Prostate-Specific Membrane Antigen Expression in Primary and Recurrent Glioblastoma

Abstract Patients with glioblastoma almost always suffer a recurrence of an aggressive treatment resistant tumour and succumb within months highlighting the need to develop treatment options for recurrent glioblastoma. One potential target is prostate specific membrane antigen (PSMA) expressed on the new vessels in primary glioblastoma tumours. This study compared the expression of PSMA in primary and recurrent glioblastoma tumours. Formalin-fixed paraffin-embedded sections of primary and matched recurrent tumours from 13 patients with glioblastoma were processed for PSMA immunohistochemical labelling. PSMA expression was scored from digitally scanned sections using a categorical system (Total PSMA expression = PSMA label intensity (0–3) x percent PSMA-positive blood vessels). Little PSMA labelling was observed in any adjacent ‘unaffected’ brain tissue. PSMA was localised to blood vessels including those displaying microvascular proliferation in 12/13 primary and 9/13 recurrent glioblastoma tumours. Total PSMA expression scores ranged from 0 – 230 (maximum score = 300) and was divided into low and high expression based on a median split (median = 57) of PSMA expression across all tumours. Regression analysis showed a significant difference in PSMA expression between primary and recurrent glioblastoma (p = 0.04) with PSMA being highly expressed in ~70% of primary and only 31% of recurrent glioblastoma. Three cases displayed high expression in both primary and recurrent glioblastoma and one case had no PSMA expression at all. In conclusion, whilst higher expression of PSMA was detected in more primary tumours, 70%+ of all primary and recurrent tumours expressed PSMA to some extent adding evidence that this may be a useful target for treatment of glioblastoma. Larger cohorts and other techniques for detecting PSMA are needed to provide further evidence for PSMA’s utility as a target and to further define which glioblastoma patients are most likely to benefit from this approach.