MP80-05\u2003PSA LEVEL PREDICTS POSITIVITY OF 18F-FLUCICLOVINE PET/CT IN PATIENTS WITH BIOCHEMICALLY RECURRENT PROSTATE CANCER

Abstract

INTRODUCTION AND OBJECTIVES: Conventional imaging has a limited yield for the detection of recurrent disease at low PSA levels. The use of PET/CT with a range of radiotracers has shown early promise in improving cancer detection in this setting. Therefore, we assess the relationship between PSA levels and 18F-Fluciclovine PET/CT positivity in patients with biochemical recurrence. METHODS: A consecutive sample of men who underwent 18F-Fluciclovine PET/CT for biochemical recurrence at a single institution were included in this retrospective study. Patients were stratified into groups according to their serum PSA level (A: <0.5 ng/mL, B: 0.5-1.0 ng/mL, C: 1-1.5 ng/mL, D: 1.5-2.0 ng/mL, E: >2.0ng/mL) and the rate of scan positivity were compared between groups. Multivariable logistic regression analysis adjusting for patient, tumor and treatment charcteristics was performed to characterise the affect of PSA on yield. We also compared the SUVmax, SUVtotal and MTV between the groups. RESULTS: A total of 76 patients with complete PSA data were included with 63.2% (n=48) having a positive 18F-Fluciclovine PET/CT. The positivity rate for each of the groups were A: 0%, B: 66.7%, C: 70%, D: 72.7%, E: 78.4% (p<0.01). PSA acted as a significant predictor of positivity (OR 2.04 per unit increase, 95%CI 1.23-4.79). There was also a significant difference in SUVmax between the groups which increased with PSA levels (p=0.009). There was no difference in SUVtotal or MTV between the groups. CONCLUSIONS: The yield of 18F-Fluciclovine PET/CT in patients with a PSA <0.5 ng/mL is low and therefore this scan should only be performed in patients with a PSA above this level. Figure. No caption available. Source of Funding: None