Population pharmacokinetic–pharmacodynamic modelling of the relationship between testosterone and prostate specific antigen in patients with prostate cancer during treatment with leuprorelin

Aims This investigation aimed to quantitatively characterize the relationship between the gonadotropin‐releasing hormone agonist leuprorelin, testosterone (T) and prostate specific antigen (PSA) concentrations over time, to aid identification of a target T concentration that optimises the balance of the benefits of T suppression whilst reducing the risk of side effects related to futile over‐suppression. Methods Data from a single dose study to investigate the effect of leuprorelin in a 6‐month depot formulation on T and PSA in prostate cancer patients were analysed using a population pharmacokinetic–pharmacodynamic modelling approach. The developed model was qualified using external data from 3 studies, in which the effect of different formulations of leuprorelin on T and PSA was evaluated in prostate cancer patients. Results The effect of leuprorelin on the relationship between T and PSA was adequately characterized by the Romero model with minor modifications, combined with a turnover model to describe the delay in response between T and PSA. The data were significantly better described when assuming a minimum PSA level that is independent on the treatment‐related reduction in T, as compared to a model with a proportional reduction in PSA and T. Conclusions The model‐based analysis suggests that on a population level, reducing T concentrations below 35 ng/dL does not result in a further decrease in PSA levels (>95% of the minimal PSA level is reached). More data are required to support this relationship in the lower T and PSA range.