Dynamic Susceptibility Perfusion Imaging for Differentiating Progressive Disease from Pseudoprogression in Diffuse...

Area of Science:

• Neuro-oncology
• Radiology
• Molecular Pathology

Background:

• Advanced adjuvant therapy for diffuse gliomas often leads to ambiguous follow-up imaging findings.
• Differentiating progressive disease (PD) from pseudoprogression (PsP) is crucial for effective treatment management.

Purpose of the Study:

• To evaluate the added value of dynamic susceptibility perfusion imaging in distinguishing PD from PsP across different molecular glioma subtypes.
• To compare the predictive performance of a multilevel classification model with a decision tree model (DTM) incorporating advanced perfusion parameters.

Main Methods:

• Retrospective analysis of 89 patients with diffuse gliomas and tumor-suspect lesions post-treatment.
• Classification of lesions based on molecular subtype (Astro-IDHwt, Astro-IDHmut, oligodendroglioma), presence of contrast-enhancing (CE) lesions, and relative cerebral blood volume (rCBV) histogram parameters.
• Comparison of a multilevel classification approach with a DTM using additional rCBV parameters for PD prediction.

Main Results:

• The overall PD rate was 69%, varying by subtype (Astro-IDHwt: 86%, Astro-IDHmut: 52%, oligodendrogliomas: 55%).
• The presence of CE lesions correlated with higher PD rates across all subtypes.
• The DTM incorporating additional rCBV parameters achieved high PD prediction accuracy (93-100% across subtypes).

Conclusions:

• Integrating molecular tumor type, CE lesion status, and rCBV parameters enhances the accuracy of PD prediction in diffuse gliomas.
• Dynamic susceptibility perfusion imaging offers valuable insights for managing patients with treated diffuse gliomas.
• This multimodal approach aids in precise diagnosis and treatment planning for glioma patients.