department of Radiology, Lucile Packard Children's Hospital, Stanford, California, USA. 3. Department of Medical Biophysics, Western University, London, Ontario, Canada. 4. Stanford School of Medicine, Stanford University, Stanford, California, USA. 5. Department of Pediatrics, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA. 6. Department of Neurosurgery, Great Ormond Street Institute of Child Health, London, UK. 7. Department of Radiology, Seattle Children's Hospital, and Harborview Medical Center, Seattle, Washington, USA. 8. Department of Radiology, Koç University School of Medicine, Istanbul, Turkey. 9. Department of Clinical Radiology & Imaging Sciences, Riley Children's Hospital, Indianapolis, Indiana, USA. 10. Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA. 11. Division of Pediatric Neurosurgery, Department of Neurosurgery, Huntsman Cancer Institute, University of Utah School of Medicine, Intermountain Healthcare Primary Children's Hospital, Salt Lake City, Utah, USA. 12. Department of Neuroradiology, Cigli Education and Research Hospital, and Tepecik Education and Research Hospital, Izmir, Turkey. 13. Department of Diagnostic Imaging, The Hospital for Sick Children, Toronto, Ontario, Canada. 14. Division of Neurosurgery, Dayton Children's Hospital, Dayton, Ohio, USA. 15. Department of Radiology, Great Ormond Street Institute of Child Health, London, UK. 16. Division of Pediatric Hematology/Oncology, Department of Pediatrics, Seattle Children's Hospital, Seattle, Washington, USA. 17. Department of Developmental Biology & Cancer, University College London Great Ormond Street Institute of Child Health, and Great Ormond Street Hospital for Children NHS Foundation Trust, London, UK. 18. Department of Neurology, Lucile Packard Children's Hospital, Stanford University, Palo Alto, California, USA. 19. Radiology Department Centre International Carthage Médicale, Monastir, Tunisia. 20. Department of Medical Imaging, Ann and Robert H. Lurie Children's Hospital of Chicago, Chicago, Illinois, USA. 21. Department of Pediatrics, Hopp Children' Cancer Center, Heidelberg, Germany. 22. Division of Neurosurgery, The Hospital for Sick Children, Toronto, Ontario, Canada. 23. Department of Neurosurgery, Lucile Packard Children's Hospital, Stanford, California, USA. 24. Division of Haematology/Oncology, Programme in Developmental and Stem Cell Biology, The Hospital for Sick Children, Toronto, Ontario, Canada. 25. Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada.
Abstract
BACKGROUND: The risk profile for posterior fossa ependymoma (EP) depends on surgical and molecular status [Group A (PFA) versus Group B (PFB)]. While subtotal tumor resection is known to confer worse prognosis, MRI-based EP risk-profiling is unexplored. We aimed to apply machine learning strategies to link MRI-based biomarkers of high-risk EP and also to distinguish PFA from PFB. METHODS: We extracted 1800 quantitative features from presurgical T2-weighted (T2-MRI) and gadolinium-enhanced T1-weighted (T1-MRI) imaging of 157 EP patients. We implemented nested cross-validation to identify features for risk score calculations and apply a Cox model for survival analysis. We conducted additional feature selection for PFA versus PFB and examined performance across three candidate classifiers. RESULTS: For all EP patients with GTR, we identified four T2-MRI-based features and stratified patients into high- and low-risk groups, with 5-year overall survival rates of 62% and 100%, respectively (P < .0001). Among presumed PFA patients with GTR, four T1-MRI and five T2-MRI features predicted divergence of high- and low-risk groups, with 5-year overall survival rates of 62.7% and 96.7%, respectively (P = .002). T1-MRI-based features showed the best performance distinguishing PFA from PFB with an AUC of 0.86. CONCLUSIONS: We present machine learning strategies to identify MRI phenotypes that distinguish PFA from PFB, as well as high- and low-risk PFA. We also describe quantitative image predictors of aggressive EP tumors that might assist risk-profiling after surgery. Future studies could examine translating radiomics as an adjunct to EP risk assessment when considering therapy strategies or trial candidacy.
BACKGROUND: The risk profile for posterior fossa ependymoma (EP) depends on surgical and molecular status [Group A (PFA) versus Group B (PFB)]. While subtotal tumor resection is known to confer worse prognosis, MRI-based EP risk-profiling is unexplored. We aimed to apply machine learning strategies to link MRI-based biomarkers of high-risk EP and also to distinguish PFA from PFB. METHODS: We extracted 1800 quantitative features from presurgical T2-weighted (T2-MRI) and gadolinium-enhanced T1-weighted (T1-MRI) imaging of 157 EP patients. We implemented nested cross-validation to identify features for risk score calculations and apply a Cox model for survival analysis. We conducted additional feature selection for PFA versus PFB and examined performance across three candidate classifiers. RESULTS: For all EP patients with GTR, we identified four T2-MRI-based features and stratified patients into high- and low-risk groups, with 5-year overall survival rates of 62% and 100%, respectively (P < .0001). Among presumed PFA patients with GTR, four T1-MRI and five T2-MRI features predicted divergence of high- and low-risk groups, with 5-year overall survival rates of 62.7% and 96.7%, respectively (P = .002). T1-MRI-based features showed the best performance distinguishing PFA from PFB with an AUC of 0.86. CONCLUSIONS: We present machine learning strategies to identify MRI phenotypes that distinguish PFA from PFB, as well as high- and low-risk PFA. We also describe quantitative image predictors of aggressive EP tumors that might assist risk-profiling after surgery. Future studies could examine translating radiomics as an adjunct to EP risk assessment when considering therapy strategies or trial candidacy.
Authors: H Zhou; R Hu; O Tang; C Hu; L Tang; K Chang; Q Shen; J Wu; B Zou; B Xiao; J Boxerman; W Chen; R Y Huang; L Yang; H X Bai; C Zhu Journal: AJNR Am J Neuroradiol Date: 2020-07 Impact factor: 3.825
Authors: Thomas E Merchant; Anne E Bendel; Noah D Sabin; Peter C Burger; Dennis W Shaw; Eric Chang; Shengjie Wu; Tianni Zhou; David D Eisenstat; Nicholas K Foreman; Christine E Fuller; Edwina Templeton Anderson; Juliette Hukin; Ching C Lau; Ian F Pollack; Fred H Laningham; Robert H Lustig; Floyd D Armstrong; Michael H Handler; Chris Williams-Hughes; Sandra Kessel; Mehmet Kocak; David W Ellison; Vijay Ramaswamy Journal: J Clin Oncol Date: 2019-02-27 Impact factor: 44.544
Authors: Vijay Ramaswamy; Thomas Hielscher; Stephen C Mack; Alvaro Lassaletta; Tong Lin; Kristian W Pajtler; David T W Jones; Betty Luu; Florence M G Cavalli; Kenneth Aldape; Marc Remke; Martin Mynarek; Stefan Rutkowski; Sridharan Gururangan; Roger E McLendon; Eric S Lipp; Christopher Dunham; Juliette Hukin; David D Eisenstat; Dorcas Fulton; Frank K H van Landeghem; Mariarita Santi; Marie-Lise C van Veelen; Erwin G Van Meir; Satoru Osuka; Xing Fan; Karin M Muraszko; Daniela P C Tirapelli; Sueli M Oba-Shinjo; Suely K N Marie; Carlos G Carlotti; Ji Yeoun Lee; Amulya A Nageswara Rao; Caterina Giannini; Claudia C Faria; Sofia Nunes; Jaume Mora; Ronald L Hamilton; Peter Hauser; Nada Jabado; Kevin Petrecca; Shin Jung; Luca Massimi; Massimo Zollo; Giuseppe Cinalli; László Bognár; Almos Klekner; Tibor Hortobágyi; Sarah Leary; Ralph P Ermoian; James M Olson; Jeffrey R Leonard; Corrine Gardner; Wieslawa A Grajkowska; Lola B Chambless; Jason Cain; Charles G Eberhart; Sama Ahsan; Maura Massimino; Felice Giangaspero; Francesca R Buttarelli; Roger J Packer; Lyndsey Emery; William H Yong; Horacio Soto; Linda M Liau; Richard Everson; Andrew Grossbach; Tarek Shalaby; Michael Grotzer; Matthias A Karajannis; David Zagzag; Helen Wheeler; Katja von Hoff; Marta M Alonso; Teresa Tuñon; Ulrich Schüller; Karel Zitterbart; Jaroslav Sterba; Jennifer A Chan; Miguel Guzman; Samer K Elbabaa; Howard Colman; Girish Dhall; Paul G Fisher; Maryam Fouladi; Amar Gajjar; Stewart Goldman; Eugene Hwang; Marcel Kool; Harshad Ladha; Elizabeth Vera-Bolanos; Khalida Wani; Frank Lieberman; Tom Mikkelsen; Antonio M Omuro; Ian F Pollack; Michael Prados; H Ian Robins; Riccardo Soffietti; Jing Wu; Phillipe Metellus; Uri Tabori; Ute Bartels; Eric Bouffet; Cynthia E Hawkins; James T Rutka; Peter Dirks; Stefan M Pfister; Thomas E Merchant; Mark R Gilbert; Terri S Armstrong; Andrey Korshunov; David W Ellison; Michael D Taylor Journal: J Clin Oncol Date: 2016-06-06 Impact factor: 44.544
Authors: Joost J M van Griethuysen; Andriy Fedorov; Chintan Parmar; Ahmed Hosny; Nicole Aucoin; Vivek Narayan; Regina G H Beets-Tan; Jean-Christophe Fillion-Robin; Steve Pieper; Hugo J W L Aerts Journal: Cancer Res Date: 2017-11-01 Impact factor: 12.701