Identification of BIRC5 and HMMR as prognostic biomarkers for immune infiltration in prostate cancer

Introduction

Prostate cancer (PCa) stands as the second most prevalent malignancy among men worldwide, following lung cancer. It is also the most commonly diagnosed cancer in men in over half of the countries globally (112 out of 185). In 2022, approximately 1.5 million new cases and 397,000 deaths were attributed to PCa globally, highlighting the profound impact of this disease (1-3). Treatment strategies for localized PCa include active surveillance, radical prostatectomy, and radiation therapy. For local recurrences post-surgery, salvage radiation therapy and/or androgen deprivation therapy (ADT) are considered. Systemic recurrences are managed with ADT combined with chemotherapy or novel agents targeting the androgen receptor (AR). Treatment plans must be customized to the patient’s condition and disease progression (4,5). Emphasizing individualized treatment is vital as emerging therapies offer new hope for advanced PCa stages.

The prognosis of PCa patients is closely tied to the tumor’s stage at diagnosis. Since 2014, PCa incidence rates have increased by 3% annually. Men diagnosed at an early, localized stage have a ten-year survival rate of 99% (4,6). Conversely, those diagnosed at an advanced stage with metastasis face a five-year survival rate of only 30% (7). A 2018 study in the United States found that over half of men with metastatic PCa (mPCa) were initially diagnosed when the disease was localized or regional (8). Nearly all mPCa patients eventually progress to castration-resistant PCa (CRPC), which is resistant to ADT (9). Neuroendocrine PCa (NEPC), an aggressive variant typically emerging in advanced stages of metastatic CRPC (mCRPC), often lacks AR expression and is highly invasive, posing significant treatment challenges (10).

Research into gene mutations and expression patterns has greatly enhanced our understanding of the molecular mechanisms driving PCa initiation and progression (11,12). Previous studies have identified several prognostic genes and genes that affect treatment responses (13-15). In this study, we focused on employing bioinformatics methods to conduct an in-depth analysis of differentially expressed genes (DEGs) at various stages of PCa progression, and to explore the functional roles of these genes in the PCa microenvironment. Our approach combined the expression profiling and high-throughput sequencing datasets, identifying some DEGs that have not been extensively explored in the context of PCa, thereby providing new insights into the biology of PCa.

Among the DEGs we identified, baculoviral inhibitor of apoptosis protein repeat containing 5 (BIRC5) and hyaluronan-mediated motility receptor (HMMR) may play important roles in the proliferation and metastasis of PCa, closely related to prognosis (16,17). Previous studies have shown that tumor-infiltrating lymphocytes (TILs) are associated with prognosis and play a crucial role in mediating the response to chemotherapy and immunotherapy across various cancers, including PCa (18,19). However, the prognostic variability in PCa, due to its biological diversity across different stages and subtypes, remains not fully understood. Therefore, we then further explored the functional roles of these DEGs and their relationship with immune cell infiltration, which may be pivotal in determining the prognosis of PCa patients. We present this article in accordance with the REMARK reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-24-359/rc).

Methods

Data collection and analysis

PCa data were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/), including four datasets: GSE3325 (n=19), GSE46602 (n=50), GSE69223 (n=30), and GSE59984 (n=14). According to the information provided by the authors in the dataset, samples were filtered based on their clinical relevance and covered different stages and periods of PCa progression. These datasets comprised 35 normal prostate tissue samples, 64 PCa tissue samples (including 22 localized and 6 mPCa samples), 12 CSPC/CRPC samples, and 2 NEPC samples. Following a strict quality control process based on the information provided for the relevant datasets, we ensured the integrity of the samples in the data analysis, confirming there were no missing values, and removed some low-expressed duplicate genes to reduce noise. Statistical analysis and visualization were performed using R (version 4.3.2). The datasets GSE3325, GSE46602, and GSE69223 shared the same probe platform (GPL570) to ensure that the samples underwent the same technical standards and processes during sequencing and data processing, thereby reducing systemic bias caused by platform differences. Batch effects were removed using the ComBat function within the SVA package, and principal component analysis (PCA) was performed to ensure that batch effects were sufficiently minimized. DEGs were analyzed using the limma package, while PCA and heatmap visualizations were generated using the ggplot2 and ComplexHeatmap packages, respectively (20).

Gene set enrichment analysis

The clusterProfiler package (21) is widely used in the field of bioinformatics and provides various functions, including enrichment analysis of biological functions and result visualization. To investigate the roles of the selected DEGs in PCa, the clusterProfiler package was employed for Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, aiming to determine if these DEGs are enriched in specific biological functions. A significance threshold of adjusted P value <0.05 was set for statistical significance, and the Benjamini-Hochberg method in the p.adjust function was employed to control the false discovery rate (FDR).

Construction of protein-protein interaction (PPI) network

The online bioinformatics tool STRING (https://string-db.org/) was used to predict PPIs, including both direct (physical) and indirect (functional) associations. Using the STRING PPI database (version 12.0), we identified the DEGs involved in PPIs. A medium confidence interaction score of 0.400 was set as the cutoff value, which is considered effective in balancing the inclusion of relevant interactions and minimizing the risk of false negatives (22,23). At the same time, to avoid potentially increasing the risk of false positives, the PPI network was visualized using Cytoscape software (version 3.9.1). Modular analysis of the PPI network was conducted with the MCODE plugin to identify key modules, aiding in the identification of densely interconnected protein subnetworks. This analysis reveals potential biological modules or functional units, providing insights into the biological significance of complex networks.

Survival analysis to screen for hub genes

Kaplan-Meier (K-M) survival curves and log-rank tests were generated using the survival and survminer packages in R. These tools facilitated the visualization of biochemical recurrence-free survival (BCRFS) among PCa patients, correlating gene expression with survival outcomes based on BCRFS data from The Cancer Genome Atlas (TCGA, n=425), helping us analyze the impact of these genes on PCa prognosis. The Cox proportional-hazards model was applied to examine the influence of covariates on survival time, and calculating hazard ratios and confidence intervals. Through the survival analysis results, hazard ratios, and confidence intervals, we screened for hub genes that significantly affect prognosis. For additional validation and identification of hub genes, the GSE70769 (n=92) dataset was downloaded from the GEO database, which has complete clinical information and exhibits good data standardization and consistency.

Analysis of gene expression

To elucidate the potential role of hub genes, gene expression analysis was conducted using the TCGA dataset obtained from the UCSC Xena data center (https://tcga.xenahubs.net). This analysis compared expression profiles between pan-cancer samples and their corresponding normal samples from the TCGA and Genotype-Tissue Expression (GTEx) projects.

Integrated analysis of immune features

The TCGA database was utilized to investigate the correlation between BIRC5, HMMR, and immune infiltration in PCa. The single sample gene set enrichment analysis (ssGSEA) function, known for its robustness in quantifying the presence of immune cell subtypes, was employed. The abundance of 28 infiltrating immune cells (24) and immune-related functions in the PRAD expression profile was explored using the ssGSEA function in the R package “GSVA” (25). This analysis focused on the relationship between BIRC5, HMMR, and immune cell infiltration. Expression data from PCa samples obtained from the GEO database were further analyzed using the xCellpackage (26) and the quanTIseq algorithm (27) from the IOBR package to evaluate immune features across different subgroups.

Immunohistochemistry (IHC)

Twelve normal prostate tissue samples and twenty-four PCa tissue samples were sourced from the Department of Urology at The First Affiliated Hospital of Wannan Medical College Yijishan Hospital, Anhui Province. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and was approved by the Ethics Committee for Scientific Research and New Technologies of Wannan Medical College Yijishan Hospital (IRB Ref 2023/04), with informed consent obtained from all patients. Tissue sections, 4 µm thick, were sliced from paraffin-embedded blocks, deparaffinized in xylene, and rehydrated in ethanol. Subsequent incubation with ethylenediaminetetraacetic acid (EDTA) at boiling temperature for 10 minutes was followed by washing with phosphate-buffered saline (PBS) and a 10-minute incubation in a 3% hydrogen peroxide solution to inhibit endogenous peroxidase activity. Blocking of non-specific binding involved a 30-minute incubation with 5% body surface area (BSA) at 37 ℃. Primary antibodies for BIRC5 (Survivin, dilution 1:100, A00379, Boster, Wuhan, China) and HMMR (dilution 1:100, PA12592, Boster, Wuhan, China) were applied and incubated overnight at 4 ℃. Following three PBS washes, slides were treated with a peroxidase-conjugated secondary antibody (Boster, Wuhan, China). 3,3'-diaminobenzidine tetrahydrochloride (DAB) is a commonly used enzyme substrate that forms a brown precipitate under the catalysis of peroxidase, exhibiting high sensitivity and good stability. Therefore, the immunocomplexes were visualized using DAB, with subsequent counterstaining using hematoxylin. Average optical density (AOD) as a standardized quantitative method eliminates the influence of sample area size on the results, enhancing the reliability and reproducibility of the results. The SQS40R scanner (Shenzhen Shengqiang Technology Co., Ltd., Shenzhen, China) scanned the IHC slides, and the AOD of the IHC images was measured using Image J software (version 4.3.2), calculated as AOD = integrated optical density (IOD)/area. Statistical analyses utilized GraphPad Prism (version 9.5.1) software.

Statistical analysis

The Shapiro-Wilk test assessed variable normality, and the Bartlett test evaluated variance homogeneity. Statistical significance for comparisons between two groups was estimated using unpaired Student’s t-test for normally distributed variables and the Mann-Whitney U test for non-normally distributed variables. Multi-group comparisons employed the Kruskal-Wallis test and analysis of variance (ANOVA) as non-parametric and parametric methods, respectively. Correlation coefficients were determined through Spearman and distance correlation analyses. Survival curves for dataset subgroups were plotted using the K-M method, with hazard ratios for univariate analysis calculated using the Cox proportional hazards model. Beyond immunohistochemistry, all statistical analyses were conducted using R software (version 4.3.2) and RStudio software, both sourced from their official websites. Significance was defined as P<0.05 (two-tailed), with levels of significance denoted as *, P<0.05; **, P<0.01; ***, P<0.001; and ****, P<0.0001.

Results

Differential expression gene screening in PCa subgroups

In the datasets GSE3325 (n=19), GSE46602 (n=50), GSE69223 (n=30), and GSE59984 (n=14), DEGs were identified within predefined PCa subgroups. PCA examined the samples in each subgroup (Figure 1A-1C), revealing 677, 4,213, and 5,158 DEGs in the respective datasets, adhering to preset cutoff criteria (|log fold change (FC)| >1; adj.P.val <0.05). The heatmap illustrates the expression profiles of the top 40 genes across different groups (Figure 1D-1F). It can be observed that MTHFD2, FZD8, and STIL are upregulated in the tumor group; STIL, GPRC5B, and FZD8 are highly expressed in mPCa compared to localized PCa; and KRT23, ATRNL1, and STIL are highly expressed in NEPC. Additionally, the volcano plots depict all DEGs (Figure 1G-1I), showing that some genes, such as HOXC6, LMOD1, and NKX2-2, are significantly upregulated or downregulated in tumor tissues.

Figure 1 Identification of DEGs in distinct subtypes of PCa. (A-C) Sample clustering analysis of different subgroups performed using principal component analysis. (D-F) Heatmap showing the top 40 DEGs from intersections of the three subgroups. (G-I) Volcano plot displaying all DEGs from the three subgroups. The red dots represent genes with logFC >1.5 and P values <0.05; the blue dots represent genes with logFC <1.5 and P values <0.05; the black dots below the horizontal dashed line indicate genes with P values >0.05, while the portion above the horizontal line represents genes with absolute logFC <1.5 and P values <0.05. PC, principal component; PCa, prostate cancer; mPCa, metastatic prostate cancer; CSPC, castration-sensitive prostate cancer; CRPC, castration-resistant prostate cancer; NEPC, neuroendocrine prostate cancer; DEGs, differentially expressed genes; FC, fold change.

The intersection of DEGs from these datasets highlighted 159 shared DEGs (Figure 2A). KEGG pathway enrichment analysis of these 159 shared DEGs indicated enrichment in five pathways (Figure 2B). Table 1 lists these significantly enriched pathways. Among them, the PI3K-Akt pathway and extracellular matrix (ECM)-receptor interaction have been confirmed in previous studies to be closely related to the proliferation, migration, and invasion of PCa (28-30). Additionally, several key molecules in the PI3K/Akt pathway, such as PTEN, mTOR, and S6K1, are also considered potential therapeutic targets for PCa (31-33), highlighting the potential roles and biological processes in which these genes may participate. Visualization of the PPI network using Cytoscape software, with hub-key genes selected via the MCODE plugin (Score 9.111), identified 10 hub genes (Figure 2C).

Figure 2 Association between the expression of BIRC5 and HMMR genes and cancer prognosis. (A) Venn diagram revealing the identification of 159 DEGs at the intersection among three subgroups of PCa. (B) KEGG pathway enrichment analysis results for the 159 DEGs. (C) Cytoscape MCODE module used to screen 10 hub-key genes. (D,E) Kaplan-Meier curves of BCRFS for BIRC5 and HMMR in TCGA-PRAD. (F,G) Kaplan-Meier curves of BCRFS for BIRC5 and HMMR in GSE70769. (H,I) Univariate Cox regression analysis for differentially expressed hub-key genes in TCGA-PRAD and GSE70769. PCa, prostate cancer; mPCa, metastatic prostate cancer; CSPC, castration-sensitive prostate cancer; CRPC, castration-resistant prostate cancer; NEPC, neuroendocrine prostate cancer; ECM, extracellular matrix; BCRFS, biochemical recurrence-free survival; HR, hazard ratio; TCGA, The Cancer Genome Atlas; PRAD, prostate adenocarcinoma; DEGs, differentially expressed genes; KEGG, Kyoto Encyclopedia of Genes and Genomes.

To evaluate the prognostic significance of these hub genes, survival analysis utilizing TCGA-PRAD (n=425) data demonstrated that nine of these genes significantly impacted survival (Figure 2D,2E; Figure S1). This finding was supported by survival analysis with the GSE70769 (n=92) dataset, where only BIRC5 and HMMR showed a significant association with survival (Figure 2F,2G; Figure S2). Univariate Cox regression analysis further confirmed the significant relationship between BIRC5 and HMMR and BCRFS (Figure 2H,2I), underscoring these genes as potential prognostic biomarkers.

Analysis of BIRC5 and HMMR expression level

The expression levels of BIRC5 and HMMR in tumor tissues were further investigated using the UCSC XENA database, which includes the TCGA-Pan Cancer dataset encompassing 33 tumor types (ACC, BLCA, BRCA, CESC, CHOL, COAD, DLBC, ESCA, GBM, HNSC, KICH, KIRC, KIRP, LAML, LGG, LIHC, LUAD, LUSC, MESO, OV, PAAD, PCPG, PRAD, READ, SARC, SKCM, STAD, TGCT, THCA, THYM, UCEC, UCS, UVM) combined with TCGA normal samples and GTEx samples as controls. The analysis demonstrated significantly elevated expression levels of BIRC5 and HMMR in tumor tissues, including prostate, bladder, breast, lung, stomach, colon, thyroid, and other tissues (Figure 3A,3B).

Figure 3 Analysis of the expression levels of BIRC5 and HMMR. (A) BIRC5 expression in tumor and normal tissues from TCGA pan-cancer and GTEx data. (B) HMMR expression in tumor and normal tissues from TCGA pan-cancer and GTEx data. (C,D) BIRC5 and HMMR expression in mPCa, localized PCa, and benign tissues from GSE3325. (E,F) BIRC5 and HMMR expression in NEPC and CSPC/CRPC tissues from GSE59984. (G) BIRC5 expression in mCRPC and localized PCa tissues from GSE35988. (H) HMMR expression in mCRPC and localized PCa tissues from GSE32269. *, P<0.05; **, P<0.01; ***, P<0.001; ns, not significant. TPM, transcripts per million; PCa, prostate cancer; mPCa, metastatic prostate cancer; CSPC, castration-sensitive prostate cancer; CRPC, castration-resistant prostate cancer; NEPC, neuroendocrine prostate cancer; PC, principal component; mCRPC, metastatic castration-resistant prostate cancer; TCGA, The Cancer Genome Atlas; GTEx, Genotype-Tissue Expression.

This elevation in expression was validated in various PCa subgroups using datasets GSE3325 (n=19), GSE59984 (n=14), GSE35988 (n=76), and GSE32269 (n=51), showing higher expression levels compared to normal tissues (Figure 3C,3D). Moreover, higher expression levels were noted in NEPC compared to CSPC/CRPC subgroups (Figure 3E,3F), and in mCRPC compared to localized PCa (Figure 3G,3H). These results suggest the significant involvement of BIRC5 and HMMR in the development and progression of PCa, particularly in CRPC and NEPC, providing insights into the molecular mechanisms of PCa progression and identifying potential therapeutic targets.

Correlation and enrichment analysis of BIRC5 and HMMR with immune cells

Using TCGA data and the ssGSEA algorithm, the correlation between BIRC5 and HMMR expression and immune infiltration scores was illustrated (Figure 4A,4B). The analysis revealed a positive correlation between the expression of activated CD4 T cells, gamma delta T cells, memory B cells, Th2 cells, and Tregs with BIRC5 and HMMR. In contrast, a negative correlation was observed with Th17 cells, plasmacytoid dendritic cells (pDCs), natural killer (NK) cells, activated B cells, and CD56dim NK cells. Enrichment analysis further indicated that immune cells such as effector memory CD4 T cells, gamma delta T cells, and activated dendritic cells are more prevalent in scenarios of high expression of BIRC5 and HMMR (Figure 4C-4H). Conversely, central memory CD4 T cells, T follicular helper (Tfh) cells, effector memory CD8 T cells, immature dendritic cells, pDCs, and NK cells are more common in low expression scenarios of BIRC5 (Figure 4C,4E,4G), while effector memory CD8 T cells, pDCs, Macrophages, and CD56dim NK cells are associated with low expression of HMMR (Figure 4D,4F,4H). These findings suggest pivotal roles for BIRC5 and HMMR in modulating immune cell activity, infiltration, and immune response regulation.

Figure 4 Correlation and enrichment analyses of BIRC5 and HMMR. (A,B) Correlation between BIRC5, HMMR, and various immune cells; (C-H) association between the expression levels of BIRC5, HMMR, and the enrichment scores of various immune cells. *, P<0.05; **, P<0.01; ***, P<0.001.

The correlation between BIRC5, HMMR, and tumor-infiltrating immune cells

To further evaluate the correlation between BIRC5, HMMR, and the infiltration of different types of immune cells in PCa subgroups, the xCell and quanTIseq algorithms analyzed samples from the GSE21034 (n=131), GSE228283 (n=119), and GSE147876 (n=24) datasets (Figure 5A-5L). Significant correlations emerged in localized PCa and mCRPC, where BIRC5 expression showed a strong association with Th2 cells (P<0.001) and a positive association with CD8⁺ T cells (Figure 5A,5B,5G,5H). In NEPC, Th1 cells positively correlated with BIRC5 (P=0.004), whereas CD4⁺ T cells exhibited a negative correlation (P=0.001; Figure 5C,5I). HMMR expression in localized PCa demonstrated significant positive associations with Th2 cells (P<0.001; Figure 5D,5E), while in mCRPC it showed significant positive associations with CD8⁺ T cells (P=0.005; Figure 5K). In NEPC, HMMR correlated positively with Th1 cells and negatively with CD4⁺ T cells (P<0.001; Figure 5F,5L). QuanTIseq analysis (Figure S3) identified a positive correlation between BIRC5 and NK cells, dendritic cells, B cells, and Tregs across different subgroups (Figure S3A-S3C,S3G), while HMMR showed a negative correlation with NK cells in localized PCa and a positive correlation with dendritic cells, B cells, and Tregs in other subgroups (Figure S3D-S3F,S3H). The expression of BIRC5 and HMMR may facilitate the accumulation of immunosuppressive cells such as Tregs and myeloid-derived suppressor cells (MDSCs) in the tumor microenvironment (TME). These findings underscore the influence of BIRC5 and HMMR on immune cell infiltration across different PCa subgroups, highlighting their significant impact on PCa immune regulation.

Figure 5 Correlation analysis between BIRC5 and HMMR expression and T helper cell and CD4⁺/CD8⁺ T cell infiltration. (A-C) Correlation between BIRC5 expression in distinct PCa subgroups and Th cell infiltration levels using the xCELL algorithm. (D-F) Correlation between HMMR expression in distinct PCa subgroups and Th cell infiltration levels using the xCELL algorithm. (G-I) Correlation between BIRC5 expression in distinct PCa subgroups and CD4⁺/CD8⁺ T cell infiltration levels using the quanTIseq algorithm. (J-L) Correlation between HMMR expression in distinct PCa subgroups and CD4⁺/CD8⁺ T cell infiltration levels using the quanTIseq algorithm. PCa, prostate cancer; mCRPC, metastatic castration-resistant PCa; NEPC, neuroendocrine prostate cancer.

Validation of the expression of BIRC5 and HMMR in PCa

IHC staining was performed on paraffin-embedded samples to assess BIRC5 and HMMR protein expression in both normal and cancerous tissues. The cancer samples were further categorized into two groups based on Gleason scores: ≤7 and >7. PCa tissues exhibited higher staining intensity for BIRC5 and HMMR compared to normal tissues (Figure 6A-6F). Additionally, a significant correlation was observed between the upregulation of BIRC5 and HMMR and the pathological grade (Figure 6G,6H). Notably, tissues with a Gleason score >7 showed stronger BIRC5 and HMMR staining intensity compared to those with a Gleason score ≤7.

Figure 6 Protein expression of BIRC5 and HMMR in PCa. (A-F) IHC images demonstrating the staining intensity of BIRC5 and HMMR in normal prostate tissues, Gleason ≤7 group, and Gleason >7 group (400×). (G,H) The AOD of IHC images of BIRC5 and HMMR used for comparisons among normal prostate tissues, the Gleason ≤7 group, and the Gleason >7 group. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001; AOD, average optical density; IHC, immunohistochemistry.

Discussion

PCa exhibits heterogeneity, with variations in biological and clinical progression. Multi-omics approaches are essential to assess clinical attributes for risk assessment and differentiate between localized and metastatically aggressive PCa, advancing personalized treatment strategies, prognostic evaluations, and treatment monitoring. As PCa progresses from localized to metastatic disease, it becomes less sensitive to androgens, leading to CRPC within 12 to 18 months of treatment initiation. Some patients may develop NEPC after CRPC, characterized by a lack of AR expression and increased neuroendocrine markers (34,35). NEPC is aggressive and resistant, presenting significant treatment challenges, as conventional PCa treatments often fail. Tailored treatment strategies, including targeting neuroendocrine pathways and immunotherapy, are crucial for NEPC patients, but an incomplete understanding of its molecular mechanisms hinders effective targeted therapies.

The interplay between PCa and immune infiltration is crucial. The immune system combats cancer cells, but cancer can evade immune surveillance, promoting growth and metastasis. Chronic inflammation correlates with PCa progression (36,37), suggesting anti-inflammatory therapy and immune cell-targeted treatments could be effective. Immunotherapy progress in PCa is slow due to low tumor mutation burden (TMB) and limited immune cell infiltration. Despite immune cells within prostate tumors, limited T cell infiltration remains a gap in understanding. Apart from the Food and Drug Administration (FDA)-approved sipuleucel-T vaccine in 2010 (38), most immune therapies, including checkpoint inhibitors, show limited success in PCa (39,40).

BIRC5 and HMMR are highly expressed in various cancers. BIRC5 is prevalent in lung, pancreatic, breast, ovarian, brain, colorectal, and renal cancers (41-44), while HMMR is linked to gastric cancer, breast cancer, glioblastoma, bladder cancer, and leukemia progression (45-49). Limited studies suggest BIRC5 promotes PCa growth and survival (50), and HMMR, regulated by the AR-mTOR-SRF axis, promotes PCa proliferation and metastasis (51). This study used the TCGA database to validate BIRC5 and HMMR expression, revealing significant upregulation in mCRPC and NEPC compared to localized PCa. Immunohistochemistry confirmed protein expression.

BIRC5 and HMMR expression levels are associated with immune cell infiltration in PCa progression. CD4⁺ T cells differentiate into Th1, Th2, Th17, and Treg cells. Th1 cells secrete interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), and lymphotoxin, while Th2 cells release interleukin-4 (IL-4), interleukin-13 (IL-13), and interleukin-5 (IL-5) (52). Th1 and Th2 cytokines inhibit each other (53). Th2 cells and their cytokines impede anti-tumor responses, correlating with unfavorable outcomes (54-56).

A positive correlation between Th2 and HMMR in localized PCa and mCRPC suggests tumor adaptation to immune pressure. A negative correlation between Th2 and BIRC5 in localized PCa indicates that a robust Th1-type response may control early tumor growth. In mCRPC, the Th2-BIRC5 relationship shifts to positive, suggesting complex immune evasion mechanisms.

CD4⁺ and CD8⁺ T cells are crucial in modulating tumor immune responses (57,58). In NEPC, negative correlations between BIRC5 and HMMR with CD4⁺ T cells, and positive correlations with Th1 cells, suggest immunosuppression and immune imbalance. BIRC5 positively correlates with CD8⁺ T cells in localized PCa and mCRPC, indicating tumor adaptability to immune pressure. In localized PCa, HMMR negatively correlates with CD8⁺ T cells, suggesting immune evasion. In mCRPC, HMMR positively correlates with CD8⁺ T cells, indicating a complex immune microenvironment and aggressive tumor phenotype.

Different immune-related patterns showcase the dynamic tumor immune landscape, resulting from tumor cell adaptation, immune cell infiltration, and TME interactions. The diagnostic and prognostic significance of multiple immune cells in PCa needs further elucidation, emphasizing personalized, stage-specific immune therapeutic strategies.

BIRC5 is an important anti-apoptotic protein that belongs to the inhibitor of apoptosis protein (IAP) family, capable of promoting cell division and tumor progression, and its expression is regulated by the PI3K/AKT axis (50,59,60). HMMR, also known as CD168, is a cell surface receptor that participates in the mTOR signaling pathway and is significantly associated with cell cycle-related genes (such as AURKA, TPX2, and CDK1) (61,62). BIRC5 inhibits cell apoptosis, potentially influencing the immune cell composition in the TME, while HMMR may affect immune cell migration and localization, thereby modulating their anti-tumor effects. This study investigates BIRC5 and HMMR expression and roles in PCa, though further research is needed to elucidate their impact on cancer progression. In IHC experiments, the DAB and AOD quantification methods may be influenced by background staining and antibody specificity issues, which could lead to false-positive signals and biased results. Messenger RNA (mRNA) level analysis from the TCGA database may not fully represent actual expression levels. PCa exhibits high heterogeneity, and immune cell infiltration may have potential variability among different patients, which could affect the generalizability of the research results. Future studies should consider the potential impact of this variability on the expression of BIRC5 and HMMR and their roles in PCa. The sample size in some datasets is relatively small, especially in the NEPC group, which reduces the statistical power of the analysis and may affect the reliability of the results. Although this study includes a significant number of validation samples and comprehensive clinical data, providing a foundation for future research, subsequent studies should still increase the sample size to enhance the credibility and applicability of the results.

Future studies should consider using larger-scale multicenter samples to verify the correlations of BIRC5 and HMMR in PCa subgroups and their association with prognosis, thereby improving the accuracy and general applicability of the results. Cell or animal experiments should elucidate how BIRC5 and HMMR regulate Th cell balance and immune responses through specific pathways, guiding more effective treatment strategies. To evaluate the clinical translational potential of BIRC5 and HMMR as therapeutic targets for PCa, we plan to update existing clinical models to provide better prognostic guidance. At the same time, we will explore the correlation between gene expression and protein levels, utilizing techniques such as mass spectrometry and enzyme-linked immunosorbent assay (ELISA) to assess the feasibility of developing specific tests targeting BIRC5 and HMMR in the future.

Conclusions

High levels of BIRC5 and HMMR in PCa are associated with poor patient outcomes, likely due to their role in modulating immune responses and affecting Th1 and Th2 cell functions. The interaction between these genes and various T cell subsets highlights the complexity of the immune environment in PCa progression. These findings position BIRC5 and HMMR as promising targets for developing personalized immune-based therapies for PCa. Future research should focus on elucidating the specific mechanisms by which these genes influence immune dynamics and T-cell behavior in PCa.

Acknowledgments

We are grateful to GEO and TCGA databases for providing us with data.

Funding: This study was supported by Special Fund for Clinical Medical Research Transformation of Anhui Province (202204295107020011), and Talent Introduction Special Fund of the First Affiliated Hospital of Wannan Medical College (Yijishan Hospital of Wannan Medical College) (KY28880624).

Footnote

Reporting Checklist: The authors have completed the REMARK reporting checklist. Available at https://tau.amegroups.com/article/view/10.21037/tau-24-359/rc

Data Sharing Statement: Available at https://tau.amegroups.com/article/view/10.21037/tau-24-359/dss

Peer Review File: Available at https://tau.amegroups.com/article/view/10.21037/tau-24-359/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tau.amegroups.com/article/view/10.21037/tau-24-359/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee for Scientific Research and New Technologies of Wannan Medical College Yijishan Hospital (IRB Ref 2023/04) and informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015;136:E359-86. [Crossref] [PubMed]
2. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2024;74:229-63. [Crossref] [PubMed]
3. Li Q, Xia C, Li H, et al. Disparities in 36 cancers across 185 countries: secondary analysis of global cancer statistics. Front Med 2024;18:911-20. [Crossref] [PubMed]
4. Rebello RJ, Oing C, Knudsen KE, et al. Prostate cancer. Nat Rev Dis Primers 2021;7:9. [Crossref] [PubMed]
5. Teo MY, Rathkopf DE, Kantoff P. Treatment of Advanced Prostate Cancer. Annu Rev Med 2019;70:479-99. [Crossref] [PubMed]
6. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
7. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2018. CA Cancer J Clin 2018;68:7-30. [Crossref] [PubMed]
8. Devasia TP, Mariotto AB, Nyame YA, et al. Estimating the Number of Men Living with Metastatic Prostate Cancer in the United States. Cancer Epidemiol Biomarkers Prev 2023;32:659-65. [Crossref] [PubMed]
9. Berish RB, Ali AN, Telmer PG, et al. Translational models of prostate cancer bone metastasis. Nat Rev Urol 2018;15:403-21. [Crossref] [PubMed]
10. Lee CF, Chen YA, Hernandez E, et al. The central role of Sphingosine kinase 1 in the development of neuroendocrine prostate cancer (NEPC): A new targeted therapy of NEPC. Clin Transl Med 2022;12:e695. [Crossref] [PubMed]
11. Pritchard CC, Mateo J, Walsh MF, et al. Inherited DNA-Repair Gene Mutations in Men with Metastatic Prostate Cancer. N Engl J Med 2016;375:443-53. [Crossref] [PubMed]
12. Januskevicius T, Vaicekauskaite I, Sabaliauskaite R, et al. Germline DNA Damage Response Gene Mutations in Localized Prostate Cancer. Medicina (Kaunas) 2023;60:73. [Crossref] [PubMed]
13. Lopez R, Ferrari AC, Goel S, et al. Genomic profiling of prostate cancer at a diverse academic center. J Clin Oncol 2020;38:64.
14. Stopsack KH, Su XA, Vaselkiv JB, et al. Transcriptomes of Prostate Cancer with TMPRSS2:ERG and Other ETS Fusions. Mol Cancer Res 2023;21:14-23. [Crossref] [PubMed]
15. San Martin R, Das P, Dos Reis Marques R, et al. Chromosome compartmentalization alterations in prostate cancer cell lines model disease progression. J Cell Biol 2022;221:e202104108. [Crossref] [PubMed]
16. Hennigs JK, Minner S, Tennstedt P, et al. Subcellular Compartmentalization of Survivin is Associated with Biological Aggressiveness and Prognosis in Prostate Cancer. Sci Rep 2020;10:3250. [Crossref] [PubMed]
17. Guo K, Liu C, Shi J, et al. HMMR promotes prostate cancer proliferation and metastasis via AURKA/mTORC2/E2F1 positive feedback loop. Cell Death Discov 2023;9:48. [Crossref] [PubMed]
18. Lu X, Horner JW, Paul E, et al. Effective combinatorial immunotherapy for castration-resistant prostate cancer. Nature 2017;543:728-32. [Crossref] [PubMed]
19. Xu Y, Song G, Xie S, et al. The roles of PD-1/PD-L1 in the prognosis and immunotherapy of prostate cancer. Mol Ther 2021;29:1958-69. [Crossref] [PubMed]
20. Gu Z, Eils R, Schlesner M. Complex heatmaps reveal patterns and correlations in multidimensional genomic data. Bioinformatics 2016;32:2847-9. [Crossref] [PubMed]
21. Wu T, Hu E, Xu S, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb) 2021;2:100141. [Crossref] [PubMed]
22. Zhao Z, Yang H, Ji G, et al. Identification of hub genes for early detection of bone metastasis in breast cancer. Front Endocrinol (Lausanne) 2022;13:1018639.
23. Deng Y, Huang H, Shi J, et al. Identification of Candidate Genes in Breast Cancer Induced by Estrogen Plus Progestogens Using Bioinformatic Analysis. Int J Mol Sci 2022;23:11892. [Crossref] [PubMed]
24. Barbie DA, Tamayo P, Boehm JS, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature 2009;462:108-12. [Crossref] [PubMed]
25. Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics 2013;14:7. [Crossref] [PubMed]
26. Aran D, Hu Z, Butte AJ. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol 2017;18:220. [Crossref] [PubMed]
27. Finotello F, Mayer C, Plattner C, et al. Molecular and pharmacological modulators of the tumor immune contexture revealed by deconvolution of RNA-seq data. Genome Med 2019;11:34. [Crossref] [PubMed]
28. Luo GC, Chen L, Fang J, et al. Hsa_circ_0030586 promotes epithelial-mesenchymal transition in prostate cancer via PI3K-AKT signaling. Bioengineered 2021;12:11089-107. [Crossref] [PubMed]
29. Zhao Y, Hu X, Yu H, et al. Alternations of gene expression in PI3K and AR pathways and DNA methylation features contribute to metastasis of prostate cancer. Cell Mol Life Sci 2022;79:436. [Crossref] [PubMed]
30. Fernandes S, Oliver-De La Cruz J, Morazzo S, et al. TGF-β induces matrisome pathological alterations and EMT in patient-derived prostate cancer tumoroids. Matrix Biol 2024;125:12-30. [Crossref] [PubMed]
31. Deng R, Guo Y, Li L, et al. BAP1 suppresses prostate cancer progression by deubiquitinating and stabilizing PTEN. Mol Oncol 2021;15:279-98. [Crossref] [PubMed]
32. Zhao G, Forn-Cuní G, Scheers M, et al. Simultaneous targeting of AMPK and mTOR is a novel therapeutic strategy against prostate cancer. Cancer Lett 2024;587:216657. [Crossref] [PubMed]
33. El Gaafary M, Morad SAF, Schmiech M, et al. Arglabin, an EGFR receptor tyrosine kinase inhibitor, suppresses proliferation and induces apoptosis in prostate cancer cells. Biomed Pharmacother 2022;156:113873. [Crossref] [PubMed]
34. Barnett E, Carbone E, Keegan NM, et al. Genomic alterations and evolution in patients with prostate cancer with histologic evidence of neuroendocrine differentiation. J Clin Oncol 2022;40:5029.
35. Liu S, Alabi BR, Yin Q, et al. Molecular mechanisms underlying the development of neuroendocrine prostate cancer. Semin Cancer Biol 2022;86:57-68. [Crossref] [PubMed]
36. Oseni SO, Naar C, Pavlović M, et al. The Molecular Basis and Clinical Consequences of Chronic Inflammation in Prostatic Diseases: Prostatitis, Benign Prostatic Hyperplasia, and Prostate Cancer. Cancers (Basel) 2023;15:3110. [Crossref] [PubMed]
37. Kustrimovic N, Bombelli R, Baci D, et al. Microbiome and Prostate Cancer: A Novel Target for Prevention and Treatment. Int J Mol Sci 2023;24:1511. [Crossref] [PubMed]
38. Kantoff PW, Higano CS, Shore ND, et al. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 2010;363:411-22. [Crossref] [PubMed]
39. Meng L, Yang Y, Mortazavi A, et al. Emerging Immunotherapy Approaches for Treating Prostate Cancer. Int J Mol Sci 2023;24:14347. [Crossref] [PubMed]
40. Xu P, Wasielewski LJ, Yang JC, et al. The Immunotherapy and Immunosuppressive Signaling in Therapy-Resistant Prostate Cancer. Biomedicines 2022;10:1778. [Crossref] [PubMed]
41. Martínez-García D, Manero-Rupérez N, Quesada R, et al. Therapeutic strategies involving survivin inhibition in cancer. Med Res Rev 2019;39:887-909. [Crossref] [PubMed]
42. Wang F, Bao MC, Xu J, et al. Scutellarin inhibits the glioma cell proliferation by downregulating BIRC5 to promote cell apoptosis. J Cell Mol Med 2023;27:1975-87. [Crossref] [PubMed]
43. Wheatley SP, Altieri DC. Survivin at a glance. J Cell Sci 2019;132:jcs223826. [Crossref] [PubMed]
44. Yuan C, Su Z, Liao S, et al. miR-198 inhibits the progression of renal cell carcinoma by targeting BIRC5. Cancer Cell Int 2021;21:390. [Crossref] [PubMed]
45. Kang HG, Kim WJ, Kang HG, et al. Galectin-3 Interacts with C/EBPβ and Upregulates Hyaluronan-Mediated Motility Receptor Expression in Gastric Cancer. Mol Cancer Res 2020;18:403-13. [Crossref] [PubMed]
46. Parnigoni A, Moretto P, Viola M, et al. Effects of Hyaluronan on Breast Cancer Aggressiveness. Cancers (Basel) 2023;15:3813. [Crossref] [PubMed]
47. Li J, Zhou Y, Wang H, et al. COX-2/sEH dual inhibitor PTUPB suppresses glioblastoma growth by targeting epidermal growth factor receptor and hyaluronan mediated motility receptor. Oncotarget 2017;8:87353-63. [Crossref] [PubMed]
48. Yang D, Ma Y, Zhao P, et al. HMMR is a downstream target of FOXM1 in enhancing proliferation and partial epithelial-to-mesenchymal transition of bladder cancer cells. Exp Cell Res 2021;408:112860. [Crossref] [PubMed]
49. Spranger S, Jeremias I, Wilde S, et al. TCR-transgenic lymphocytes specific for HMMR/Rhamm limit tumor outgrowth in vivo. Blood 2012;119:3440-9. [Crossref] [PubMed]
50. Frazzi R. BIRC3 and BIRC5: multi-faceted inhibitors in cancer. Cell Biosci 2021;11:8. [Crossref] [PubMed]
51. Sun Y, Li Z, Song K. AR-mTOR-SRF Axis Regulates HMMR Expression in Human Prostate Cancer Cells. Biomol Ther (Seoul) 2021;29:667-77. [Crossref] [PubMed]
52. Saravia J, Chapman NM, Chi H. Helper T cell differentiation. Cell Mol Immunol 2019;16:634-43. [Crossref] [PubMed]
53. Shang Q, Yu X, Sun Q, et al. Polysaccharides regulate Th1/Th2 balance: A new strategy for tumor immunotherapy. Biomed Pharmacother 2024;170:115976. [Crossref] [PubMed]
54. Spinner CA, Lamsoul I, Métais A, et al. The E3 Ubiquitin Ligase Asb2α in T Helper 2 Cells Negatively Regulates Antitumor Immunity in Colorectal Cancer. Cancer Immunol Res 2019;7:1332-44. [Crossref] [PubMed]
55. Li Z, Wu Z, You X, et al. Pan-cancer analysis reveals that TK1 promotes tumor progression by mediating cell proliferation and Th2 cell polarization. Cancer Cell Int 2024;24:329. [Crossref] [PubMed]
56. Piro G, Simionato F, Carbone C, et al. A circulating T(H)2 cytokines profile predicts survival in patients with resectable pancreatic adenocarcinoma. Oncoimmunology 2017;6:e1322242. [Crossref] [PubMed]
57. Wang C, Zhang Y, Gao WQ. The evolving role of immune cells in prostate cancer. Cancer Lett 2022;525:9-21. [Crossref] [PubMed]
58. Apusiga K. Immune cell infiltration-based prognosis in prostate cancer: a review of current knowledge. Bull Natl Res Cent 2023;47:131.
59. Liang J, Zhao W, Tong P, et al. Comprehensive molecular characterization of inhibitors of apoptosis proteins (IAPs) for therapeutic targeting in cancer. BMC Med Genomics 2020;13:7. [Crossref] [PubMed]
60. Siragusa G, Tomasello L, Giordano C, et al. Survivin (BIRC5): Implications in cancer therapy. Life Sci 2024;350:122788. [Crossref] [PubMed]
61. Yang C, Li C, Zhang P, et al. Redox Responsive Hyaluronic Acid Nanogels for Treating RHAMM (CD168) Over-expressive Cancer, both Primary and Metastatic Tumors. Theranostics 2017;7:1719-34. [Crossref] [PubMed]
62. Shabir A, Qayoom H, Haq BU, et al. Exploring HMMR as a therapeutic frontier in breast cancer treatment, its interaction with various cell cycle genes, and targeting its overexpression through specific inhibitors. Front Pharmacol 2024;15:1361424. [Crossref] [PubMed]