Bone alkaline phosphatase serum level predicts the response to antiandrogen withdrawal

To analyze if the biochemical response to antiandrogen withdrawal correlates with total and percent free serum prostate specific antigen (PSA) and bone alkaline phosphatase (BAP).

A group of 46 patients in whom maximal androgen blockade (MAB) failed was included in this study. Flutamide was used in 32 patients and bicalutamide in 14. Total and free PSA and BAP were determined in serum the same day in which antiandrogen was withdrawn. Thereafter, serum PSA was determined every 15 days. A biochemical response was established when a decrease in serum PSA greater than 50% was observed. The duration of biochemical response was considered until the first of two consecutive PSA serum elevations.

The rate of biochemical responses was 23.9%. The mean duration of responses was 5.2 months. The mean serum PSA in patients who responded to the antiandrogen withdrawal was 86.4 ng/ml, while it was 98.6 ng/ml in those who did not respond, P > 0.05. The mean of percent free PSA was 13.9 and 17.7%, respectively, P > 0.05. However, the mean BAP in responder patients was significantly lower, 18.9 ng/ml versus 100.6 ng/ml, P < 0.03. The rate of responses to flutamide withdrawal was 25% and to bicalutamide withdrawal 21.4%, P > 0.05. Other factors as age, Gleason score, initial clinical stage and the period of antiandrogen exposure were also analyzed. However, none of them had statistical significance. Serum BAP was the only predictor of response in the logistic regression analysis and a cut-off point of 50 ng/ml provided a relative risk of 1.46 (95% CI 1.17-1.83).

Lower serum levels of BAP seems to correlate with a better response to antiandrogen withdrawal. Moreover, a level of BAP higher than 50 ng/ml would predict strongly the absence of response.