P15.02.A EVALUATION OF [68GA]GA-PSMA-617 PET AS A DIAGNOSTIC AGENT IN RECURRENT GLIOBLASTOMA PATIENTS: RESULTS OF THE GENESIS GBM 001 PHASE I/II STUDY

Abstract

Abstract BACKGROUND PSMA is a cell-surface protein that is highly expressed in the endothelium of the tumour neovasculature of several solid tumours, including glioblastoma (GBM), making it an excellent target for antibody-drug conjugates or peptide receptor radionuclide therapy. Several [68Ga]Ga-PSMA-based radioligands have shown utility in the diagnosis and treatment response assessment of primary GBM patients in a number of case studies. The small-molecule ligand [68Ga]Ga-PSMA-617 has the advantage of being suitable for radiolabelling with different radioisotopes, including 68Ga, 177Lu, 111In and 90Y, which raised the possibility of using it in a theranostic application in glioblastoma. The aim of this phase I/II prospective study was to evaluate the potential of [68Ga]Ga-PSMA-617 as a PET imaging biomarker and as a candidate ligand for targeted radionuclide therapy in patients with recurrent GBM. MATERIAL AND METHODS Ten patients with recurrent glioblastoma were recruited for the study between October 2018 and September 2021. Patients underwent [68Ga]Ga-PSMA-617 and [18F]F-FET PET brain scans done on two separate days. Two patients had full-body [68Ga]Ga-PSMA-617 scans. Tumour selectivity of [68Ga]Ga-PSMA-617 was assessed by measuring tumour volume on [68Ga]Ga-PSMA-617 PET scans and comparing it with tumour volume measured on [18F]F-FET PET scans, which are more widely used clinically. Tumour specificity of [68Ga]Ga-PSMA-617 was assessed by comparing its standard uptake value (SUV) and tumour-to-brain-ratio (TBR) with those of [18F]F-FET TBR, and by measuring [68Ga]Ga-PSMA-617 tumour-to-liver ratio (TLR). RESULTS PSMA-defined tumour volume was on average 1.87±1.10 times larger than the FET-defined tumour volume (p=0.084), and the two volumes had a DICE similarity coefficient of 0.578±0.175. Mean and maximum TBR was significantly (p=0.002) higher (~10x) for PSMA than for FET. Mean SUV in the tumour was significantly higher for FET than for PSMA (p=0.002), while maximum SUV was slightly higher for PSMA than for FET (p=0.08). For the two patients with full-body [68Ga]Ga-PSMA PET scan available, PSMA mean TLR was 0.91 and 0.68, and maximum TLR was 1.20 and 1.07, respectively. CONCLUSION [68Ga] Ga-PSMA PET delineates a larger tumour volumes compared to [18F]F-FET PET, suggesting there are areas of the tumour with emerging tumour neovasculature that are not yet metabolically active. The poor spatial overlap between PSMA and FET tumour volume suggests the complementary diagnostic role of the two tracers. The higher TBR of PSMA compared to FET demonstrates the higher tumour specificity of the former, and thus its potential ability as a diagnostic agent for recurrent tumour volume delineation. Relatively low values of PSMA SUV and TLR suggest that PSMA specific uptake is not sufficient to use this ligand for targeted radionuclide therapy in glioblastoma when conjugated to [177Lu].