Advancements in pCASL Imaging for Glioma Management Post-Surgery

pCASL reliably differentiates tumor recurrence from pseudoprogression in postoperative glioma surveillance.This noncontrast perfusion technique provided measurable separation between recurrence and treatment-related change, a distinction that can prompt earlier therapeutic escalation or avoid unnecessary interventions.

In a retrospective cohort of 120 postoperative glioma patients (65 with recurrence, 55 with pseudoprogression), mean cerebral blood flow (CBF) from pCASL maps was the primary quantitative parameter and differed significantly between groups (Mann–Whitney U, p<0.001). On receiver-operating-characteristic analysis, CBF alone yielded an area under the curve (AUC) of 0.879 (95% CI: 0.817–0.941), with sensitivity 84.6% and specificity 83.6%—indicating clear diagnostic separation based on perfusion metrics in this sample.

Radiomics-derived features (seven retained) were then combined with machine-learning classifiers; the SVM model performed best, achieving an AUC of 0.971, overall accuracy ≈88.7%, sensitivity 95.0%, and specificity 81.3%. Adding radiomics to pCASL perfusion increased discrimination from an AUC of 0.879 to 0.971, markedly improving classification performance beyond perfusion alone.

Key Takeaways:

• Noncontrast pCASL metrics plus radiomics substantially improved noninvasive differentiation of recurrence versus pseudoprogression, increasing diagnostic confidence over perfusion-only assessment.
• Postoperative glioma patients under surveillance—particularly those with early or equivocal contrast-enhancing lesions—may be reclassified toward recurrence when combined perfusion–radiomics signals are high, prompting earlier multidisciplinary review.
• These findings support incorporating targeted pCASL sequences and radiomics workflows into follow-up protocols to prioritize cases for expedited diagnostic or therapeutic decision-making when imaging suggests recurrence.