Is the detection rate of 18F-choline PET/CT influenced by androgen-deprivation therapy?

To evaluate if the detection rate (DR) of (18)F-choline (18F-CH) PET/CT is influenced by androgen-deprivation therapy (ADT) in patients with prostate cancer (PC) already treated with radical intent and presenting biochemical relapse.

We have retrospectively evaluated (18)F-CH PET/CT scans of 325 consecutive PC patients enrolled in the period November 2009 to December 2012 previously treated with radical intent and referred to our centre to perform (18)F-CH PET/CT for biochemical relapse. Two different groups of patients were evaluated. group A included the whole sample of 325 patients (mean age 70 years, range: 49-86) who presented trigger PSA between 0.1 and 80 ng/ml (mean 5.5 ng/ml), and group B included 187 patients (mean age 70 years, range 49-86) with medium-low levels of trigger PSA ranging between 0.5 and 5 ng/ml (mean PSA 2.1 ng/ml); group B was chosen in order to obtain a more homogeneous group of patients in terms of PSA values also excluding both very low and very high PSA levels avoiding the "a priori" higher probability of negative or positive PET scan, respectively. At the time of examination, 139 patients from group A and 72 patients from group B were under ADT: these patients were considered to be hormone-resistant PC patients because from their oncologic history (>18 months) an increase of PSA levels emerged despite the ongoing ADT. The relationship between (18)F-CH PET/CT findings and possible clinical predictors was investigated using both univariate and multivariate binary logistic regression analyses, including trigger PSA and ADT.

Considering the whole population, overall DR of (18)F-CH PET was 58.2 % (189/325 patients). In the whole sample of patients (group A), both at the univariate and multivariate logistic regression analysis, trigger PSA and ADT were significantly correlated with the DR of (18)F-CH PET (p < 0.05). Moreover, the DR in patients under ADT (mean PSA 7.8 ng/ml) was higher than in patients not under ADT (mean PSA 3.9 ng/ml), (DR was 70.5 % and 48.9 %, respectively; p < 0.001), therefore, demonstrating the existence of a significant correlation between the DR of (18)F-CH PET and ADT. In group B patients only trigger PSA resulted a reliable predictor of the (18)F-CH positivity, since ADT was not correlated to the DR of (18)F-CH PET (p = 0.061). Also in group B the DR of (18)F-CH PET in patients under ADT was higher than in patients not under ADT (65.3 % and 51.3 %, respectively) but the difference was not significant without a statistically significant correlation in the Mann Whitney test (p = 0.456) therefore, suggesting the lack of correlation between DR (18)F-CH PET/CT and ADT.

Similarly to previous published studies, in our series the overall DR of (18)F-CH PET/CT was 58 % and was significantly correlated to trigger PSA. The most important finding of the present study is that ADT does not negatively influence DR of (18)F-CH PET/CT in PC patients with biochemical relapse; therefore, it can be suggested that it is not necessary to withdraw ADT before performing (18)F-CH PET/CT.