The Multifaceted Role of Androgen Receptor Signaling in Immunity: Implications for Oncology

Abstract

While the androgen receptor (AR) is canonically known for its role in the prostate and testis, AR signaling exerts broad immunomodulatory effects through direct and indirect signaling in multiple immune cell compartments and contributes significantly to sex differences in autoimmunity, infection, and cancer. Mouse model perturbations of androgen signaling through castration, testicular feminization, and cell type-specific AR knockout have provided important insights into cell-intrinsic and -extrinsic mechanisms by which AR signaling affects innate and adaptive immunity. However, the precise molecular underpinnings of these effects remain largely unknown. Moreover, despite convincing epidemiological and correlative observations that highlight the importance of AR signaling in human immune function, it remains unclear how reliably findings in mice will translate to humans. A better understanding of how to augment immune function through androgen signaling modulation could have significant clinical relevance for the treatment of cancer, as well as other disease states involving immune dysregulation. In this review, we discuss the current evidence for the functional effects of AR signaling within the major immune cell compartments of the innate and adaptive immune systems. We also review ongoing clinical efforts that modify AR signaling for the purpose of enhancing antitumor immunity.

Introduction

The androgen receptor (AR) is a member of the nuclear hormone receptor family that is activated by binding to testosterone and dihydrotestosterone (DHT) ligands and plays a crucial role in male sexual differentiation. While AR signaling has been most extensively studied for its physiologic role in the prostate and testis and its pathologic role in driving and sustaining prostate cancer, its contribution toward regulating the function and sex-specific differences across many different tissues and cell types is being increasingly recognized (1). The connection between AR signaling and the immune system is of particular clinical relevance. Androgens are considered generally immunosuppressive, as females are more predisposed to autoimmune diseases while males are more susceptible to cancer (2). Multiple cell types within both the innate and adaptive arms of the immune system express the AR including neutrophils, macrophages, B lymphocytes, T lymphocytes, and dendritic cells (3–5), and androgen signaling exerts at least indirect effects on multiple other immune compartments. However, much remains to be learned about the precise mechanisms by which AR signaling modulates immune cell function and how the AR-regulated transcriptional programs in immune cells differ from the canonical AR cistrome commonly studied in the context of prostate cancer. Deeper insights into these questions could pave the way for designing AR-targeted immune modulating therapies, for example to boost antitumor immunity in the treatment of cancer. With the rapid application and development of novel immunotherapeutic agents across multiple tumor types, the potential to augment immunotherapy efficacy through perturbation of AR signaling represents an appealing and relatively unexplored opportunity. In this review, we discuss the currently known mechanisms and consequences of AR signaling within different immune cell compartments, summarize clinical efforts to modulate AR signaling to enhance antitumor immunity, and outline directions for future avenues of investigation.

Effects of androgens and AR signaling on components of the innate immune system

The immune system is broadly divided into the innate, or general, non-specific immune system and the adaptive, or specialized immune system. The innate immune system comprises cells organized to provide barrier functions, others that produce enzymes and other factors that inhibit pathogens, as well as specialized cell types that carry out diverse functions spanning phagocytosis to regulating the adaptive immune system. A range of experiments have demonstrated a role for AR signaling in modulating innate immune system components (Figure 1). While the studies are compelling, the vast majority of work has involved experiments in mouse models – and the validation of key findings in humans remains an important area of investigation.

Figure 1. Effects of androgen receptor (AR) signaling on the immune system.

AR signaling exerts diverse effects on the proliferation, differentiation, and function of different immune cell compartments. In macrophages, AR suppresses inflammatory response to infection, impairs wound healing, and modulates M1 vs M2 polarization, although the latter effect is variable across different tissues. In NK cells, androgen signaling suppresses proliferation and decreases activity. Within the dendritic cell compartment, AR signaling in particular affects plasmacytoid DC’s by suppressing interferon and inflammatory cytokine production. AR effects on MDSCs are less well characterized, although one study showed that AR abrogation enhanced their immunosuppressive activity. Within neutrophils, AR signaling actually promotes granulopoiesis and neutrophil homing, however once neutrophils reach their tissue destination androgen signaling dampens bactericidal and phagocytic activity. AR signaling indirectly decreases eosinophil infiltration, although AR expression in eosinophils has not been confirmed. In mast cells, the evidence is not convincing that AR signaling enhances mast cell degranulation, although perinatal androgen exposure may reduce severity of mast-cell mediated anaphylaxis. In B lymphocytes, AR signaling impairs B lymphopoiesis, disrupts follicle center organization, and reduces antibody responses to vaccines. In T lymphocytes, AR signaling inhibits the function and differentiation of Th1 and Th2 cells, while enhancing Treg function. Androgen inhibition reverses CD8 T cell exhaustion and may improve intratumoral T cell infiltration. Created in BioRender. Lee, P. (2025) https://BioRender.com/7ve97k3

Neutrophils.

Neutrophils are the most abundant immune cell type in humans and play a key role in the innate immune system. Upon activation, neutrophils home to sites of tissue infection or injury and enact antimicrobial activity through phagocytosis, degranulation, or deployment of neutrophil extracellular traps. Androgens appear to enhance neutrophil production and localization while impairing cytotoxic function. Genetically engineered mouse models (GEMMs) with germline/systemic AR knockout (KO) exhibit neutropenia with impaired neutrophil maturation despite normal numbers of myeloid progenitor cells (6). AR overexpression in fluorescence activated cell sorter (FACS)-isolated AR KO granulocyte and macrophage progenitor cells restored neutrophil maturation, suggesting these effects are mediated through neutrophil-specific AR signaling. Mechanistically, AR KO mice have decreased granulopoietic responses to granulocyte colony-stimulating factor (G-CSF). Conversely, administration of exogenous androgen stimulated granulopoiesis in castrated male mice (6). This phenomenon is clinically relevant, as neutropenia is an uncommon but increasingly recognized side effect of AR pathway inhibitors in prostate cancer patients (7).

Androgens also influence neutrophil recruitment and function. In a rat model of bacterial prostatitis, testosterone treatment enhanced neutrophil recruitment to the site of infection. However, testosterone impaired bactericidal function as evidenced by decreased myeloperoxidase activity and increased expression of the anti-inflammatory cytokines IL-10 and TGF-β (8,9). Similarly, in a mouse model of urinary tract infection, androgen exposure decreased phagocytic activity and degranulation of kidney neutrophils (9). Interestingly, androgen exposure also enriched for a population of aged but immature (CD49d+, CD101−) neutrophils from the kidneys, but not from bone marrow or spleen. Thus, despite a positive effect on granulopoiesis in the bone marrow, androgens may adversely affect neutrophil maturation after tissue homing. Evidence for androgen-mediated suppression of neutrophil function is more limited. In vitro testosterone treatment of neutrophils from healthy male donors decreased superoxide production and microbiocidal activity, although there were mixed effects of neutrophil function at varying concentrations of testosterone (10). Further work will be needed to clarify the functional consequences of androgen signaling in human neutrophils.

Neutrophils also play a significant role in both pro- and anti-tumor responses, and androgen signaling has been shown to modulate these interactions. Androgens and AR signaling were determined to promote liver metastasis in mouse models of melanoma and colorectal cancer through the hepatic accumulation of neutrophils via multiple mechanisms (11). Androgen treatment promoted neutrophil production in the bone marrow and increased the absolute and relative number of circulating neutrophils, consistent with prior reports of androgen stimulating granulopoiesis (6). Neutrophils from sham-castrated male mice also exhibited increased expression of genes related to neutrophil functioning in metastasis compared to females and castrated males, and livers from sham males expressed higher levels of chemokines important for neutrophil recruitment. Other studies have shown that androgen signaling enhances the anti-tumor function of neutrophils. For example, in a murine syngeneic melanoma model, castration increased tumor metastatic burden, which was associated with decreased intratumoral neutrophil infiltration compared to sham males (12). The proportion of isolated tumor-infiltrating neutrophils that produced reactive oxygen species after stimulation was lower in castrated compared to sham males as well. Additionally, in vitro co-culture systems of prostate cancer patient-derived neutrophils and prostate cancer cell lines determined that treatment with the AR signaling inhibitor (ARSI) enzalutamide suppressed neutrophil cytotoxicity (13). This study further identified Type 1 TGF-β Receptor (TβRI) as a possible mediator of AR inhibition-mediated cytotoxic dysfunction. Thus, the effects of androgen signaling on neutrophil anti-tumor function appear to be complex and likely context-dependent. For example, both the liver metastasis and melanoma models demonstrated that androgens promoted tissue accumulation of neutrophils. However, in the liver these neutrophils facilitated metastasis while in the melanoma model they controlled tumor burden.

Macrophages.

Macrophages comprise a diverse subset of immune cells that are responsible for many aspects of tissue homeostasis including but not limited to orchestrating inflammatory responses to infection and cancer, wound healing, fibrosis, bone remodeling, metabolism, cardiovascular health, and brain development. Macrophages also contribute to the pathogenesis of several malignancies and have been shown to influence invasion, metastasis, and treatment resistance (14). Bone-marrow derived monocytes differentiate into macrophages upon establishment of tissue residence and adopt highly tissue-specific functions. While macrophage activation states encompass a diverse range of cellular programs that continue to be defined, they are generally divided into two major phenotypes: M1 macrophages, which produce proinflammatory cytokines, nitric oxide, and reactive oxygen species and are associated with antimicrobial immunity, and M2 macrophages, which tend to produce anti-inflammatory cytokines and are associated with tissue healing and production of extracellular matrix components (15). Prior evaluation of human bone marrow samples has confirmed AR protein expression in macrophages (3). Given the broad range of macrophage functions in tissue homeostasis, in this review we will focus on how androgens affect macrophage function in immunity specifically.

Androgens suppress the acute inflammatory response of macrophages to infection. In vitro treatment of murine macrophages with testosterone reduced the expression and ligand sensitivity of Toll-like receptor 4 (TLR4) – a key pattern recognition receptor of the innate immune response that binds lipopolysaccharide (16). The authors subsequently observed that castrated mice were more susceptible to endotoxic shock, and macrophages isolated from these mice also exhibited increased TLR4 expression. Another in vitro experiment determined that testosterone treatment also reduced macrophage production of nitric oxide and TNFα (17). In a mouse model of hemorrhagic shock followed by sepsis, treatment with the AR antagonist flutamide increased splenic macrophage production of the pro-inflammatory cytokines IL-1 and IL-6 and improved survival (18).

The AR also modulates macrophage polarization, although it appears to have opposing effects in different tissue contexts. In a mouse model of allergic lung inflammation, AR KO in alveolar macrophages impaired M2 polarization (19), whereas in a myocarditis mouse model, treatment with an AR degrader enhanced M2 polarization (20). However, in the latter case, the possibility of AR-mediated effects on other cell types which in turn secrete factors to promote M2 polarization were not excluded. For example, AR KO in renal tubules increased macrophage recruitment and promoted M2 polarization, which the authors posit could be mediated via modulation of macrophage colony-stimulating factor 1 (CSF-1) (21). Thus, both macrophage-intrinsic AR signaling and AR effects on surrounding tissue can each contribute to macrophage polarization, sometimes in opposing ways. In the context of prostate cancer, current evidence favors that AR signaling promotes a pro-tumorigenic, M2-like phenotype. AR signaling in the macrophage-like cell line THP-1 promotes prostate cancer cell invasiveness through increased TREM-1 signaling, and AR signaling in murine bone marrow-derived macrophages facilitates TREM-2-mediated M2-like macrophage polarization (22,23). Conversely, systemic androgen deprivation therapy (ADT) is associated with increased M1 macrophage intratumoral infiltration in patients with localized prostate cancer, although whether this is due to macrophage-intrinsic vs tumor-intrinsic AR signaling remains unknown (24).

Myeloid-derived suppressor cells.

Myeloid-derived suppressor cells (MDSCs) are pathologically activated myeloid cells which exhibit strong immunosuppressive activity, in particular on T lymphocyte function. They are categorized by their cell of origin: granulocytic/polymorphonuclear MDSCs (PMN-MDSCs), which are CD11b⁺CD14⁻CD15⁺/CD66b⁺ in humans, and monocytic MDSCs (M-MDSCs), which are CD14⁺CD15⁻HLA-DR^lo/− in humans (25). PMN-MDSCs employ reactive oxygen species, arginase 1, peroxynitrite, and prostaglandin E2 to exert their immunosuppressive effects, while M-MDSCs rely on nitric oxide, immunosuppressive cytokines, and immune checkpoints (25). To date, the effects of AR signaling on MDSCs are incompletely characterized. One study determined that AR abrogation enhanced the immunosuppressive properties of MDSCs via suppression of mitochondrial respiration, leading to increased tumor growth (26). This was assessed both with exposure to the AR antagonist enzalutamide in MC-38 murine colon cancer cells with AR deletion, and in the TRAMP C2 prostate cancer model with a myeloid specific AR deletion, suggesting that the enhanced tumor growth is myeloid-mediated.

NK cells.

NK cells are innate lymphoid cells with cytotoxic capabilities that are able to detect virally infected or cancerous cells by recognition of multiple features such as loss of major histocompatibility complex (MHC) class I or upregulation of cellular stress ligands. While not extensively studied, androgens have been shown to affect NK cell proliferation and function. In a study of 12 healthy human males treated with androgen deprivation therapy, the percentage of circulating NK cells significantly increased, suggesting a role for androgens in suppressing NK cell proliferation (27). In vitro studies have revealed that AR inhibition with ARSIs such as enzalutamide or darolutamide appears to activate NK cells, with increased secretion of IFN-γ and Granzyme B, and increased NK cell-mediated killing when co-cultured with prostate cancer cell lines (bioRxiv. 2023:2023.11.15.567201). While the direct effects of AR inhibition on prostate cancer cells may contribute to this observation, increased cytotoxicity was observed regardless of the AR status of the cell line model. Assessments of circulating NK cells in metastatic prostate cancer patients starting ADT quantitated increased numbers of granzyme B⁺ and perforin⁺ NK cells with castration (bioRxiv. 2023:2023.11.15.567201).

Dendritic cells.

Dendritic cells (DCs) are sentinel cells located throughout the body that activate both the innate immune system in response to pathogen detection, and the adaptive immune system through T lymphocyte priming. They can be divided into conventional DCs (cDCs) and non-conventional DCs, which include monocyte-derived DCs (MoDCs) that respond and home to inflammatory stimuli, and plasmacytoid DCs (pDCs) which excel in type I interferon production in response to infection. Androgens can influence multiple DC subtypes. In pDCs, in vitro studies suggest that androgens suppress the production of type I interferons and proinflammatory cytokines (28). In addition, pDCs isolated from HIV1-infected women produce more interferon alpha when challenged with HIV-1-encoded TLR7 ligands compared to pDCs from men (29). Androgens may also impair the function of MoDCs, as in vitro treatment of MoDCs with the AR antagonist bicalutamide induced an immunogenic and mature phenotype with increased CD86 and HLA-DR and a decrease in anti-inflammatory cytokines IL-10 and TGF-β (30). While this evidence generally argues for a suppressive role of androgens in dendritic cell function, one study in hypogonadal men determined that testosterone treatment increased LAMP-2 expression, a marker of antigen-presenting cell activation, however the increase was more notable in monocytes as opposed to dendritic cells (31).

Eosinophils.

Eosinophils are granulocytes that play a prominent role in fighting parasitic infections and in mediating allergic responses through degranulation and release of eosinophil peroxidase, major basic protein, and ribonucleases and through regulation of cytokines and chemoattractants. While prior IHC characterization has not demonstrated AR expression in eosinophils (3), androgen signaling still has effects on eosinophil mobilization and recruitment. In a PCOS (polycystic ovarian syndrome)-like mouse model, DHT-treated female mice exhibited a large reduction in the eosinophil populations within uterine tissue, with decreases also seen in visceral adipose tissue and the peripheral blood (32). These effects could be mitigated by the AR antagonist flutamide. Androgen signaling has also been shown to indirectly decrease eosinophil infiltration. In several different mouse models of allergic airway inflammation, androgen treatment consistently decreases eosinophil airway infiltration, through multiple mechanisms such as diminished Th2-associated cytokines and chemokines, attenuation of Th17 cells, enhancement of Treg suppressive function, and M2 polarization of macrophages (19,33,34).

Mast cells.

Mast cells are granulocytes which mediate allergic and anaphylactic responses through release of histamine and multiple other mediators. Mast cells have been shown to express AR by IHC (35). In vitro experiments assessing the effect of androgens on mast cell function have been somewhat conflicting, with reports of androgens inhibiting (36) or having no significant effect (35) on mast cell degranulation. Interestingly, an in-depth study in rodent models implicated a role for perinatal androgens in regulating sex differences in mast cell-mediated anaphylaxis (37). Androgenization of female mice reduced the severity of mast cell-mediated anaphylaxis and decreased the amount of pre-stored mast cell mediators, while mast cells from testicular feminization mutant rats had higher histamine content compared to WT males. While further validation experiments are needed, the demonstration of AR expression in mast cells and the clear phenotypic sex differences in allergic predisposition support an important role for AR signaling in mast cell function.

Maintenance of immunological barriers.

While not a function of one specific immune cell, the maintenance of physical barriers to inhibit pathogen entry, heal disrupted barriers and also establish sites of immune privilege is a key aspect of immune regulation. Canonical sites of immune privilege include the central nervous system (CNS), eyes, testis, ovary, and placenta. To date, androgens have a clearly defined role in establishing testicular immune privilege, in particular. This occurs through Sertoli cells, the “nurse” cells which are responsible for maintaining an appropriate microenvironment within the seminiferous tubules to foster spermatogenesis, including establishment of the blood-testis barrier through a network of specialized tight junctions. In a Sertoli cell-specific AR ablation mouse model, AR loss resulted increased the permeability of the blood-testis barrier, possibly through claudin-3 downregulation. This phenotype was associated with a rich inflammatory testicular infiltrate of macrophages, neutrophils, plasma cells, and eosinophils and with the development of germ cell antigen autoantibodies (38). However, claudin-3 KO does not phenocopy the AR mutant model and other tight junction components are dysregulated in AR Sertoli mutants, suggesting a more extensive program by which the AR enforces blood-testis barrier integrity (39). There is also some evidence that testosterone maintains tight junction integrity within the prostate, as low testosterone correlates with reduced prostatic expression of Claudin 4 and Claudin 8 and impaired tight junction structure by electron microscopy (40).

AR signaling may also influence immunity in the brain, although evidence for this connection is still preliminary. Castration of male mice can increase blood-brain barrier permeability with evidence of tight junction disorganization and decreased expression of tight junction proteins claudin-5 and ZO-1, changes which are reversed with testosterone supplementation (41). Additional studies will be needed to further validate and explore these findings. Interestingly, in contrast to its mostly immunosuppressive role throughout the body, AR signaling may promote antitumor immunity within the brain. Mice with intracranially implanted glioblastoma or other non-CNS tumors experienced accelerated tumor growth after castration (Res Sq 2024 Mar 29:rs.3.rs-4014556). This phenomenon occurred only in intracranial as opposed to subcutaneous tumors, suggestive of a CNS-specific effect, and was not observed in immunodeficient mice, suggestive of an immune-mediated mechanism. Mechanistically, the authors determined that castration in the setting of an intracranial tumor resulted in exaggerated hypothalamic-pituitary-adrenal axis activation and increased circulating glucocorticoid levels, which could plausibly induce immunosuppression.

Wound healing also represents an important barrier function of innate immunity, and AR signaling heavily influences this process primarily through macrophages, which play a central role in the regulation of tissue repair and homeostasis. Males exhibit delayed cutaneous wound healing compared to females, and castration and ARSI treatment of male mice accelerates wound healing (42). Cell-type specific mouse models confirm that myeloid-specific, but not keratinocyte- or fibroblast-specific, AR KO recapitulated the accelerated wound healing phenotype, suggesting that macrophages are key mediators of this process (43). AR signaling achieves suppression of wound healing through multiple mechanisms including increased tissue macrophage expression of TNF-α, higher levels of circulating inflammatory monocytes, and increased bone-marrow derived monocyte/macrophage chemotaxis via AR-mediated CCR2 upregulation (42,43).

Androgens also influence the repair and remodeling of vasculature through regulation of macrophage function. Global loss of AR attenuates the development of abdominal aortic aneurysm (AAA), although this effect is probably mediated by multiple cell types as AR KO in both macrophages and smooth muscle cells reduced AAA incidence (44,45). Interestingly, these studies also determined that AR KO mice exhibited decreased expression of proinflammatory cytokines such as IL-1α, IL-6, IL-17, and TGF-β1 within aortic tissue. This could be contradictory to prior evidence that androgens impair macrophage inflammatory cytokine production. However, it is also possible these cytokines could be produced by other cell types such as smooth muscle cells which do have AR expression, or that decreased cytokine levels could be a reflection of the reduced macrophage infiltration.

Effects of androgens and AR signaling on components of the adaptive immune system

Compared to the innate immune system, the adaptive immune system generates antigen-specific responses against pathogens, allowing for long-term memory and protection. The two main arms of the adaptive immune system are B and T lymphocytes. B lymphocytes have the ability to recognize foreign extracellular or surface antigens through the B cell receptor, which consists of a uniquely recombined heavy chain and light chain sequence allowing for recognition of a vast number of unique epitopes. Activated B cells can subsequently differentiate into plasma cells to secrete antibodies or into memory B cells to maintain immunological memory. T lymphocytes also express a uniquely recombined surface receptor – the T cell receptor (TCR) – that has the ability to recognize intracellular or extracellular-derived peptide fragments when they are presented on MHC class I or II. The two major subsets of T cells include CD8+ cytotoxic T cells, whose TCR recognizes peptide bound to MHC class I and can exert direct cytolytic activity on infected or cancerous cells, and CD4+ helper T cells, which recognize extracellular peptides presented on MHC class II by professional antigen presenting cells such as dendritic cells, macrophages, and B cells and subsequently produce cytokines to coordinate a broader immune response. The major subtypes of CD4+ T helper cells (T_H) each coordinate different responses: T_H1 cells coordinate cell-mediated responses against intracellular pathogens, T_H2 cells facilitate antibody production and defense against helminths, T_H17 cells drive inflammatory and allergic responses, and T regulatory cells (Tregs) exert immunosuppressive activity. The importance of the adaptive immune system, in particular T lymphocytes, in enforcing antitumor immunity has motivated deeper investigation into the effect of androgens on adaptive immunity (Figure 1).

B Lymphocytes.

Androgens broadly suppress B lymphopoiesis. Castrated male mice exhibit increased circulating B lymphocytes, increased spleen weight, and increased B cell progenitors in the bone marrow, and these effects are reversed with androgen supplementation (46,47). Similar trends are observed in humans (48). The AR was found to be expressed in immature B cells and bone marrow stromal cells but not mature B cells, indicating that androgens affect earlier stages of B cell development (47). Several potential mechanisms for this observation have been proposed. Chimeric mice with either AR-null bone marrow stromal cells or AR-null lymphoid cells were created by reciprocal bone marrow transplantation: androgen administration impaired B lymphopoiesis only in mice with AR-null bone marrow stroma, highlighting that AR activity in stromal cells is necessary for androgen-mediated suppression of B lymphopoiesis (49). A subsequent study showed that both germline AR KO and conditional B lymphocyte-specific AR KO conferred B lymphocytes with increased resistance to apoptosis and increased proliferative capacity, suggesting the presence of at least some degree of B cell-intrinsic AR signaling (50). Building upon prior investigations of the effects of bone marrow stromal cells, AR-deletion in the osteoblast lineage also caused an increase in bone marrow B cells (51). However, since this osteoblast AR KO model did not recapitulate the increase in splenic B cell populations, additional mechanisms were identified whereby testosterone was determined to regulate the production of the B cell survival factor BAFF by fibroblastic reticular cells in the spleen (52).

Androgens may also influence B lymphocyte function. Activated male B lymphocytes and testosterone-treated B lymphocytes exhibit less efficient follicle-center localization in secondary lymphoid organs compared to female and castrated male B lymphocytes (53). Androgens may also play a role in immunoglobulin class switching. In male lupus patients, longer AR CAG repeat length, which corresponds to progressive insensitivity to androgens, resulted in higher IgG levels and lower IgM levels (54), and ex vivo testosterone treatment of lupus patient peripheral blood mononuclear cells reduced IgG production (55) Lastly, females exhibit more robust antibody responses to vaccination than males do. In a study of serum responses to H1N1 vaccination, females exhibited higher neutralizing antibody titers compared to males, and within the cohort of adult males, testosterone concentration negatively correlated with antibody titer (56).

T lymphocytes.

AR signaling exerts diverse effects on the function of multiple T lymphocyte subsets. However, reports on whether T lymphocytes express AR at the protein level have been mixed. AR has been detected by Western blot in mouse T lymphocytes (57) but was not seen in lymphoid cells in an analysis of human bone marrow biopsies (3). While it is possible that AR expression levels may vary depending on T lymphocyte subtype and activation status, perturbation of AR signaling clearly affects the function of multiple subtypes. CD8⁺ T cells are of particular clinical interest due to their role in cancer immunity and immunotherapy, and AR signaling plays a significant role in modulating their function. In murine models, tumors consistently grow faster in males than females, which has been shown to be CD8⁺ T cell-dependent, and multiple mechanisms have been proposed to explain this phenomenon (57). Studies of genetically engineered mouse models including a CD8⁺ T cell-specific AR deletion, determined that androgen signaling increased the number of intratumoral TCF1+ progenitor exhausted CD8 T cells, and the AR was shown to directly regulate TCF1 expression (57). However, other studies have reported that androgen signaling induces a shift from stem cell-like to terminally exhausted CD8 T cell phenotypes (58). Moreover, Zhang and colleagues also determined that AR can potentially inhibit NF-kB signaling in CD8 T cells through upregulation of USP18 (59).

Regardless of mechanism, targeting AR signaling in CD8⁺ T cells appears to have functional consequences for antitumor immunity. This concept was evaluated in a mouse model whereby implanted prostate carcinoma cells developed resistance to ADT and PD-1-directed therapy. Treatment was then initiated with the combination of enzalutamide – to further repress AR signaling - and an anti-PD-L1 antibody which led to significant tumor regression, increased the number of CD8⁺ tumor-infiltrating lymphocytes (TILs), and enhanced the production of effector cytokines (5). AR was shown to bind to an open chromatin region associated with interferon gamma (IFNG), highlighting a possible mechanism of AR-mediated IFNG repression. However, a caveat to the effects of androgen signaling on CD8⁺ T cell function is a report demonstrating the AR antagonist flutamide negatively affected CD8⁺ T cell priming, which appeared to be independent of androgen signaling, and was correlated to a possible off target effect on GABA-A signaling (60). Thus, the specific method and pharmacologic agent used to modulate androgen signaling for the purpose of augmenting CD8⁺ T cell function should be carefully considered.

AR signaling is also consequential for the function of several CD4⁺ T cell subsets. For example, testosterone impairs T_H1 differentiation through inhibition of IL-12-induced STAT4 phosphorylation (61). Additionally, androgens enhance Treg function. In models of allergic airway inflammation, androgens enhance the immunosuppressive effects of Tregs while suppressing cytokine production from Th2 cells (34). In a rat experimental autoimmune orchitis model, testosterone treatment reduced the overall number of CD4⁺ T lymphocytes but notably increased the number of regulatory T cells (Tregs) (62). In vitro treatment of human CD4⁺ T lymphocytes with DHT also induced Treg expansion (63). Conversely, a study of androgen deprivation therapy in healthy human males significantly decreased the number of circulating Tregs (27). A study designed to directly identify the molecular underpinnings of these observations determined that the AR binds to the genomic region encoding FOXP3, the Treg master transcriptional factor, and changes H4 acetylation (63).

The characterized effects of AR signaling perturbation on intratumoral T lymphocyte infiltration in humans is mostly limited to prostate cancer in which androgen deprivation and AR inhibition are standard of care. In one clinical trial, treatment with ADT, with or without a granulocyte-macrophage colony-stimulating factor-secreting cellular vaccine, for two weeks in high-risk localized prostate cancer patients resulted in significantly increased CD8⁺ T lymphocyte infiltration at the time of prostatectomy by IHC (64). However, it should be noted that there was also a proportional increase in Tregs such that the CD8/Treg ratio remained similar in treated compared to untreated patients, which could mitigate the immunogenicity of this response. A separate study employing multiplex immunofluorescence immune quantification of ADT-treated prostatectomy specimens similarly showed significant increases in total T lymphocytes, CD4⁺ T lymphocytes, and Tregs, with a trend towards increased CD8⁺ T lymphocytes (24). Another trial of neoadjuvant androgen inhibition prior to prostatectomy showed a predominantly CD4⁺ T lymphocyte infiltrating response as opposed to CD8⁺, although treatment consisted of ADT plus the AR antagonist flutamide as opposed to ADT alone in the former study (65). While the distribution of different helper T cell subsets was not elucidated, 65% of mononuclear cells expressed interferon gamma, suggestive of an effector phenotype.

Androgens and Thymic Function.

The thymus plays an essential role in T cell development through the coordination of positive selection, which enriches for T cells which are able to effectively engage MHC class I or II, and negative selection, which eliminates T cells which are reactive against self-antigens. In this context, androgens play a key role in regulating the growth and involution of the thymus (Figure 2). Multiple rodent model studies have demonstrated that castration or AR KO produces thymic enlargement and regeneration, with increases in early thymic progenitors, CD4⁺CD8⁺ T cells, and recent thymic emigrants (RTE’s), with testosterone supplementation producing generally opposite effects (66–68). These findings are observed in humans as well: hypogonadal men treated with testosterone exhibited a decrease in circulating naïve T cell and T-cell receptor excision circles (TRECs) (69), while ADT in prostate cancer patients increased naïve T cell and TREC levels (68). However, it is important to also consider the kinetics and persistence of this androgen inhibition-induced thymic regeneration. In one study, mouse thymic weight and cellularity peaked at 5 weeks after castration, although a persistent increase in these parameters was still noted after 12 weeks (70). However another study in mice noted a peak in thymic cellularity at 20 days post-castration with a return to pre-castration cellularity by 34 days (71). Further studies will be needed to define the kinetics of these changes in mice, and to determine if such changes are generalizable to humans.

Figure 2. Ablation of AR signaling augments thymopoiesis through multiple mechanisms.

AR signaling plays a key role in suppressing multiple aspects of thymic development, and ablation of androgen signaling results in significant enhancement of thymopoiesis. AR inhibition increases mobilization of T cells from the bone marrow that subsequently home to the thymus. Suppression of androgen signaling stimulates thymic epithelial cells (TECs) to proliferate and secrete CCL25, which further attracts T cell precursors. AR loss subsequently plays a key role in modulating positive and negative selection of maturing T cells. In cortical TECs, AR deletion increases positive selection of T cells. AR loss in cortical TECs has also been shown to increase expression of the Notch ligand Dll4 which also improves thymopoiesis. In medullary TECs, AR deletion decreases expression of Aire, which could reduce negative selection and thus contribute to enhanced thymopoiesis. All of these effects lead to an increase in thymic emigrants upon suppression of AR signaling, which has been consistently observed in both mouse and human studies. Created in BioRender. Lee, P. (2025) https://BioRender.com/a07tgsj

Extensive efforts have been dedicated to understand mechanistically how AR signaling induces these dramatic changes in thymic function. While all subsets of thymocytes express the AR (72), cell type-specific AR KO experiments have revealed that AR signaling in thymic epithelial cells (TECs), as opposed to thymocytes themselves, is responsible for the dramatic AR-mediated effects on thymopoiesis (73,74). Multiple interesting mechanisms have been described that may contribute to this phenotype. Androgen withdrawal increases bone marrow mobilization of T cell precursors which subsequently migrate to the thymus (75). Within the thymus, AR loss increases TEC proliferation and secretion of CCL25, which attracts early T cell precursors to the thymus (67). Within the thymic epithelial cell compartment, AR deletion in cortical TECs enhances positive selection of double-positive thymocytes thus promoting T cell survival (73). AR loss in cortical TECs has been shown to upregulate the Notch ligand Dll4 which can facilitate thymopoiesis (76). Within medullary TECs, ablation of AR signaling decreases the expression of Aire, a transcription factor that facilitates the expression of diverse tissue-specific antigens to allow for elimination of self-reactive T cells (77). Thus, the presence of androgen signaling could impose more stringent T cell selection processes.

Androgen inhibition-mediated thymic regeneration may also carry therapeutic potential in cancer and other disease states. For example, recent thymic emigrants generated by ADT treatment in mice were able to traffic to tumors and produce effector cytokines, suggesting they can participate in antitumor immunity (70). A pilot study using the LHRH agonist goserelin to suppress androgens also improved T cell reconstitution following bone marrow transplantation in both mice (73) and humans (78).

Fibroblast androgen receptor signaling and immune cell recruitment

While not generally considered a component of the immune system, cancer-associated fibroblasts (CAFs) play a key role in shaping the composition and phenotype of immune cells within the tumor microenvironment. Often associated with pro-tumorigenic functions, CAFs recruit immunosuppressive cells such as Tregs, MDSCs, tumor-associated macrophages, and tumor-associated neutrophils, express immune checkpoints, and release immunosuppressive cytokines such as TGF-β, IL-17, and indoleamine 2,3-deoxygenase (IDO) that inhibit T lymphocytes and NK cells (79). Since CAFs in several cancers, particularly prostate and breast, express the AR, the effects of androgens on CAF function may contribute to the net effects of androgen inhibition on antitumor immunity. In orthotopic mouse models of squamous cell cancer and melanoma admixed with human dermal fibroblasts (HDFs), AR silencing activates HDFs and upregulates genes enriched for cytokine activity and immune response, although it is difficult to extrapolate clear immune recruitment effects in the setting of an immunodeficient mouse model (80). In a mouse model of prostate intraepithelial neoplasia (PIN), fibroblast- and smooth muscle cell-specific AR knockout decreased cytokine secretion from stromal cells and resulted in reduced T lymphocyte and macrophage infiltration (81). This was associated with reduced incidence of PIN, suggesting that loss of AR in prostate stromal cells results in a form of immune exclusion that may have antitumorigenic effects. However, prostate cancer-derived fibroblasts can also secrete immunosuppressive cell-recruiting cytokines such as CCL2 and CXCL8 upon AR inhibition (82). While current evidence does not define a clear pro- or anti-tumorigenic immune effect of AR inhibition in CAFs, androgen perturbations clearly modulate CAF cytokine secretion, the effects of which should be considered when evaluating effects of androgens on tumor immunity.

Androgen receptor signaling and regulation of tumor-intrinsic immune evasion

Emerging evidence indicates that AR signaling within cancer cells themselves may enhance tumor-intrinsic mechanisms of immune evasion. A recent pan-cancer computational analysis of TCGA data determined that cancer cell AR activity, inferred from bulk RNA-seq data, inversely correlated with immune system processes, including the IFN-γ pathway, and with immune cell infiltration across cancers (83). Moreover, low AR activity correlated with improved response to checkpoint blockade in melanoma, lung cancer, and mixed cancer cohorts. While much work remains to be done, several potential mechanisms have been described.

AR has been reported to repress MHC class I expression in a genome-wide CRISPRi screen in a cell line model of prostate cancer (84). Furthermore, AR expressing prostate cancer cells pre-treated with an AR inhibitor showed increased sensitivity to antigen-specific CD8⁺ T cell-mediated cytotoxicity. AR effects on MHC class I are maintained in prostate cancer patients, as neoadjuvant ADT prior to radical prostatectomy increased both MHC class I and II (24). The relationship between AR and PD-L1 has also been explored in several cancers. Interestingly, augmentation of AR signaling in vitro through androgen treatment or AR overexpression suppresses PD-L1 expression in prostate cancer (85), bladder cancer (86), papillary thyroid cancer (87), and hepatocellular carcinoma (88), possibly through binding of AR to androgen response elements in the PD-L1 promoter. However, one other study in bladder cancer identified an opposite trend (89). AR signaling may also facilitate immune evasion from NK cells, as androgens promote shedding of the NK cell inhibitory ligands MICA and MICB in melanoma cell lines (90). In breast cancer, AR expression and presence of tumor associated neutrophils (TANs) appear to be somewhat mutually exclusive, with AR expression more common in Luminal A/B subtypes and TANs more common in triple-negative disease, but differences in estrogen or progesterone receptor expression and other subtype-specific features could certainly contribute to this observation (91).

While limited evidence thus far suggests that androgens promote tumor-intrinsic immune evasion, adrenocortical carcinoma (ACC) is one example in which androgens have been shown to facilitate increased immune infiltration. Mouse models of ACC driven by Znrf3 knockout showed significantly increased myeloid cell infiltration in males compared to females, which was driven by an androgen-amplified senescence-associated secretory phenotype (SASP) within tumor cells. Male mice were less likely to develop malignant tumors, and this observation mirrors the disproportionate incidence of ACC in human females compared to males. In patient ACC samples from The Cancer Genome Atlas (TCGA), increased myeloid infiltration scores corresponded with improved prognosis, highlighting the clinical relevance of these observations (92).

Clinical efforts to promote antitumor immunity through perturbation of androgen signaling

The diverse and generally immunosuppressive effects of AR signaling, particularly in T lymphocytes, has garnered significant interest for the application of antiandrogen therapies to augment antitumor immune responses. This concept is further supported by the observation that men tend to have decreased overall survival benefit from immunotherapy compared to women across multiple cancers (93), although this finding has not been consistent across studies. Moreover, numerous peripheral blood flow cytometry assessments conducted in cancer patients treated with androgen deprivation or antiandrogen therapies have shown changes in peripheral immune composition, most consistently an increase in circulating T cells (94). Unfortunately, clinical efforts thus far have not shown convincing benefit with the combination of antiandrogens and immunotherapy (Table 1). Several phase III trials in metastatic prostate cancer including IMbassador-250, KEYNOTE-991, and KEYNOTE-641 have studied the AR inhibitor enzalutamide in combination with anti-PD-1 or PD-L1 therapy without evidence of improvement in overall survival or progression-free survival (95–97). A phase I trial in prostate cancer patients with biochemical recurrence after surgery or radiation evaluated the combination of bicalutamide and the CTLA-4 antibody tremelimumab, which prolonged PSA doubling time in 3 of 11 patients (98).

Table 1.

Notable clinical trials with published results in prostate cancer and other cancers studying the combination of androgen-modulating therapies and immunotherapy.

Clinical Trial ID	Phase	Cancer	Combination Therapy	Clinical Outcome	Key Immune Findings	Reference
IMbassador250 (NCT03016312)	3	Prostate (mCRPC)	Atezolizumab + enzalutamide vs enzalutamide	Primary endpoint of OS not met	Longer PFS seen with high PD-L1, CD8 expression, and immune gene signatures	(103)
KEYNOTE-991 (NCT04191096)	3	Prostate (mHSPC)	Pembrolizumab + enzalutamide vs enzalutamide	Dual primary endpoint of rPFS and OS not met. Stopped for futility.	Not reported	(96)
KEYNOTE-641	3	Prostate (chemotherapy-naïve mCRPC)	Pembrolizumab + enzalutamide vs enzalutamide	Primary endpoints of rPFS and OS not met. Stopped for futility.	Not reported	(97)
STARVE-PC (NCT02601014)	2	Prostate (AR-V7-expressing mCPRC)	Nivolumab + ipilimumab + enzalutamide (cohort 2)	Primary endpoint of safety and PSA response rate, which was 0% in cohort 2	Lower serum IL-6 and IL-7 and higher IL-17 correlate with improved OS	(104)
NCT00702923	1	Prostate (biochemical recurrence)	Tremelimumab + bicalutamide	Primary endpoint of safety. PSA doubling time prolonged in 3/11 patients	Not reported	(98)
NCT01377389	2	Prostate (mHSPC)	Finite ADT + ipilimumab	Primary endpoint: proportion of patients with PSA ≤0.2 at 7 months. Study stopped early due to side effects	Increased CTLA-4+ T cells and increased ratio of CTLA-4+ T cells to PD-1+ CD3 T cells associated with clinical benefit	(105)
Debio 8200-IMM-101	1	Melanoma	Triptorelin (ADT), bicalutamide, and nivolumab	Primary endpoint: safety and tolerability. PR in 5/14 patients	Increase in TRECs in 2/14 patients, increase in TILs in 2/4 patients with paired biopsies	(100)
NCT03650894	2	AR+ metastatic breast cancer	Bicalutamide, nivolumab, and ipilimumab	Primary endpoint: 24 week clinical benefit rate, which was 17%	Therapy associated with CD8+ T cell expansion; TREC expansion noted in patients under 50 years old	(101)
BTCRC-HN17–111 (NCT03942653)	2	AR+ salivary gland cancers	ADT and pembrolizumab	Primary endpoint: objective response rate 22%	Not reported	(102)

ADT, androgen deprivation therapy; mCRPC, metastatic castration-resistant prostate cancer; mHSPC, metastatic hormone-sensitive prostate cancer; OS, overall survival; PFS, progression-free survival; PR, partial response; PSA, prostate serum antigen; rPFS, radiographic progression-free survival; TILs, tumor-infiltrating lymphocytes; TREC, T cell receptor excision circle.

While prostate cancer is the most accessible setting to study immune effects of antiandrogen therapies since they are already standard of care, it may also be one of the more difficult cancers in which to augment antitumor immunity, based on its known immunosuppressive microenvironment and lack of significant benefit with checkpoint inhibitor treatment to date. Despite the generally positive effects of AR inhibition on immune cell function, long-term ARSI treatment eventually predisposes to more aggressive disease that may results from a more suppressive microenvironment. Enzalutamide treatment in the Myc-CaP syngeneic mouse model showed initial tumor control followed by disease recurrence by 2 months, which was accompanied by immunosuppressive tumor immune microenvironment changes including decreased CD8⁺ T lymphocytes and increased MDSCs and macrophages (85). Nevertheless, several novel antiandrogen and immunotherapy combinations are currently being explored in prostate cancer including ipilimumab plus abiraterone (NCT01688492); abiraterone, apalutamide, and ipilimumab (NCT02703623); and abiraterone, cabozantinib, and nivolumab (NCT04477512). Moreover, novel STEAP1-targeted immunotherapeutics, including STEAP1 CAR T-cell therapy (NCT06236139) (99) and AMG-509, a STEAP1-CD3 bispecific antibody (NCT04221542), are being studied in combination with androgen receptor pathway inhibitors. Testing androgen perturbation with these novel cell-based and immune cell-engaging therapies may shed light on novel synergies not previously seen with checkpoint blockade. Other trials are studying these combinations in earlier lines such as neoadjuvant therapy prior to radical prostatectomy (NCT03753243). Localized prostate cancer may represent a more immunogenic substrate compared to metastatic, castration-resistant disease.

Antiandrogen therapy for the purpose of immune reconstitution is also being evaluated in non-prostate cancers (Figure 3, Table 1). A phase I study of ADT in combination with anti-PD-1 therapy in 14 metastatic melanoma patients who had previously received anti-PD-1 therapy showed a disease control rate of 42.8%. However, this combination did not produce significant increases in TILs or T cell receptor excision circles (TRECs), which would have been indicative of thymic rejuvenation (100). A phase II trial in AR⁺ metastatic breast cancer determined that the combination of the AR antagonist bicalutamide, together with ipilimumab and nivolumab produced a 24-week clinical benefit rate of 27% (4 of 15 patients) in the triple-negative breast cancer subgroup, although this did not meet criteria for trial expansion (101). However, this regimen did produce a statistically significant increase in circulating CD8⁺ T cells, and some degree of TREC expansion, a marker of thymic regeneration. Another phase II trial combined ADT with pembrolizumab in salivary gland tumors, which are often AR-positive, with an initial objective response rate of 22% in 9 patients (102). Ongoing studies without reported results as of yet include a trial of leuprolide and anti-PD-1 therapy in advanced lung cancer (NCT06512207) and pembrolizumab plus flutamide in advanced head and neck squamous cell cancer (NCT06694350).

Figure 3. Summary of clinical trials exploring the application of androgen deprivation therapy or androgen receptor pathway inhibition to enhance antitumor immunity.

Several ongoing or recently completed trials are exploring the combination of androgen deprivation therapy/AR inhibition with immunotherapeutic agents across multiple cancer types. For sake of brevity, previously completed trials of ARSIs and checkpoint inhibitors in metastatic prostate cancer are omitted from this graph but are discussed within the text and in other reviews. Instead, we highlight novel combinations in prostate cancer, such as ARSI plus STEAP1 targeted immunotherapies (NCT06236139, NCT04221542), or applications in different clinical contexts, such as the use of neoadjuvant pembrolizumab and enzalutamide prior to radical prostatectomy (NCT03753243). We also summarize combinations of immunotherapy and androgen receptor in other cancers including melanoma, salivary gland tumors, head and neck squamous cell cancers, breast cancer, and lung cancer. Created in BioRender. Lee, P. (2025) https://BioRender.com/cd5opde

Conclusions and Future Directions

Androgens and androgen receptor signaling are relevant mediators of function across multiple immune cell subtypes, in which they generally exert immunosuppressive effects. However, much remains to be discovered regarding the transcriptional programs and precise mechanisms by which AR signaling exerts these diverse responses. Much of the difficulty in parsing out these effects is due to the pleotropic activity of androgen signaling in both immune and non-immune cells. Moving forward, cell type-specific perturbations of AR activity will be instructive to distinguish cell-intrinsic versus secondary consequences of AR pathway manipulation. Most of the current mechanistic knowledge of AR-specific effects within immune cells is derived from murine models, and while corresponding analyses within human clinical studies are steadily increasing our understanding, much remains to be learned (Table 2). Moreover, our current knowledge of the AR regulatory network has been developed primarily in the context of the prostate and prostate cancer, and it is likely that AR programs vary substantially across different cells and tissues. Better characterization of how the AR cistrome varies amongst different immune cell types will be paramount to clarifying exactly how the AR alters immunological responses. Determining the mechanistic roles for the AR on immunity is timely for the treatment of cancer, as immunotherapy is standard of care across many different cancers and continues to evolve. As our understanding grows of how cancers evade immune surveillance, immunotherapeutics have been developed to engage multiple different arms of the immune system. Modulation of androgen signaling represents an intriguing therapeutic opportunity to enhance antitumor immunity and synergize with pre-existing immunotherapies. Defining the precise and unique AR regulatory programs within each immune compartment can help to identify the most promising cell types in which AR perturbation can augment their antitumor function. Such advances can help guide selection of the optimal cancer subtypes and patient contexts for therapeutically targeting the AR.

Table 2.

Summary of notable AR signaling effects on different immune cells, organized by pre-clinical (both animal and human) and clinical evidence.

Immune Cell	Biologic Process	Effect of AR signaling
		Preclinical – animal models	Preclinical – human models	Clinical
Neutrophils	Granulopoiesis	AR KO mice exhibit neutropenia (6)	-	AR pathway inhibitors can cause neutropenia (7)
	Cytotoxic Function	Testosterone impairs bactericidal function in rat prostatitis model (8)	Testosterone decreased superoxide production in neutrophils from healthy male donors (10)	Neutrophils from prostate cancer patients on ADT exhibit signs of constitutive activation (12)
	Antitumor function	Mixed – androgens seem to promote tissue accumulation of neutrophils, but this has resulted in both tumorigenic and antitumor effects (11,12)	Enzalutamide suppresses patient-derived neutrophil cytotoxicity when cocultured with prostate cancer cell line (13)	Neutrophils from prostate cancer patients on ADT exhibit decreased positivity of CD16, a marker associated with antitumor activity (12)
Macrophages	Response to infection	Androgens suppress TLR4 and inflammatory cytokine production (17,18)	-	-
	M1 vs M2 polarization in cancer	Ablation of the TREM2-AR signaling axis reduces intratumoral M2 macrophages in prostate cancer mouse model (23)	AR signaling promotes M2-like polarization in the THP-1 cell line (23)	ADT promotes intratumoral M1 macrophage infiltration in localized prostate cancer (24)
NK cells	Activation	-	ARSIs increase secretion of IFN-γ and Granzyme B (bioRxiv. 2023:2023.11.15.567201.)	Increased number of circulating granzyme B+ and perforin+ NK cells in prostate cancer patients who started ADT (bioRxiv. 2023:2023.11.15.567201).
B lymphocytes	B lymphopoiesis	Castrated mice show increased B lymphocytes in bone marrow, spleen, and circulation (46,47)	-	Increased B lymphocytes in hypogonadism patients treated with gonadotropin (48)
	Antibody production	-	In vitro testosterone treatment of PBMCs from lupus patients suppressed IgG production (55)	Females exhibit higher neutralizing antibody titers compared to males, and testosterone level correlates with titer in males (56)
T lymphocytes	CD8+ T lymphocyte function	Enzalutamide or castration synergizes with PD-1:PD-L1 blockade to enhance CD8+ T lymphocyte function and prevent exhaustion (5,58)	-	Male tumor-infiltrating CD8+ T lymphocytes show less stemness and more exhaustion compared to females (58)
	Th1 cell differentiation	Testosterone impairs Th1 differentiation (61)	-	-
	Treg cells	Androgens enhance Treg function in autoimmune inflammation models (34,62)	DHT treatment of human CD4+ T lymphocytes causes Treg expansion (63)	ADT in healthy males decreases circulating Tregs (27)
	Tumor infiltrating lymphocytes (TILs)	ADT in the Myc-CaP mouse model induces transient increase in CD4+, CD8+, and Treg subtypes, which decrease at time of castration resistance (106)	-	ADT in localized prostate cancer generally increases tumor-infiltrating CD4+, CD8+, and Treg subtypes (24,64,65)

ADT, androgen deprivation therapy; ARSI, androgen receptor signaling inhibitor; DHT, dihydrotestosterone; KO, knockout; PBMCs, peripheral blood mononuclear cells; Treg, T regulatory cell.