TGF-β Conveys Undesirable Side Effects of Androgen Depletion

Androgens are anabolic steroids with pleotropic actions. Testosterone and its metabolite dihydrotestosterone bind to the androgen receptor (AR) to transcriptionally activate target genes in the prostate gland. The pioneering work of Dr Charles Huggins in the 1940s defined the therapeutic benefit of androgen deprivation therapy (ADT) via surgical or hormonal castration in the treatment of androgen-dependent metastatic prostate cancer, which had a transformative impact in the management of patients with hormone-sensitive advanced or recurrent disease (1). Today, ADT remains the mainstay of treatment and is deployed for men with metastatic and high-risk localized prostate cancer and is frequently utilized with chemotherapy (2). Virtually all patients, however, eventually progress to therapeutic resistance with biochemical and radiographic progression despite castrate levels of serum testosterone. The continued clinical progression to castration-resistant prostate cancer (CRPC) is driven by sustained AR signaling, under seemingly “starved” androgen conditions (2). Furthermore, this androgen starvation (via ADT) is associated with significant side effects that include erectile dysfunction and severely compromised metabolic function. The latter is classified as sarcopenia, which leads to loss of muscle mass and function, and bone mass resulting in obese frailty, a clinical disability syndrome (3, 4). Approximately 400 000 men in the United States receive ADT each year, with more than 50% of these men age 75 or older, who are already experiencing some degree of age-related sarcopenia before ADT. Thus, a population of more than 200 000 men treated annually for advanced prostate cancer with ADT suffer from this remarkable side effect (4). Clearly, a thorough understanding of the mechanisms involved in sarcopenia induction post-ADT is required to better control this undesirable side effect in prostate cancer patients.

The study published in the current issue of Endocrinology by Pan et al (5) used aged mice and successfully developed a mouse-model of ADT-induce sarcopenia, recapitulating the clinical castration-induced metabolic syndrome and associated features found in prostate cancer patients after ADT treatment; a loss of lean muscle and a loss of mass of individual muscles, coordinated with loss of function. The investigators found that the onset of sarcopenia occurred in the older cohort of mice at approximately 6 weeks after castration-induced ADT and is mediated by multiple members of the TGF-β superfamily, including myostatin and activins. TGF-β is a multifunctional cytokine belonging to a protein superfamily, members of which elicit diverse cellular responses in cell type- and age-dependent context during normal development and tumorigenesis (6). The article defines a central role for the activins, as TGF-β primary candidate ligands that dictate the effects of ADT on muscle in aged mice.

An actionable cross talk between the androgen axis and TGF-β-Smad signaling has previously been functionally linked to significant consequences on cellular response of target cells. Diverse cell types including carcinoma-associated stromal cells, endothelial cells, lymphocytes, and cancer epithelial cells comprise the dynamic prostate microenvironment, under the control by TGF-β to promote tumor growth and progression (7). Androgens regulate epithelial cell proliferation and apoptosis in the prostate via interactions with the TGF-β signaling, an effect that is biphasic in nature (7). Paradoxically, although targeting AR signaling by first and second generation antiandrogens, Casodex and enzalutamide, results in prostate cancer growth suppression, this approach also promotes prostate cancer cell invasion by engaging TGF-β family members. TGF-β family members regulate angiogenesis, invasion and migration, immune surveillance, and epithelial-mesenchymal-transition (EMT) via navigating an array of interactions between epithelial cells and myofibroblasts/carcinoma-associated stromal cells maintaining a reactive microenvironment. The TGF-β Smad intracellular effectors regulate formation of fibrotic collagenous micronodules and Smad3 phosphorylation hinders recovery of skeletal muscle fiber in models of skeletal atrophy. The defining balance of activities by TGF-β superfamily members in terms of their functional consequences across a spectrum of target cells must be exploited in the context of the microenvironment, that being the prostate gland, bone, muscle, or adipose tissue. In that regard, the study by Pan et al (5) provides only an initial insight into the mechanistic platform where each specific TGF-β family member engages distinct pathway/partners to take executive action against muscle and bone integrity and obesity under regulators transcriptionally suppressed by ADT.

There is evidence linking a subgroup of TGF-β superfamily members, including myostatin and activins, to the negative regulation of skeletal muscle mass and strength in normal, aging, and cachectic mouse models (8). Targeting myostatin with neutralizing antibodies has been previously developed to therapeutically manage muscle loss; however, the approach has not been proven functionally effective as some of these have restored muscle volume in human clinical trials but without affecting/restoring muscle strength. Activins (ligand) bind to a heterodimeric activin receptor complex Activin Receptor Complex II (ActRII) that binds and phosphorylates a type I receptor. Remarkably, Pan et al (5) made the important observation that the sarcopenic phenotype was fully reversed by treatment with an antibody targeting ActRII in older mice. Clearly a more specific blocking regimen, with less severe and fewer comorbidities is needed to preserve both muscle volume and strength for sarcopenic patients after ADT. Although targeting TGF-β signaling in prostate cancer progression to metastasis from the angle of the bifunctional role of this cytokine during tumor progression (from initiation, to early and late stage disease) is of value, one must consider the significance of the potential redundant roles of TGF-β superfamily members in the metabolic syndrome during emergence of CRPC in patients under ADT. The article by Pan et al (5) delivers an intimate profile of TGF-β ligands framed by the response to ADT, consequently providing a “golden fleece” to block 2 or more of these induced ligands towards maximizing therapeutic benefit and minimizing comorbidities. But the functional integration of the androgen axis/AR under ADT in dictating the actions of TGF-β family members in bone muscle loss and acquired obesity was not interrogated in the study and will be required in future research to encompass the entirety of molecular changes that occur during prostate cancer progression.

There are some limitations in using the ageing mouse model centered on the various TGF-β family members performing a distinct role within the prostate tumor microenvironment, which might impact the translation of the findings in human conditions. The authors used the temporally defined sarcopenic muscle samples to examine the expression of plausible candidate drivers of muscle loss in extracts of the muscle tissue (5). Biological interrogation of the presented TGF-β networks must be executed in additional preclinical models, in order to assess the potential of cotargeting or combination therapies, linking stratification of TGF-β family members and their intracellular signature effectors after ADT, not only to therapeutic response but also to obese frailty in patients with advanced disease. Attractive options include transgenic mouse models of aberrantly active TGF-β, promoting EMT to mesenchymal-epithelial-transition interconversions, inflammation, and progression to bone metastasis in an age-dependent manner (8). Also, mechanisms controlling temporal TGF-β programming EMT to mesenchymal-epithelial-transition while processing contributions by androgen signaling are being actively explored. TGF-β may exert its tumor suppression function through an EMT-mediated disruption of a lineage-specific network, enforcing a differentiation-survival state before apoptosis induction of target epithelial cells as recently shown in pancreatic cancer (9).

The timing of directed therapies is of paramount significance in effectively impairing the action of individual TGF-β ligands and their signaling pathways in respective target cells. A potential combination of ADT with inhibitors of TGF-β signaling may empower the patients to fight cancer and overcome resistance while taking control of their skeletal muscle function and rejecting obesity. However, a new dimension emerges from recent evidence suggesting that patients with CRPC upon receiving ADT after androgen administration exhibit a sensitivity to subsequent ADT with a therapeutic benefit (10). Controversy withheld, such a bipolar strategy will empower monitoring the dual action of androgens against an intact TGF-β superfamily. A more rational therapy needs to be based on the identification of expression profiles and kinetics of muscle modulating ligands (including specific loss of genes encoding the cytokines) in prostate cancer patients undergoing ADT. Further exploration of the functional exchange between distinct TGF-β family proteins and the androgen axis in individual patients after ADT will potentially lead to new platforms for therapy of advanced prostate cancer and confer a benefit of simultaneous or sequential targeting of the drivers of sarcopenia and muscle loss.