A Multiscale, Mechanism-Driven, Dynamic Model for the Effects of 5α-Reductase Inhibition on Prostate Maintenance

Michael G Zager,Hugh A Barton,Franky L Chan

doi:10.1371/journal.pone.0044359

Abstract

A systems-level mathematical model is presented that describes the effects of inhibiting the enzyme 5α-reductase (5aR) on the ventral prostate of the adult male rat under chronic administration of the 5aR inhibitor, finasteride. 5aR is essential for androgen regulation in males, both in normal conditions and disease states. The hormone kinetics and downstream effects on reproductive organs associated with perturbing androgen regulation are complex and not necessarily intuitive. Inhibition of 5aR decreases the metabolism of testosterone (T) to the potent androgen 5α-dihydrotestosterone (DHT). This results in decreased cell proliferation, fluid production and 5aR expression as well as increased apoptosis in the ventral prostate. These regulatory changes collectively result in decreased prostate size and function, which can be beneficial to men suffering from benign prostatic hyperplasia (BPH) and could play a role in prostate cancer. There are two distinct isoforms of 5aR in male humans and rats, and thus developing a 5aR inhibitor is a challenging pursuit. Several inhibitors are on the market for treatment of BPH, including finasteride and dutasteride. In this effort, comparisons of simulated vs. experimental T and DHT levels and prostate size are depicted, demonstrating the model accurately described an approximate 77% decrease in prostate size and nearly complete depletion of prostatic DHT following 21 days of daily finasteride dosing in rats. This implies T alone is not capable of maintaining a normal prostate size. Further model analysis suggests the possibility of alternative dosing strategies resulting in similar or greater effects on prostate size, due to complex kinetics between T, DHT and gene occupancy. With appropriate scaling and parameterization for humans, this model provides a multiscale modeling platform for drug discovery teams to test and generate hypotheses about drugging strategies for indications like BPH and prostate cancer, such as compound binding properties, dosing regimens, and target validation.

Highlights

With the availability of information describing many individual components of biological systems, there is increasing focus on developing multiscale computational models that capture the overall systems behavior [1]
Finasteride exposure estimates following oral dosing were needed. These changes in prostate kinetics and dynamics occur in the broader context of the whole body, including pharmacokinetics of endogenous T and DHT described with physiologically based pharmacokinetic (PBPK) models and androgen-luteinizing hormone (LH) signaling between the brain and testes forming the testicular-pituitary axis (Figure 2)
Changes in prostate size are of concern from both a clinical perspective and from the perspective of toxicological studies, which are typically conducted using rats

Summary

Introduction

With the availability of information describing many individual components of biological systems, there is increasing focus on developing multiscale computational models that capture the overall systems behavior [1]. Testosterone (T) is metabolized by 5aR into the more potent 5a-dihydrotestosterone (DHT) [6,7], the driving force of prostate gene-regulation [8,9,10,11,12,13,14]. One isoform (labeled 5aR1 in this article) is widely distributed in the body and is expressed abundantly in the liver of rats, a major tissue for T metabolism. The other isoform (labeled 5aR2 in this article) has been reported to be expressed mainly in androgen-dependent tissues and is abundant in the prostate of rats, the other major site for T metabolism [6]. In human cancerous prostate tissue, expression levels for each isoform have been shown to increase [20,21]

Methods

Results

Conclusion