This study [1] examined individual variations in serum testosterone levels every 6 months for 3 years in men receiving androgen-deprivation therapy (ADT). The authors also examined whether variations can be explained by variability between assays rather than a true difference, the rate of breakthrough above castrate levels, and whether these could be predicted.

The authors found breakthrough increases of >50 ng/dL at the first determination in 12.3%, 24.7% at the third, with subsequent results remaining stable. Breakthrough increases of 20–50 ng/dL were 27.4% at the first determination, 31.5% at the second, and then stable thereafter. A first determination of <20 ng/dL provided an 11.4% probability for future increases of >50 ng/dL, with a 5.7% probability if two consecutive determinations were <20 ng/dL and a null probability when three consecutive determinations were <20 ng/dL.

These data highlight the issue of failure to achieve castrate levels in patients receiving ADT, which is not an isolated occurrence [2]. The authors admit that one limitation of their study was that patients did not receive the same LHRH analogue. It would be interesting to examine which particular LHRH preparation those patients with non-castrate hormone levels were receiving, as previous reports of failure to achieve castration have concerned leuprolide acetate [3]. In fact, we observed apparent hormone-escaped disease in several men with metastatic prostate cancer receiving leuprolide acetate [4]. Further investigations in these patients revealed non-castrate levels of both testosterone and LH and no progression of the disease. Alteration to an alternative agent in these men resulted in a return to castrate hormone levels and subsequent reduction in PSA levels. Following previous reports [5], we suggested that the cause of these failures might be related to a granulomatous reaction to biodegradable microcapsules in the leuprolide preparation, leading to drug dispersion, non-castrate hormone levels and a rising serum PSA-level.

Further interesting work with the present data would be to examine whether castration failure translates into an increase in PSA levels or a worsening of symptoms, as occurred in our series [4]. I agree with the authors’ comment that serum testosterone should be determined at least three times during ADT, and would suggest it should also be measured at any time there is an apparent failure of treatment. It might be that with longer-term work on this subject we find that hormone levels should be checked routinely throughout the duration ADT to ensure its effectiveness.