RESUMO
OBJECTIVES: To compare the ability of 68Ga -PSMA PET/CT (PSMA PET/CT) and multiparametric MRI (mpMRI) to exclude lymph node invasion (LNI) in patients who undergo radical prostatectomy (RP). MATERIALS AND METHODS: A multicenter cohort of patients who underwent PSMA PET/CT and pelvic mpMRI prior to RP with pelvic lymph node dissection (PLND) was analyzed. Increased Ga68-PSMA uptake on PET/CT and enlarged (> 10 mm) or abnormal lymph nodes on mpMRI were considered positive findings. The final surgical pathology served as the standard of reference. The negative predictive value (NPV) was calculated for each modality separately, as well as the combined value. RESULTS: Included were 89 patients with D'Amico intermediate (45%) or high-risk (55%) prostate cancer. The median number of extracted LN was 9 (IQR 6-14). LNI was found in 12 (13.5%) patients. The NPV of mpMRI, PSMA PET/CT, and the two tests combined were 87%, 89%, and 90%, in the entire cohort, 95%, 97%, and 97% in patients with intermediate-risk disease, and 80%, 82%, and 83% in patients with high-risk disease, respectively. The median diameter of LN missed by both imaging and the median intranodal tumor diameter was 5.5 (IQR 3-10) mm and 1 (IQR 1-3) mm, respectively. CONCLUSIONS: PSMA PET/CT and mpMRI demonstrated similar performance in excluding pelvic LNI with NPV of approximately 90%. The combination of both tests does not improve NPV significantly. Therefore, even in the era of advanced imaging, PLND is still recommended for accurate staging, especially in the high-risk population.