Introduction

To evaluate the diagnostic performance of 18F-fluciclovine (FACBC) - positron emission tomography/computer tomography (PET/CT) in nodal staging prior to pelvic lymph node dissection (PLND) during radical prostatectomy (RP) in primary prostate cancer (PCa) patients.

Methods

Overall, 60 patients were consecutively and prospectively enrolled and underwent 18F-FACBC-PET/CT. Inclusion criteria were: intermediate and high-risk biopsy-proven PCa according to D’Amico risk group classification; standard staging workup (including 11C-Choline-PET/CT); absence of previous or ongoing androgen deprivation therapy; eligible for PLND in course of RP. Any nodal uptake greater than surrounding background was interpreted with a visual 5-point-scale (confidence of disease): 1-2 probably negative; 4-5 probably positive; 3 equivocal. Lesions’ number, site and size, SUV max and target-to-background-ratio were also recorded. PET/CT results using both tracers were compared, on a per-patient basis analysis, with histopathology from extended PLND (60 patients).

Results

Main characteristics of the population are shown in Table 1. Overall, 1342 lymph nodes were retrieved and 32 (2%) LNMs were counted. At final pathologic results, we found 9 and 15 false positive (FP) results with FACBC and Choline, respectively; while 5 false negative (FN) results were found with both tracers. The diagnostic performance of both tracers in detecting LNMs are summarized in Table 2: sensitivity, specificity and accuracy were 58%, 69% and 67% vs. 58%, 81% and 77% for 11C-Choline PET/CT vs. 18F-fluciclovine PET/CT, respectively._x000D_

Conclusions

Conclusion: 18F-fluciclovine demonstrated suboptimal sensitivity, equal to Choline, suggesting that neither tracer is ideal for PCa nodal staging. However, our preliminary data show encouraging slightly better overall diagnostic performance and practical/technical advantages in favor of 18F-fluciclovine as compared with the routine tracer. Enrolment and further analyses (region-based/semi-quantitative/uni-multivariate-logistic-regression to evaluate predictive factors) are ongoing.

Funding

none