TY - JOUR
T1 - Utility of pre-treatment FDG PET/CT–derived machine learning models for outcome prediction in classical Hodgkin lymphoma
AU - Frood, Russell
AU - Clark, Matt
AU - Burton, Cathy
AU - Tsoumpas, Charalampos
AU - Frangi, Alejandro F.
AU - Gleeson, Fergus
AU - Patel, Chirag
AU - Scarsbrook, Andrew
N1 - Funding Information:
R Frood, F Gleeson, C Patel and A Scarsbrook received funding from Innovate UK. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Publisher Copyright:
© 2022, The Author(s).
PY - 2022/10/1
Y1 - 2022/10/1
N2 - Objectives: Relapse occurs in ~20% of patients with classical Hodgkin lymphoma (cHL) despite treatment adaption based on 2-deoxy-2-[18F]fluoro-d-glucose positron emission tomography/computed tomography response. The objective was to evaluate pre-treatment FDG PET/CT–derived machine learning (ML) models for predicting outcome in patients with cHL. Methods: All cHL patients undergoing pre-treatment PET/CT at our institution between 2008 and 2018 were retrospectively identified. A 1.5 × mean liver standardised uptake value (SUV) and a fixed 4.0 SUV threshold were used to segment PET/CT data. Feature extraction was performed using PyRadiomics with ComBat harmonisation. Training (80%) and test (20%) cohorts stratified around 2-year event-free survival (EFS), age, sex, ethnicity and disease stage were defined. Seven ML models were trained and hyperparameters tuned using stratified 5-fold cross-validation. Area under the curve (AUC) from receiver operator characteristic analysis was used to assess performance. Results: A total of 289 patients (153 males), median age 36 (range 16–88 years), were included. There was no significant difference between training (n = 231) and test cohorts (n = 58) (p value > 0.05). A ridge regression model using a 1.5 × mean liver SUV segmentation had the highest performance, with mean training, validation and test AUCs of 0.82 ± 0.002, 0.79 ± 0.01 and 0.81 ± 0.12. However, there was no significant difference between a logistic model derived from metabolic tumour volume and clinical features or the highest performing radiomic model. Conclusions: Outcome prediction using pre-treatment FDG PET/CT–derived ML models is feasible in cHL patients. Further work is needed to determine optimum predictive thresholds for clinical use. Key points: • A fixed threshold segmentation method led to more robust radiomic features. • A radiomic-based model for predicting 2-year event-free survival in classical Hodgkin lymphoma patients is feasible. • A predictive model based on ridge regression was the best performing model on our dataset.
AB - Objectives: Relapse occurs in ~20% of patients with classical Hodgkin lymphoma (cHL) despite treatment adaption based on 2-deoxy-2-[18F]fluoro-d-glucose positron emission tomography/computed tomography response. The objective was to evaluate pre-treatment FDG PET/CT–derived machine learning (ML) models for predicting outcome in patients with cHL. Methods: All cHL patients undergoing pre-treatment PET/CT at our institution between 2008 and 2018 were retrospectively identified. A 1.5 × mean liver standardised uptake value (SUV) and a fixed 4.0 SUV threshold were used to segment PET/CT data. Feature extraction was performed using PyRadiomics with ComBat harmonisation. Training (80%) and test (20%) cohorts stratified around 2-year event-free survival (EFS), age, sex, ethnicity and disease stage were defined. Seven ML models were trained and hyperparameters tuned using stratified 5-fold cross-validation. Area under the curve (AUC) from receiver operator characteristic analysis was used to assess performance. Results: A total of 289 patients (153 males), median age 36 (range 16–88 years), were included. There was no significant difference between training (n = 231) and test cohorts (n = 58) (p value > 0.05). A ridge regression model using a 1.5 × mean liver SUV segmentation had the highest performance, with mean training, validation and test AUCs of 0.82 ± 0.002, 0.79 ± 0.01 and 0.81 ± 0.12. However, there was no significant difference between a logistic model derived from metabolic tumour volume and clinical features or the highest performing radiomic model. Conclusions: Outcome prediction using pre-treatment FDG PET/CT–derived ML models is feasible in cHL patients. Further work is needed to determine optimum predictive thresholds for clinical use. Key points: • A fixed threshold segmentation method led to more robust radiomic features. • A radiomic-based model for predicting 2-year event-free survival in classical Hodgkin lymphoma patients is feasible. • A predictive model based on ridge regression was the best performing model on our dataset.
KW - Hodgkin disease: positron emission tomography computed tomography
KW - Machine learning, progression-free survival
KW - Humans
KW - Middle Aged
KW - Neoplasm Recurrence, Local
KW - Male
KW - Machine Learning
KW - Hodgkin Disease/diagnostic imaging
KW - Young Adult
KW - Adolescent
KW - Aged, 80 and over
KW - Adult
KW - Fluorodeoxyglucose F18/metabolism
KW - Aged
KW - Retrospective Studies
KW - Positron Emission Tomography Computed Tomography/methods
KW - Positron-Emission Tomography/methods
UR - http://www.scopus.com/inward/record.url?scp=85137013428&partnerID=8YFLogxK
U2 - 10.1007/s00330-022-09039-0
DO - 10.1007/s00330-022-09039-0
M3 - Article
C2 - 36006428
AN - SCOPUS:85137013428
SN - 0938-7994
VL - 32
SP - 7237
EP - 7247
JO - European Radiology
JF - European Radiology
IS - 10
ER -
