Advancing prostate cancer local recurrence detection: An analysis of optimal imaging modalities.

Abstract

e17146 Background: Prostate-specific membrane antigen positron emission tomography-computed tomography (PSMA/PET CT) has become the gold standard for staging primary prostate cancer and restaging biochemical recurrent cancer, surpassing magnetic resonance imaging (MRI). However, the role of PSMA/PET CT after radical prostatectomy in localizing the local recurrence of prostate cancer remains unclear. Two studies suggest that MRI yields a higher sensitivity for detecting local recurrence. We aim to further investigate the detectability of the local recurrence of prostate cancer after prostatectomy with PSMA/PET CT versus MRI. Methods: Cases of post-prostatectomy patients with and without local recurrent prostate cancer who received both MRI and PSMA/PET CT imaging were collected. Four radiologists reviewed each case in two phases. In the first phase MRI images were reviewed, while PSMA/PET CT images were reviewed in the second. Each phase was randomized and conducted one month apart to minimize recall. The accuracy, sensitivity, specificity, positive and negative predictive value (PPV, NPV) were calculated. Fleiss’ Kappa scores were used to determine inter-rater reliability. Generalized estimating equations were performed. Results: 37 cases of post-prostatectomy patients who received both PSMA/PET CT and MRI imaging were collected; 23 cases were considered positive for the local recurrence of prostate cancer by our institution’s Tumor Board. When using MRI, radiologists had an accuracy of 73.7%, with a sensitivity of 77.2%, a specificity of 73.1%, a PPV of 83.5%, and a NPV of 64.4%. With PSMA PET/CT, the diagnostic accuracy decreased to 66.2% (p = 0.17). PSMA/PET CT had a lower sensitivity of 65.2%, but an equal specificity of 73.1%. The PPV (81.1%) and NPV (54.3%) were lower with PSMA/PET CT. Of the four radiologists participating in our study, two had a higher accuracy with MRI, while another had a higher accuracy with PSMA/PET CT, and the fourth had an equal accuracy with both imaging methods. PSMA/PET CT showed greater inter-rater reliability than MRI (Fleiss’ Kappa score of 0.36 versus 0.22). Conclusions: MRI showed a non-statistically significant superior diagnostic performance compared to PSMA/PET CT in identifying the local recurrence of prostate cancer in patients treated with prostatectomy. However, only two of four radiologists achieved a higher accuracy when reviewing MRI studies. PSMA/PET CT also had an equal specificity and a greater inter-rater reliability, suggesting both imaging techniques may be valuable in identifying prostate cancer recurrence.