BACKGROUND: Prostate-specific membrane antigen (PSMA) is overexpressed on the endothelial cells of tumor neo-vessels of several solid malignancies, including differentiated thyroid cancer (DTC). We aimed to test the potential role of PSMA as a biomarker for DTC aggressiveness and outcome prediction. We retrospectively screened all patients who underwent thyroidectomy between 1 January 2010 and 31 December 2017 in our institution. Applying the inclusion (histological diagnosis of thyroid cancer and tissue availability) and exclusion criteria (no clinical or follow-up data or diagnosis of medullary thyroid cancer), a cohort of 59 patients was selected. The monoclonal mouse anti-human PSMA antibody was used to stain tissue sections. A 3-point scale was used to score PSMA positivity: 0-5% expression was considered as negative (score 0), 6-50% as moderately positive (score 1), and 51-100% as highly positive (score 2). A cumulative score (0-10%, 11-79%, and 80-100%) was also explored. Univariate and multivariate logistic regression analyses were performed to predict the presence of distant metastases, chosen as endpoint of aggressiveness. The area under the curve (AUC) was calculated. Cox models were built to predict patient outcome in terms of recurrence, iodine refractoriness, and status at last follow-up, which were calculated using the Kaplan-Meier failure function. RESULTS: At immunostaining, 12, 25, and 22 patients had scores of 0, 1, and 2, respectively. According to the cumulative score, PSMA expression was ≤ 10% in 17 cases, 11-79% in 31 cases, and ≥ 80% in 11 cases. At multivariate analysis, age, sex, histotype, vascular invasion, T and N parameters, and PSMA positivity were significant predictors of distant metastases. The AUC was 0.92. Recurrence or progression occurred in 19/59 patients. Twelve patients developed radioiodine (RAI) refractoriness, after a median time of 17 months (range 2-32). One patient died of DTC; 46 of the 58 patients alive at last follow-up were disease free. Median DFS was 23 months (range 3-82). The final multivariate model to predict RAI refractoriness included as covariates the stage, high PSMA expression (≥ 80%), and the interaction between moderate PSMA expression (11-79%) and stage. CONCLUSIONS: PSMA, a marker of neovasculature formation expressed by DTC, contributes in the prediction of tumor aggressiveness and patient outcome.