Development and Optimization of a Machine-Learning Prediction Model for Acute Desquamation After Breast Radiation Therapy in the Multicenter REQUITE Cohort

Mahmoud Aldraimli, Sarah Osman, Diana Grishchuck, Samuel Ingram, Robert Lyon, Anil Mistry, Jorge Oliveira, Robert Samuel, Leila E.A. Shelley, Daniele Soria, Miriam V. Dwek, Miguel E. Aguado-Barrera, David Azria, Jenny Chang-Claude, Alison Dunning, Alexandra Giraldo, Sheryl Green, Sara Gutiérrez-Enríquez, Carsten Herskind, Hans van HulleMaarten Lambrecht, Laura Lozza, Tiziana Rancati, Victoria Reyes, Barry S. Rosenstein, Dirk de Ruysscher, Maria C. de Santis, Petra Seibold, Elena Sperk, R. Paul Symonds, Hilary Stobart, Begoña Taboada-Valadares, Christopher J. Talbot, Vincent J.L. Vakaet, Ana Vega, Liv Veldeman, Marlon R. Veldwijk, Adam Webb, Caroline Weltens, Catharine M. West, Thierry J. Chaussalet, Tim Rattay*

*Corresponding author for this work

Research output: Contribution to journalArticle (journal)peer-review

13 Downloads (Pure)

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.
Original languageEnglish
Pages (from-to)100890
JournalAdvances in Radiation Oncology
Volume7
Issue number3
Early online date10 Feb 2022
DOIs
Publication statusE-pub ahead of print - 10 Feb 2022

Keywords

  • breast cancer
  • radiation therapy

Fingerprint

Dive into the research topics of 'Development and Optimization of a Machine-Learning Prediction Model for Acute Desquamation After Breast Radiation Therapy in the Multicenter REQUITE Cohort'. Together they form a unique fingerprint.

Cite this