Introduction
Prostate cancer (PCa) is the second most common malignant tumor in men worldwide, with 1,466,680 new cases diagnosed in 2022, accounting for ~397,000 fatalities (1). Although its mortality rate is comparatively low, it continues to be the primary cause of cancer-related death in numerous nations across the Caribbean, sub-Saharan Africa, Central and South America, as well as Sweden in Europe. Increasing incidence of PCa was noted in China, Baltic countries and Eastern Europe, which might be a result of widely adopted prostate-specific antigen (PSA) testing and more frequent transurethral resections (2-4).
Given the global trend of increasing life expectancy and the established correlation between PCa incidence and age, the number of men newly diagnosed with this type of cancer will likely rise in the foreseeable future (5). Prediction of patient prognosis following prostatectomy is currently based on classic clinicopathological parameters, however incorporating data related to specific tissue biomarkers might be a valuable addition in this context, especially for patients with aggressive disease prone to early recurrence.
Neuropilin-1 (NRP-1), a member of neuropilin family of receptor proteins, was initially identified as modulator of nervous system development, functioning as semaphorin coreceptor alongside plexin. Later, it was found to be receptor of vascular endothelial growth factor (VEGF) family (6). NRP-1, together with NRP-2, transfers signals for angiogenesis, cellular communication and migration (7). NRP-1has emerged as a focal point in tumor biology research due to its participation in vascular and lymphatic development, as well as its role in tumor cells (6,8,9).
Elevated expression of NRP-1 was demonstrated not only in PCa, but also in breast, colon, lung and bladder cancers as well as in neuroblastoma (10-14). NRP-1 is involved in tumor cell proliferation, migration, survival and therapy resistance (6,14,15).
As a transmembrane co-receptor NRP-1, alongside with VEGFR2 (VEGF tyrosine kinase receptor), activates VEGF-A-mediated vascular development (16,17). Targeted disruption of a single allele of VEGF gene in murine models is sufficient to induce vascular abnormalities and results in embryonic lethality, highlighting that the in vivo level of VEGF is crucial for proper vascular development (18,19). Furthermore, by interacting with VEGF165 NRP-1 activates c-MET receptor tyrosine kinase pathway, which leads to the subsequent activation of downstream effectors such as Src kinase and signal transducer and activator of transcription 3 (STAT3), culminating in the upregulation of anti-apoptotic proteins like myeloid cell leukemia-1 (Mcl-1). Mcl-1 is an established survival factor in PCa cells, which is upregulated in patients with PCa with bone metastasis (20). b1/b2 binding site of NRP-1 also interacts with extracellular domain of epidermal growth factor receptor (EGFR) leading to its phosphorylation and subsequent activation of the downstream PI3K/AKT/mTOR signaling pathway. This activation promotes proliferation and migration of PCa cells. Additionally, hypoxic conditions within the tumor microenvironment (TME) induce the accumulation of hypoxia-inducible factor 1-alpha (HIF1-α), which transcriptionally upregulates NRP-1 expression, further amplifying these oncogenic signaling cascades (21,22).
Proangiogenic function of NRP-1 is dependent on its b1/b2 binding site, whereas a1/a2 binding sites respond for semaphorin3A (SEMA3A) recognition (23). In addition to binding VEGF, NRP-1 is a co-receptor for SEMA3A, a protein involved in the regulation of sympathetic ganglion development and axon pathfinding (24). SEMA3A also acts as a tumor suppressor by promoting apoptosis, suppressing cell migration and angiogenesis, at least partly in NRP-1-dependent manner. Binding to NRP-1 initiates signaling cascades that inhibit endothelial cell lamellipodia formation, adhesion, migration and survival, thereby effectively suppressing angiogenesis. VEGF165 competes with SEMA3A for NRP-1 binding on endothelial cells and promotes the internalization of cell-surface NRP-1. Interestingly, SEMA3A requires a lower concentration to achieve similar effects, suggesting a higher affinity for NRP-1 (25,26). Notably, both VEGF165 and SEMA3A induce vascular permeability through interaction with their shared receptor (27). Hence, different NRP-1 functions in PCa may result from different downstream pathways activated by individual ligands (24,28-32).
Additionally, recent research indicates that NRP-1 expression is dynamically regulated during the development of castration-resistant PCa (CRPC), particularly in response to androgen-targeted therapies (ATTs) and suggests its upregulation in metastatic compared with localized PCa. PCa cells with NRP-1 knockdown exhibited a less invasive morphology, characterized by reduced vimentin and enhanced E-cadherin expression, which was associated with decreased metastatic dissemination in a zebrafish xenograft model. Yet, the exact signaling mechanism between the androgen receptor (AR) and NRP-1 remains unclear (33).
Previous studies on potential prognostic and predictive utility of NRP-1 in PCa are encouraging, however, immunohistochemistry (IHC)-based data are limited and inconclusive (20,21,32-34). Moreover, the significance of NRP-1 levels in lymph node metastases of PCa has not been investigated. The present study systematically analyzed clinicopathological correlations of NRP-1 immunoreactivity in primary PCa and matched lymph node metastases.
Discussion
Despite significant advancements in PCa diagnostics in recent years, it continues to pose a considerable challenge for urologists and uropathologists with regard to accurate diagnosis and effective treatment strategies. Considering the global increase in life expectancy and the well-documented link between PCa incidence and aging, it is projected that the incidence of PCa will increase in the near future (5). The current reliance on traditional clinicopathological parameters, such as PSA levels, GGG, and pT stage, underscores the need to identify novel, specific biomarkers that could improve the accuracy of survival prognosis and the assessment of cancer dissemination risk in patients following prostatectomy and to incorporate them into current urological guidelines for PCa management.
The present study provided new insights into the role of NRP-1in PCa progression, particularly in the context of lymph node metastases. None of previous studies on NRP-1 in PCa investigated its expression in lymph node metastases, highlighting the novelty of the present study (20,21,33,39-42).
Although the literature on the clinical implications of NRP-1 expression in PCa is limited, several studies reported a positive association between NRP-1 levels in PCa tissue and poor prognosis. In a study by Hu et al (42) NRP-1 expression correlated with OS; however, only in Caucasian-American patients and not in Chinese and African American men. In another study, IHC analysis of TMAs demonstrated a positive correlation between elevated NRP-1 expression and higher Gleason grade, pathological T stage, lymph node involvement and primary treatment failure. Moreover, multivariate analyses identified NRP-1 expression at the time of radical prostatectomy as an independent prognostic factor for biochemical recurrence following radiotherapy, as well as for metastatic progression and cancer-specific mortality. Interestingly, NRP-1 was identified as a gene repressed by androgen signaling, with its expression upregulated during the adaptive response to ATTs, ultimately contributing to the progression toward metastatic CRPC (33).
Notably, roles of NRP-1 observed in animal and in vitro models of PCa, may not reflect clinical conditions (20,21,32,34). Moreover, other studies reporting high biomarker expression in PCa also identified certain methodological inconsistencies related to selection bias and result assessment methods (20,21).
In one study, NRP-1 overexpression was positively associated with PCa progression and bone metastasis. However, a bias in patient selection was noted, as only two patients with a Gleason score of 6 were included in the analysis, compared with 43 cases with Gleason scores of 8-10(20).
Another study reported significantly higher NRP-1 levels in PCa specimens from patients with poor prognosis (higher Gleason score, high PSA, advanced T stage lymph node metastasis). Further, NRP-1 expression was associated with worse disease-free survival, shorter OS and cancer-specific survival (21). Nonetheless, IHC staining in that study was evaluated using the immunoreactivity score, which has certain limitations and lacks the broader a dynamic range provided by the H-score used in the analysis of the present study. Although the H-score is more time-consuming and requires a higher level of expertise from the pathologist, it offers a more detailed assessment of the material, accounting for intra-tumoral heterogeneity in antigen expression and potentially improving accuracy (43,44).
In addition, some studies failed to corroborate earlier findings or yielded unclear results (39-41). Talagas et al (39) did not find any association between NRP-1 and unfavorable clinicopathological parameters in patients with clinically localized PCa after radical prostatectomy. In turn, elevated endothelial expression of NRP-1 was an independent predictor for the absence of distant relapse in PCa.
Another study reported a 10-fold increase in NRP-1 mRNA levels in tumor specimens compared with benign tissue samples. Interestingly, a comparison between benign prostate hyperplasia (BPH) and malignant tissues revealed that NRP-1 IHC staining was elevated in PCa specimens; however, it was confined to the stroma and absent in malignant epithelial cells, which contradicts previous studies. In addition, the stroma exhibited positive staining in all samples, with no significant variation in intensity across different tumor specimens. Consequently, staining intensity could not be associated with Gleason score or recurrence risk (40).
Finally, Yacoub et al (41) hypothesized in their study, that the function of NRP-1 in PCa varies depending on the specific ligand to which it is attached (SEMA3A/VEGF). In contrast to VEGF, SEMA3A presented lower protein expression in hormone-refractory PCa (HRPC) compared with localized cancer. In localized cancer, SEMA3A was linked to lower pT stage and showed a close correlation with NRP-1 expression.
No significant relationship was observed between NRP-1 and VEGF in localized disease, and no significant correlation was found between NRP-1 and SEMA3A in HRPC. The expression of both SEMA3A and NRP-1 was linked to predictors of favorable prognosis, such as lower preoperative PSA levels, a low Gleason score for NRP-1, and a lower pT stage for SEMA3A, while VEGF expression correlated with elevated PSA levels.
Consequently, the function of NRP-1 in PCa may vary depending on VEGF/SEMA3A ratio. When SEMA3A is the primary ligand, as observed in clinically localized PCa, NRP-1 expression could be indicative of a favorable prognosis due to its anti-tumoral effects. Conversely, in cases of metastatic HRPC where VEGF is the predominant ligand, NRP-1 expression might be associated with a more aggressive disease profile due to its pro-tumoral nature (16,17).
The analysis of the present study revealed that patients with high NRP-1 expression in PCa samples exhibited a significantly lower percentage of affected lymph nodes compared with those with lower expression (9.5% vs. 15.0%, P=0.027). The percentage of metastatic lymph nodes, commonly referred to as lymph node density, is widely recognized as a reliable predictor of clinical recurrence. Its prognostic value has been demonstrated in multiple studies, emphasizing its importance in guiding postoperative management and risk stratification in patients with PCa (45,46).
Additionally, the group with higher NRP-1 staining exhibited a significantly lower postoperative PSA level (0.02 vs. 0.21 ng/ml, P=0.039), which is also considered a positive prognostic factor. Postoperative PSA levels are strongly correlated with an increased risk of biochemical recurrence and overall mortality in patients following prostatectomy (47-50). These observations could be explained by the hypothesis proposed by Yacoub et al (41), suggesting that the function of NRP-1may depend on the ligand to which it is bound (SEMA3A or VEGF). SEMA3A may mediate anti-angiogenic and pro-apoptotic effects through NRP-1, potentially counteracting the pro-tumoral influence of VEGF, as documented in several studies (26,30,31).
Notably, all patients in the present cohort study had lymph node involvement; however, their disease was clinically localized and had been treated with curative intent. Consequently, these cases represent hormone-sensitive PCa. These findings highlight the importance of further exploring NRP-1expression as a prognostic biomarker in this specific patient subgroup.
Despite a thorough outcome analysis, no statistically significant associations were identified between high NRP-1 expression and key clinicopathological parameters, including pT stage, Gleason score, surgical margin status, or extracapsular extension. Moreover, NRP-1 expression was not found to influence 5-year survival rates, a result consistent with earlier studies (39,41). These findings suggest that, while NRP-1 may play a role in specific TME interactions, its utility as a standalone prognostic marker for long-term outcomes in PCa remains limited.
While the present study highlighted the potential of NRP-1 as a biomarker, several important limitations must be acknowledged. Most notably, the retrospective design and single-center setting represent significant constraints that may limit the generalizability and external validity of the present study's findings. Although these factors are common in exploratory studies, the results should be interpreted with caution and the findings underscore the need for prospective validation, preferably focusing on hormone-sensitive patients with PCa and lymph node involvement. Secondly, the lack of data on specific VEGF/SEMA3A ratios in patient samples is another limitation-studies incorporating quantitative analysis of VEGF and SEMA3A expression will help clarify the ligand-dependent role of NRP-1 in tumor biology. The patient groups in the present study were relatively small-future research should focus on validating these findings in larger, multicenter cohorts and exploring the molecular mechanisms underlying NRP-1's role in PCa. Additionally, investigating its interaction with other pathways critical to PCa progression, such as AR signaling, could also yield valuable information. TMAs were selected for the present study to enable efficient, standardized immunohistochemical analysis across a large number of patient samples. TMAs allow simultaneous staining and evaluation of multiple cases under uniform conditions, minimizing inter-assay variability and conserving valuable tissue. This approach is particularly advantageous when assessing biomarker expression semi-quantitatively across matched tumor regions. However, it is important to note that TMAs represent only selected areas of the tumor and do not include its full architecture, such as surgical margins, which limits their use for margin assessment. Finally, a comparative analysis of NRP-1 staining between cancerous and non-cancerous tissues (such as BPH tissue) could further enhance the study's robustness. In conclusion the present study evaluated the role of NRP-1in PCa, particularly in lymph node metastases-a novel area of focus. Positive NRP-1 immunohistochemical staining was detected in both PCa and metastatic lymph nodes. Patients with high NRP-1 expression demonstrated significantly lower percentage of affected lymph nodes and lower postoperative PSA level. NRP-1 expression was not associated with other clinicopathological parameters or survival rates. These findings highlight a complex ligand-dependent role of NRP-1, where SEMA3A may mediate anti-tumoral effects in clinically localized PCa, counteracting VEGF's pro-tumoral influence in more advanced cancers. While the results of the present study suggested that NRP-1 holds promise as a prognostic biomarker in clinically localized, hormone-sensitive PCa, its clinical utility remains to be established. Additional studies are necessary to validate these observations in larger patient cohorts and to explore the molecular mechanisms that regulate NRP-1 function in PCa. In particular, assessing NRP-1 expression in the context of VEGF/SEMA3A balance and AR signaling pathways may provide critical insights for its future use in establishing prognosis and therapy.