Head-to-Head Comparison of (68)Ga-Prostate-Specific Membrane Antigen PET/CT and Ferumoxtran-10-Enhanced MRI for the Diagnosis of Lymph Node Metastases in Prostate Cancer Patients

Item does not contain fulltext
Accurate assessment of lymph node (LN) metastases in prostate cancer (PCa) patients is critical for prognosis and patient management. Both prostate-specific membrane antigen (PSMA) PET/CT and ferumoxtran-10 nanoparticle-enhanced MRI (nano-MRI) are imaging modalities with high potential to identify LN metastases in PCa patients. The aim of this study was to compare the results of these imaging technologies in terms of characteristics and anatomic localization of suspicious LNs in order to assess the feasibility of their complementary use for imaging in PCa patients. Methods: In total, 45 patients with either primary PCa (n = 8) or recurrence (n = 36) were included in this retrospective study. All patients underwent both (68)Ga-PSMA PET/CT and nano-MRI between October 2015 and July 2017 within 3 wk. Both scans were performed at the same institution according to local clinical protocols. All scans were analyzed independently by experienced nuclear medicine physicians and radiologists. The size, anatomic location, and level of suspicion were determined for all visible LNs. Subsequently, the findings from (68)Ga-PSMA PET/CT and nano-MRI were compared without respect to a reference standard. Results: In total, 179 suspicious LNs were identified. Significantly more suspicious LNs per patient were detected by nano-MRI (P < 0.001): 160 were identified in 33 patients by nano-MRI, versus 71 in 25 patients by (68)Ga-PSMA PET/CT. Of all suspicious LNs, 108 were identified only by nano-MRI (60%), 19 (11%) only by (68)Ga-PSMA PET/CT, and 52 (29%) by both methods. The mean size of the suspicious LNs as identified by nano-MRI was significantly smaller (5.3 mm) than that by (68)Ga-PSMA PET/CT (6.0 mm; P = 0.006). The median level of suspicion did not differ significantly. Both modalities identified suspicious LNs in all anatomic regions of the pelvis. Conclusion: Both modalities identified suspicious LNs that were missed by the other. Both modalities identified suspicious LNs in all