Association of [<sup>18</sup>F]-FDG PET/CT-Derived Radiomic Features with Clinical Outcomes and Genomic Profiles in Patients with Chronic Lymphocytic Leukemia

Association of [<sup>18</sup>F]-FDG PET/CT-Derived Radiomic Features with Clinical Outcomes and Genomic Profiles in Patients with Chronic Lymphocytic Leukemia

<b>Background:</b> The role of PET/CT imaging in chronic lymphoproliferative syndromes (CLL) is debated. This study examines the potential of PET/CT radiomics in predicting outcomes and genetic profiles in CLL patients. <b>Methods:</b> A retrospective analysis was conducted o...

Full description

Saved in:
Bibliographic Details
Main Authors: Fabiana Esposito, Luigi Manco, Guglielmo Manenti, Livio Pupo, Andrea Nunzi, Roberta Laureana, Luca Guarnera, Massimiliano Marinoni, Elisa Buzzatti, Paola Elda Gigliotti, Andrea Micillo, Giovanni Scribano, Adriano Venditti, Massimiliano Postorino, Maria Ilaria Del Principe
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Diagnostics
Subjects:
chronic lymphocytic leukemia
radiomics
PET/CT
machine learning
genetic profile
Online Access:https://www.mdpi.com/2075-4418/15/10/1281
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:<b>Background:</b> The role of PET/CT imaging in chronic lymphoproliferative syndromes (CLL) is debated. This study examines the potential of PET/CT radiomics in predicting outcomes and genetic profiles in CLL patients. <b>Methods:</b> A retrospective analysis was conducted on 50 CLL patients treated at Policlinico Tor Vergata, Rome, and screened, at diagnosis, with [<sup>18</sup>F]-FDG PET/CT. Potentially pathological lymph nodes were semi-automatically segmented. Genetic mutations in <i>TP53</i>, <i>NOTCH1</i>, and <i>IGVH</i> were assessed. Eight hundred and sixty-five radiomic features were extracted, with the cohort split into training (70%) and validation (30%) sets. Four machine learning models, each with Random Forest, Stochastic Gradient Descent, and Support Vector Machine learners, were trained. <b>Results:</b> Progression occurred in 10 patients. The selected radiomic features from CT and PET datasets were correlated with four models of progression and mutations (<i>TP53</i>, <i>NOTCH1</i>, <i>IGVH</i>). The Random Forest models outperformed others in predicting progression (AUC = 0.94/0.88, CA = 0.87/0.75, TP = 80.00%/87.50%, TN = 72.70%/87.50%) and the occurrence of <i>TP53</i> (AUC = 0.94/0.96, CA = 0.87/0.80, TP = 87.50%/90.21%, TN = 85.70%/90.90%), and <i>NOTCH1</i> (AUC = 0.94/0.85, CA = 0.87/0.67, TP = 80.00%/88.90%, TN = 80.00%/83.30%)mutations. The IGVH models showed poorer performance. <b>Conclusions:</b> ML models based on PET/CT radiomic features effectively predict outcomes and genetic profiles in CLL patients.
ISSN:2075-4418

Similar Items