Histogram-based analysis of cerebral blood flow using arterial spin labeling MRI in de novo brain gliomas: relationship to histopathologic grade and molecular markers

Purpose We developed multiple histogram-based CBF indices and evaluated their association with histopathologic grade in de novo brain tumor patients. Furthermore, the associations between these advanced CBF indices and molecular markers, including IDH1 mutation, ATRX loss, and 1p/19q co-deletion were also investigated. Methods Thirteen de novo brain tumor patients (age 21-68 years, 9 M/4F) who were enrolled in our prospective study were scanned on 3 T MRI using a pCASL perfusion sequence following IRB-approved written informed consent. All patients have since undergone surgical intervention with tissue sampling for histopathologic tumor grading and molecular marker assessment. Tumor region of interest (ROI) were manually delineated on FLAIR images including the full extent of the tumor and peritumoral edema. Fourteen rCBF indices were derived from the histogram of the voxels with the ROI. Multi-linear regression was then used to compare rCBF indices with histopathologic tumor grade and molecular markers. Results Averaged rCBF in top 10 and top 20 voxels (p < 0.004), but not the entire tumor ROI, was positively associated with WHO tumor grade. After accounting for tumor grade, the presence of 1p/19q co-deletion was associated with higher rCBF in top voxels, as well as with standard deviation of rCBF in the tumor ROI (p < 0.001). ATRX retention was related to higher rCBF, and this effect appears to be present in both higher-perfusion (p < 0.004) and low-perfusion (p < 0.05) voxels. IDH mutation was not significantly associated with any of the CBF indices investigated. Conclusion ASL MRI may provide useful supplemental noninvasive imaging assessment of brain tumor grade and molecular marker status.

Automated summary

This study developed multiple histogram-based CBF indices and evaluated their association with histopathologic grade in de novo brain tumor patients. The study also investigated the associations between these CBF indices and molecular markers, such as IDH1 mutation, ATRX loss, and 1p/19q co-deletion. The results showed that ASL MRI could provide useful noninvasive imaging assessment of brain tumor grade and molecular marker status, with specific correlations found between certain CBF indices and tumor grade, as well as molecular markers like 1p/19q co-deletion and ATRX retention.