Diagnostic performance of MRI for prediction of recurrent prostate cancer after high-intensity focused ultrasound: a systematic review and meta-analysis

PURPOSE: This article aims to evaluate the pooled diagnostic performance control MRI for prediction of recurrent prostate cancer (PCa) after high-intensity focused ultrasound (HIFU). MATERIALS AND METHODS: MEDLINE, EMBASE, and Cochrane library databases up to December 31, 2021, were searched. We included studies providing 2×2 contingency table for diagnostic performance of MRI in predicting recurrent PCa after HIFU, using control biopsy as reference standard. The quality of the included studies was assessed using Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2). Sensitivity and specificity were pooled and displayed in a summary receiver operating characteristics (SROC) plot. Meta-regression analysis using clinically relevant covariates was performed for the causes of heterogeneity. RESULTS: Nineteen studies (703 patients) were included. All included studies satisfied at least four of the seven QUADAS-2 domains. Pooled sensitivity was 0.81 (95% CI 0.72–0.90) with specificity of 0.91 (95% CI 0.86–0.96), with area under the SROC curve of 0.81. Larger studies including more than 50 patients showed relatively poor sensitivity (0.68 vs. 0.84) and specificity (0.75 vs. 0.93). The diagnostic performance of studies reporting higher nadir serum prostate-specific antigen levels (>1 ng/mL) after HIFU was inferior, and differed significantly in sensitivity (0.54 vs. 0.78) rather than specificity (0.85 vs. 0.91). CONCLUSIONS: Although MRI showed adequate diagnostic performance in predicting PCa recurrence after HIFU, these results may have been exaggerated.