Advances in Radiation Oncology (May 2022)
Development and Optimization of a Machine-Learning Prediction Model for Acute Desquamation After Breast Radiation Therapy in the Multicenter REQUITE Cohort
- Mahmoud Aldraimli, PhD,
- Sarah Osman, PhD,
- Diana Grishchuck, MSc,
- Samuel Ingram, MSc,
- Robert Lyon, PhD,
- Anil Mistry, MSc,
- Jorge Oliveira, PhD,
- Robert Samuel, MBChB,
- Leila E.A. Shelley, PhD,
- Daniele Soria, PhD,
- Miriam V. Dwek, PhD,
- Miguel E. Aguado-Barrera, MD, PhD,
- David Azria, MD,
- Jenny Chang-Claude, PhD,
- Alison Dunning, PhD,
- Alexandra Giraldo, MD,
- Sheryl Green, MD,
- Sara Gutiérrez-Enríquez, PhD,
- Carsten Herskind, PhD,
- Hans van Hulle, MD,
- Maarten Lambrecht, MD,
- Laura Lozza, MD,
- Tiziana Rancati, MSc,
- Victoria Reyes, MD,
- Barry S. Rosenstein, PhD,
- Dirk de Ruysscher, MD,
- Maria C. de Santis, MD,
- Petra Seibold, PhD,
- Elena Sperk, MD,
- R. Paul Symonds, MD,
- Hilary Stobart,
- Begoña Taboada-Valadares, MD,
- Christopher J. Talbot, PhD,
- Vincent J.L. Vakaet, MD,
- Ana Vega, PhD,
- Liv Veldeman, MD, PhD,
- Marlon R. Veldwijk, PhD,
- Adam Webb, PhD,
- Caroline Weltens, MD,
- Catharine M. West, PhD,
- Thierry J. Chaussalet, PhD,
- Tim Rattay, MBChB, PhD
Affiliations
- Mahmoud Aldraimli, PhD
- Health Innovation Ecosystem, University of Westminster, London, United Kingdom
- Sarah Osman, PhD
- Patrick G. Johnston Centre for Cancer Research, Queen's University Belfast, Belfast, United Kingdom
- Diana Grishchuck, MSc
- Imperial College Healthcare NHS Trust, London, United Kingdom
- Samuel Ingram, MSc
- Division of Cancer Sciences, Faculty of Biology, Medicine and Health, University of Manchester, Manchester, United Kingdom
- Robert Lyon, PhD
- Department of Computer Science, Edge Hill University, Ormskirk, Lancashire, United Kingdom
- Anil Mistry, MSc
- Guy's and St. Thomas’ NHS Foundation Trust, London, United Kingdom
- Jorge Oliveira, PhD
- Mirada Medical, Oxford, United Kingdom
- Robert Samuel, MBChB
- University of Leeds, Leeds Cancer Centre, St. James's University Hospital, Leeds, United Kingdom
- Leila E.A. Shelley, PhD
- Edinburgh Cancer Centre, Western General Hospital, Edinburgh, United Kingdom
- Daniele Soria, PhD
- School of Computing, University of Kent, Canterbury, United Kingdom
- Miriam V. Dwek, PhD
- School of Life Sciences, University of Westminster, London, United Kingdom
- Miguel E. Aguado-Barrera, MD, PhD
- Fundación Publica Galega Medicina Xenomica, Santiago de Compostela, Spain; Instituto de Investigación Sanitaria de Santiago (IDIS), Servicio Galego de Saúde (SERGAS), Santiago de Compostela, Spain
- David Azria, MD
- University of Montpellier, Montpellier, France
- Jenny Chang-Claude, PhD
- Division of Cancer Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany; UKE University Cancer Center Hamburg, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
- Alison Dunning, PhD
- Centre for Cancer Genetic Epidemiology, University of Cambridge, Strangeways Research Laboratory, Worts Causeway, Cambridge, United Kingdom
- Alexandra Giraldo, MD
- Radiation Oncology Department, Vall d'Hebron Hospital Universitari, Vall d'Hebron Hospital Campus, Barcelona, Spain
- Sheryl Green, MD
- Department of Radiation Oncology, Icahn School of Medicine at Mount Sinai, New York, New York
- Sara Gutiérrez-Enríquez, PhD
- Hereditary Cancer Genetics Group, Vall d'Hebron Institute of Oncology (VHIO), Vall d'Hebron Hospital Campus, Barcelona, Spain
- Carsten Herskind, PhD
- Department of Radiation Oncology, Universitätsmedizin Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
- Hans van Hulle, MD
- Department of Human Structure and Repair, Ghent University, Ghent, Belgium
- Maarten Lambrecht, MD
- Department of Radiation Oncology, University Hospital, Leuven, Belgium
- Laura Lozza, MD
- Department of Radiation Oncology 1, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
- Tiziana Rancati, MSc
- Prostate Cancer Program, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
- Victoria Reyes, MD
- Radiation Oncology Department, Vall d'Hebron Hospital Universitari, Vall d'Hebron Hospital Campus, Barcelona, Spain
- Barry S. Rosenstein, PhD
- Icahn School of Medicine at Mount Sinai, New York, New York
- Dirk de Ruysscher, MD
- Maastricht University Medical Center, Department of Radiation Oncology (Maastro), GROW, Maastricht, The Netherlands
- Maria C. de Santis, MD
- Department of Radiation Oncology 1, Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
- Petra Seibold, PhD
- Division of Cancer Epidemiology, German Cancer Research Center (DKFZ), Heidelberg, Germany
- Elena Sperk, MD
- Department of Radiation Oncology, Universitätsmedizin Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
- R. Paul Symonds, MD
- Cancer Research Centre, University of Leicester, Leicester, United Kingdom
- Hilary Stobart
- Independent Cancer Patients’ Voice, London, United Kingdom
- Begoña Taboada-Valadares, MD
- Fundación Publica Galega Medicina Xenomica, Santiago de Compostela, Spain; Department of Radiation Oncology, Complexo Hospitalario Universitario de Santiago, Servicio Galego de Saúde (SERGAS), Santiago de Compostela, Spain
- Christopher J. Talbot, PhD
- Department of Genetics and Genome Biology, University of Leicester, Leicester, United Kingdom
- Vincent J.L. Vakaet, MD
- Department of Human Structure and Repair, Ghent University, Ghent, Belgium
- Ana Vega, PhD
- Fundación Publica Galega Medicina Xenomica, Santiago de Compostela, Spain; Instituto de Investigación Sanitaria de Santiago (IDIS), Servicio Galego de Saúde (SERGAS), Santiago de Compostela, Spain; Biomedical Network on Rare Diseases (CIBERER), Madrid, Spain
- Liv Veldeman, MD, PhD
- Department of Human Structure and Repair, Ghent University, Ghent, Belgium
- Marlon R. Veldwijk, PhD
- Department of Radiation Oncology, Universitätsmedizin Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany
- Adam Webb, PhD
- Department of Genetics and Genome Biology, University of Leicester, Leicester, United Kingdom
- Caroline Weltens, MD
- Department of Radiation Oncology, University Hospital, Leuven, Belgium
- Catharine M. West, PhD
- University of Manchester, Christie Hospital, Manchester, United Kingdom
- Thierry J. Chaussalet, PhD
- Health Innovation Ecosystem, University of Westminster, London, United Kingdom
- Tim Rattay, MBChB, PhD
- Cancer Research Centre, University of Leicester, Leicester, United Kingdom; Corresponding author
- Journal volume & issue
-
Vol. 7,
no. 3
p. 100890
Abstract
Purpose: Some patients with breast cancer treated by surgery and radiation therapy experience clinically significant toxicity, which may adversely affect cosmesis and quality of life. There is a paucity of validated clinical prediction models for radiation toxicity. We used machine learning (ML) algorithms to develop and optimise a clinical prediction model for acute breast desquamation after whole breast external beam radiation therapy in the prospective multicenter REQUITE cohort study. Methods and Materials: Using demographic and treatment-related features (m = 122) from patients (n = 2058) at 26 centers, we trained 8 ML algorithms with 10-fold cross-validation in a 50:50 random-split data set with class stratification to predict acute breast desquamation. Based on performance in the validation data set, the logistic model tree, random forest, and naïve Bayes models were taken forward to cost-sensitive learning optimisation. Results: One hundred and ninety-two patients experienced acute desquamation. Resampling and cost-sensitive learning optimisation facilitated an improvement in classification performance. Based on maximising sensitivity (true positives), the “hero” model was the cost-sensitive random forest algorithm with a false-negative: false-positive misclassification penalty of 90:1 containing m = 114 predictive features. Model sensitivity and specificity were 0.77 and 0.66, respectively, with an area under the curve of 0.77 in the validation cohort. Conclusions: ML algorithms with resampling and cost-sensitive learning generated clinically valid prediction models for acute desquamation using patient demographic and treatment features. Further external validation and inclusion of genomic markers in ML prediction models are worthwhile, to identify patients at increased risk of toxicity who may benefit from supportive intervention or even a change in treatment plan.