The biological context of angiogenesis and prostate cancer (PCa) inspired a plethora of research, specifically in metastatic PCa and more specifically in castration-resistant disease (CRPC), the clinical stage in which the majority of clinical trials on angiogenesis inhibition was performed (1). Metastatic PCa is an androgen-driven and -dependent cancer (2), with androgen deprivation therapy (ADT) being the primary treatment. Despite high response rates – practically 90% of patients initially respond to hormone therapy – the vast majority will end up relapsing (3) in a predictable and irreversible manner. There has been a fair amount of research to try to analyze the mechanisms of progression to CRPC, which is the lethal phenotype of metastatic PCa – and current evidence suggest a function of clonal selection and adaptation by androgen receptor (AR)-dependent and independent mechanisms (4).

Indeed, ADT together with next generation hormonal agents such as Abiraterone (5) and Enzalutamide (6) still represent the foundation of systemic PCa treatment. Beyond hormone therapy, approved chemotherapy regimens mainly consist of Docetaxel and Cabazitaxel as microtubule inhibitors (7–9). Regarding bone as a favorite localization of PCa metastasis (10–12), therapeutic (combination) approaches include Radium-223 (13). In recent years, PCa treatment has rapidly developed towards precision oncology by addressing two novel target pathways: DNA repair and Prostate-specific membrane antigen (PSMA)-related signaling. Regarding DNA repair, cancers with mutations in BRCA1/2 (Breast Cancer Associated Genes 1 and 2) can be treated with PARP (Poly-ADP-Ribose-Polymerase) inhibitors originally established in Ovarian Cancer (14, 15). For PSMA, strategies include radioligand therapy as a theragnostic approach performed by nuclear medicine specialists (16).

Beyond these established and approved cancer therapies, this review aims to address an obvious treatment gap – given the crucial role of angiogenesis for PCa development and progression. Despite this fundamental promise reflected by in vitro and preclinical evidence, phase III trials with angiogenesis inhibitors failed to meet clinical endpoints.

About 50 years ago, Folkman and colleagues highlighted the importance of angiogenesis and neovascularization for tumor growth – reasoning that targeting tumor blood vessels might prove beneficial for patients with cancer (17). Meanwhile, state-of-the-art techniques highlighted the crucial but not completely understood link between angiogenesis (endothelial cells) and tumor immunity (18). For PCa, histopathology pinpoints high micro-vessel density and increased VEGF (Vascular Endothelial Growth Factor) expression compared to non-neoplastic conditions. Moreover, VEGF levels are associated with higher tumor stages as well as advanced grading and plasma VEGF is increased in metastatic PCa versus localized disease (19–21). Higher VEGF expression evaluated by immunohistochemistry has also been associated with reduced disease-specific survival in patients with PCa (22). In addition, levels of urinary VEGF were associated with worse survival (23) and elevated plasma VEGF/sVCAM-1, a vascular cell adhesion molecule, correlated with worse outcome (24).

In principle, many drugs and angiogenic target structures known from other solid and hematological malignancies are available for PCa (25–30). As a consequence, clinical trials combined antiangiogenic agents with Taxanes in mCRPC (31); however, not a single drug combined with Docetaxel showed a statistically significant success in terms of outcome (32). Therefore, clinicians started trials in less symptomatic patients, investigating compounds as single agents. Unfortunately, all of these phase III trials with thousands of patients were collectively negative for OS – despite promising biological preclinical as well as promising phase II trials. Despite efforts studying more than 1,000 patients, the combination of Bevacizumab or Aflibercept with chemotherapy showed no improvement compared to chemotherapy alone (33, 34). Sunitinib as a single agent compared to prednisone showed no improvement, either (35).

Making it even worse, Lenalidomide treatment resulted in a sobering scenario (36): While effective in several hematologic conditions (37–40), combination treatment of patients with PCa (Lenalidomide + Docetaxel + Prednisone) led to a significantly worse OS compared to treatment with Docetaxel and Prednisone (36). Another surprising and quite sobering example is Cabozantinib, an oral inhibitor of Tyrosine Kinases including MET and VEGFR2, two major drivers of malignant progression in several neoplasia (41–47), which did not guarantee an OS advantage in patients with PCa (48). Indeed, Cabozantinib showed anti-angiogenic and antitumor effects in a wide range of preclinical tumor models (49–51), also blocking progression of PCa xenografts in soft tissue and bone (52–54). Additionally, Cabozantinib affected key actors of the bone niche – with reduction in osteoclasts and biphasic effects osteoblasts, while altering bone remodeling with increased volume in mice (55). MET and VEGFR2 cooperate to promote tumor survival, thereby boosting angiogenesis via improved tumor blood flow and improved oxygenation. Moreover, MET promotes migration and invasion, also facilitating the escape from hypoxic areas. Consequently, bone metastases are associated with high levels of MET expression. In specific, MET expression increased with androgen deprivation in preclinical models and with progression and metastasis in bone and lymph nodes (56). Promising early phase II trial results from bone scans upon combined Docetaxel and Cabozantinib treatment showed activity in 300 patients (48, 57). Soft tissue effects were also present, with objective response and significant progression-free survival (PFS) benefit (48). Improvement in pain and reduction of narcotics corroborated these initial results (58). These data were paralleled by a reduction of circulating tumor cells (57), while keeping activity in subjects heavily pretreated with Docetaxel, Abiraterone and/or Enzalutamide (48, 57). The lowest effective dose of these studies was 40 mg/day (59). Nevertheless, within phase III trial, Cabozantinib did not perform better than Prednisone (60). The dose and the stage of disease could have been the cause for this failure.

To determine the status quo of clinical trials investigating anti-angiogenesis in PCa, we performed a database research on clinicaltrials.gov. As of September 2021, a total sum of 866 actively recruiting interventional trials were registered for patients suffering from PCa. As outlined in Table 1, only a minority of clinical trials investigated the effects of angiogenesis inhibitors/Tyrosine kinase inhibitors. Specifically, we identified 20 clinical trials addressing angiogenesis inhibition. While some studies aim to identify predictive biomarkers for future clinical stratification in entity-independent trials (NCT02465060, NCT03878524), others combine angiogenesis inhibition with immune checkpoint blockade – e. g. CONTACT-02 trial investigating Cabozantinib in combination with Atezolizumab in patients with mCRPC (NCT04446117). Of note, other studies include patients in different stages, such as metastatic castration sensitive disease (CABIOS phase I trial, NCT04477512) and even localized disease in a neoadjuvant setting before Radical Prostatectomy (SPARC phase II trial, NCT03964337).

Beyond this relatively small number of trials directly aiming at tumor vessels, we found several studies investigating compounds known to have additional anti-angiogenic effects (bottom part of Table 1). Curcumin, Green Tea Catechins and Metformin were among the substances identified. For Metformin, a tumor suppressive role was shown in several cancer entities (61). Moreover, adjuvant Metformin intake was associated with improved outcome in Clear Cell Renal Cell Carcinoma patients treated with Tyrosine Kinase inhibitors in two independent cohorts (62, 63). One reason for this protective effect could be the role of Metformin as a mitochondrial inhibitor. Interestingly, recent evidence implies a prominent role for mitochondrial signaling not only in Clear Cell Renal Cell Carcinoma (64), but also in high-grade PCa (65). Potentially, angiogenesis inhibition could be more effective in patients suffering from PC when combined with adjuvants such as Metformin.

From a histopathological and preclinical perspective, there is convincing evidence for a significant role of angiogenesis in PCa development and progression. For example, VEGFR2 was shown to mark PCa cases with a high risk of progression (30, 66). In addition, angiogenesis-related microRNAs such as let-7, miR-195 and miR-205 (67) are also deregulated and play prominent roles in PCa (68–70). However, no angiogenesis-specific inhibitor has met its clinical endpoint in phase III trials (see Figure 1A). Consequently, angiogenesis inhibitors currently do not play a role in PCa treatment guidelines. As shown by our database search on clinicaltrials.gov, several clinical trials are currently recruiting patients with PCa to address the discrepancy between promising preclinical findings and sobering clinical trial results.

Targeting angiogenesis with specific inhibitors unfortunately has failed to impact OS in patients with mCRPC despite promising early data – and despite convincing clinical activity in several other malignancies. This discrepancy highlights the importance of the microenvironment niche, as PCa is characterized by substantial inter- and intra-patient heterogeneity and adaptive biology. Therapeutic strategies to overcome this challenge include biomarker-guided screening for patient subgroups most likely to benefit from anti-angiogenesis. Moreover, several trials investigate combinatorial approaches. Beyond specific angiogenesis inhibitors, approved compounds such as antiandrogens, PARP inhibitors and PSMA-targeting approaches probably also have a substantial anti-angiogenic impact in PCa biology.

Conceptualization: AS and MK. Methodology: AS and MK. Writing – draft preparation: AS, CK, and MK. Writing – review and editing: AS, CK, and MK. All authors contributed to the article and approved the submitted version.

This project was supported in part by the Apulian Regional Project Medicina di Precisione to A.G.S. Moreover, M.K. was funded by a personal grant from Else-Kröner-Foundation (Else Kröner Integrative Clinician Scientist College for Translational Immunology, University Hospital Würzburg, Germany). This publication was supported by the Open Access Publication Fund of the University of Würzburg.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.