Diagnostic performance of 68Gallium-PSMA-11 PET/CT to detect significant prostate cancer and comparison with 18FEC PET/CT

Abstract

// Manuela A. Hoffmann 1, 2, 3 , Matthias Miederer 2 , Helmut J. Wieler 3 , Christian Ruf 4 , Frank M. Jakobs 5 and Mathias Schreckenberger 2 1 Supervisory Center for Medical Radiation Protection, Bundeswehr Medical Service Headquarters, Koblenz, Germany 2 Department of Nuclear Medicine, Johannes Gutenberg-University, Mainz, Germany 3 Department of Nuclear Medicine, Bundeswehr Central Hospital, Koblenz, Germany 4 Department of Urology, Bundeswehr Central Hospital, Koblenz, Germany 5 Department of Epidemiology, German Air Force Center for Aerospace Medicine, Fürstenfeldbruck, Germany Correspondence to: Manuela A. Hoffmann, email: manuhoffmann@web.de Keywords: prostate cancer; positron emission tomography/computed tomography; prostate-specific membrane antigen; choline; detection of significant cancer Received: August 16, 2017      Accepted: October 27, 2017      Published: November 14, 2017 ABSTRACT Background : Radiolabeled prostate-specific membrane antigen (PSMA) has proven to be a highly accurate method to detect recurrence and metastases of prostate cancer, but only sparse data is available about its performance in the diagnosis of clinically significant primary prostate cancer. Methods : We compared 68 Ga-PSMA-11 PET/CT in 25 patients with 18 FEC PET/CT in 40 patients with suspected prostate carcinoma based on an increased PSA level. The PET/CT results were compared with the histopathologic Gleason Score (GS) of biopsies. Results : The 68 Ga-PSMA-11 PET/CT revealed highly suspect prostatic lesions (maximum standardized uptake value/SUV max >2.5) in 21/25 patients (84%), associated with GS≥6 (low-grade/high-grade carcinoma). Two histopathologic non-malignancy-relevant cases (GS<6) had PSMA-SUV max ≤2.5; all histopathologic high-grade cases (GS≥7b) showed PSMA-SUV max >12.0 which further increased with rising GS. There were 2 false positives and no false negative findings for high-grade prostate cancer using a cut off-level for SUV max of 2.5. In contrast, the 18 FEC PET/CT showed suspected malignant lesions in 38/40 patients (95%), which included 3 lesions with GS<6. The mean SUV max values did not differ with different GS. There were 11 false positives and 1 false negative for detection of high-grade prostate cancer (cut off 2.5). By means of ROC analysis a SUV max of 5.4 was found to be an optimal cut off-level to distinguish between low- and high-grade carcinoma in 68 Ga-PSMA-11 PET/CT (AUC=0.9692; 95% CI 0.9086;1.0000;SD(AUC)=0.0309)). Choosing a cut off-level of SUV max 5.4, 68 Ga-PSMA-11 PET/CT was able to distinguish between GS ≤7a/≥7b with a sensitivity of 84%, a specificity of 100%, a negative predictive value (NPV) of 67%, and an efficiency of 88% ( p <0.001). The ROC analysis revealed a SUV max 6.5 as an optimal cut off-level to distinguish between low- and high-grade carcinoma in 18 FEC PET/CT (AUC=0.7470; 95% CI 0.5919;0.9020;SD(AUC)=0.0791) with a sensitivity of 61% and a specificity of 92%; but the efficiency was only 70% and the NPV 50% ( p =0.01). Conclusion : 68 Ga-PSMA-11 PET/CT guided biopsy of the prostate increases diagnostic precision and is likely to help to reduce overtreatment of low-grade malignant disease as well as detect the foci of the highest Gleason pattern. Both methods ( 68 Ga-PSMA-11, 18 FEC) were suitable to detect primary prostate cancer, but the excellent image quality, the higher specificity and the good correlation of positive scans with GS are advantages of 68 Ga-PSMA-11.