Discovery of Pre-Treatment FDG PET/CT-Derived Radiomics-Based Models for Predicting Outcome in Diffuse Large B-Cell Lymphoma

Cancers

Fergus Gleeson¹  Cathy Burton²  Charalampos Tsoumpas³  Andrew F. Scarsbrook⁴  Russell Frood⁴  Chirag Patel⁴  Matthew Clark⁵  Alejandro F. Frangi⁶ 

[1] Centre for Computational Imaging and Simulation Technologies in Biomedicine (CISTIB), School of Computing and School of Medicine, University of Leeds, Leeds LS2 9JT, UK;Department of Haematology, Leeds Teaching Hospitals NHS Trust, Leeds LS2 9JT, UK;Department of Nuclear Medicine and Molecular Imaging, University Medical Center of Groningen, University of Groningen, 9713 AV Groningen, The Netherlands;Department of Nuclear Medicine, Leeds Teaching Hospitals NHS Trust, Leeds LS2 9JT, UK;Department of Radiology, Leeds Teaching Hospitals NHS Trust, Leeds LS2 9JT, UK;Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, Leeds LS2 9JT, UK;

关键词: diffuse large B-cell lymphoma;    lymphoma;    predictive modelling;    radiomics;    machine learning;   

DOI  :  10.3390/cancers14071711

来源: DOAJ

【 摘 要 】

Background: Approximately 30% of patients with diffuse large B-cell lymphoma (DLBCL) will have recurrence. The aim of this study was to develop a radiomic based model derived from baseline PET/CT to predict 2-year event free survival (2-EFS). Methods: Patients with DLBCL treated with R-CHOP chemotherapy undergoing pre-treatment PET/CT between January 2008 and January 2018 were included. The dataset was split into training and internal unseen test sets (ratio 80:20). A logistic regression model using metabolic tumour volume (MTV) and six different machine learning classifiers created from clinical and radiomic features derived from the baseline PET/CT were trained and tuned using four-fold cross validation. The model with the highest mean validation receiver operator characteristic (ROC) curve area under the curve (AUC) was tested on the unseen test set. Results: 229 DLBCL patients met the inclusion criteria with 62 (27%) having 2-EFS events. The training cohort had 183 patients with 46 patients in the unseen test cohort. The model with the highest mean validation AUC combined clinical and radiomic features in a ridge regression model with a mean validation AUC of 0.75 ± 0.06 and a test AUC of 0.73. Conclusions: Radiomics based models demonstrate promise in predicting outcomes in DLBCL patients.

【 授权许可】

Unknown