Texture analysis of 18F-FDG PET/CT and CECT: Prediction of refractoriness of Hodgkin lymphoma with mediastinal bulk involvement

  • 0Nuclear Medicine Unit, Department of Experimental and Clinical Biomedical Sciences 'Mario Serio', University of Florence, Florence, Italy.

|

|

Summary

This summary is machine-generated.

Identifying patients with refractory Hodgkin Lymphoma (HL) is crucial. This study found that combining PET scan texture analysis (HISTO_EntropyPET) with clinical staging (GHSG stage) accurately predicts treatment refractoriness in newly diagnosed classical HL.

Area Of Science

  • Oncology
  • Radiology
  • Medical Imaging

Background

  • Accurate prognostication is vital for newly diagnosed classical Hodgkin Lymphoma (cHL) patients, especially those with mediastinal bulk, to identify high-risk individuals needing intensified therapy.
  • Current prognostic models require enhancement with novel parameters to improve the stratification of patients likely to experience refractory disease.

Purpose Of The Study

  • To investigate the predictive value of metabolic and texture features from baseline 18F-FDG PET/CT and CECT scans, alongside clinical data, for first-line therapy refractoriness in cHL.
  • To identify independent predictors of refractoriness and assess their combined prognostic power.

Main Methods

  • Retrospective review of 69 cHL patients who underwent staging PET and CECT scans.
  • Extraction of metabolic and texture features using LIFEx software; analysis of clinical data, including GHSG staging.
  • Statistical evaluation using ROC, Cox regression, and Kaplan-Meier analyses to determine prognostic significance and independent predictors.

Main Results

  • GHSG stage was a significant clinical predictor of refractoriness.
  • Second-order texture features from CECT (GLCM, GLRLM) and various PET features (SUVmean, shape, GLCM, GLRLM, NGLDM) showed prognostic power.
  • HISTO_EntropyPET from PET scans and GHSG stage emerged as significant independent predictors at multivariate analysis.

Conclusions

  • Radiomic features from PET/CT, particularly HISTO_EntropyPET, offer valuable prognostic information for mediastinal bulky cHL.
  • Combining HISTO_EntropyPET with GHSG stage significantly improves the prediction of treatment refractoriness, identifying distinct risk groups.