Sage Journals: Discover world-class research

Abstract

To explore the signature function of the tumor mutational burden (TMB) and potential biomarkers in prostate cancer (PCa), transcriptome profiles, somatic mutation data, and clinicopathologic feature information were downloaded from The Cancer Genome Atlas (TCGA) database. R software package was used to generate a waterfall plot to summarize the specific mutation information and calculate the TMB value of PCa. Least absolute shrinkage and selection operator (LASSO) Cox regression analysis was used to select the hub genes related to the TMB from the ImmPort network to build a risk score (RS) model to evaluate prognostic values and plot Kaplan–Meier (K-M) curves to predict PCa patients survival. The results showed that PCa patients with a high TMB exhibited higher infiltration of CD8+ T cells and CD4+ T cells and better overall survival (OS) than those with a low TMB. The anti-Mullerian hormone (AMH), baculoviral IAP repeat-containing 5 (BIRC5), and opoid receptor kappa 1 (OPRK1) genes were three hub genes and their copy number variation (CNV) was relatively likely to affect the infiltration of immune cells. Moreover, PCa patients with low AMH or BIRC5 expression had a longer survival time and lower cancer recurrence, while elevated AMH or BIRC5 expression favored PCa progression. In contrast, PCa patients with low OPRK1 expression had poorer OS in the early stage of PCa and a higher recurrent rate than those with high expression. Taken together, these results suggest that the TMB may be a promising prognostic biomarker for PCa and that AMH, OPRK1, and BIRC5 are hub genes affecting the TMB; AMH, OPRK1, and BIRC5 could serve as potential immunotherapeutic targets for PCa treatment.

Keywords

tumor mutational burden immune cell infiltration prostate cancer hub gene biomarker

Abbreviations

AMH, anti-Mullerian hormone; AMHRII, AMH type II receptors; BIRC5, baculoviral IAP repeat-containing 5; BPH, benign prostatic hyperplasia; CNV, copy number variation; DEGs, differentially expressed genes; DFS, disease-free survival; GEPIA, Gene Expression Profiling Interactive Analysis; GS, Gleason score; ICGC, International Cancer Genome Consortium; K-M, Kaplan–Meier; LASSO, least absolute shrinkage and selection operator; OPRK1, opioid receptor kappa 1; OS, overall survival; PCa, prostate cancer; PSA, prostate-specific antigen; rAMH, recombinant AMH; ROC, receiver operating characteristic; RS, risk score; SNP, single-nucleotide polymorphism; TCGA, The Cancer Genome Atlas; TGF-β, transforming growth factor beta; TIICs, tumor-infiltrating immune cells; TIM, tumor immune microenvironment; TMB, tumor mutational burden; TIMER, tumor Immune Estimation Resource.

Introduction

Prostate cancer (PCa) is the second most common malignant tumor in males and the fifth leading cause of cancer-related death worldwide,¹ although in recent years, some research progresses have been made in its pathogenesis and treatment.^2–6 The latest report on cancer statistics estimated 174 650 new cases of PCa and 31 620 PCa-related deaths in the United States in 2019.⁷ Currently, prostate-specific antigen (PSA) is the only commonly used circulating biomarker for the early diagnosis of PCa in the clinic.⁸ However, benign prostatic hyperplasia (BPH) and prostatitis can also increase the serum PSA level, which creates certain limits for PSA screening.⁹ Furthermore, the most common treatment for patients under 75 years old is radical prostatectomy and neoadjuvant endocrine therapy (castration or androgen blockade),¹⁰ but up to 50% of patients suffer from biochemical recurrence after prostate removal.^11,12 Therefore, it is of great significance to explore novel biomarkers and new therapeutic strategies for PCa treatment.

The immune system plays an essential role in cancer recognition and control. On the one hand, the immune system can protect the host from virus-induced tumors by clearing or suppressing viral infections. On the other hand, the immune system can specifically recognize tumor-specific antigens and destroy tumor cells before these cells cause damage, a process known as tumor-specific surveillance.¹³ Thus, discovering the molecular determinants of immunotherapeutic responsiveness has been the focus of many efforts.¹⁴ Currently, the neoantigen load and tumor-infiltrating lymphocytes are considered acknowledged molecular determinants.^15,16 In addition, combinations of different immunotherapies have been explored to increase the efficiency of tumor immunotherapy.¹⁷ The tumor mutational burden (TMB) has been considered a good predictor of immunological response and tumor behavior.¹⁸ In recent studies, highly mutated tumors were shown to produce many novel peptides and thus more neoantigens that lead to increased tumor immunogenicity, rendering tumors more susceptible immune cell targets and resulting in an improved clinical response to immunotherapy.^19,20 For example, patients with high-TMB ovarian cancer, skin cutaneous melanoma, or bladder urothelial carcinoma show a prolonged survival time, supporting the conclusion that a high-TMB may be a harbinger of good clinical outcomes.²¹ Therefore, the TMB plays a vital role in immunotherapy and clinical prognosis in cancers. However, the function of the TMB in PCa patients and the identification of hub genes that affect the TMB are remain elusive.

In recent years, bioinformatics has provided a vital and effective tool for discovery in tumor-related research. Abundant somatic mutation data for tumor samples have been identified using whole-exome sequencing techniques.²² Here, to investigate the correlations of the TMB with clinical prognosis and antitumor immunity in PCa, we compared and analyzed the gene sets of high-TMB and low-TMB groups of PCa patients based on The Cancer Genome Atlas (TCGA).

Materials and Methods

Data Collection

In this study, we downloaded somatic mutation data, a gene expression matrix, and clinical information for PCa samples from the TCGA. For the somatic mutation cohort, we first used the R GenVisR package to generate a waterfall plot to summarize the mutational data of the top 10 genes in the samples. We further used the “maftools” in R package to analyze the general characteristics of the somatic mutation data and comutation of the top 25 mutated genes. We subsequently calculated the TMB value (the total No. of mutations divided by the size of the target coding area) of each specimen. Finally, the mRNA expression matrix was obtained from the TCGA for later analysis.

Identification of Differentially Expressed Genes of the Tumor Mutational Burden Subgroups

We first divided the samples into a high-TMB group and a low-TMB group based on the median value of the TMB. Then, the “limma” R package was utilized to analyze differentially expressed genes (DEGs) between the two TMB groups. Specifically, the adjusted P value and |log₂fold change| (|log FC|) were used to screen significant DEGs. We defined an adjusted P value < 0.05 and a |logFC| > 1 as the cutoff criteria.

Prediction of Immune Infiltration in Prostate Cancer Samples

CIBERSORT is an analytical tool that can provide precise values for different types of tumor-infiltrating immune cells (TIICs), which is based on a “gene signature matrix” of 547 genes.²³ In this study, we used DEGs expression to predict the content of 22 subtypes of immune cells per sample via the reference matrix provided by CIBERSORT. CIBERSORT was also used to calculate a reference P value to judge the accuracy of the TIIC data for each sample. A P value ≥ 0.05 was considered to indicate relatively low accuracy for prediction and filtered out of the matrix. On the contrary, a P value < 0.05 was viewed as accurate for prediction and retained.

Relationships between the Tumor Mutational Burden and Immune Cells

For the 22 subtypes of TIICs, we used violin plots and the “limma” R package to evaluate the difference in each immune cell type between the high-TMB group and the low-TMB group and created a violin plot to visualize the distributions of TIIC subtypes in the two groups. A P value < 0.05 was considered statistically significant.

Prognostic Analysis of the Tumor Mutational Burden

The International Cancer Genome Consortium (ICGC) database (https://icgc.org/) is a free public database. Data for 188 specimens (132 live patients and 56 dead patients) were downloaded from the ICGC. After calculating the TMB value and matching it with clinical survival data, we combined the TCGA clinical data with the ICGC data. Kaplan–Meier (K-M) curves were constructed to analyze the potential effects of the TMB on survival rates. We considered a P value < 0.05 statistically significant. We also utilized the “limma” R package to analyze the correlations between the TMB value and clinicopathologic features (eg age, T stage, and N stage).

Overlap between Differentially Expressed Genes and Immune Genes

The ImmPort ecosystem is an open-access dataset that includes more than 300 studies and is freely available at the Shared Data portal (www.immport.org/immport-open), which allows research data to be repurposed to accelerate the development of new insights into discoveries.²⁴ A total of 2498 immune-related genes were downloaded from ImmPort, and we identified overlapping genes between the DEGs and immune genes.

Construction of a Prostate Cancer Risk Score Model

To investigate the prognostic role of overlapping gene regulators in PCa, we utilized least absolute shrinkage and selection operator (LASSO) Cox regression analysis to screen hub genes from the overlapping genes. The hub genes and their coefficients were determined. Samples were separated into a high-risk group and a low-risk group based on the median risk score (RS) value and overall survival (OS) was evaluated. The OS prediction accuracy of the RS model was verified using a receiver operating characteristic (ROC) curve.

Influence of the Copy Number Variation of Hub Genes on the Immune Infiltrate in Prostate Cancer

tumor Immune Estimation Resource (TIMER, https://cistrome.shinyapps.io/timer/) is a comprehensive resource for systematic analysis of immune infiltrates across diverse cancer types and mainly contains six types of immune cells (eg B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells) analyzed according to the TIMER algorithm to estimate cell infiltration.²⁵ We included hub genes in TIMER to explore the relationship between somatic copy number variation (CNV) and the abundance of TIICs in PCa.

Hub Genes Related to Clinical Features

After screening by LASSO Cox regression analysis, some hub genes were identified. We further probed the correlations of the hub genes with clinicopathologic features (eg age, T stage, and N stage) to study the influence of hub gene expression on clinical progression. With respect to the ability of hub genes to predict the OS and disease-free survival (DFS), we employed Gene Expression Profiling Interactive Analysis (GEPIA), which is a web-based tool that delivers fast and customizable functionalities based on TCGA data.²⁶

Results

Mutation Signature of The Cancer Genome Atlas Datasets

According to waterfall and maftools analysis by R package, we characterized various basic features of PCa somatic mutation data from the TCGA database (https://portal.gdc.cancer.gov). As identified by a waterfall plot, the top 10 mutated genes were TTN, TP53, SPOP, KMT2D, SYNE1, MUC16, FOXA1, KMT2C, SPTA1, and ATM. The clinical information showed that among 182 PCa patients, 62 (34.1%) were more than 65 years old, 154 (84.6%) were White, 24 (13.2%) were Black or African-American, and 4 (2.2%) were Asian (Figure 1A). The summary plot showed that the main variant classification was missense mutation, the major variant type was single-nucleotide polymorphism (SNP), and the most common type of simple nucleotide variation (SNV) class was cytosine changed into thymine (Figure 1B).

Figure 1.

PCa mutation cohort in the TCGA. (A) Waterfall plot of the top 10 mutated genes in the TCGA PCa cohort. (B) Overview of mutations in the TCGA PCa cohort. (C) Volcano plot of DEGs between the high- and low-TMB groups. Upregulated genes are shown in green, downregulated genes are shown in red, and non-DEGs are shown in black.

Identification of Differentially Expressed Genes in Prostate Cancer

To explore the DEGs between high-TMB and low-TMB groups, the “limma” R package was applied for TCGA cohort analysis. The results demonstrated that a total of 185 DEGs were screened from the high-TMB and low-TMB groups, including 90 upregulated genes and 95 downregulated genes (Figure 1C).

Distribution of Tumor-infiltrating Immune Cells

To confirm the 22 subtypes of TIICs in PCa, we used CIBERSORT as an analytical tool to calculate the TIIC distribution in each PCa patient. The results revealed that the top three subtypes in the high-TMB group were resting memory CD4+ T cells, CD8+ T cells, and naive B cells, while in the low-TMB group, the top three subtypes were resting memory CD4+ T cells, naive B cells, and CD8+ T cells, respectively (Figure 2). Statistical analysis showed that compared to the low-TMB group, the high-TMB group had higher infiltration of CD8+ T cells (P < 0.05) and activated memory CD4+ T cells (Figure 2, P < 0.05), which may have vital roles in PCa.

Figure 2.

Differential analysis of 22 types of TIICs between the high- and low-TMB groups. CD8+ T cells and activated memory CD4+ T cells showed higher infiltration in the high-TMB group than in the low-TMB group (P < 0.05). There were no significant differences in the other 20 types of TIICs between the 2 TMB groups.

Survival Analysis and Clinicopathologic Features of the Tumor Mutational Burden

After processing with the survival analysis in R package, we built a K-M curve to determine the survival rate. The results showed significant differences between PCa patients with a low-TMB value and high-TMB value, and the patients with a low-TMB had poor OS (Figure 3A, P < 0.01). Moreover, compared with PCa patients younger than 65 years old, those older than 65 years old had a significantly higher TMB (Figure 3B, P < 0.01). In addition, we found that with increasing T and N stages, patients also had increasing TMB values (Figure 3C and D, P < 0.01). According to the K-M plot, a higher TMB favored longer survival, and we speculated that the high infiltration of CD8+ and CD4+ T cells might produce an antitumor effect. These data also suggested that a high TMB is closely related to an old age, a high T stage, and a high N stage, which coincides with the immune viewpoint that the more gene mutations there are, the more likely malignancy is.

Figure 3.

TMB-related clinical feature analysis. (A) Overall survival (OS) of the high- and low-TMB groups. The high-TMB group had longer survival than the low-TMB group (P < 0.001). (B) R software analysis of patients stratified by age. PCa patients older than 65 years had a significantly higher TMB than those younger than 65 years (P < 0.001). (C) R software analysis of patients stratified by T stage. The higher the T stage was, the higher the TMB value. (D) R software analysis of patients stratified by N stage. The TMB value of N1 patients was higher than that of N0 patients (P < 0.001).

Prostate Cancer Risk Score Model

After identifying the overlapping genes between DEGs and immune genes, we obtained 21 overlapping genes in total. We next used LASSO Cox regression analysis to screen the 21 overlapping genes and estimate the adjustment parameter λ by the cross-validation method. The results showed that when log (λ) = 5.5, the error rate was minimized (Figure 4A and B). After screening, three hub genes, AMH, OPRK1, and BIRC5, were retained to construct the RS model, in which the coefficients of AMH, OPRK1, and BIRC5 were 0.1857, −0.0442, and 0.0468, respectively. Then, we separated PCa patients into a high-risk group and a low-risk group based on the RS of each sample to investigate the prognostic role of the RS model. The results revealed that the high-risk PCa patients were predicted to have shorter OS times than the low-risk patients (Figure 4C, P < 0.05). Finally, a receiver ROC curve analysis was used to measure the accuracy of the RS model by determining the area under the ROC curve (AUC). The RS model precisely predicted the prognosis of the PCa patients (Figure 4D, AUC = 0.801).

Figure 4.

Construction of the RS model for evaluating the prognostic value of three hub genes. (A) RS modeling by the LASSO Cox regression algorithm to screen 21 overlapping genes. (B) Distribution of the LASSO coefficients of the 21 overlapping genes. (D) Kaplan–Meier (K-M) plots of overall survival (OS) were generated with the RS model. (D) ROC curve evaluating the predictive efficiency of the RS model.

Influence of the Copy Number Variation of Hub Genes on Infiltrating Immune Cells in Prostate Cancer

The CNV of the three hub genes was investigated using the TIMER database. We found that deep deletion of AMH resulted in a low level of infiltrating CD8+ T cells in PCa compared with the normal copy number (Figure 5A, P < 0.05). Moreover, both the deep deletion and high amplification of AMH were associated with a low level of infiltrating CD4+ T cells in PCa compared with the normal copy number (Figure 5A, P < 0.05). For BIRC5, CD8+ T cells were found at a low infiltration level in PCa with a high amplification copy number compared with normal PCa (Figure 5B, P < 0.05), and CD4+ T cells were found to have a low infiltration level in PCa with a high amplification copy number compared with normal PCa (Figure 5B, P < 0.05), which indicated that the CNV of AMH, and BIRC5 could influence the infiltration level of CD8+ T cells and CD4+ T cells in PCa patients. However, the CNV of OPRK1 did not influence CD8+ or CD4+ T cell infiltration levels in PCa (Figure 5C, P > 0.05). The above analyses suggest that the CNV of AMH, BIRC5, and OPRK1 were relatively likely to affect the infiltration of immune cells in PCa, indicating that AMH, BIRC5, and OPRK1 could be potential immunotherapeutic targets for PCa treatment.

Figure 5.

Effects of the CNV of AMH, BIRC5, and OPRK1 genes on immune cell infiltration. (A) The effect of the CNV of the AMH gene on six types of infiltrating immune cells in PCa. The CNV of the AMH gene relatively stronger affected the infiltration of CD8+ T cells, CD4+ T cells, macrophages, and neutrophils. (B) The effect of the CNV of the BIRC5 gene on six types of infiltrating immune cells in PCa. All six immune cell types were influenced by the CNV of the BIRC5 gene. (C) The effect of the CNV of the OPRK1 gene on six types of infiltrating immune cells in PCa. The CNV of the OPRK1 gene had a strong impact on macrophage infiltration.

Relationships between Hub Genes and Clinicopathologic Features

To clarify the clinical significance of the three hub genes in PCa prognosis, the OS and DFS were plotted as the survival curves by the GEPIA (http://gepia.cancer-pku.cn/) webtool. The OS results suggested that patients with low AMH expression had longer survival times than those with high AMH expression (Figure 6A, P < 0.05), and the DFS analysis indicated that elevated AMH expression favored disease progression in PCa (Figure 6B, P < 0.01). Similarly, PCa patients with low BIRC5 expression had longer OS (Figure 6C, P < 0.05) and DFS than those with high BIRC5 expression (Figure 6D, P < 0.01). Conversely, the survival analysis indicated that PCa patients with low OPRK1 expression had poorer OS (Figure 6E, P < 0.05) and longer DFS than those with high OPRK1 expression (Figure 6F, P < 0.05). These results indicate that AMH, OPRK1, and BIRC5 can serve as potential prognostic biomarkers in PCa.

Figure 6.

K-M plots of the OS and DFS of PCa patients stratified by AMH, BIRC5, or OPRK1. (A–B) OS and DFS of AMH analyses based on AMH showed that high expression of AMH could shorten survival time and promote disease progression. (C–D) OS and DFS analyses based on BIRC5 showed that elevated BIRC5 expression could decrease the survival rates and promote disease progression. (E–F) OS analysis based on OPRK1 showed that high expression of OPRK1 might result in an elevated death rate in the early stage of PCa. DFS analysis based on OPRK1 showed that elevated OPRK1 expression might decrease the survival rate and increase tumor progression.

With respect to clinicopathologic features, no significant difference in age was found for any of the three hub genes (Figure 7A-C, P > 0.05). However, elevated expression of these three hub genes was associated with a high T stage (Figure 7D-F, P < 0.01). Similarly, AMH (Figure 7G, P < 0.01), OPRK1 (Figure 7H, P < 0.01), and BIRC5 (Figure 7I, P < 0.01) were more highly expressed in a high N stage than in a high T stage.

Figure 7.

The relationship of the expression of the three hub genes and clinicopathologic features in PCa. (A–C) There were no significant differences in the expression of AMH, BIRC5, or OPRK1 between the two age groups (all P values > 0.05). (D–F) The expression levels of AMH, BIRC5, and OPRK1 were higher in T3/T4 than in T1/T2 (all P values < 0.001), which showed that the expression of the three hub genes might promote tumor progression. (G–I) The expression levels of AMH, BIRC5, and OPRK1 were higher in the N1 stage than in the N0 stage (all P values < 0.01), which revealed that the expression of the three hub genes might promote tumor metastasis.

Discussion

Despite great progress in diagnostic methods and therapeutic regimens, PCa remains a fatal condition for those who develop advanced disease.²⁷ TIICs are essential components of the tumor immune microenvironment (TIM), which can change the immune status of the tumor. Currently, studies have demonstrated that the TMB plays an important role in survival prognosis in several cancer types and may have a significant impact on the TIM.^21,28 Targeted immunotherapeutic strategies have proven effects against various kinds of tumors.^29–32 Therefore, exploration of whether the TMB can predict the prognosis of tumors and potential immunotherapeutic targets in PCa is urgently needed.

To our knowledge, this study is the first to correlate the TMB with clinical outcomes and immunotherapeutic targets in PCa by bioinformatics. In our study, mutation analysis of a TCGA dataset revealed that missense mutations, SNPs, and cytosine-to-thymine changes were the most common mutation types in PCa. We found that TP53 and SPOP were the most commonly mutated genes via a waterfall plot. Recent reports have suggested that TP53 mutation significantly correlates with a high Gleason score (GS) and PCa recurrence³³ and the accumulation of TP53 mutations increases T cell density in patients with PCa.³⁴ GS is one of the most important standards to determinate the tumor malignancy. In this study, we analyzed the relationship between TMB and GS, and found a significant positive correlation between them (Supplementary Figure). Additionally, SPOP mutation can promote prostate tumorigenesis via the PI3 K/mTOR and AR signaling pathways,³⁵ and SPOP mutation screening is a valuable personalized medicine tool for effective antitumor treatment.³⁶ Hence, both TP53 mutations and SPOP mutations may play important roles in PCa.

According to the somatic mutation profile, we identified and analyzed DEGs between the high-TMB group and the low-TMB group, and a total of 185 DEGs were identified. Based on these DEGs, 22 subtypes of TIICs were predicted per sample via the reference matrix provided by CIBERSORT. We found that CD8+ T cells and activated memory CD4+ T cells were more highly infiltrated in the high-TMB group than that in the low-TMB group. Traditionally, T lymphocytes, especially CD8+ cytotoxic T lymphocytes, are considered the main immunologic effector cells of antitumor immunity, as they can cause tumor cell death without affecting normal cells. tumor-associated T cells have been confirmed to improve the survival rate in multiple solid tumors independently.^37,38 Moreover, T cell infiltration has a stronger independent effect on prognosis than current clinicopathological criteria, such as tumor size, depth of invasion, degree of differentiation, and lymph node status.³⁹ CD4+ T helper cells include Th1 cells, Th2 cells, Th17 cells, and regulatory T cells (Tregs) and are present in solid tumors at a frequency equal to or greater than that of CD8+ T cells. Meng et al.⁴⁰ suggested a positive association between activated memory CD4+ T cells and CD8+ T cells in PCa patients. Currently, increasing evidence emphasizes that CD4+ T cells play vital roles in antitumor immunity.⁴¹ tumor infiltration by Th1 cells is associated with an improved survival rate in different types of cancer.⁴² In addition, Th1 cells can generate cytokines, especially IL-2 and IFN-γ, that activate and promote the proliferation of CD8+ T cells and NK cells.⁴³ In the present study, based on the deconvolution algorithm CIBERSORT, we calculated the densities of 22 TIICs and compared them between the high-TMB group and the low-TMB group. We observed that CD8+ T cells and activated CD4+ T cells were at higher infiltrated levels in the high-TMB group. Additionally, according to our OS analysis based on the TMB, we found that a high-TMB had benefits for OS in PCa patients. Hence, we speculated that a high-TMB means more novel tumor antigens are generated, which makes the tumor vulnerable to attack and infiltration by immune cells¹⁹ to lead to that PCa patients with an elevated TMB exhibit prolonged survival. It should be noted that tumor patients with high TMB are generally expected to have worse survival, whereas it is different from our findings in this study. For our result, one possible explanation is that higher TMB value signifies a higher heterogeneity of the tumor, which activates the body immune system to attack the tumor cells to benefit the patients. Some studies support our explanation, for example, Bi et al.⁴⁴ reported that a high TMB is associated with better clinical outcomes of ovarian cancer, and immune microenvironment analysis indicated the correlations between TMB and infiltrating immune cells. Zhang et al.⁴⁵ found that in bladder urothelial, carcinoma CD8+ T cell and memory-activated CD4+ T cell subsets not only revealed higher infiltrating abundance in high-TMB group but correlated with prolonged OS and lower risk of tumor recurrence, respectively. Together, our results coincided with the above immunological view and suggested that the TMB might be a potential prognostic factor in PCa that has potential value in targeted immunotherapy and may serve as a promising prognostic biomarker in PCa.

Anti-Mullerian hormone (AMH), a member of the transforming growth factor beta (TGF-β) family, is a potential therapeutic agent for tumor treatment.^46,47 Both in vivo and in vitro studies have revealed that AMH can induce apoptosis and inhibit cancer cell growth in ovarian, breast, and PCa.^48,49 AMH mainly induces cell apoptosis through specific AMH type II receptors (AMHRII), which are overexpressed in many cancer cells. Hence, recombinant AMH (rAMH) is viewed as a new potential anticancer agent, especially in diseases such as PCa.⁵⁰ AMH can also induce NF-κB signaling in PCa cells,⁴⁸ while NF-κB activity induced by dominant-negative IκB-α ablates AMH-mediated molecular events and inhibits PCa cell growth, suggesting that the prostate is a candidate target for the action of AMH.⁵¹ In this study, high expression of AMH was observed in PCa with a high T stage, and overexpression of AMH was detected in PCa with a high N stage, which suggested that AMH had a high correlation with the clinical progression of PCa. We also found that the elevated AMH expression observed in PCa samples provided a better prognostic value for PCa patients. Collectively, the cancer pathogenesis and molecular mechanisms of AMH in PCa need to be assessed with in-depth studies.

Opioid receptor kappa 1 (OPRK1) is a member of the opioid receptors, G protein-coupled receptors that bind opioid ligands, including enkephalins and endorphins. As a tumor suppressor, an OPRK1 agonist was shown to reduce the growth of lung cancer. Xenograft experiments using mice showed that loss of OPRK1 enhanced melanoma and lung cancer tumor growth by suppressing angiogenesis.⁵² Chen et al.⁵³ found that downregulation of the expression of the tumor suppressor κ-opioid receptor predicts a poor prognosis in hepatocellular carcinoma patients. Furthermore, a study on PCa performed by Hironobu et al. revealed that OPRK1 is an androgen-repressed gene that may suppress the growth of PCa.⁵⁴ Our study showed that a high level of OPRK1 in PCa samples was correlated with high-grade malignancy. Moreover, elevated OPRK1 expression levels in PCa showed potential roles as a prognostic biomarker in this cancer. Further studies should be carried out to explore the mechanisms underlying the function of OPRK1 during cancer pathogenesis.

Baculoviral IAP repeat-containing 5 (BIRC5) plays a crucial role in the occurrence and progression of cancer.⁵⁵ BIRC5 functions are involved in the tumorigenesis of colorectal tumors.⁵⁶ Moreover, Hu et al. considered BIRC5 to be a promising prognostic molecular biomarker in PCa.⁵⁷ Our study showed that elevated expression of BIRC5 not only had prognostic value in PCa but also presented close associations with clinical stage and tumor progression. Hence, the molecular mechanisms underlying the role of BIRC5 in human cancers require further research.

Footnotes

Acknowledgments

The authors would like to thank The Cancer Genome Atlas (TCGA), the International Cancer Genome Consortium (ICGC), the Shared Data portal, the ImmPort, the Tumor Immune Estimation Resource (TIMER), etc., databases listed in the manuscript for providing the data.

Authors’ Note

L.W. and Y.Y. contributed to the conceptualization and methodology, data collection, and writing the original draft. C.X. and X.W. contributed to statistical analyses, construction of the nomogram, and provision of the study materials. D.W. contributed to critical evaluation of the results. Z.H. contributed to revision of the manuscript and agreed to be accountable for all aspects of the work. All authors read and approved the final manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the grants from the Shanghai Sailing Program (20YF1443800), the Science Foundation of Tongji Hospital (SKW2007), the National Natural Science Foundation of China (81671318), New Frontier Technology Joint Research Project of Shanghai Municipal Hospital (SHDC 12019112), and Clinical Research Plan of SHDC (SHDC2020CR3074B).

Ethical Statement

Our study did not require an ethical board approval because it did not contain human or animal trials.

ORCID iD

Zhe Hong

Supplementa1 Material

Supplementary material for this article is available online.

References

Bray

Ferlay

Soerjomataram

, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394-424.

Antonarakis

Wang

, et al. AR-V7 and resistance to enzalutamide and Abiraterone in prostate cancer. N Engl J Med. 2014;371(11):1028-1038.

Asangani

Dommeti

Wang

, et al. Therapeutic targeting of BET bromodomain proteins in castration-resistant prostate cancer. Nature. 2014;510(7504):278-282.

Litwin

Tan

. The diagnosis and treatment of prostate cancer: a review. JAMA. 2017;317(24):2532-2542.

Hong

Zhang

Ding

, et al. DNA damage promotes TMPRSS2-ERG oncoprotein destruction and prostate cancer suppression via signaling converged by GSK3beta and WEE1. Mol Cell. 2020;79(6):1008-1023. e1004.

Hong

Xiang

Zhang

, et al. Histone acetyltransferase 1 upregulates androgen receptor expression to modulate CRPC cell resistance to enzalutamide. Clin Transl Med. 2021;11(7):e495.

Siegel

Miller

Jemal

. Cancer statistics, 2019. CA Cancer J Clin. 2019;69(1):7-34.

Cava

Bertoli

Colaprico

, et al. In-silico integration approach to identify a key miRNA regulating a gene network in aggressive prostate cancer. Int J Mol Sci. 2018;19(3):910

McDonald

Vira

Vidal

, et al. Association between systemic inflammatory markers and serum prostate-specific antigen in men without prostatic disease - the 2001-2008 national health and nutrition examination survey. Prostate. 2014;74(5):561-567.

10.

Jemal

Ward

Thun

. Declining death rates reflect progress against cancer. PLoS One. 2010;5(3):e9584.

11.

Clendinen

Gaul

Monge

, et al. Preoperative metabolic signatures of prostate cancer recurrence following radical prostatectomy. J Proteome Res. 2019;18(3):1316-1327.

12.

Paller

Antonarakis

. Management of biochemically recurrent prostate cancer after local therapy: evolving standards of care and new directions. Clin Adv Hematol Oncol. 2013;11(1):14-23.

13.

Smyth

Swann

Hayakawa

. Innate tumor immune surveillance. Adv Exp Med Biol. 2007;590:103-111.

14.

Gibney

Weiner

Atkins

. Predictive biomarkers for checkpoint inhibitor-based immunotherapy. Lancet Oncol. 2016;17(12):e542-e551.

15.

Efremova

Finotello

Rieder

Trajanoski

. Neoantigens generated by individual mutations and their role in cancer immunity and immunotherapy. Front Immunol. 2017;8:1679.

16.

Zito Marino

Ascierto

Rossi

, et al.

Are tumor-infiltrating lymphocytes protagonists or background actors in patient selection for cancer immunotherapy?

Expert Opin Biol Ther. 2017;17(6):735-746.

17.

Wang

Zhang

Jankovic

, et al. Combination cancer immunotherapy targeting PD-1 and GITR can rescue CD8(+) T cell dysfunction and maintain memory phenotype. Sci Immunol. 2018;3(29):eaat7061

18.

Goodman

Kato

Bazhenova

, et al. Tumor mutational burden as an independent predictor of response to immunotherapy in diverse cancers. Mol Cancer Ther. 2017;16(11):2598-2608.

19.

Chalmers

Connelly

Fabrizio

, et al. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017;9(1):34.

20.

Rizvi

Hellmann

Snyder

, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science. 2015;348(6230):124-128.

21.

Wang

. Correlate tumor mutation burden with immune signatures in human cancers. BMC Immunol. 2019;20(1):4.

22.

Liu

Wang

Chen

. Whole-exome sequencing reveals recurrent somatic mutation networks in cancer. Cancer Lett. 2013;340(2):270-276.

23.

Newman

Liu

Green

, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12(5):453-457.

24.

Bhattacharya

Dunn

Thomas

, et al. Immport, toward repurposing of open access immunological assay data for translational and clinical research. Sci Data. 2018;5:180015.

25.

Fan

Wang

, et al. TIMER: a web server for comprehensive analysis of tumor-infiltrating immune cells. Cancer Res. 2017;77(21):e108-e110.

26.

Tang

Kang

, et al. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017;45(W1):W98-W102.

27.

Taylor

Watt

. Unsuspected protumorigenic signaling role for the oncometabolite GABA in advanced prostate cancer. Cancer Res. 2019;79(18):4580-4581.

28.

Klebanov

Artomov

Goggins

, et al. Burden of unique and low prevalence somatic mutations correlates with cancer survival. Sci Rep. 2019;9(1):4848.

29.

Chaganty

BKR

Qiu

Gest

, et al. Trastuzumab upregulates PD-L1 as a potential mechanism of trastuzumab resistance through engagement of immune effector cells and stimulation of IFNgamma secretion. Cancer Lett. 2018;430:47-56.

30.

O’Connor

Fessele

Steiner

, et al. Speed of adoption of immune checkpoint inhibitors of programmed cell death 1 protein and comparison of patient ages in clinical practice vs pivotal clinical trials. JAMA Oncol. 2018;4(8):e180798.

31.

Wei

Yang

Wang

, et al. Crosstalk between cancer cells and tumor associated macrophages is required for mesenchymal circulating tumor cell-mediated colorectal cancer metastasis. Mol Cancer. 2019;18(1):64.

32.

Zhang

Cui

, et al. Phase I/II clinical trial of a Wilms’ tumor 1-targeted dendritic cell vaccination-based immunotherapy in patients with advanced cancer. Cancer Immunol Immunother. 2019;68(1):121-130.

33.

Sun

Zhang

Cai

, et al. Identification of critical pathways and hub genes in TP53 mutation prostate cancer by bioinformatics analysis. Biomark Med. 2019;13(10):831-840.

34.

Kaur

Guedes

, et al. TP53 Missense mutation is associated with increased tumor-infiltrating T cells in primary prostate cancer. Hum Pathol. 2019;87:95-102.

35.

Blattner

Liu

Robinson

, et al. SPOP mutation drives prostate tumorigenesis in vivo through coordinate regulation of PI3 K/mTOR and AR signaling. Cancer Cell. 2017;31(3):436-451.

36.

Ostertag

Hutwelker

Plettenburg

Sattler

Popowicz

. Structural insights into BET client recognition of endometrial and prostate cancer-associated SPOP mutants. J Mol Biol. 2019;431(11):2213-2221.

37.

Chew

Chen

Lee

, et al. Chemokine-driven lymphocyte infiltration: an early intratumoural event determining long-term survival in resectable hepatocellular carcinoma. Gut. 2012;61(3):427-438.

38.

Martinet

Garrido

Filleron

, et al. Human solid tumors contain high endothelial venules: association with T- and B-lymphocyte infiltration and favorable prognosis in breast cancer. Cancer Res. 2011;71(17):5678-5687.

39.

Galon

Pages

Marincola

, et al. The immune score as a new possible approach for the classification of cancer. J Transl Med. 2012;10:1.

40.

Meng

Liu

Guan

, et al. The establishment of immune infiltration based novel recurrence predicting nomogram in prostate cancer. Cancer Med. 2019;8(11):5202-5213.

41.

Spitzer

Carmi

Reticker-Flynn

, et al. Systemic immunity is required for effective cancer immunotherapy. Cell. 2017;168(3):487-502. e415.

42.

Costa-Nunes

Cachot

Bobisse

, et al. High-throughput screening of human tumor antigen-specific CD4 T cells, including neoantigen-reactive T cells. Clin Cancer Res. 2019;25(14):4320-4331.

43.

Knutson

Disis

. Tumor antigen-specific T helper cells in cancer immunity and immunotherapy. Cancer Immunol Immunother. 2005;54(8):721-728.

44.

Chen

Yang

. Significance of tumor mutation burden combined with immune infiltrates in the progression and prognosis of ovarian cancer. Cancer Cell Int. 2020;20:373.

45.

Zhang

Shen

Wang

Luo

. Multi-omics analysis of tumor mutation burden combined with immune infiltrates in bladder urothelial carcinoma. J Cell Physiol. 2020;235(4):3849-3863.

46.

Kushnir

Seifer

Barad

Sen

Gleicher

. Potential therapeutic applications of human anti-mullerian hormone (AMH) analogues in reproductive medicine. J Assist Reprod Genet. 2017;34(9):1105-1113.

47.

MacLaughlin

Donahoe

. Mullerian inhibiting substance/anti-mullerian hormone: a potential therapeutic agent for human ovarian and other cancers. Future Oncol. 2010;6(3):391-405.

48.

Hoshiya

Gupta

Segev

, et al. Mullerian inhibiting substance induces NFkB signaling in breast and prostate cancer cells. Mol Cell Endocrinol. 2003;211(1-2):43-49.

49.

Stephen

Pearsall

Christian

, et al. Highly purified mullerian inhibiting substance inhibits human ovarian cancer in vivo. Clin Cancer Res. 2002;8(8):2640-2646.

50.

Rak

Trofimov

Pigareva

, et al. Purification of human recombinant anti-mullerian hormone and its derivatives. Biomed Chromatogr. 2020;34(5):e4782.

51.

Segev

Hoshiya

, et al. Mullerian-inhibiting substance regulates NF-kappa B signaling in the prostate in vitro and in vivo. Proc Natl Acad Sci U S A. 2002;99(1):239-244.

52.

Yamamizu

Furuta

Hamada

, et al. Small ka, cyrillic opioids inhibit tumor angiogenesis by suppressing VEGF signaling. Sci Rep. 2013;3:3213.

53.

Chen

Yan

, et al. Down-regulation of the tumour suppressor kappa-opioid receptor predicts poor prognosis in hepatocellular carcinoma patients. BMC Cancer. 2017;17(1):553.

54.

Yamashita

Shuman

Warrick

Raman

Degraff

. Androgen represses opioid growth factor receptor (OGFR) in human prostate cancer LNCaP cells and OGFR expression in human prostate cancer tissue. Am J Clin Exp Urol. 2018;6(4):164-171.

55.

Altieri

. Survivin, cancer networks and pathway-directed drug discovery. Nat Rev Cancer. 2008;8(1):61-70.

56.

Wang

Zhang

Wang

, et al. Investigation of cell free BIRC5 mRNA as a serum diagnostic and prognostic biomarker for colorectal cancer. J Surg Oncol. 2014;109(6):574-579.

57.

Jiang

Luo

, et al. Development of an autophagy-related gene expression signature for prognosis prediction in prostate cancer patients. J Transl Med. 2020;18(1):160.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.07 MB

Exploration of the Tumor Mutational Burden as a Prognostic Biomarker and Related Hub Gene Identification in Prostate Cancer

Abstract

Keywords

Abbreviations

Introduction

Materials and Methods

Data Collection

Identification of Differentially Expressed Genes of the Tumor Mutational Burden Subgroups

Prediction of Immune Infiltration in Prostate Cancer Samples

Relationships between the Tumor Mutational Burden and Immune Cells

Prognostic Analysis of the Tumor Mutational Burden

Overlap between Differentially Expressed Genes and Immune Genes

Construction of a Prostate Cancer Risk Score Model

Influence of the Copy Number Variation of Hub Genes on the Immune Infiltrate in Prostate Cancer

Hub Genes Related to Clinical Features

Results

Mutation Signature of The Cancer Genome Atlas Datasets

Identification of Differentially Expressed Genes in Prostate Cancer

Distribution of Tumor-infiltrating Immune Cells

Survival Analysis and Clinicopathologic Features of the Tumor Mutational Burden

Prostate Cancer Risk Score Model

Influence of the Copy Number Variation of Hub Genes on Infiltrating Immune Cells in Prostate Cancer

Relationships between Hub Genes and Clinicopathologic Features

Discussion

Footnotes

Acknowledgments

Authors’ Note

Declaration of Conflicting Interests

Funding

Ethical Statement

ORCID iD

Supplementa1 Material

References

Supplementary Material