The goal of the article is to determine whether the dog is an appropriate model of human prostate cancer. Dogs are the only species other than man that spontaneously develop prostate cancer. In humans, prostate cancer is initially regulated by the steroid hormone receptor - the androgen receptor, which mediates the effects of testosterone and dihydrotestosterone. Hence, human patients with disseminated or advanced prostate cancer are treated with androgen deprivation therapy alone or together with treatment that targets the androgen receptor, which is initially effective, but patients eventually become resistant to these treatments. Unlike humans, dogs are castrated at a young age, and hence may be good models of advanced prostate cancer in patients who are resistant to initially androgen deprivation. The is article discusses various current therapies and compares and contrasts their benefits are discussed.
1. Introduction
The initiation and progression of prostate cancer (PCa) in humans is initially reliant on androgen receptor (AR) signaling
[1][2][3]. Directly targeting androgen ligands with androgen-inhibiting drugs (e.g., chemical castration) or decreasing their production via surgical castration has been utilized over the last 80 years in an attempt to suppress AR signaling and PCa-tumor progression
[4][5][6]. Unfortunately, androgen-deprivation therapies (ADT) eventually fail for a subset of patients, and despite the presence of castrate levels of androgens, PCa progresses to an incurable form, termed castration-resistant prostate cancer (CRPC)
[7][8][9]. This form of prostate cancer often continues AR signaling that is not reliant on androgen ligands (i.e., androgen independent) by way of multiple mechanisms, including mutations in the receptor or copy-number variations, and is typically treated with androgen-receptor inhibitors (ARIs)
[10][11][12][13]. Eventually, new driver mutations develop in various genes in CRPC tumors, which leads to the abandonment of AR signaling altogether. Once PCa progresses, irrespective of AR signaling, it is termed androgen-receptor-indifferent PCa, a highly aggressive lethal form of the disease with poor outcomes
[14][15]. Recently, the molecular characterization of CPRC and androgen-indifferent prostate cancer (AIPC) have improved our understanding of the drivers of these variants, which is critical for the identification of novel therapeutics. However, animal models for advanced forms of PCa are lacking and are wrought with limitations
[16][17], which may make the approval of novel therapeutics challenging and the translation of results between species inconsistent.
The prostate of the dog, unlike that of rodents, is morphologically, histologically, and physiologically similar to that of humans and is under the control of androgens via androgen-receptor signaling
[18][19][20]. Dogs also naturally develop other pathologic prostatic conditions with age, such as benign prostatic hyperplasia (BPH), in contrast to rodents, which typically develop prostatic atrophy with age
[21][22]. More importantly, dogs are the among the only animals that spontaneously develop PCa, and often present with highly aggressive metastatic disease. The neutering or castration of male dogs is common in developed countries, with the majority of the male dogs in the United States undergoing this procedure
[23][24]. It has been shown that castration influences PCa progression as castrated dogs develop PCa at higher rates and experience more metastases than intact dogs
[25][26][27][28][29][30].
Initially, PCa is reliant on androgens and AR signaling for tumor growth and progression and is characterized by a rising prostate-specific antigen (PSA) level in humans, a marker not typically expressed in dog PCa
[31][32]. This requirement for androgen-dependent growth is clinically exploited to combat advanced or recurrent PCa (following initial treatment with surgery and radiation) with the use of androgen-deprivation therapies (ADT) (i.e., androgen-ablation therapies, chemical castration) to inhibit the production or actions of androgens
[33][34]. Invariably, PCa becomes resistant to ADT, and is then termed CRPC, where AR continues to signal irrespective of the presence of androgens
[35][36]. Because of this continued signaling, AR targeting remains a valid treatment strategy using androgen-signaling inhibitors (ASIs) (e.g., androgen-synthesis inhibitors, androgen-receptor inhibitors). Dog PCa is most often low or null for AR expression, as well as for androgen-regulated proteins, such as NKX3.1
[37][38] (
Table 1). However, opposing reports show some AR expression in intact dogs with PCa and cytoplasmic sequestration or the loss of AR in castrated dogs with PCa
[39][40], making intact dogs with PCa potential models for androgen-dependent disease. As noted over 50 years ago in landmark studies performed by Huggins et al., dogs have similar pathophysiologies with respect to the androgen-dependent growth of the prostate and may still be of value to researchers
[6][41].
Table 1. Summary of molecular characteristics of dog and human PCa.
2. Androgen-Receptor Structure
The AR in humans is a protein of 919 amino acids consisting of several functional domains, including an N-terminal domain (NTD), DNA binding domain (DBD), and a ligand binding domain (LBD) at the C-terminus
[42][43]. In dogs, the AR is approximately 907 amino acids, and has homologous DBD and LBD, which are highly conserved across evolution in various species and are activated upon binding to androgen ligands
[43][44].
The NTD of the AR is essential for function and the least evolutionarily conserved region of AR; however, there are still similarities between humans and dogs. The polyglutamine (i.e., polyQ, CAG) repeat region of the NTD has an average of 21–23 Qs in humans. Longer polyQ repeats are related to decreased AR transcriptional activity, while shorter polyQ repeats are related to increased AR transcriptional activity and are often associated with increased PCa risk
[45][46][47]. This finding has also been recapitulated in dogs in vitro, where the introduction of AR with fewer polyQ repeats resulted in higher AR activity
[48]. However, the association between shorter polyQ repeats regions and PCa is unclear in dogs: while some studies reveal that a shorter polyQ length does not predispose dogs to PCa, others show that certain breeds with shorter polyQ lengths are predisposed to developing PCa
[25][49][50][51].
3. Androgen-Receptor Co-Chaperones
In the absence of the agonist ligand, the AR is bound to heat-shock proteins (HSP40, HSP70 and HSP90) and other co-chaperone proteins in a complex known as the foldosome
[52][53]. Many small co-chaperone proteins with tetratricopeptide repeats (TPR), such as CYP60, PP5, FKBP51, FKB52, PP5, CHIP, and SGTA, have been shown to interact with the AR-foldosome complex
[54]. The small glutamine-rich tetratricopeptide repeat-containing protein α (SGTA), is a co-chaperone of interest in PCa, and is known to stabilize the apo-AR structure in the cytoplasm prior to ligand binding. In human PCa, the SGTA, a steroid-receptor molecular co-chaperone that influences hormone action, is known to regulate AR function. The AR:SGTA ratio is increased compared to patient-matched BPH, and it is also increased when metastatic PCa tumors are compared with their primary tumors
[55][56]. It is hypothesized that AR thereby overwhelms the capacity for SGTA to limit AR response to ligands and ensure the appropriate cellular localization of AR in vivo. In addition, in vitro work from this
res
earchtudy showed that SGTA overexpression blunted the AR’s response to androgen ligands
[56].
This concept has also been explored in multiple studies in dogs, which have shown that the overexpression of SGTA in vitro abates AR signaling
[57][58]. Therefore, androgen-independent disease was hypothesized to be attributed to SGTA overexpression in dog-PCa-patient tumor samples by some researchers, who also subsequently showed that interference with SGTA dimerization in vitro rescues AR signaling
[59].