Similarity of Biological Information Captured by [68]Ga-PSMA-11 PET and ASL Perfusion MRI in Glioblastoma.

BACKGROUND AND PURPOSE: The overexpression of prostate-specific membrane antigen (PSMA) in the microvasculature of glioblastoma (GBM), in contrast to its negligible expression in normal brain parenchyma, provides a compelling rationale for using PSMA-targeted radioligands in the diagnosis and radiation therapy of GBM. Arterial spin-labeling (ASL) perfusion imaging, an advanced MRI technique, allows noninvasive assessment of tumor microvascular density and perfusion. This study aimed to investigate the relationship between gallium 68 [68]Ga-PSMA-11 PET and ASL perfusion imaging, as well as the correlation between PSMA tracer uptake and immunohistochemical microvascular indices.

MATERIALS AND

METHODS

Twenty-six patients with newly diagnosed or recurrent GBM who underwent conventional MRI, ASL, and PSMA PET were prospectively enrolled. We investigated the correlation between the semiquantitative parameters of ASL and PSMA PET. Spatial similarity and overlap between tumor volumes delineated by contrast-enhanced (CE) MRI, ASL, and PSMA PET were evaluated using Dice Similarity Coefficient (DSC) and overlap volume (OV). Additionally, microvascular density values and PSMA expression levels in the tumor tissue samples were assessed using immunohistochemistry, and their correlation with PSMA tracer uptake was analyzed.

RESULTS

A total of 28 lesions was found in 26 patients, all of whom showed high perfusion, moderate-to-high tracer uptake, and contrast enhancement. In 23 lesions (82%), the areas with the highest tracer uptake on PSMA PET were precisely localized to the regions of the highest perfusion on ASL. A strong positive correlation was found between the parameters derived from ASL and PSMA PET (P < .001). PSMA PET exhibited moderate-to-high spatial similarity and overlap with ASL (mean DSC, 0.62 [SD, 0.17]; mean OV, 0.78 [SD, 0.12]) and with CE MRI (mean DSC, 0.65 [SD, 0.14]; mean OV, 0.79 [SD, 0.15]). Furthermore, a positive correlation was observed between PSMA tracer uptake and both microvascular density and PSMA expression.

CONCLUSIONS

Our results highlight the similarity of biologic information captured by PSMA PET and ASL perfusion imaging in GBM. The tracer uptake observed in PSMA PET can also reliably reflect the spatial heterogeneity of the microvascular distribution of GBM. These findings provide compelling support for PSMA-targeted radiation therapy as an anti-angiogenic therapeutic strategy for patients with GBM, while also advancing our understanding of the mechanisms underlying PSMA tracer uptake.