Imaging modalities with high accuracy are essential for proper selection of patients for salvage lymph node dissection (sLND). Unlike nodal staging before radical prostatectomy, data on prostate-specific membrane antigen (PSMA) positron emission tomography (PET) before sLND are scarce. We analyzed 23 patients undergoing 68Ga-PSMA-PET before sLND and compared the imaging findings with histopathology at sLND. Altogether, 29/109 resected lymph node (LN) fields harbored histologically confirmed LN metastases (26.6%)., The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 89.5%, 74.1%, 70.8%, 90.9%, and 80.4% in side-based analysis, and 75.9%, 87.5%, 68.8%, 90.9%, and 84.4%, respectively, in LN field-based analysis. In contrast to existing studies, sensitivity in particular was lower. This might be because of referral for sLND on the basis of PSMA-PET from 13 different nuclear medicine centers. However, this could still provide more realistic data if sLND should become widely used. Second, our pathologists routinely used immunohistochemistry, resulting in a higher yield of smaller metastases, increasing the accuracy of histopathology. Despite its acceptable accuracy in a real-world scenario, sLND should not only focus on PSMA-PET-positive fields, as small PSMA-PET-negative metastases may be present. PATIENT SUMMARY: We analyzed the accuracy of prostate-specific membrane antigen positron emission tomography (PSMA-PET) before salvage lymph node dissection in a real-world situation. Although the accuracy is reasonable, resection should not exclusively focus on PSMA-PET-positive fields.
Keywords: Accuracy; Nodal recurrence; Positron emission tomography; Prostate-specific membrane antigen; Radical prostatectomy; Salvage lymph node dissection.
Copyright © 2018. Published by Elsevier B.V.