Identification of genes that predict the biochemical recurrence of prostate cancer

Predicting prostate cancer recurrence: Introducing PCRPS, an advanced online web server

Background

Prostate cancer (PCa) is one of the leading causes of cancer death in men. About 30% of PCa will develop a biochemical recurrence (BCR) following initial treatment, which significantly contributes to prostate cancer-related deaths. In clinical practice, accurate prediction of PCa recurrence is crucial for making informed treatment decisions. However, the development of reliable models and biomarkers for predicting PCa recurrence remains a challenge. In this study, the aim is to establish an effective and reliable tool for predicting the recurrence of PCa.

Methods

We systematically screened and analyzed potential datasets to predict PCa recurrence. Through quality control analysis, low-quality datasets were removed. Using meta-analysis, differential expression analysis, and feature selection, we identified key genes associated with recurrence. We also evaluated 22 previously published signatures for PCa recurrence prediction. To assess prediction performance, we employed nine machine learning algorithms. We compared the predictive capabilities of models constructed using clinical variables, expression data, and their combinations. Subsequently, we implemented these machine learning models into a user-friendly web server freely accessible to all researchers.

Results

Based on transcriptomic data derived from eight multicenter studies consisting of 733 PCa patients, we screened 23 highly influential genes for predicting prostate cancer recurrence. These genes were used to construct the Prostate Cancer Recurrence Prediction Signature (PCRPS). By comparing with 22 published signatures and four important clinicopathological features, the PCRPS exhibited a robust and significantly improved predictive capability. Among the tested algorithms, Random Forest demonstrated the highest AUC value of 0.72 in predicting PCa recurrence in the testing dataset. To facilitate access and usage of these machine learning models by all researchers and clinicians, we also developed an online web server (https://urology1926.shinyapps.io/PCRPS/) where the PCRPS model can be freely utilized. The tool can also be used to (1) predict the PCa recurrence by clinical information or expression data with high accuracy. (2) provide the possibility of PCa recurrence by nine machine learning algorithms. Furthermore, using the PCRPS scores, we predicted the sensitivity of 22 drugs from GDSC2 and 95 drugs from CTRP2 to the samples. These predictions provide valuable insights into potential drug sensitivities related to the PCRPS score groups.

Conclusion

Overall, our study provides an attractive tool to further guide the clinical management and individualized treatment for PCa.

Prediction of prostate cancer biochemical recurrence by using discretization supports the critical contribution of the extra-cellular matrix genes

Due to its complexity, much effort has been devoted to the development of biomarkers for prostate cancer that have acquired the utmost clinical relevance for diagnosis and grading. However, all of these advances are limited due to the relatively large percentage of biochemical recurrence (BCR) and the limited strategies for follow up. This work proposes a methodology that uses discretization to predict prostate cancer BCR while optimizing the necessary variables. We used discretization of RNA-seq data to increase the prediction of biochemical recurrence and retrieve a subset of ten genes functionally known to be related to the tissue structure. Equal width and equal frequency data discretization methods were compared to isolate the contribution of the genes and their interval of action, simultaneously. Adding a robust clinical biomarker such as prostate specific antigen (PSA) improved the prediction of BCR. Discretization allowed classifying the cancer patients with an accuracy of 82% on testing datasets, and 75% on a validation dataset when a five-bin discretization by equal width was used. After data pre-processing, feature selection and classification, our predictions had a precision of 71% (testing dataset: MSKCC and GSE54460) and 69% (Validation dataset: GSE70769) should the patients present BCR up to 24 months after their final treatment. These results emphasize the use of equal width discretization as a pre-processing step to improve classification for a limited number of genes in the signature. Functionally, many of these genes have a direct or expected role in tissue structure and extracellular matrix organization. The processing steps presented in this study are also applicable to other cancer types to increase the speed and accuracy of the models in diverse datasets.

Deep Learning-Based Multi-Omics Integration Robustly Predicts Relapse in Prostate Cancer

Objective

Post-operative biochemical relapse (BCR) continues to occur in a significant percentage of patients with localized prostate cancer (PCa). Current stratification methods are not adequate to identify high-risk patients. The present study exploits the ability of deep learning (DL) algorithms using the H2O package to combine multi-omics data to resolve this problem.

Methods

Five-omics data from 417 PCa patients from The Cancer Genome Atlas (TCGA) were used to construct the DL-based, relapse-sensitive model. Among them, 265 (63.5%) individuals experienced BCR. Five additional independent validation sets were applied to assess its predictive robustness. Bioinformatics analyses of two relapse-associated subgroups were then performed for identification of differentially expressed genes (DEGs), enriched pathway analysis, copy number analysis and immune cell infiltration analysis.

Results

The DL-based model, with a significant difference (P = 6e-9) between two subgroups and good concordance index (C-index = 0.767), were proven to be robust by external validation. 1530 DEGs including 678 up- and 852 down-regulated genes were identified in the high-risk subgroup S2 compared with the low-risk subgroup S1. Enrichment analyses found five hallmark gene sets were up-regulated while 13 were down-regulated. Then, we found that DNA damage repair pathways were significantly enriched in the S2 subgroup. CNV analysis showed that 30.18% of genes were significantly up-regulated and gene amplification on chromosomes 7 and 8 was significantly elevated in the S2 subgroup. Moreover, enrichment analysis revealed that some DEGs and pathways were associated with immunity. Three tumor-infiltrating immune cell (TIIC) groups with a higher proportion in the S2 subgroup (p = 1e-05, p = 8.7e-06, p = 0.00014) and one TIIC group with a higher proportion in the S1 subgroup (P = 1.3e-06) were identified.

Conclusion

We developed a novel, robust classification for understanding PCa relapse. This study validated the effectiveness of deep learning technique in prognosis prediction, and the method may benefit patients and prevent relapse by improving early detection and advancing early intervention.

Identification and construction of a 13-gene risk model for prognosis prediction in hepatocellular carcinoma patients

We attempted to screen out the feature genes associated with the prognosis of hepatocellular carcinoma (HCC) patients through bioinformatics methods, to generate a risk model to predict the survival rate of patients. Gene expression information of HCC was accessed from GEO database, and differentially expressed genes (DEGs) were obtained through the joint analysis of multi-chip. Functional and pathway enrichment analyses of DEGs indicated that the enrichment was mainly displayed in biological processes such as nuclear division. Based on TCGA-LIHC data set, univariate, LASSO, and multivariate Cox regression analyses were conducted on the DEGs. Then, 13 feature genes were screened for the risk model. Also, the hub genes were examined in our collected clinical samples and GEPIA database. The performance of the risk model was validated by Kaplan-Meier survival analysis and receiver operation characteristic (ROC) curves. While its universality was verified in GSE76427 and ICGC (LIRI-JP) validation cohorts. Besides, through combining patients' clinical features (age, gender, T staging, and stage) and risk scores, univariate and multivariate Cox regression analyses revealed that the risk score was an effective independent prognostic factor. Finally, a nomogram was implemented for 3-year and 5-year overall survival prediction of patients. Our findings aid precision prediction for prognosis of HCC patients.

Predicting biochemical-recurrence-free survival using a three-metabolic-gene risk score model in prostate cancer patients

BACKGROUND

Biochemical recurrence (BCR) after initial treatment, such as radical prostatectomy, is the most frequently adopted prognostic factor for patients who suffer from prostate cancer (PCa). In this study, we aimed to construct a prognostic model consisting of gene expression profiles to predict BCR-free survival.

METHODS

We analyzed 70 metabolic pathways in 152 normal prostate samples and 494 PCa samples from the UCSC Xena dataset (training set) via gene set enrichment analysis (GSEA) to select BCR-related genes and constructed a BCR-related gene risk score (RS) model. We tested the power of our model using Kaplan-Meier (K-M) plots and receiver operator characteristic (ROC) curves. We performed univariate and multivariate analyses of RS using other clinicopathological features and established a nomogram model, which has stronger prediction ability. We used GSE70770 and DFKZ 2018 datasets to validate the results. Finally, we performed differential expression and quantitative real-time polymerase chain reaction analyses of the UCSC data for further verification of the findings.

RESULTS

A total of 194 core enriched genes were obtained through GSEA, among which 16 BCR-related genes were selected and a three-gene RS model based on the expression levels of CA14, LRAT, and MGAT5B was constructed. The outcomes of the K-M plots and ROC curves verified the accuracy of the RS model. We identified the Gleason score, pathologic T stage, and RS model as independent predictors through univariate and multivariate Cox analyses and constructed a nomogram model that presented better predictability than the RS model. The outcomes of the validation set were consistent with those of the training set. Finally, the results of differential expression analyses support the effectiveness of our model.

CONCLUSION

We constructed an RS model based on metabolic genes that could predict the prognosis of PCa patients. The model can be easily used in clinical applications and provide important insights into future research on the underlying mechanism of PCa.

A Novel Approach for the Discovery of Biomarkers of Radiotherapy Response in Breast Cancer

Radiotherapy (RT) is an important treatment modality for the local control of breast cancer (BC). Unfortunately, not all patients that receive RT will obtain a therapeutic benefit, as cancer cells that either possess intrinsic radioresistance or develop resistance during treatment can reduce its efficacy. For RT treatment regimens to become personalised, there is a need to identify biomarkers that can predict and/or monitor a tumour's response to radiation. Here we describe a novel method to identify such biomarkers. Liquid chromatography-mass spectrometry (LC-MS) was used on conditioned media (CM) samples from a radiosensitive oestrogen receptor positive (ER⁺) BC cell line (MCF-7) to identify cancer-secreted biomarkers which reflected a response to radiation. A total of 33 radiation-induced secreted proteins that had higher (up to 12-fold) secretion levels at 24 h post-2 Gy radiation were identified. Secretomic results were combined with whole-transcriptome gene expression experiments, using both radiosensitive and radioresistant cells, to identify a signature related to intrinsic radiosensitivity. Gene expression analysis assessing the levels of the 33 proteins showed that 5 (YBX3, EIF4EBP2, DKK1, GNPNAT1 and TK1) had higher expression levels in the radiosensitive cells compared to their radioresistant derivatives; 3 of these proteins (DKK1, GNPNAT1 and TK1) underwent in-lab and initial clinical validation. Western blot analysis using CM samples from cell lines confirmed a significant increase in the release of each candidate biomarker from radiosensitive cells 24 h after treatment with a 2 Gy dose of radiation; no significant increase in secretion was observed in the radioresistant cells after radiation. Immunohistochemistry showed that higher intracellular protein levels of the biomarkers were associated with greater radiosensitivity. Intracellular levels were further assessed in pre-treatment biopsy tissues from patients diagnosed with ER⁺ BC that were subsequently treated with breast-conserving surgery and RT. High DKK1 and GNPNAT1 intracellular levels were associated with significantly increased recurrence-free survival times, indicating that these two candidate biomarkers have the potential to predict sensitivity to RT. We suggest that the methods highlighted in this study could be utilised for the identification of biomarkers that may have a potential clinical role in personalising and optimising RT dosing regimens, whilst limiting the administration of RT to patients who are unlikely to benefit.

Limitations of Explainability for Established Prognostic Biomarkers of Prostate Cancer

High-throughput technologies do not only provide novel means for basic biological research but also for clinical applications in hospitals. For instance, the usage of gene expression profiles as prognostic biomarkers for predicting cancer progression has found widespread interest. Aside from predicting the progression of patients, it is generally believed that such prognostic biomarkers also provide valuable information about disease mechanisms and the underlying molecular processes that are causal for a disorder. However, the latter assumption has been challenged. In this paper, we study this problem for prostate cancer. Specifically, we investigate a large number of previously published prognostic signatures of prostate cancer based on gene expression profiles and show that none of these can provide unique information about the underlying disease etiology of prostate cancer. Hence, our analysis reveals that none of the studied signatures has a sensible biological meaning. Overall, this shows that all studied prognostic signatures are merely black-box models allowing sensible predictions of prostate cancer outcome but are not capable of providing causal explanations to enhance the understanding of prostate cancer.

Biomarker Identification through Multiomics Data Analysis of Prostate Cancer Prognostication Using a Deep Learning Model and Similarity Network Fusion

Simple Summary

Around 30% of men treated with adjuvant therapy experience recurrences of prostate cancer (PC). Current monitoring of the relapse of PC requires regular postoperative prostate-specific antigen (PSA) value follow-up. Our study aims to identify potential multiomics biomarkers using modern computational analytic methods, deep learning (DL), similarity network fusion (SNF), and the Cancer Genome Atlas (TCGA) prostate adenocarcinoma (PRAD) dataset. Six significantly intersected omics biomarkers from the two models, TELO2, ZMYND19, miR-143, miR-378a, cg00687383 (MED4), and cg02318866 (JMJD6; METTL23) were collected for multiomics panel construction. The difference between the Kaplan–Meier curves of high and low recurrence-risk groups generated from the multiomics panels and clinical information achieve p-value = 2.97 × 10⁻¹⁵ and C-index = 0.713, and the prediction performance of five-year recurrence achieves AUC = 0.789. The results show that the multiomics panel provided valuable biomarkers for the early detection of high-risk recurrent patients, and integrating multiomics data gave us the power to detect the complex mechanisms of cancer among the interactions of different genetic and epigenetic factors.

Abstract

This study is to identify potential multiomics biomarkers for the early detection of the prognostic recurrence of PC patients. A total of 494 prostate adenocarcinoma (PRAD) patients (60-recurrent included) from the Cancer Genome Atlas (TCGA) portal were analyzed using the autoencoder model and similarity network fusion. Then, multiomics panels were constructed according to the intersected omics biomarkers identified from the two models. Six intersected omics biomarkers, TELO2, ZMYND19, miR-143, miR-378a, cg00687383 (MED4), and cg02318866 (JMJD6; METTL23), were collected for multiomics panel construction. The difference between the Kaplan–Meier curves of high and low recurrence-risk groups generated from the multiomics panel achieved p-value = 5.33 × 10⁻⁹, which is better than the former study (p-value = 5 × 10⁻⁷). Additionally, when evaluating the selected multiomics biomarkers with clinical information (Gleason score, age, and cancer stage), a high-performance prediction model was generated with C-index = 0.713, p-value = 2.97 × 10⁻¹⁵, and AUC = 0.789. The risk score generated from the selected multiomics biomarkers worked as an effective indicator for the prediction of PRAD recurrence. This study helps us to understand the etiology and pathways of PRAD and further benefits both patients and physicians with potential prognostic biomarkers when making clinical decisions after surgical treatment.

Upregulation of glucosamine-phosphate N-acetyltransferase 1 is a promising diagnostic and predictive indicator for poor survival in patients with lung adenocarcinoma

Lung adenocarcinoma, a type of non-small cell lung cancer, is the leading cause of cancer death worldwide. Great efforts have been made to identify the underlying mechanism of adenocarcinoma, especially in relation to oncogenes. The present study by integrating computational analysis with western blotting, aimed to understand the role of the upregulation of glucosamine-phosphate N-acetyltransferase 1 (GNPNAT1) in carcinogenesis. In the present study, publicly available gene expression profiles and clinical data were downloaded from The Cancer Genome Atlas to determine the role of GNPNAT1 in lung adenocarcinoma (LUAD). In addition, the association between LUAD susceptibility and GNPNAT1 upregulation were analyzed using Wilcoxon signed-rank test and logistic regression analysis. In LUAD, GNPNAT1 upregulation was significantly associated with disease stage [odds ratio (OR)=2.92, stage III vs. stage I], vital status (dead vs. alive, OR=1.89), cancer status (tumor status vs. tumor-free status, OR=1.85) and N classification (yes vs. no, OR=1.75). Cox regression analysis and the Kaplan-Meier method were utilized to evaluate the association between GNPNAT1 expression and overall survival (OS) time in patients with LUAD. The results demonstrated that patients with increased GNPNAT1 expression levels exhibited a reduced survival rate compared with those with decreased expression levels (P=8.9×10⁻⁵). In addition, Cox regression analysis revealed that GNPNAT1 upregulation was significantly associated with poor OS time [hazard ratio (HR): 1.07; 95% confidence interval (CI): 1.04–1.10; P<0.001]. The gene set enrichment analysis revealed that ‘cell cycle’, ‘oocyte meiosis’, ‘pyrimidine mediated metabolism’, ‘ubiquitin mediated proteolysis’, ‘one carbon pool by folate’, ‘mismatch repair progesterone-mediated oocyte maturation’ and ‘basal transcription factors purine metabolism’ were differentially enriched in the GNPNAT1 high-expression samples compared with GNPNAT1 low-expression samples. The aforementioned pathways are involved in the pathogenesis of LUAD. The findings of the present study suggested that GNPNAT1 upregulation may be considered as a promising diagnostic and prognostic biomarker in patients with LUAD. In addition, the aforementioned pathways may be pivotal pathways perturbed by the abnormal expression of GNPNAT1 in LUAD. The findings of the present study demonstrated the therapeutic value of the regulation of GNPNAT1 in lung adenocarcinoma.

The Impact of Cand1 in Prostate Cancer

Evidence has accumulated asserting the importance of cullin-RING (really interesting new gene) ubiquitin ligases (CRLs) and their regulator Cullin-associated neural-precursor-cell-expressed developmentally down-regulated 8 (NEDD8) dissociated protein 1 (Cand1) in various cancer entities. However, the role of Cand1 in prostate cancer (PCa) has not been intensively investigated so far. Thus, in the present study, we aimed to assess the relevance of Cand1 in the clinical and preclinical setting. Immunohistochemical analyses of radical prostatectomy specimens of PCa patients showed that Cand1 protein levels are elevated in PCa compared to benign areas. In addition, high Cand1 levels were associated with higher Gleason Scores, as well as higher tumor recurrence and decreased overall survival. In line with clinical findings, in vitro experiments in different PCa cell lines revealed that knockdown of Cand1 reduced cell viability and proliferation and increased apoptosis, therefore underlining its role in tumor progression. We also found that the cyclin-dependent kinase inhibitor p21 is significantly upregulated upon downregulation of Cand1. Using bioinformatic tools, we detected genes encoding for proteins linked to mRNA turnover, protein polyubiquitination, and proteasomal degradation to be significantly upregulated in Cand1^high tumors. Next generation sequencing of PCa cell lines resistant to the anti-androgen enzalutamide revealed that Cand1 is mutated in enzalutamide-resistant cells, however, with little functional and clinically relevant impact in the process of resistance development. To summarize the present study, we found that high Cand1 levels correlate with PCa aggressiveness.

Assessment of biochemical recurrence of prostate cancer (Review)

The assessment of the risk of biochemical recurrence (BCR) is critical in the management of males with prostate cancer (PC). Over the past decades, a comprehensive effort has been focusing on improving risk stratification; a variety of models have been constructed using PC-associated pathological features and molecular alterations occurring at the genome, protein and RNA level. Alterations in RNA expression (lncRNA, miRNA and mRNA) constitute the largest proportion of the biomarkers of BCR. In this article, we systemically review RNA-based BCR biomarkers reported in PubMed according to the PRISMA guidelines. Individual miRNAs, mRNAs, lncRNAs and multi-gene panels, including the commercially available signatures, Oncotype DX and Prolaris, will be discussed; details related to cohort size, hazard ratio and 95% confidence intervals will be provided. Mechanistically, these individual biomarkers affect multiple pathways critical to tumorigenesis and progression, including epithelial-mesenchymal transition (EMT), phosphatase and tensin homolog (PTEN), Wnt, growth factor receptor, cell proliferation, immune checkpoints and others. This variety in the mechanisms involved not only validates their associations with BCR, but also highlights the need for the coverage of multiple pathways in order to effectively stratify the risk of BCR. Updates of novel biomarkers and their mechanistic insights are considered, which suggests new avenues to pursue in the prediction of BCR. Additionally, the management of patients with BCR and the potential utility of the stratification of the risk of BCR in salvage treatment decision making for these patients are briefly covered. Limitations will also be discussed.

Differentially Expressed Genes Associated With Prognosis in Locally Advanced Lymph Node-Negative Prostate Cancer

Older age is one of the main risk factors for cancer development. The incidence of prostate cancer, as a multifactorial disease, also depends upon demographic factors, race, and genetic predisposition. Prostate cancer most frequently occurs in men over 60 years of age, indicating a clear association between older age and disease onset. Carcinogenesis is followed by the deregulation of many genes, and some of these changes could serve as biomarkers for diagnosis, prognosis, prediction of drug therapy efficacy, as well as possible therapeutic targets. We have performed a bioinformatic analysis of a The Cancer Genome Atlas (TCGA) data and RNA-Seq profiling of a Russian patient cohort to reveal prognostic markers of locally advanced lymph node-negative prostate cancer (lymph node-negative LAPC). We also aimed to identify markers of the most common molecular subtype of prostate cancer carrying a fusion transcript TMPRSS2-ERG. We have found several genes that were differently expressed between the favorable and unfavorable prognosis groups and involved in the enriched KEGG pathways based on the TCGA (B4GALNT4, PTK6, and CHAT) and Russian patient cohort data (AKR1C1 and AKR1C3). Additionally, we revealed such genes for the TMPRSS2-ERG prostate cancer molecular subtype (B4GALNT4, ASRGL1, MYBPC1, RGS11, SLC6A14, GALNT13, and ST6GALNAC1). Obtained results contribute to a better understanding of the molecular mechanisms behind prostate cancer progression and could be used for further development of the LAPC prognosis marker panel.