Decoding GDF15: Impact on prostate cancer metabolism, chemoresistance, and clinical applications

Abstract

Prostate cancer (PCa) presents a significant global health challenge, and cachexia in end-stage PCa further reduces survival time and deteriorates patient quality of life. This necessitates nuanced approaches to prevention, diagnosis, and treatment. This review analyzes the complex role of growth differentiation factor 15 (GDF15) in various aspects of PCa, including metabolism, chemoresistance, metastasis, and clinical implications. Mechanically, the interactions of GDF15 with PCa lipid metabolism and stromal activation are hypothesized to drive PCa progression and promote cachexia. GDF15’s role in PCa bone metastasis is mediated through interactions with osteoblasts and osteoclasts. Furthermore, GDF15’s role in chemoresistance emphasizes its impact on drug responses and suggests a potential therapeutic target. Clinically, GDF15 has shown potential as a biomarker, offering diagnostic precision and prognostic value, particularly when combined with established markers such as prostate-specific antigen. The review concludes with a discussion on several monoclonal antibodies targeting GDF15 in clinical trials, such as AV-380, NGM120, Visugromab, and AZD8853, highlighting promising strategies for novel therapies and precision medicine.

Introduction

Cancer continues to pose a formidable global health challenge, with high morbidity and mortality rates necessitating ongoing attention and research efforts.^[1] Despite a general decline in cancer incidence over two decades, prostate cancer (PCa) morbidity has increased by approximately 3% annually over the last five years. This increase is attributed to inflammation, senescence, and other factors.^[2] In China, the prevalence of advanced PCa at initial diagnosis, increased metastatic rates, and lower five-year overall survival (OS) rates are notably higher than in the United States.^[1,3] Despite significant advances in PCa treatment, the continued prevalence of this disease highlights the urgent need for further advancements and targeted solutions.

Growth differentiation factor 15 (GDF15), also known as macrophage inhibitory cytokine 1 (MIC-1) or nonsteroidal anti-inflammatory drug-activated gene-1 (NAG-1), is a distinct member of the transforming growth factor-beta (TGF-β) superfamily.^[4] It is expressed in various human tissues, with the highest levels found in the liver in mice.^[5] GDF15 is implicated in a range of physiological and pathological processes, demonstrating significant relevance in tumorigenesis, ischemic disease, metabolic alterations, and neuronal degeneration.^[6]

Despite its role in various cancer types manifesting as either antitumor or protumor effects,^[6–9] the specific role of GDF15 in PCa remains ambiguous. For example, Zhang et al^[10] observed fluctuating levels of GDF15 mRNA and protein, initially increasing and then decreasing from benign prostate tissue to primary tumors and metastatic castration-resistant prostate cancer (CRPC). Similarly, Rasiah et al^[11] confirmed a consistent increase in GDF15 levels during the progression from benign prostate tissue through low-grade/high-grade prostatic intraepithelial neoplasia to PCa. Conversely, Srour et al^[12] found a negative correlation between GDF15 and PCa risk. These conflicting findings extend to correlations with Gleason score, recurrence risk, and invasion capacity.^[11,13] However, it is relatively consistent that research by Noorali et al^[14] supports the hypothesis that GDF15 acts as a dual growth factor, capable of promoting cell proliferation or inducing growth arrest and differentiation in prostate cells.

Given these discrepancies, this review aims to critically assess the existing literature, clarifying the complex relationship between GDF15 and PCa from the perspectives of immunity, metabolism, and clinical management. By synthesizing diverse findings, this review seeks to enhance understanding of the potential dual role of GDF15 in PCa pathogenesis.

GDF15 and PCa Characteristics

Initially, GDF15 was identified as an autocrine inhibitor of macrophage primary activation in response to secreted proinflammatory cytokines,^[15] including interleukin 1 beta (IL-1β), tumor necrosis factor alpha (TNF-α), interleukin 2 (IL-2), and macrophage colony-stimulating factor 1 (MCSF-1). In the human genome, GDF15 is encoded by two simple exons located on chromosome 19p13.1.^[16] The exon encodes an RNA transcript of approximately 1200 bp, which is translated into a 308 amino acid sequence. This includes a signal peptide sequence of 29 amino acids, a pro-peptide sequence of 167 amino acids, and a mature polypeptide sequence of 112 amino acids.^[17] Although intracellular processing occurs, GDF15 is primarily secreted as a proprotein with its pro-domain remaining attached to the extracellular matrix (ECM). Consequently, extracellular GDF15 enables the rapid release of large amounts of mature GDF15 upon proteolytic cleavage.^[18] Like other members of this family, mature GDF15 contains a conserved interchain disulfide bond and is secreted as a dimeric protein with a molecular weight of approximately 25 kDa. Physiologically, GDF15 expression is most pronounced in the placenta and the prostate, where androgen regulation of GDF15 has been demonstrated.^[19] Functionally, its roles in appetite regulation, metabolism, cell and tissue survival, and immune tolerance have been documented.

GDF15 has been linked to cancer development and progression across various tumor types.^[20] In PCa research, it has attracted significant attention due to its association with high-grade PCa. Protein profiling on microdissected tissues from matched normal prostate, high-grade prostatic intraepithelial neoplasia (HGPIN), and PCa showed GDF15 expression in both HGPIN and cancer cells, but not in normal cells.^[21] This suggests a correlation between early prostate carcinogenesis and GDF15 expression. Further studies using human PCa cell lines, including LNCaP, PC3, MDAPCa2b, and DU145, revealed that GDF15 expression was restricted to cell lines expressing the androgen receptor (LNCaP and MDAPCa2b).^[22,23] Subsequent high-throughput RNA-sequencing data have identified GDF15 as a prognostic biomarker for PCa,^[24] underscoring its significant role in this field.

GDF15 and PCa microenvironment

GDF15 serves as a dual-edged sword, originally identified as a negative regulator of anti-tumor T cell activity, which suppresses immune responses and hinders T cell recruitment into the tumor microenvironment (TME).^[25] Tumor cells express and release large amounts of GDF15 within the TME, enabling evasion of recognition by circulating immune cells and anti-tumor immune responses. Research across several cancers has shown that GDF15 mediates a suppressive TME.^[25–27] Notably, Li et al^[25] discovered that ablation of GDF15 transformed the immunosuppressively “cold” TME into an inflammatory state, highlighting a potential strategy to enhance the efficacy of current immunotherapies in PCa.

In PCa, however, the role of GDF15 varies. Studies on transgenic adenocarcinoma of the mouse prostate (TRAMP) mice, which represent PCa patients lacking GDF15 expression, have yielded inconsistent results.^[28] Husaini et al^[28] reported that, compared to TRAMP mice, the hybrid offspring of TRAMP and GDF15 overexpressing mice exhibited longer survival, smaller tumors, and lower histopathological grades, but had more distant metastatic sites. This suggests that GDF15 expression may limit local tumor growth yet could promote distant metastasis during PCa progression. Interestingly, the tumor-protective function of GDF15 in spontaneously developed PCa in TRAMP mice relies on a complete adaptive immune response. Husaini et al^[29] demonstrated the crucial role of recently activated CD8-positive T cells in this process. Blocking these cells with a monoclonal antibody targeting CD8, resulting in T-cell exhaustion, reversed the protective effects of GDF15. In addition, infusing GDF15 into mice with orthotopic TRAMP tumors significantly reduced tumor growth, an effect amplified by simultaneous administration of PD-1 antibodies.^[29] Consistent with other findings, T cells influence the progression of Pca in TRAMP models.^[30] The complex role of GDF15 is further demonstrated in studies indicating that its absence leads to increased tumor growth and lower survival rates, while its overexpression enhances local invasion and metastasis during PCa progression.^[31] Another mechanism involving GDF15 is mediated by p53.^[32] As a direct transcriptional target of p53, GDF15 is induced and secreted following p53 activation, akin to p21.^[33] In ovarian cancer cells, wild-type p53 induced GDF15 expression upon cisplatin treatment, contributing to chemoresistance by modulating the TME.^[34] Furthermore, Recarte et al^[35] revealed that activation of peroxisome proliferator activated receptor beta/delta (PPARβ/δ) increased p53 levels, which subsequently enhanced GDF15 expression in glucose and lipid metabolism.

The complex regulation of immunity by GDF15 in the TME involves macrophages, particularly the M2 type. Bonaterra et al^[36] provided novel data demonstrating a significant association between GDF15-positive cells and the density of M1/M2 macrophages. During prostate tissue carcinogenesis, an increase in M2 macrophages is noted, which exerts an inhibitory effect on anti-tumor immune responses.^[37] Furthermore, in PCa, high infiltration of M2-like macrophages in tumor tissue is associated with tumor recurrence and metastasis.^[37] Sadasivan et al^[38] suggested that GDF15 expression is associated with the presence of M2 macrophages, particularly when its expression is more pronounced in normal-appearing glands adjacent to tumors than in the tumor region itself.

The relationship between GDF15 and PCa can also be viewed through the lens of inflammation, a known contributing factor to PCa.^[39] For instance, immunohistochemical staining of prostatectomy specimens revealed a negative correlation between GDF15 levels and prostatic inflammation, likely due to the suppression of nuclear factor kappa B (NF-κB) activity.^[40] This finding is supported by the study of Rybicki et al,^[41] who noted that GDF15 may inhibit chronic inflammation in benign prostatic hyperplasia and play a role opposite to that of NF-κB in prostate tumorigenesis. In addition, GDF15 has an anti-inflammatory effect by inhibiting granulocytes, macrophages, dendritic cells, and other cell types. Mechanically, Haake et al^[42] reported that GDF15 impaired lymphocyte function-associated antigen-1 (LFA-1)/β2-integrin-mediated T cell adhesion to activated endothelial cells, and neutralizing GDF15 enhanced T cell trafficking and therapeutic efficiency in mouse tumor models. This multifaceted role of GDF15 in immune regulation is further emphasized by its mediation of the inhibitory effects of nonsteroidal anti-inflammatory drugs (NSAIDs) on PCa cell migration through the p38 mitogen-activated protein kinase (MAPK)-p75 neurotrophin receptor (NTR) pathway.^[43]

In addition, the GDF15 signaling pathway extends its impact to cancer-associated fibroblasts (CAFs), contributing to a robust tumor-promoting effect and altered tumor matrix composition. Bruzzese et al^[44] verified that stromal cells are significant sources of GDF15 in PCa tissue. Their findings identified fibroblast-derived GDF15 as a candidate regulator of PCa progression. Data from both in vitro coculture conditions and mouse xenograft models demonstrated a robust tumor-promoting effect of GDF15, even stimulating the growth of indolent cancer cells at a distance.^[44] Interestingly, GDF15-secreting fibroblasts exhibited characteristics similar to myofibroblasts, with increased expressions of factors encoding ECM components and matrix remodeling enzymes.^[44] In addition, the altered tumor matrix, characterized by a higher abundance of collagens mediated by GDF15-secreting fibroblasts, may partly explain increased tumor stiffness, thereby facilitating increased PCa invasion.^[45]Figure 1 summarizes the interactions between GDF15 and cells in the TME.

Figure 1.

The interplays between GDF15 and cells in TME. GDF15 functions as a dual-edged regulator within the TME. On one hand, it suppresses anti-tumor immunity by impairing T cell recruitment and adhesion through inhibition of LFA-1/β2-integrin-mediated interactions with activated endothelial cells, blocking dendritic cell and granulocyte infiltration, and promoting the immunosuppressive activity of M2 macrophages. These effects collectively establish a “cold” TME that facilitates tumor immune evasion. On the other hand, GDF15 exhibits context-dependent tumor-protective effects where its expression limited local tumor growth in a CD8⁺ T cell-dependent manner, although it simultaneously promoted distant metastasis. Beyond immune regulation, GDF15 also remodels the stromal compartment: it stimulates cancer-associated fibroblasts (CAFs) to adopt a myofibroblast phenotype and enhance collagen secretion, thereby increasing tumor stiffness and invasiveness. Moreover, fibroblasts are an important source of GDF15 that further reinforces tumor progression. In the bone microenvironment, GDF15 enhances osteoblast activity, promotes CCL2 and RANKL secretion, recruits osteoclasts, and activates osteoclastogenesis, ultimately facilitating prostate cancer bone metastasis. Together, these findings underscore the complex and sometimes paradoxical functions of GDF15 in prostate cancer progression, immune regulation, chemoresistance, and metastasis. CCL2: C-C Motif Chemokine Ligand 2; EC: Endothelial cell; GDF15: Growth differentiation factor 15; LFA-1: Lymphocyte function-associated antigen-1; PCa: Prostate cancer; RANK: Receptor activator of nuclear factor kappa-B; RANKL: Receptor activator of nuclear factor kappa-B ligand; TME: Tumor microenvironment. Solid lines represent the direct effects of GDF15, while dashed lines indicate that the corresponding cell types can secrete GDF15 to form feedback.

In summary, the complex interplay between GDF15 and the PCa microenvironment highlights its multifaceted role in immune regulation, inflammation, and the contribution of CAFs to PCa progression. This comprehensive exploration of GDF15 within the PCa microenvironment reveals intricate regulatory mechanisms and potential avenues for targeted therapeutic interventions.

GDF15, PCa metabolism, and cachexia

In the context of PCa metabolism and cachexia, the intricate relationship between GDF15 and lipid metabolism emerges as a key player in PCa progression. Unlike other normal tissues, the healthy prostate primarily relies on glycolysis to produce citrate by interrupting the Krebs cycle-related oxidative phosphorylation.^[46] However, PCa cells undergo a metabolic transition from glycolysis to efficient fatty acid oxidation (FAO).^[47] This adaptability in energy production, particularly in lipid oxidation, is considered a hallmark of PCa tumorigenesis and metastasis.^[47] Flaig et al^[48] demonstrated that reduced lipid oxidation, mediated by carnitine palmitoyltransferase 1 (CPT1), diminished PCa proliferation and invasion, emphasizing the significance of lipid oxidation in supporting PCa growth. Itkonen et al^[49] reported that the expression of the lipid metabolism-related enzyme Enoyl-CoA delta isomerase 2 (ECI2) could promote PCa survival and predict mortality. Notably, in PCa mouse models with a high-fat diet, excessive fat intake stimulated PCa cell growth and invasion by upregulating GDF15 signaling.^[50] This also promoted the secretion of pro-inflammatory mediators (IL-6 and IL-8) in surrounding fibroblasts. Furthermore, upregulated adipose lipolysis and free fatty acid release from periprostatic adipocytes also stimulated GDF15 and IL-8 secretion. More importantly, elevated serum GDF15 showed a significant correlation with the activation of PCa stroma, elevated serum levels of IL-8, IL-6, and lipase activity, and advanced PCa progression.^[50] These findings indicate an inseparable connection between the activation of lipid metabolism and changes in the PCa microenvironment, potentially key mechanisms in inducing PCa progression through metabolic pathways. While direct evidence linking GDF15 to PCa initiation and development through lipid metabolism is lacking, the close association between GDF15 expression and lipid metabolism warrants further exploration. An important study highlighted the crucial role of GDF15 in regulating triglyceride metabolism,^[51] suggesting its essential impact in coordinating tolerance to inflammatory damage by modulating triglyceride metabolism,^[51] indicating its potential role in PCa initiation and progression. In addition, results from a large randomized clinical trial showed significant associations between changes in serum GDF15 in metformin recipients and weight loss.^[52,53] Subsequently, it was observed that antagonist of GDNF family receptor alpha like (GFRAL, a receptor of GDF15) reversed the weight-lowering effect of metformin in high-fat-fed and obese mouse models.^[53] Recombinant GDF15 increases the phosphorylation of extracellular signal-related kinase (ERK), RAC-alpha serine–threonine-protein kinase (AKT), and phosphoinositide phospholipase C-γ1 (PLCγ1).^[54]

In advanced PCa, radical surgery or radiotherapy is usually less favorable. Cisplatin, a first-line chemotherapy treatment for various cancers,^[55–57] effectively inhibits cancer cell growth and division by interfering with DNA synthesis. Compared with docetaxel, the first-line treatment, there is increasing evidence of the anti-tumor activities of cisplatin and its derivatives in aggressive variant PCa (AVPC) and PCa with neuroendocrine differentiation (NEPC).^[58] Although effective, cisplatin and its derivatives pose challenges due to platinum-related adverse reactions. In addition, cachexia accompanying advanced PCa imposes a significant burden on both treatment and survival, manifesting debilitating symptoms such as uncontrollable fatigue, weight loss, muscle atrophy, nausea, vomiting, impaired social function, and increased mortality.^[59] Recently, studies have suggested that targeting GDF15 may alleviate the side effects of platinum-based treatments and cachexia in PCa through metabolic pathways. Elevated GDF15 abundance is closely correlated with anorexia and weight loss in advanced PCa.^[60] Studies by Breen et al^[61] demonstrated a link between increased circulating GDF15, weight loss, and platinum-based chemotherapy. Monoclonal antibodies targeting GDF15 attenuate treatment-induced anorexia and weight loss, improving survival in both mice and nonhuman primate models.^[61] Treatment targeting GDF15 also improved nausea and vomiting.^[62] Both Johnen et al^[60] and Lerner et al^[63] identified GDF15 as a key driver of cancer cachexia in PCa xenograft mouse models. Central signaling regulation through GFRAL is pivotal in this process. Four laboratories identified GFRAL as a regulator in the metabolic response to stress by signaling through the coreceptor ret proto-oncogene (RET), and suggested GFRAL as a potential treatment for anorexia/cachexia syndrome.^[64] Mechanically, Suriben et al^[65] revealed mechanisms of GDF15–GFRAL in cancer cachexia. They found that activation of the GFRAL–RET pathway induced expressions of lipid-metabolic genes in adipose tissue, and GDF15 initiated a lipolytic response in adipose tissue through the peripheral sympathetic axis, ultimately reducing fatty and muscle tissue in tumor-beating mouse.^[65] They further developed a therapeutic monoclonal antibody, 3P10, which targets GFRAL and inhibits RET, reversing lipid oxidation and cancer cachexia even under calorie-restricted conditions.^[65] These results about the GDF15–GFREAL pathway represent a previously unexplored mechanism in mitigating the side effects of platinum-based therapy and PCa-associated cachexia [Figure 2]. Accordingly, combining cisplatin with a GDF15/GFRAL antagonist, or even using GDF15/GFRAL antagonist monotherapy, emerges as a promising therapeutic strategy to enhance treatment tolerability and improve quality of life for patients with advanced PCa, particularly AVPC and NEPC.

Figure 2.

GDF15/GFRAL pathway-related mechanism of cachexia development. Circulating GDF15 specifically binds to its receptor GFRAL, which is restricted to neurons of the area postrema and the nucleus tractus solitarii (NTS) in the brainstem. Upon ligand binding, GFRAL recruits the co-receptor RET, leading to receptor dimerization and activation of downstream signaling cascades, including ERK, AKT, and PLCγ1 phosphorylation. This central pathway orchestrates systemic metabolic responses, such as modulation of lipid metabolism, anorexia, and cachexia. In the context of prostate cancer, elevated GDF15 not only contributes to tumor progression through metabolic reprogramming but also exacerbates chemotherapy-associated cachexia. AP: Area pos­trema; GDF15: Growth differentiation factor 15; GFRAL: GDNF family receptor alpha like; NTS: Nucleus tractus solitarii; RET: Ret proto-oncogene.

GDF15, PCa chemoresistance, and metastasis

Castration resistance and chemoresistance present challenges in the prolonged treatment of PCa patients ineligible for radical therapy. Despite initial success with androgen deprivation therapy (ADT), a significant proportion of patients progresses to castration-resistant PCa (CRPC).^[66] Epidermal growth factor receptor (EGFR)-driven PCa progression and metastasis are well-established.^[67,68] In LNCaP cells during long-term ADT, GDF15 expression was initially increased but later lost its inhibitory effect on the EGFR signaling pathway.^[69] Wang et al^[69] suggested that this loss was due to N70 glycosylation. Such variable glycosylation may explain GDF15’s controversial role in PCa and present a potential therapeutic strategy for CRPC. Once CRPC occurs, docetaxel is a primary treatment option to improve symptoms and survival rates. However, it has been estimated that docetaxel chemotherapy is ineffective in nearly 50% of cases. The remaining patients do not show significant changes in prostate-specific antigen (PSA) levels and may be exposed to severe toxicity without direct benefits.^[70] Thus, GDF15 may influence PCa drug resistance. Data from androgen-independent PCa cell lines indicate that GDF15 is upregulated in the docetaxel-resistant PC3 cell line (PC3-Rx).^[71] Furthermore, recombinant GDF15 was shown to induce resistance in docetaxel-sensitive cells, while its knockout restored docetaxel sensitivity in resistant cells.^[71] Similar results were reported by Huang et al,^[72] who found that exposure to docetaxel increased GDF15 expression in PC3 and DU145 cell lines, both of which then exhibited resistance to docetaxel. Notably, GDF15 may inhibit the proliferation of DU145 cells but does not affect the cell-protective effects mediated by GDF15 against docetaxel/mitoxantrone.^[72,73] In the androgen-sensitive LNCaP cell line, GDF15 was expressed and supported cell growth.^[74] Similar phenomena are observed in PCa patients. Studies revealed significant upregulation of GDF15 in patients undergoing neoadjuvant docetaxel and mitoxantrone chemotherapy, possibly contributing to chemoresistance.^[72] Therefore, GDF15 inhibition might offer a potential therapeutic strategy for improving the current efficacy of docetaxel, as demonstrated in in vitro cell line experiments.^[75] A potential mechanistic explanation is the close relationship between GDF15 and the aging–cancer axis. GDF15 has been identified as a senescence-associated secretory phenotype (SASP) or candidate biomarker of cellular senescence.^[76–78] It is reported that GDF15 secreted by senescent cells contributes to cancer proliferation, progression, and invasion in gastric and colorectal cancer.^[79,80] The relationship between PCa progression and cellular senescence is well-documented, suggesting that GDF15’s role in PCa progression and chemoresistance may be similar.

GDF15 and PCa bone metastasis are closely associated. Distant metastasis, especially to the bone,^[81] is a critical event in advanced PCa, leading to severe physical complications such as pain, bone fractures, and hypercalcemia, and a significant decline in the 5-year survival rate.^[82] However, studies on PCa bone metastasis yield complex results: while pathological fractures, indicating osteolytic damage, are common in patients with bone-metastatic PCa, radiological findings show that osteoblastic changes predominantly occur at metastatic sites.^[83] A more convincing hypothesis is that interactions between PCa cells and osteoblasts, as well as osteoclasts, jointly promote PCa bone metastasis.^[84] As a member of the TGF-β superfamily, GDF15 is believed to exhibit characteristics similar to TGF-β, which has long been known to enhance PCa bone metastasis.^[85,86] Moreover, GDF15 plays a significant role in the bone metastasis of PCa itself. Siddiqui et al^[87] found that GDF15 enhanced the functionality of stromal cells (primarily osteoblasts), increasing C-C motif chemokine ligand 2 (CCL2) and receptor activator of nuclear factor kappa-B ligand (RANKL) secretion, recruiting osteoclasts, and activating osteoclastogenesis, which ultimately facilitated the growth of metastatic PCa within the bone microenvironment. In addition, metastatic PCa cells directly interact with osteocytes, stimulating the secretion of GDF15, thereby promoting the proliferation, migration, and invasion of PCa. This hypothesis has been validated by Wang et al^[88] in various PCa cell lines (including PC3, DU145, and LNCaP), with mechanistic insights revealing early growth response protein 1 (ERG1) as a downstream molecule regulated by GDF15. Further evidence demonstrating the role of GDF15 in PCa metastasis was reported by Senapati et al.^[89] Their studies in PC3 and LNCaP cell lines suggested that increased GDF15 expression enhanced the movement and metastasis capacities of PCa metastatic variants through direct control of actin structural rearrangements and cell motility.^[89]

GDF15 and PCa Diagnosis, Surveillance, and Prognosis

From a clinical perspective, GDF15 emerges as a promising biomarker for the diagnosis, surveillance, and prognosis of PCa. Given the protracted nature of PCa progression, a specific and reliable biomarker is essential for accurate diagnosis and prognosis across different disease stages.

GDF15 as a biomarker in PCa

Although PSA is the primary indicator in current clinical practice, its lack of tumor specificity can lead to false positives, as elevated PSA levels may also occur in benign prostatic hyperplasia and prostatitis. In this context, GDF15 serves as an alternate marker. Compared to normal prostate tissues, serum GDF15 levels are lower in benign disease and localized PCa,^[90] but higher in metastatic PCa.^[91] More importantly, serum GDF15 can be combined with PSA to improve diagnosis and enhance the utilization of information derived from PSA. The algorithm minimum inhibitory concentration-prostate-specific antigen (MIC-PSA) score, developed by combining GDF15 and PSA, can more specifically diagnose PCa and reduce the number of unnecessary biopsies by 27% compared with PSA alone.^[90] Another study suggested that combining GDF15 with the prostate health index (phi), serum fucosylated PSA (Fuc-PSA), and syndecan-1 (SDC1) achieved the best performance in predicting aggression in low-risk PCa (area under the curve [AUC]_{phi+Fuc-PSA+SDC1+GDF-15} = 0.942 vs. AUC_phi = 0.872).^[92] Multiparametric magnetic resonance imaging (MRI) compensates for the uncertainties in serological indicators and has become a powerful tool for PCa diagnosis and therapy. However, not all prostate tumors are visible on MRI.^[93] RNA-seq data indicated that GDF15 was a predictor for MRI visibility, and even for prognostic indicators including progression-free survival and metastatic deposits.^[94] Moreover, GDF15 could independently predict poorer cancer-specific survival rates, providing valuable prognostic insights.^[95] In patients diagnosed with localized PCa, GDF15 significantly improved the discriminatory power in distinguishing indolent from lethal PCa compared to established prognostic markers such as clinical staging, pathological grading, and PSA levels.^[95] Beyond serum, GDF15 in urine or seminal plasma holds the potential for PCa diagnosis and prognosis evaluation.^[36] Although no statistical significance in differential GDF15 levels was found in the urine of PCa patients compared with their normal counterparts, GDF15 levels in the urine of PCa patients were much higher than in others, and the enrichment of the C-terminal portion in PCa-associated crystalloids suggests diagnostic utility.^[96]

Therapeutic targeting of GDF15 in PCa

Excitingly, numerous monoclonal antibodies targeting GDF15 have entered clinical trials at various phases [Table 1]. Ponsegromab, a highly selective humanized monoclonal antibody, blocks the binding of GDF15 to GFRAL in the brain. Previously, ponsegromab was evaluated in two Phase I single-dose clinical trials. According to a study by Crawford et al^[97], the drug showed preliminary efficacy in phase I clinical trials (NCT04299048). Patients with cancer-related cachexia showed a decrease in serum GDF15 levels compared to healthy volunteers after receiving ponsegromab, and their weight also increased significantly. Furthermore, ponsegromab was well tolerated with no treatment-related adverse reactions reported.^[97] Recently, a phase 2 randomized, double-blind, placebo-controlled study on GDF15 is exploring its efficacy, safety, and tolerability in patients with cancer cachexia (NCT05546476). AV-380, the first highly efficient humanized inhibitory immunoglobulin gamma 1 (IgG1) antibody targeting GDF15 from AVEO (Delaware, USA) for the treatment of cancer cachexia, was found by Lerner et al^[98] to rehabilitate weight, muscle, fat, and normal organ sizes in HT-1080 human fibrosarcoma xenografts with increased plasma GDF15 levels and cachexia, reversing cachexia-related phenotypes and metabolic alterations. The drug was well tolerated and showed no dose-limiting toxicities. AV-380 has completed a phase 1 clinical trial in healthy volunteers (NCT04815551) and is recruiting patients with metastatic cancer for a phase 1 dose escalation study (NCT05865535). Although PCa patients were not included in the study, the findings still have relevant guiding significance. Notably, therapy involving an antibody targeting GFRAL (also known as 3P10) could reverse excessive lipid oxidation and prevent cancer cachexia in tumor-bearing mouse models, even under calorie-restricted conditions.^[65] Inhibition of the GDF15-GFRAL signaling pathway can enhance the immune system’s ability to kill solid tumors and also prevent or treat cancer cachexia. This drug performs particularly well in the treatment of advanced PCa, with proven safety and tolerability in both healthy volunteers and patients with advanced cancer (NCT03392116, NCT04068896).^[99] Another novel GFRAL antagonist, NGM120, developed by NGM Biopharmaceuticals (California, USA), is under evaluation in clinical trials. NGM120 showed encouraging signs of anti-cancer activity in a phase 1a trial of patients with advanced PCa.^[100] At 62 weeks, two of five PCa patients achieved disease control (32% and 99% reduction of PSA, respectively), including one patient who achieved a partial response (PR).^[100] Moreover, the PCa patient with PR exhibited up to a 90% reduction in serum GDF15 level and a 9.7% increase in body weight. Analysis of circulating T cell showed an increase in CD8 central memory T cells and proliferative CD8 cytotoxic T cells, as well as a decrease in mature regulatory T cells (Treg)/naïve Treg cells.^[100] Considering the immune role of GDF15 in PCa, visugromab, another GDF15 monoclonal antibody, would be an ideal drug. By neutralizing GDF15, it helps enhance the response to programmed cell death protein 1/programmed cell death protein ligand 1 (PD1/PDL1) immunotherapy by improving immune cell infiltration in the TME. In addition, visugromab is designed to improve the activation of T cells by dendritic cells (DCs) and promote the cytotoxic effect of T cells and natural killer (NK) cells, potentially addressing the issue of the “cold” immune microenvironment in PCa. Results from a phase 2 clinical trial of visugromab-nivolumab (a PD-1 inhibitor) combination therapy supported its promising clinical activity (NCT04725474). Bermejo et al^[101] reported significant clinical activity of the combination of visugromab and nivolumab among 79 patients with advanced/metastatic solid tumors across several expansion cohorts (including bladder cancer, non-small cell lung cancer, melanoma), manifesting persistent tumor regressions. Another clinical trial focuses on the potential of immunotherapy in combination with the visugromab (NCT06059547). AZD8853, also a GDF15 monoclonal antibody targeting solid tumors, showed significant anti-tumor activity in a mouse model with anti-PD-L1 refractory solid tumors, with complete tumor regression in 50% of the animals.^[102] After two doses of AZD8853, an increase in activated T cells and DC in the TME was observed. T cell depletion with anti-CD4 and anti-CD8 in mice ablated the antitumor activity observed with anti-GDF15 mAb treatment.^[102] The significant anti-tumor activity of AZD8853 in preclinical studies prompted a phase 1 clinical study to evaluate its potential in cancers refractory to immunotherapy (NCT05397171).^[42] These developments suggest directions for future clinical research and treatment of PCa, especially for those patients who are ineligible for receiving radical treatment.

Table 1.

Summary of clinical research related to GDF15 monoclonal antibody.

ID	Antibody	Study type	Duration	Aim	Phase	Intervention	Sample size	Main findings
NCT04299048	Ponsegromab	Open-label clinical trial	12 weeks	Evaluation of safety, tolerability, pharmacokinetics, and pharmacodynamics in patients with cancer and cachexia	Phase 1	Open-label 200 mg Ponsegromab, administered subcutaneously by site staff, every 3 weeks for 12 weeks (five doses total)	11	Good safety tolerance; the results showed that serum GDF-15 levels were inhibited, and there was preliminary evidence of improving weight, appetite, and physical activity
NCT05546476	Ponsegromab	RCT, double-blinded	12 weeks	Evaluation of safety, tolerability, pharmacokinetics, and pharmacodynamics following repeated subcutaneous administrations in patients with cancer and cachexia	Phase 2	Ponsegromab/placebo low dose subcutaneous injection every 4 weeks for 12 weeks	177	Main endpoints were achieved: the weight change in the treatment group (ponsegromab) was significantly improved compared with placebo at 12 weeks; the medium/high dose group improved more; the secondary endpoints such as appetite, quality of life, and physical activity were also improved. The side effects rate is similar to that of placebo
NCT04815551	AV-380	RCT, double-blinded	An average of 60 days	Evaluation of safety, tolerability, pharmacokinetics, pharmacodynamics, and immunogenicity in healthy subjects	Phase 1	IV infusion of AV-380 at dose level 4/8/13/20 mg/kg; Subcutaneous injection of AV-380 at dose level 4/2/1 mg/kg; Placebo Comparator	51	Mainly safety/pharmacokinetics/pharmacodynamics; no published results on cachexia or tumors
NCT05865535	AV-380	Non-randomized, open-label clinical trial	Up to 4 months per cohort	Evaluation of safety, pharmacokinetics, pharmacodynamics, and immunogenicity in patients with cancer and cachexia in metastatic cancer patients with cachexia	Phase 1	Dose escalation study: In each cohort (total 5 cohorts): IV infusion of AV-380. 7 doses will be given—the 2nd dose will be 28 days after the first dose, the remaining 5 doses will be given every 2 weeks	30 (ongoing)	Ongoing; abstracts have indicated that the study is evaluating safety, pharmacokinetics/pharmacodynamics/immunogenicity. Preliminary data show that AV-380 has the effect of reversing cachexia in animal models. Wait for more human data
NCT03392116	NGM120	RCT, double-blinded	28 (single dose)/84 (multiple dose) days	Evaluation of safety, tolerability, and pharmacokinetics in healthy subjects	Phase 1	Subcutaneous injection of single/multiple dose of NGM120 versus placebo	92	The primary focus was on safety, tolerability, and pharmacokinetics; no serious dose-limiting toxicities were observed; no large-scale results regarding tumor efficacy or cachexia improvement have been reported
NCT04068896	NGM120	RCT	12 weeks	Dose exploration of NGM120 in subjects with advanced solid tumors and pancreatic cancer using combination therapy	Phase 1/2	Subcutaneous injection of NGM120 from dose 1 to dose 6 or placebo	75	Preliminary findings: NGM120 demonstrated good tolerability as monotherapy or in combination therapy. Antitumor activity signals were observed in certain cases (e.g., significant reduction in CA19-9 markers in some patients). However, overall efficacy and improvement in cachexia remain inconclusive at this stage
NCT04725474	Visugromab	Non-randomized, open-label clinical trial	At least 2 months	Dose escalation of Visugromab as monotherapy and in combination with an approved CPI in patients with advanced solid tumors	Phase 1/2	Up to 5 dose levels with Visugromab administered as IV monotherapy and in combination with a CPI	274 (ongoing)	Key Findings: This combination induces deep and durable antitumor responses in these advanced solid tumors; increased T-cell infiltration, T-cell proliferation, and characteristic upregulation of interferon-γ were detected in the tumor microenvironment. Well tolerated
NCT06059547	Visugromab	Non-randomized, stratified, and single-blinded, clinical trial	At least 18 months	Evaluation of neoadjuvant immunotherapy in combination with Visugromab for the treatment of muscle invasive bladder cancer	Phase 2	Visugromab in combination with CIN (nivolumab) or placebo	30	No major therapeutic results have been fully disclosed. Mainly focused on endpoints such as pathological complete response rate, radioactive response rate, safety, event incidence rate, and overall survival rate
NCT05397171	AZD8853	Non-randomized, open-label clinical trial	Approximately 6 months	Evaluation of the safety, pharmacokinetics, pharmacodynamics, and preliminary efficacy of AZD8853 in participants with selected advanced/metastatic solid tumors	Phase 1/2	AZD8853 monotherapy dose escalation; AZD8853 monotherapy safety expansion at dose levels and indications determined to be safe before; AZD8853 monotherapy safety and preliminary efficacy expansion at dose levels and indications determined to be safe before	16	This study was terminated early, based on overall risk-benefit profile observed to date. No safety concerns reported

CA19-9: Carbohydrate antigen 19-9; CPI: Checkpoint inhibitor; GDF15: Growth differentiation factor 15; IV: Intravenous injection; RCT: Randomized controlled trial.

Conclusions

In conclusion, this review synthesizes the emerging insights into GDF15’s role in PCa, highlighting its critical involvement in metabolic reprogramming, chemoresistance, bone metastasis, and clinical applications. The established links between GDF15 and various aspects of PCa underline its potential as a key factor in disease progression and therapeutic responses. The clinical utility of GDF15 as a diagnostic and prognostic biomarker introduces new possibilities for personalized patient care. The development of monoclonal antibodies against GDF15 in clinical trials offers potential for novel therapeutic approaches. As research continues to unravel the complexities of GDF15 in PCa, it may become a significant target in PCa research. These insights facilitate a more profound understanding of the disease and foster the creation of targeted treatments to enhance patient outcomes.

Acknowledgments

We appreciated the BioRender (https://www.biorender.com/) for its assistance in drawing.

Funding

This study was supported by the grants from the National Natural Science Foundation of China (Nos. 82170785, 81974099, 81974098, and 82170784), National Key Research and Development Program of China (No. 2021YFC2009303), the program from Science and Technology Department of Sichuan Province (No. 2021YFH0172), Young Investigator Award of Sichuan University 2017 (No. 2017SCU04A17), Technology Innovation Research and Development Project of Chengdu Science and Technology Bureau (No. 2019-YF05-00296-SN), and Sichuan University—Panzhihua Science and Technology Cooperation Special Fund (No. 2020CDPZH-4).

Conflicts of interest

None.