Diagnostic value of multiparametric MRI in detecting residual or recurrent prostate cancer after high-intensity focused ultrasound

Purpose To evaluate the diagnostic performance of follow-up multiparametric MRI for prediction of recurrent prostate cancer after high-intensity focused ultrasound (HIFU), and to find other, if any, clinical or radiological predictors. Materials and methods Post-HIFU MRIs of 110 consecutive patients who underwent follow-up biopsies between August 2019 and April 2021 were retrospectively analyzed and the likelihood of recurrence was assessed on a five-point Likert scale by two board-certified uroradiologists. Diagnostic performance of the Likert scale assigned to the post-HIFU MRI was assessed using the follow-up biopsy results as a reference standard. Among the clinical and radiological variables, predictors of the recurrence were examined through logistic regression. Results In per-patient and per-sector analyses, Likert scale on post-HIFU MRI showed a sensitivity and specificity of 0.37 and 0.97, and 0.42 and 0.87, respectively, in predicting recurrence. Two patients with high suspicion on MRI required additional treatment to regain biochemical control despite negative biopsies. High suspicion on post-HIFU MRI (odds ratio = 1.74; p < 0.01), and more cancer-positive cores on initial biopsy (odds ratio = 1.25; p = 0.03) were independent predictors of recurrence. Conclusion Albeit with low sensitivity, high suspicion on post-HIFU MRI may be clinically important because of its high specificity, especially when considering the possibility of sampling error in biopsies. Patients with a high number of cancer-positive cores at diagnosis should avoid HIFU as they have an increased risk of recurrence.

Automated summary

This study evaluated the diagnostic performance of follow-up multiparametric MRI after high-intensity focused ultrasound (HIFU) for predicting recurrent prostate cancer and identified other clinical or radiological predictors. The post-HIFU MRIs of 110 patients who underwent follow-up biopsies were analyzed, and the likelihood of recurrence was assessed using a Likert scale. Results showed that high suspicion on post-HIFU MRI and a higher number of cancer-positive cores on initial biopsy were independent predictors of recurrence. Although the sensitivity of post-HIFU MRI was low, its high specificity suggests clinical significance, especially considering the possibility of sampling error in biopsies.