Detection of positive resection margins in surgical procedures of high-risk prostate cancer is key for minimizing the risk of recurrence. This study aimed at evaluating the accuracy of functional tumor-volume segmentation in intraoperative ex vivo PET/CT for margin assessment in prostate cancer patients undergoing radical prostatectomy. Methods: Seven high-risk prostate cancer patients received [18F]PSMA-1007 before radical prostatectomy. After removal of the prostate gland, ex vivo imaging on the AURA 10 PET/CT system was performed, and functional tumor volume was segmented using 4 semiautomatic segmentation methods. Resection margins and volumes were compared with histopathology. Additionally, a supportive phantom study was conducted to assess segmentation accuracy at low radiopharmaceutical activity. Results: Clinically, 18 lesions were analyzed in intraoperative PET/CT. Sensitivity, specificity, and positive and negative predictive values of margin detection were 83%, 100%, 100%, and 92%, respectively, using an iterative thresholding method. In 1 patient, a biochemical recurrence was observed within 1 y of prostate-specific antigen follow-up, and 1 patient underwent adjuvant radiotherapy. The remaining 5 patients were still undergoing prostate-specific antigen follow-up with no evidence of biochemical recurrence. On the basis of a phantom-deduced minimal segmentable activity concentration of approximately 2 kBq/mL, we propose an administered [18F]PSMA-1007 activity of at least 1.9 and 0.4 MBq/kg for preoperative and intraoperative injections, respectively. Conclusion: Intraoperative ex vivo PET/CT is a promising modality for intraoperative margin assessment. Prospective trials are needed to further investigate the value of specimen PET/CT-based radioguided surgery in high-risk prostate cancer.