Machine Learning-Based Identification of Co-expressed Genes in Prostate Cancer and CRPC and Construction of Prognostic Models

Machine Learning-Based Identification of Co-expressed Genes in Prostate Cancer and CRPC and Construction of Prognostic Models | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Machine Learning-Based Identification of Co-expressed Genes in Prostate Cancer and CRPC and Construction of Prognostic Models Zhiheng Huang, Han Xu, Tianhe Zhang, Haiyang Wei, Junfeng Gao, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4203768/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Feb, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Objective The objective of this study was to employ machine learning to identify shared differentially expressed genes (DEGs) in prostate cancer (PCa) initiation and castration resistance, aiming to establish a robust prognostic model and enhance understanding of patient prognosis for personalized treatment strategies. Methods mRNA transcriptome data associated with Castration-Resistant Prostate Cancer (CRPC) were obtained from the GEO database. Differential expression analysis was conducted using the limma R package to compare normal prostate samples with PCa samples, and PCa samples with CRPC samples. Next, we applied LASSO regression, univariate, and multivariate COX regression analyses to pinpoint genes linked to prognosis and build prognostic models. Validation was performed using the TCGA_PRAD dataset to confirm expression differences of hub genes and explore their correlation with clinical variables and prognostic significance. Results We successfully established a prostate cancer risk prediction model containing seven genes (KIF4A, UBE2C, FAM72D, CCDC78, HOXD9, LIX1 and SLC5A8) and verified its accuracy on an independent data set. The results of calibration curve and decision curve show that the model has potential clinical application value. The nomogram can accurately predict the prognosis of patients. Additionally, elevated expression of KIF4A, UBE2C, and FAM72D, or reduced expression of LIX1, correlated with advanced pathological T and N stages, clinical T stage, prostate-specific antigen (PSA) level, age at diagnosis, Gleason score, and shorter progression-free interval (PFI) (P < 0.05). Conclusion By integrating bioinformatics analysis and clinical data, we not only established a reliable prognostic model for prostate cancer but also identified key genes pivotal in disease progression and treatment resistance. These findings provide novel insights and methodologies for assessing prognosis and tailoring treatment strategies for prostate cancer patients. Health sciences/Urology/Prostate Health sciences/Oncology/Cancer Biological sciences/Molecular biology prostate cancer castration resistance TCGA differentially expressed genes prognostic modeling Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Introduction Prostate cancer ranks as the predominant malignancy affecting males in Western developed nations, with the second highest mortality rate after lung cancer [1] . Androgen deprivation therapy (ADT) represents the cornerstone treatment for intermediate and advanced prostate cancer [2] . However, nearly all patients develop resistance to ADT within 18-36 months, resulting in castration-resistant prostate cancer (CRPC) [3] , characterized by a median survival of approximately 13 months [4] . Timely identification of biomarkers associated with the onset and progression of prostate cancer, combined with early intervention, can improve patient quality of life and prolong survival. Therefore, there is an urgent need for reliable biomarkers to predict the prognosis of PCa patients and identify potential therapeutic targets. Due to the heterogeneous nature of prostate cancer, traditional single prognostic markers often lack predictive accuracy. Therefore, it is critical to search for new biomarkers and establish effective prognostic models to improve the prognosis of PCa patients. In this study, we screened common differentially expressed genes in prostate cancer tissues, CRPC tissues, and benign tissues employing datasets retrieved from the NCBI Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) repositories in the United States. We examined their association with clinical features and prognostic outcomes, and constructed a new prognostic model accordingly. This effort aims to provide a theoretical framework for predicting disease progression and guiding precise treatment strategies. 1. Information and methodology 1.1 Data acquisition mRNA microarray data from the GEO database (https://www.ncbi.nlm.nih.gov) were obtained for the training set, which comprised samples of normal prostate tissue, prostate cancer tissue, and castration-resistant prostate cancer tissue from the GSE35988 dataset [5] . This dataset included 12 samples of normal prostate tissue, 49 samples of prostate cancer tissue, and 27 samples of castration-resistant prostate cancer tissue. Additionally, the GSE66187 [6] dataset was utilized as the validation set, consisting of 24 LuCaP-PCa xenografts and 71 CRPC metastatic tumors. 1.2 Identification of shared differentially expressed genes We conducted differential expression analysis separately for normal tissues versus prostate cancer tissues and prostate cancer tissues versus Castration-Resistant Prostate Cancer tissues using the limma package [7] on the GSE35988 dataset. Significant DEGs were obtained after setting the screening criteria (p.adj1), where log2FC>1 was set as "up" for up-regulated genes among differentially expressed genes, and log2FC<-1 was set as "down" for down-regulated genes among differentially expressed genes. After identifying significantly differentially expressed genes, ggplot2 was loaded to plot the volcano map of the dataset, and the heatmap package was loaded to obtain the corresponding heatmap of gene expression. The computational principle of the Venn diagram was employed to identify commonly dysregulated genes across the two stages of progression. 1.3 Functional analysis of differential genes: GO and KEGG signaling pathway analysis The enrichplot R packages and clusterProfiler R packages were utilized for enrichment analysis and visualization of functional analysis results. Figures were generated using the barplot R packages and dotplot R packages [8] . 1.4 Establishment of risk prediction model In the TCGA-PRAD dataset, the survival package was used to conduct univariate Cox regression analysis on the aforementioned common differentially expressed genes, aiming to further identify genes associated with PFI (P<0.05). The LASSO method [9] and ten-fold cross-validation were employed for variable selection in the Cox regression model to determine the penalty parameter (λ). After eliminating gene covariates and reducing the number of genes, multivariate Cox regression analysis was performed. Based on the regression coefficients and the optimized gene expression levels, patients' PFI risk scores were calculated using the formula RS = EXPgene1*β1 + EXPgene2 *β2 + EXPgene3*β3 +.. + EXP gene n*βn (where EXP represents gene expression and βn is the regression coefficient in multivariate Cox regression) [10] . Prostate cancer patients were stratified into high and low-risk groups based on their risk scores, with evaluation conducted using Kaplan-Meier and ROC analyses [11] . Calibration curves and decision curve analysis were utilized to assess the predictive model's accuracy and clinical value. Univariate and multivariate Cox regression analyses were performed to ascertain whether the risk score served as an independent prognostic factor for PFI in PRAD patients, considering covariates such as age at diagnosis, Gleason score, prostate-specific antigen (PSA) level, clinical stage, and pathological stage. 1.5 Validation of the predictive model's accuracy GSE116918 [12] data from the GEO database was obtained for additional validation of the established model. After calculating each patient's risk score using the training set's formula, we grouped patients into low-risk and high-risk categories according to the median score. To analyze survival disparities, we utilized Kaplan-Meier (KM) curves between these groups, while assessing feature prediction accuracy via receiver operating characteristic (ROC) curves. 1.6 Construction of column line plots and calibration curves We integrated clinical data such as age, clinical T-stage, pathological T and N-stage, PSA level, Gleason score, and risk scores. The RMS package in R software was utilized to create a column chart to forecast individual survival probability. Additionally, calibration curves were generated to assess the predicted survival rates for PRAD patients at 1, 3, and 5 years. The clinical relevance of these graphical representations was evaluated using decision curve analysis (DCA), providing insights into their practical utility. 1.7 Validation of Differential Expression of Hub Genes TCGA_PRAD RNAseq data in TPM format were retrieved from the Tumor and Cancer Genome Atlas (TCGA) database. Statistical calculations and visualization of TCGA_PRAD were performed using the R package 3.6.3. Hub gene expression differences between cancerous and normal tissues were analyzed. Similarly, GSE66187 was analyzed to compare the expression differences of hub genes between castration-resistant prostate cancer and primary prostate cancer tissues. 1.8 Clinical Characteristics and Prognostic Analysis of Hub Genes in Prostate Cancer Patients Selected hub genes may have clinical significance in the prognosis of prostate cancer. The expression levels of target genes were individually analyzed for their correlation with clinical variables [pathological stage, clinical stage, age at diagnosis, prostate-specific antigen (PSA) level, Gleason score], and their association with progression-free interval (PFI) was evaluated. 1.9 Statistical processing In this study, SPSS 25, R language (R 4.3.2), and R studio (2023.12.0 Build 372) were employed for data processing. Measurement data were expressed as mean±standard deviation (x±s) if they followed a normal distribution, and compared using t-tests; if not normally distributed, non-parametric tests were employed. Count data were expressed as rates (%), and compared using chi-square tests. 2. Results 2.1 Exploration of Differentially Expressed Genes (DEGs) In the comparison between normal and prostate cancer groups, 494 genes showed differential expression, including 192 up-regulated genes and 302 down-regulated genes (Figure 1a). Similarly, 4867 genes showed differential expression between the hormone-sensitive and castration-resistant groups, comprising 1900 up-regulated genes and 2967 down-regulated genes (Figure 1b). Heatmaps showed 494 differentially expressed genes in the normal and prostate cancer groups (Figure 1c), and the top 500 differentially expressed genes in log2FC in the hormone-sensitive and castration-resistant groups (Figure 1d). Venn plot analysis unveiled 182 common DEGs shared between the two datasets, including 30 co-regulated up-regulated genes and 152 co-regulated down-regulated genes (Figure 1e,f). 2.2 GO and KEGG Enrichment Analysis Results The GO analysis highlighted the involvement of differentially expressed genes in various biological processes, including the mitotic cell cycle, cell division, supramolecular complexes, microtubule cytoskeleton, cytoskeleton protein binding, and microtubule binding (Figure 2a–c). Additionally, examination through the KEGG database unveiled enrichment in metabolic pathways such as focal adhesion, the Hippo signaling pathway, vascular smooth muscle contraction, and the TGF-β signaling pathway (Figure 2d). 2.3 Risk Prediction Model for Prostate Cancer Prognosis For the construction of a prognostic model in prostate cancer, common differentially expressed genes from both datasets underwent univariate Cox regression analysis, leading to the discovery of 67 genes linked to prognosis. Subsequently, Lasso regression analysis was performed, which revealed that the optimal model comprised seven genes: KIF4A, UBE2C, FAM72D, LIX1, CCDC78, HOXD9, and SLC5A8 (Figure 3a,b). Further refinement of the model was achieved through multivariate Cox regression analysis, providing regression coefficients for each gene and a constant term. Utilizing these coefficients, a risk prediction model was developed, integrating gene expression values (Figure 3c). The corresponding regression coefficients β1-β7 are 0.486295137, -0.084724889, 0.20856452, -0.480372962, 0.177600963, 0.608003224, and -0.250531262, with a constant term of -2.421738301. According to the risk assessment model scoring formula, the prostate cancer prognostic model Risk Score is calculated as -2.421738301 + 0.486295137EXP(KIF4A) - 0.084724889EXP(UBE2C) + 0.20856452EXP(FAM72D) - 0.480372962EXP(LIX1) + 0.177600963EXP(CCDC78) + 0.608003224EXP(HOXD9) - 0.250531262*EXP(SLC5A8), where EXP(gene) represents the gene expression value. Using this risk prediction model, we assessed its effectiveness through Kaplan-Meier survival analysis, ROC curve analysis, and decision curve analysis (DCA). Patients were stratified into high-risk and low-risk categories according to their calculated risk scores. The high-risk group demonstrated elevated rates of disease recurrence or mortality and shorter disease-free intervals compared to the low-risk group. Notably, higher risk scores correlated with inferior prognoses in patients diagnosed with prostate adenocarcinoma (PRAD). Examination of gene expression patterns unveiled elevated expression of KIF4A, UBE2C, FAM72D, CCDC78, and HOXD9 in the high-risk group, whereas LIX1 and SLC5A8 exhibited downregulation (Figure 4a). Kaplan-Meier curves corroborated the poorer prognosis associated with the high-risk group (P < 0.001, HR = 4.72, Fig. 4b). As shown in Fig. 4c, the risk model exhibits good predictive value at 1 year (AUC = 0.754), 3 years (AUC = 0.776), and 5 years (AUC = 0.706). Moreover, the risk model demonstrated strong predictive ability at 1, 3, and 5 years, as evidenced by the AUC values. DCA analysis depicted the model's significant net benefit (Figure 4d), while the calibration curve affirmed its consistent predictive accuracy (Figure 4e). To assess the standalone prognostic importance of the risk score and clinical characteristics, we carried out both univariate and multivariate Cox regression analyses. Univariate analysis identified risk score, pathological T and N stages, clinical T stage, Gleason score, and PSA level as prognostic indicators for TCGA-PRAD, with both risk score and PSA level emerging as independent prognostic factors in multivariate analysis. These results highlight the robustness and practicality of our established prognostic model as a biomarker for prostate cancer prognosis (Figure 4f). 2.4 Validation of the Risk Model Using the GEO Database In the GSE116918 validation dataset, Kaplan-Meier analysis (Figure 5a) confirmed that patients categorized as low-risk exhibited better prognoses than those classified as high-risk (P=0.020; HR=1.92, 95% CI=1.11-3.31), mirroring the findings from the training dataset. Additionally, the area under the curve (AUC) of the survival ROC curve illustrated the model's high sensitivity and specificity in predicting patient prognosis (Figure 5b), with corresponding values at 1 year, 3 years, and 5 years of 0.728, 0.526, and 0.599, respectively. 2.5 Creation and Evaluation of Column Charts We devised a column-line diagram that provides clinicians with a quantitative approach to predict the prognosis of PRAD patients, incorporating Gleason score, PSA level, tumor clinical T stage, pathological T and N stages, along with risk scores. This diagram highlighted the significance of risk scores among various clinical parameters (Figure 6a). Additionally, calibration curves illustrated the alignment between the column-line plots and actual survival outcomes of PRAD patients (Figure 6c). Compared to conventional prognostic scoring systems, our model exhibited a higher AUC value (AUC = 0.775, Fig. 6b). 2.6 Validation of Hub Gene Expression Levels These findings were corroborated in the TCGA-PRAD dataset, where KIF4A, UBE2C, FAM72D, and CCDC78 were highly expressed in prostate cancer, while LIX1, SLC5A8, and HOXD9 were expressed at lower levels (Figure 7a). Using the raw microarray data from GSE66187, scatter plots were generated to illustrate the expression differences of Hub genes (Figure 7b). Results indicated that KIF4A, UBE2C, FAM72D, and CCDC78 exhibited high expression levels in Castration-Resistant Prostate Cancer, whereas LIX1, SLC5A8, and HOXD9 showed low expression levels, consistent with the risk prediction model results. 2.7 Clinical Significance and Survival Analysis of Hub Genes Through integration with clinical prognostic information from the TCGA database, the screened Hub genes underwent clinical prognostic analysis (Figure 8). High expression of KIF4A, UBE2C, FAM72D, or low expression of LIX1 was associated with higher pathological T and N stages, clinical T stage, age, PSA level, Gleason score, and poorer PFI in prostate cancer patients. Similarly, low expression of SLC5A8 was linked to higher pathological T and N staging, clinical T and M staging, age, Gleason score, and poorer PFI. Additionally, high CCDC78 expression correlated with higher pathological T and N staging, age, Gleason score, and poorer PFI. Furthermore, increased expression of HOXD9 correlated with higher Gleason scores and worse progression-free interval (PFI). 3. Discussion The rising occurrence of prostate cancer presents an increasingly formidable obstacle due to factors like the aging population, improved standards of living, and heightened healthcare awareness [13] . Amidst the array of treatment options for advanced prostate cancer, androgen deprivation therapy (ADT) emerges as a cornerstone approach. Despite achieving short-term relief from symptoms, the disease often persists and evolves, culminating in recurrence and the development of Castration-Resistant Prostate Cancer (CRPC) [14] . Consequently, there is a pressing need to identify biomarkers indicative of prostate cancer initiation and progression to CRPC, enabling early intervention to inform clinical diagnosis and treatment strategies. The objective of this study was to pinpoint shared fundamental genes linked to both prostate cancer (PCa) and CRPC, enabling early detection of high-risk patients and development of a prognostic model rooted in these core genes, shedding light on their potential impact on tumor prognosis. Initially, GSE35988 was selected as the training dataset to conduct differential expression analysis between the normal versus prostate cancer group and the prostate cancer versus Castration-Resistant Prostate Cancer (CRPC) group, with the aim of pinpointing commonly differentially expressed genes. Following this, a prognostic risk model was established through Cox proportional hazard modeling and Lasso Cox regression analysis, integrating seven genes (KIF4A, UBE2C, FAM72D, CCDC78, HOXD9, LIX1, and SLC5A8). Notably, Kinesin family member 4A (KIF4A), belonging to the kinesin 4 subfamily, plays a pivotal role in regulating chromosome cohesion and segregation during mitosis [15] . Its overexpression has been linked to adverse outcomes in lung, breast, and colon cancers, highlighting its significance in cancer biology. Understanding KIF4A's role in mitosis and its correlation with cancer prognosis could inform potential therapeutic approaches for these malignancies. Our findings revealed a connection between elevated KIF4A expression and advanced pathological stage and higher Gleason score in prostate cancer patients. This association suggests potential implications for disease progression and clinical outcomes in these individuals. Consistent with prior studies, heightened KIF4A expression was linked to enhanced proliferation and migration abilities in prostate cancer cells. Moreover, downregulation of KIF4A was demonstrated to counteract the progression of endocrine therapy-resistant CRPC through modulation of the androgen receptor (AR) [16] . Ubiquitin-binding enzyme E2 C (UBE2C) serves as a critical regulator in eukaryotic protein degradation pathways, contributing to the disruption of mitotic cycle proteins and affecting cell cycle progression. Elevated UBE2C expression has been linked to the development and advancement of several cancers, such as lung, esophageal adenocarcinoma, hepatocellular carcinoma, nasopharyngeal carcinoma, and breast cancer. Moreover, its overexpression is frequently correlated with poor prognoses in breast, thyroid, cervical, bile duct, and gastric cancers [17] . In our study, we observed significant up-regulation of UBE2C during prostate cancer formation, with increased expression observed in castration-resistant Prostate Cancer (CRPC). Elevated UBE2C expression was found to correlate with higher pathological T and N stages, clinical T stage, patient age, PSA level, Gleason score, and poorer Progression-Free Interval (PFI) in CRPC patients. On the other hand, FAM72D, located on Chromosome 1q21.1, has been reported to be up-regulated in high-risk multiple myeloma and is associated with enhanced MM cell proliferation, indicative of a poor prognosis [18] . However, little is known about its function in prostate cancer and further exploration is needed. Limb Expression 1 (LIX1) is localized in mitochondria, where it regulates mitochondrial shape and redox signaling. It is predominantly expressed in gastrointestinal mesenchymal tumors, often indicating an unfavorable prognosis. Knockdown of LIX1 has been shown to inhibit the MAPK pathway in GIST cells and enhance the anti-tumor effect of imatinib [19] . Our findings reveal a significant downregulation of LIX1 expression in both prostate cancer and Castration-Resistant Prostate Cancer (CRPC), indicating a potential correlation with poorer prognoses. Conversely, Homeobox D9 (HOXD9), known for its pivotal role in governing cellular processes, exhibits a typical pattern of overexpression in various cancers including gastric, cervical, pancreatic, colorectal, and hepatocellular carcinomas, often associated with unfavorable clinical outcomes [20] . Surprisingly, our investigation unveils a decreased expression of HOXD9 in prostate cancer, which further diminishes in CRPC compared to normal prostate tissues. Furthermore, this downregulation aligns with higher Gleason scores and diminished disease-free survival in patients. SLC5A8, a protein-coding gene, is known to exert tumor-suppressive effects, particularly in colon and thyroid cancer cells. This gene promotes apoptosis, facilitating the programmed cell death of tumor cells, while simultaneously inhibiting their proliferation, thereby impeding tumor progression [21] . Our study findings align with this role of SLC5A8 in prostate cancer. This study also has some limitations. The mechanism of action of these genes on prostate cancer cell proliferation, invasion, and apoptosis remains unclear and requires further exploration. Additionally, the sample size in this study is small, highlighting the need for subsequent multicenter, large-sample, prospective studies to validate the findings. Conclusion This study established a prognostic model utilizing KIF4A, UBE2C, FAM72D, CCDC78, HOXD9, LIX1, and SLC5A8, accurately forecasting the outcomes of prostate adenocarcinoma (PRAD) patients. We systematically investigated how these genes correlate with clinical characteristics in prostate cancer patients, indicating their promise as diagnostic and prognostic markers. Furthermore, these genes may offer valuable therapeutic targets for individualized treatment strategies. Declarations Author Contribution Z.H.H. contributed to the study design, data analysis, result interpretation and manuscript drafting. H.X., T.H.Z., H.Y.W. and C.H.F. were involved in the reviewing the manuscript. Z.H.H., H.X., J.F.G., C.B.X., and C.H.F. were involved in the manuscript revision. All authors read and approved the final manuscript. Data Availability TCGA gene expression profile (FPKM value) and corresponding clinical information during the current study are publicly available in the TCGA-PRAD repository (https:// portal.gdc.cancer.gov/repository/). The data on microarray expression profiling and corresponding survival information from the GEO database are publicly available in GSE35988, GSE66187 and GSE116918 repository (https://www.ncbi.nlm.nih.gov/geo/). Further inquiries can be directed to the corresponding author. References GLOBAL BURDEN OF DISEASE CANCER COLLABORATION, FITZMAURICE C, AKINYEMIJU T F, et al. Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic Analysis for the Global Burden of Disease Study[J/OL]. JAMA oncology, 2018, 4(11): 1553-1568. https://doi.org/10.1001/jamaoncol.2018.2706. HOFFMAN K E, PENSON D F, ZHAO Z, et al. Patient-Reported Outcomes Through 5 Years for Active Surveillance, Surgery, Brachytherapy, or External Beam Radiation With or Without Androgen Deprivation Therapy for Localized Prostate Cancer[J/OL]. JAMA, 2020, 323(2): 149-163. https://doi.org/10.1001/jama.2019.20675. WATSON P A, ARORA V K, SAWYERS C L. Emerging mechanisms of resistance to androgen receptor inhibitors in prostate cancer[J/OL]. Nature Reviews. Cancer, 2015, 15(12): 701-711. https://doi.org/10.1038/nrc4016. EL FAKIRI M, GEIS N M, AYADA N, et al. PSMA-Targeting Radiopharmaceuticals for Prostate Cancer Therapy: Recent Developments and Future Perspectives[J/OL]. Cancers, 2021, 13(16): 3967. https://doi.org/10.3390/cancers13163967. GRASSO C S, WU Y M, ROBINSON D R, et al. The mutational landscape of lethal castration-resistant prostate cancer[J/OL]. Nature, 2012, 487(7406): 239-243. https://doi.org/10.1038/nature11125. ZHANG X, COLEMAN I M, BROWN L G, et al. SRRM4 Expression and the Loss of REST Activity May Promote the Emergence of the Neuroendocrine Phenotype in Castration-Resistant Prostate Cancer[J/OL]. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 2015, 21(20): 4698-4708. https://doi.org/10.1158/1078-0432.CCR-15-0157. limma powers differential expression analyses for RNA-sequencing and microarray studies - PubMed[EB/OL]. [2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/25605792/. YU G, WANG L G, HAN Y, et al. clusterProfiler: an R package for comparing biological themes among gene clusters[J/OL]. Omics: A Journal of Integrative Biology, 2012, 16(5): 284-287. https://doi.org/10.1089/omi.2011.0118. GOEMAN J J. L1 penalized estimation in the Cox proportional hazards model[J/OL]. Biometrical Journal. Biometrische Zeitschrift, 2010, 52(1): 70-84. https://doi.org/10.1002/bimj.200900028. Comprehensive investigation of a novel differentially expressed lncRNA expression profile signature to assess the survival of patients with colorectal adenocarcinoma - PubMed[EB/OL]. [2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/28187432/. A nonparametric test for the association between longitudinal covariates and censored survival data - PubMed[EB/OL]. [2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/30796830/. JAIN S, LYONS C A, WALKER S M, et al. Validation of a Metastatic Assay using biopsies to improve risk stratification in patients with prostate cancer treated with radical radiation therapy[J/OL]. Annals of Oncology: Official Journal of the European Society for Medical Oncology, 2018, 29(1): 215-222. https://doi.org/10.1093/annonc/mdx637. SHOWALTER T N. Commentary: In search of answers regarding the benefits and harms of short term ADT for intermediate-risk prostate cancer[J]. The Canadian Journal of Urology, 2017, 24(1): 8663. CHOWDHURY S, BJARTELL A, LUMEN N, et al. Real-World Outcomes in First-Line Treatment of Metastatic Castration-Resistant Prostate Cancer: The Prostate Cancer Registry[J/OL]. Targeted Oncology, 2020, 15(3): 301-315. https://doi.org/10.1007/s11523-020-00720-2. KAHM Y J, KIM I G, JUNG U, et al. Impact of KIF4A on Cancer Stem Cells and EMT in Lung Cancer and Glioma[J/OL]. Cancers, 2023, 15(23): 5523. https://doi.org/10.3390/cancers15235523. CHEN J, LI M, FANG S, et al. KIF4A: A potential biomarker for prediction and prognostic of prostate cancer[J/OL]. Clinical and Investigative Medicine. Medecine Clinique Et Experimentale, 2020, 43(3): E49-59. https://doi.org/10.25011/cim.v43i3.34393. ONG K H, LAI H Y, SUN D P, et al. Ubiquitin-conjugating enzyme E2C (UBE2C) is a prognostic indicator for cholangiocarcinoma[J/OL]. European Journal of Medical Research, 2023, 28(1): 593. https://doi.org/10.1186/s40001-023-01575-9. CHATONNET F, PIGNARRE A, SÉRANDOUR A A, et al. The hydroxymethylome of multiple myeloma identifies FAM72D as a 1q21 marker linked to proliferation[J/OL]. Haematologica, 2020, 105(3): 774-783. https://doi.org/10.3324/haematol.2019.222133. S R D, E T, S D, et al. LIX1 Controls MAPK Signaling Reactivation and Contributes to GIST-T1 Cell Resistance to Imatinib[J/OL]. International journal of molecular sciences, 2023, 24(8)[2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/37108337/. LI J, YANG P, HONG L, et al. BST2 promotes gastric cancer metastasis under the regulation of HOXD9 and PABPC1[J/OL]. Molecular Carcinogenesis, 2024. https://doi.org/10.1002/mc.23679. YANG Y, LIAO C, YANG Q, et al. Role of hypermethylated SLC5A8 in follicular thyroid cancer diagnosis and prognosis prediction[J/OL]. World Journal of Surgical Oncology, 2023, 21(1): 367. https://doi.org/10.1186/s12957-023-03240-1. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 16 Feb, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 26 Dec, 2024 Reviews received at journal 25 Dec, 2024 Reviewers agreed at journal 23 Dec, 2024 Reviews received at journal 22 Apr, 2024 Reviewers agreed at journal 11 Apr, 2024 Reviewers agreed at journal 11 Apr, 2024 Reviewers invited by journal 09 Apr, 2024 Editor assigned by journal 09 Apr, 2024 Editor invited by journal 09 Apr, 2024 Submission checks completed at journal 09 Apr, 2024 First submitted to journal 02 Apr, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4203768","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":290301860,"identity":"08934019-e3aa-4e98-80e0-49d7d1c9031c","order_by":0,"name":"Zhiheng Huang","email":"","orcid":"","institution":"Second Affiliated Hospital of Zhengzhou University","correspondingAuthor":false,"prefix":"","firstName":"Zhiheng","middleName":"","lastName":"Huang","suffix":""},{"id":290301861,"identity":"4dd3ece4-11f7-49b6-a163-ae0394268e7d","order_by":1,"name":"Han Xu","email":"","orcid":"","institution":"Zhengzhou University","correspondingAuthor":false,"prefix":"","firstName":"Han","middleName":"","lastName":"Xu","suffix":""},{"id":290301862,"identity":"864258a8-c72e-4a3a-9ef2-9ea9f5faafcc","order_by":2,"name":"Tianhe Zhang","email":"","orcid":"","institution":"Second Affiliated Hospital of Zhengzhou University","correspondingAuthor":false,"prefix":"","firstName":"Tianhe","middleName":"","lastName":"Zhang","suffix":""},{"id":290301863,"identity":"9334fa65-128c-42b1-86d0-ff72ec9f8d85","order_by":3,"name":"Haiyang Wei","email":"","orcid":"","institution":"Second Affiliated Hospital of Zhengzhou University","correspondingAuthor":false,"prefix":"","firstName":"Haiyang","middleName":"","lastName":"Wei","suffix":""},{"id":290301864,"identity":"fcaa82bd-748f-48c3-994c-44d7abc5761b","order_by":4,"name":"Junfeng Gao","email":"","orcid":"","institution":"Second Affiliated Hospital of Zhengzhou University","correspondingAuthor":false,"prefix":"","firstName":"Junfeng","middleName":"","lastName":"Gao","suffix":""},{"id":290301865,"identity":"ca99729c-dbb2-4973-9a4b-e50fbce84e37","order_by":5,"name":"Changbao Xu","email":"","orcid":"","institution":"Second Affiliated Hospital of Zhengzhou University","correspondingAuthor":false,"prefix":"","firstName":"Changbao","middleName":"","lastName":"Xu","suffix":""},{"id":290301866,"identity":"011ff289-2f0a-4552-8205-f7695e3b51c4","order_by":6,"name":"Changhui Fan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAz0lEQVRIiWNgGAWjYBACxvb+h4///mGr52dvIFILc88ZZgPeBr4EyZ4DRGphn5HDJsHbIJdgcCOBSC28M3KPSUjuMMtjuPl44w2GGptogloke94lWxieSStmnJ1WbMFwLC23gZAWw/YEwxsJbMcYm6VzzCQYGw4T1mJ/IMFA4gDbf8Y2yTNEamHsyDGSbGxjS+yR4CFWS8+xZGOGM2zGEjxAvyQQ4xfG9uaDjxkq2OTsjx/eeONDjQ1hLcjAQCKBFOUQLaTqGAWjYBSMgpEBAI1mQcq1jlLBAAAAAElFTkSuQmCC","orcid":"","institution":"Second Affiliated Hospital of Zhengzhou University","correspondingAuthor":true,"prefix":"","firstName":"Changhui","middleName":"","lastName":"Fan","suffix":""}],"badges":[],"createdAt":"2024-04-02 04:32:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4203768/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4203768/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-90444-y","type":"published","date":"2025-02-16T15:58:13+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":54866165,"identity":"dd0d9c1e-4418-46db-8de6-5dea57ca4807","added_by":"auto","created_at":"2024-04-17 20:50:41","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1871754,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of differentially expressed genes in GSE35988: (a) Volcano plot showing 494 genes differentially expressed between the normal and primary prostate cancer groups; (b) Volcano plot showing 4867 genes differentially expressed between the hormone-sensitive and castration-resistant groups; (c) Heatmap of differentially expressed genes between the normal and primary prostate cancer groups; (d) Heatmap of the top 500 differentially expressed genes ranked by log2FC between the hormone-sensitive and castration-resistant groups; (e) Venn diagram illustrating commonly upregulated genes between the TN and WAT groups; (f) Venn diagram illustrating commonly downregulated genes between the TN and WAT groups.\u003c/p\u003e","description":"","filename":"Figure.1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/f67e2ef3249dbeed69eff707.jpg"},{"id":54866170,"identity":"d1f45b80-d14a-44b5-87b0-5b1504d04fb7","added_by":"auto","created_at":"2024-04-17 20:50:42","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":701350,"visible":true,"origin":"","legend":"\u003cp\u003eGO and KEGG analysis of common differentially expressed genes: (a) Biological processes. (b) Cellular components. (c) Molecular functions. (d) KEGG pathways: Kyoto Encyclopedia of Genes and Genomes.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/fa1adea59b68bf379238db81.jpg"},{"id":54866166,"identity":"1fdfdf0b-2cef-4b87-a44d-ca21e4208c42","added_by":"auto","created_at":"2024-04-17 20:50:42","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":906507,"visible":true,"origin":"","legend":"\u003cp\u003eIdentification of co-expressed genes associated with the progression-free interval (PFI) in prostate cancer. (a) The partial likelihood deviance of different variable numbers identified by the LASSO regression model is depicted. Blue dots indicate the partial likelihood deviance values, while grey lines represent the partial likelihood deviance ± standard error (SE). The two vertical lines on the left and right denote optimal values determined by minimum criteria and 1-SE criteria, respectively. The appropriate log (Lambda) value was selected through 10-fold cross-validation using minimum criteria. LASSO stands for the least absolute shrinkage and selection operator method. (b) The LASSO coefficient profiles of the 67 genes related to PFI are presented. (c) Multivariate Cox regression analysis was conducted on the genes identified through LASSO regression.\u003c/p\u003e","description":"","filename":"Figure.3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/b5d1605564380189128c7918.jpg"},{"id":54866169,"identity":"f075300f-0b98-4900-8fc5-a5a35ac7484b","added_by":"auto","created_at":"2024-04-17 20:50:42","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1335189,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction of the prognostic model for predicting Prostate Cancer PFI in the TCGA training cohort. (a) Distribution of prostate cancer patients in the TCGA training cohort based on prognostic risk model scores, recurrence status, and expression patterns of seven genes. (b) Kaplan-Meier curve illustrating PFI based on prognostic risk model scores in the TCGA training cohort. (c) Time-dependent ROC curve assessing the predictability of 1, 3, and 5-year PFI in the TCGA training cohort. (d) DCA curve depicting risk scores in the TCGA training cohort. (e) Calibration curves predicting 1, 3, and 5-year PFI for patients. (f) Univariate and multivariate Cox regression analyses of risk scores, patient age at diagnosis, Gleason score, PSA level, pathological tumor stage, and clinical tumor stage in the TCGA cohort. PSA ≤ 4 was defined as 0, while PSA \u0026gt; 4 was defined as 1.\u003c/p\u003e","description":"","filename":"Figure.4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/2ee7097d5ad8a50a83516f99.jpg"},{"id":54866168,"identity":"5d4fe18c-b9ef-483b-a7ec-6483d80c85c6","added_by":"auto","created_at":"2024-04-17 20:50:42","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":365285,"visible":true,"origin":"","legend":"\u003cp\u003eValidation of Risk Model in GSE116918 Dataset (a) Kaplan-Meier curves of PFI based on risk assessment model scores (b) Time-dependent ROC curves predicting patient PFI at 1, 3, and 5 years.\u003c/p\u003e","description":"","filename":"Figure.5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/a1ee5ff6614df1789d5a3f09.jpg"},{"id":54866171,"identity":"40060b83-9d1d-4589-a960-eee642cfe881","added_by":"auto","created_at":"2024-04-17 20:50:42","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":902922,"visible":true,"origin":"","legend":"\u003cp\u003eConstruction and Evaluation of Nomogram for Predicting 1-year, 3-year, and 5-year PFI in PCa (a) Nomogram for predicting 1-year, 3-year, and 5-year PFI in PCa patients (b) Calibration curves predicting 1-year, 3-year, and 5-year PFI incidence rates (c) ROC curves comparing risk scores and other variables for predicting PFI.\u003c/p\u003e","description":"","filename":"Figure.6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/4579d24e588bd5a481c750d1.jpg"},{"id":54866172,"identity":"98634b2e-94b1-4eb1-8be6-2ac8c42affbe","added_by":"auto","created_at":"2024-04-17 20:50:42","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":117391,"visible":true,"origin":"","legend":"\u003cp\u003eExpression Levels of Hub Genes Validated in TCGA-PRAD and GSE66187 (a) Expression of hub genes in the TCGA-PRAD cohort (b) Expression of hub genes in GSE66187.\u003c/p\u003e","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/593e80c1a143311491c6281b.jpg"},{"id":54866173,"identity":"73343c92-9772-4d8e-a721-2ee2ca877276","added_by":"auto","created_at":"2024-04-17 20:50:43","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":4678513,"visible":true,"origin":"","legend":"\u003cp\u003eRelationship between Hub Gene Expression and Clinical Characteristics (a) KIF4A (b) SLC5A8 (c) LIX1 (d) UBE2C (e) FAM72D (f) CCDC78 (g) HOXD9\u003c/p\u003e","description":"","filename":"Figure.8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/935cc1877c208ed489df5f87.jpg"},{"id":76488278,"identity":"75ed2d1f-bcac-478c-a515-a90b86edbc3d","added_by":"auto","created_at":"2025-02-17 16:13:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":11288039,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4203768/v1/5a906811-7343-4029-946b-4b1c57b77a20.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Machine Learning-Based Identification of Co-expressed Genes in Prostate Cancer and CRPC and Construction of Prognostic Models","fulltext":[{"header":"Introduction","content":"\u003cp\u003eProstate cancer ranks as the predominant malignancy affecting males in Western developed nations, with the second highest mortality rate after lung cancer\u003csup\u003e[1]\u003c/sup\u003e. Androgen deprivation therapy (ADT) represents the cornerstone treatment for intermediate and advanced prostate cancer\u003csup\u003e[2]\u003c/sup\u003e. However, nearly all patients develop resistance to ADT within 18-36 months, resulting in castration-resistant prostate cancer (CRPC)\u003csup\u003e[3]\u003c/sup\u003e,\u0026nbsp;characterized by a median survival of approximately 13 months\u003csup\u003e[4]\u003c/sup\u003e. Timely identification of biomarkers associated with the onset and progression of prostate cancer, combined with early intervention, can improve patient quality of life and prolong survival. Therefore, there is an urgent need for reliable biomarkers to predict the prognosis of PCa patients and identify potential therapeutic targets. Due to the heterogeneous nature of prostate cancer, traditional single prognostic markers often lack predictive accuracy. Therefore, it is critical to search for new biomarkers and establish effective prognostic models to improve the prognosis of PCa patients. In this study, we screened common differentially expressed genes in prostate cancer tissues, CRPC tissues, and benign tissues employing datasets retrieved from the NCBI Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) repositories in the United States. We examined their association with clinical features and prognostic outcomes, and constructed a new prognostic model accordingly. This effort aims to provide a theoretical framework for predicting disease progression and guiding precise treatment strategies.\u003c/p\u003e"},{"header":"1. Information and methodology","content":"\u003cp\u003e1.1 Data acquisition\u003c/p\u003e\n\u003cp\u003emRNA microarray data from the GEO database (https://www.ncbi.nlm.nih.gov) were obtained for the training set, which comprised samples of normal prostate tissue, prostate cancer tissue, and castration-resistant prostate cancer tissue from the GSE35988 dataset\u003csup\u003e[5]\u003c/sup\u003e. This dataset included 12 samples of normal prostate tissue, 49 samples of prostate cancer tissue, and 27 samples of castration-resistant prostate cancer tissue. Additionally, the GSE66187\u003csup\u003e[6]\u003c/sup\u003e dataset was utilized as the validation set, consisting of 24 LuCaP-PCa xenografts and 71 CRPC metastatic tumors.\u003c/p\u003e\n\u003cp\u003e1.2\u0026nbsp;Identification of shared differentially expressed genes\u003c/p\u003e\n\u003cp\u003eWe conducted differential expression analysis separately for normal tissues versus prostate cancer tissues and prostate cancer tissues versus Castration-Resistant Prostate Cancer tissues using the limma package\u003csup\u003e[7]\u003c/sup\u003e on the GSE35988 dataset. Significant DEGs were obtained after setting the screening criteria (p.adj\u0026lt;0.01, |log2FC|\u0026gt;1), where log2FC\u0026gt;1 was set as \"up\" for up-regulated genes among differentially expressed genes, and log2FC\u0026lt;-1 was set as \"down\" for down-regulated genes among differentially expressed genes. After identifying significantly differentially expressed genes, ggplot2 was loaded to plot the volcano map of the dataset, and the heatmap package was loaded to obtain the corresponding heatmap of gene expression. The computational principle of the Venn diagram was employed to identify commonly dysregulated genes across the two stages of progression.\u003c/p\u003e\n\u003cp\u003e1.3 Functional analysis of differential genes: GO and KEGG signaling pathway analysis\u003c/p\u003e\n\u003cp\u003eThe enrichplot R packages and clusterProfiler R packages were utilized for enrichment analysis and visualization of functional analysis results. Figures were generated using the barplot R packages and dotplot R packages\u003csup\u003e[8]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e1.4 Establishment of risk prediction model\u003c/p\u003e\n\u003cp\u003eIn the TCGA-PRAD dataset, the survival package was used to conduct univariate Cox regression analysis on the aforementioned common differentially expressed genes, aiming to further identify genes associated with PFI (P\u0026lt;0.05). The LASSO method\u003csup\u003e[9]\u003c/sup\u003e and ten-fold cross-validation were employed for variable selection in the Cox regression model to determine the penalty parameter (λ). After eliminating gene covariates and reducing the number of genes, multivariate Cox regression analysis was performed. Based on the regression coefficients and the optimized gene expression levels, patients' PFI risk scores were calculated using the formula RS = EXPgene1*β1 + EXPgene2 *β2 + EXPgene3*β3 +.. + EXP gene n*βn (where EXP represents gene expression and\u0026nbsp;βn is the regression coefficient in multivariate Cox regression)\u003csup\u003e[10]\u003c/sup\u003e. Prostate cancer patients were stratified into high and low-risk groups based on their risk scores, with evaluation conducted using Kaplan-Meier and ROC analyses\u003csup\u003e[11]\u003c/sup\u003e. Calibration curves and decision curve analysis were utilized to assess the predictive model's accuracy and clinical value. Univariate and multivariate Cox regression analyses were performed to ascertain whether the risk score served as an independent prognostic factor for PFI in PRAD patients, considering covariates such as age at diagnosis, Gleason score, prostate-specific antigen (PSA) level, clinical stage, and pathological stage.\u003c/p\u003e\n\u003cp\u003e1.5 Validation of the predictive model's accuracy\u003c/p\u003e\n\u003cp\u003eGSE116918\u003csup\u003e[12]\u003c/sup\u003e data from the GEO database was obtained for additional validation of the established model. After calculating each patient's risk score using the training set's formula, we grouped patients into low-risk and high-risk categories according to the median score. To analyze survival disparities, we utilized Kaplan-Meier (KM) curves between these groups, while assessing feature prediction accuracy via receiver operating characteristic (ROC) curves.\u003c/p\u003e\n\u003cp\u003e1.6 Construction of column line plots and calibration curves\u003c/p\u003e\n\u003cp\u003eWe integrated clinical data such as age, clinical T-stage, pathological T and N-stage, PSA level, Gleason score, and risk scores. The RMS package in R software was utilized to create a column chart to forecast individual survival probability. Additionally, calibration curves were generated to assess the predicted survival rates for PRAD patients at 1, 3, and 5 years. The clinical relevance of these graphical representations was evaluated using decision curve analysis (DCA), providing insights into their practical utility.\u003c/p\u003e\n\u003cp\u003e1.7 Validation of Differential Expression of Hub Genes\u003c/p\u003e\n\u003cp\u003eTCGA_PRAD RNAseq data in TPM format were retrieved from the Tumor and Cancer Genome Atlas (TCGA) database. Statistical calculations and visualization of TCGA_PRAD were performed using the R package 3.6.3. Hub gene expression differences between cancerous and normal tissues were analyzed. Similarly, GSE66187 was analyzed to compare the expression differences of hub genes between castration-resistant prostate cancer and primary prostate cancer tissues.\u003c/p\u003e\n\u003cp\u003e1.8 Clinical Characteristics and Prognostic Analysis of Hub Genes in Prostate Cancer Patients\u003c/p\u003e\n\u003cp\u003eSelected hub genes may have clinical significance in the prognosis of prostate cancer. The expression levels of target genes were individually analyzed for their correlation with clinical variables [pathological stage, clinical stage, age at diagnosis, prostate-specific antigen (PSA) level, Gleason score], and their association with progression-free interval (PFI) was evaluated.\u003c/p\u003e\n\u003cp\u003e1.9 Statistical processing\u003c/p\u003e\n\u003cp\u003eIn this study, SPSS 25, R language (R 4.3.2), and R studio (2023.12.0 Build 372) were employed for data processing. Measurement data were expressed as mean±standard deviation (x±s) if they followed a normal distribution, and compared using t-tests; if not normally distributed, non-parametric tests were employed. Count data were expressed as rates (%), and compared using chi-square tests.\u003c/p\u003e"},{"header":"2. Results","content":"\u003cp\u003e2.1 Exploration of Differentially Expressed Genes (DEGs)\u003c/p\u003e\n\u003cp\u003eIn the comparison between normal and prostate cancer groups, 494 genes showed differential expression, including 192 up-regulated genes and 302 down-regulated genes (Figure 1a). Similarly, 4867 genes showed differential expression between the hormone-sensitive and castration-resistant groups, comprising 1900 up-regulated genes and 2967 down-regulated genes (Figure 1b). Heatmaps showed 494 differentially expressed genes in the normal and prostate cancer groups (Figure 1c), and the top 500 differentially expressed genes in log2FC in the hormone-sensitive and castration-resistant groups (Figure 1d). Venn plot analysis unveiled 182 common DEGs shared between the two datasets, including 30 co-regulated up-regulated genes and 152 co-regulated down-regulated genes (Figure 1e,f).\u003c/p\u003e\n\u003cp\u003e2.2 GO and KEGG Enrichment Analysis Results\u003c/p\u003e\n\u003cp\u003eThe GO analysis highlighted the involvement of differentially expressed genes in various biological processes, including the mitotic cell cycle, cell division, supramolecular complexes, microtubule cytoskeleton, cytoskeleton protein binding, and microtubule binding (Figure 2a\u0026ndash;c). Additionally, examination through the KEGG database unveiled enrichment in metabolic pathways such as focal adhesion, the Hippo signaling pathway, vascular smooth muscle contraction, and the TGF-\u0026beta; signaling pathway (Figure 2d).\u003c/p\u003e\n\u003cp\u003e2.3 Risk Prediction Model for Prostate Cancer Prognosis\u003c/p\u003e\n\u003cp\u003eFor the construction of a prognostic model in prostate cancer, common differentially expressed genes from both datasets underwent univariate Cox regression analysis, leading to the discovery of 67 genes linked to prognosis. Subsequently, Lasso regression analysis was performed, which revealed that the optimal model comprised seven genes: KIF4A, UBE2C, FAM72D, LIX1, CCDC78, HOXD9, and SLC5A8 (Figure 3a,b). Further refinement of the model was achieved through multivariate Cox regression analysis, providing regression coefficients for each gene and a constant term. Utilizing these coefficients, a risk prediction model was developed, integrating gene expression values (Figure 3c). The corresponding regression coefficients \u0026beta;1-\u0026beta;7 are 0.486295137, -0.084724889, 0.20856452, -0.480372962, 0.177600963, 0.608003224, and -0.250531262, with a constant term of -2.421738301. According to the risk assessment model scoring formula, the prostate cancer prognostic model Risk Score is calculated as -2.421738301 + 0.486295137EXP(KIF4A) - 0.084724889EXP(UBE2C) + 0.20856452EXP(FAM72D) - 0.480372962EXP(LIX1) + 0.177600963EXP(CCDC78) + 0.608003224EXP(HOXD9) - 0.250531262*EXP(SLC5A8), where EXP(gene) represents the gene expression value.\u003c/p\u003e\n\u003cp\u003eUsing this risk prediction model, we assessed its effectiveness through Kaplan-Meier survival analysis, ROC curve analysis, and decision curve analysis (DCA). Patients were stratified into high-risk and low-risk categories according to their calculated risk scores. The high-risk group demonstrated elevated rates of disease recurrence or mortality and shorter disease-free intervals compared to the low-risk group. Notably, higher risk scores correlated with inferior prognoses in patients diagnosed with prostate adenocarcinoma (PRAD). Examination of gene expression patterns unveiled elevated expression of KIF4A, UBE2C, FAM72D, CCDC78, and HOXD9 in the high-risk group, whereas LIX1 and SLC5A8 exhibited downregulation (Figure 4a). Kaplan-Meier curves corroborated the poorer prognosis associated with the high-risk group (P \u0026lt; 0.001, HR = 4.72, Fig. 4b). As shown in Fig. 4c, the risk model exhibits good predictive value at 1 year (AUC = 0.754), 3 years (AUC = 0.776), and 5 years (AUC = 0.706). Moreover, the risk model demonstrated strong predictive ability at 1, 3, and 5 years, as evidenced by the AUC values. DCA analysis depicted the model\u0026apos;s significant net benefit (Figure 4d), while the calibration curve affirmed its consistent predictive accuracy (Figure 4e).\u003c/p\u003e\n\u003cp\u003eTo assess the standalone prognostic importance of the risk score and clinical characteristics, we carried out both univariate and multivariate Cox regression analyses. Univariate analysis identified risk score, pathological T and N stages, clinical T stage, Gleason score, and PSA level as prognostic indicators for TCGA-PRAD, with both risk score and PSA level emerging as independent prognostic factors in multivariate analysis. These results highlight the robustness and practicality of our established prognostic model as a biomarker for prostate cancer prognosis (Figure 4f).\u003c/p\u003e\n\u003cp\u003e2.4 Validation of the Risk Model Using the GEO Database\u003c/p\u003e\n\u003cp\u003eIn the GSE116918 validation dataset, Kaplan-Meier analysis (Figure 5a) confirmed that patients categorized as low-risk exhibited better prognoses than those classified as high-risk (P=0.020; HR=1.92, 95% CI=1.11-3.31), mirroring the findings from the training dataset. Additionally, the area under the curve (AUC) of the survival ROC curve illustrated the model\u0026apos;s high sensitivity and specificity in predicting patient prognosis \u0026nbsp;(Figure 5b), with corresponding values at 1 year, 3 years, and 5 years of 0.728, 0.526, and 0.599, respectively.\u003c/p\u003e\n\u003cp\u003e2.5 Creation and Evaluation of Column Charts\u003c/p\u003e\n\u003cp\u003eWe devised a column-line diagram that provides clinicians with a quantitative approach to predict the prognosis of PRAD patients, incorporating Gleason score, PSA level, tumor clinical T stage, pathological T and N stages, along with risk scores. This diagram highlighted the significance of risk scores among various clinical parameters (Figure 6a). Additionally, calibration curves illustrated the alignment between the column-line plots and actual survival outcomes of PRAD patients (Figure 6c). Compared to conventional prognostic scoring systems, our model exhibited a higher AUC value (AUC = 0.775, Fig. 6b).\u003c/p\u003e\n\u003cp\u003e2.6 Validation of Hub Gene Expression Levels\u003c/p\u003e\n\u003cp\u003eThese findings were corroborated in the TCGA-PRAD dataset, where KIF4A, UBE2C, FAM72D, and CCDC78 were highly expressed in prostate cancer, while LIX1, SLC5A8, and HOXD9 were expressed at lower levels (Figure 7a). Using the raw microarray data from GSE66187, scatter plots were generated to illustrate the expression differences of Hub genes (Figure 7b). Results indicated that KIF4A, UBE2C, FAM72D, and CCDC78 exhibited high expression levels in Castration-Resistant Prostate Cancer, whereas LIX1, SLC5A8, and HOXD9 showed low expression levels, consistent with the risk prediction model results.\u003c/p\u003e\n\u003cp\u003e2.7 Clinical Significance and Survival Analysis of Hub Genes\u003c/p\u003e\n\u003cp\u003eThrough integration with clinical prognostic information from the TCGA database, the screened Hub genes underwent clinical prognostic analysis (Figure 8). High expression of KIF4A, UBE2C, FAM72D, or low expression of LIX1 was associated with higher pathological T and N stages, clinical T stage, age, PSA level, Gleason score, and poorer PFI in prostate cancer patients. Similarly, low expression of SLC5A8 was linked to higher pathological T and N staging, clinical T and M staging, age, Gleason score, and poorer PFI. Additionally, high CCDC78 expression correlated with higher pathological T and N staging, age, Gleason score, and poorer PFI. Furthermore, increased expression of HOXD9 correlated with higher Gleason scores and worse progression-free interval (PFI).\u003c/p\u003e"},{"header":"3. Discussion","content":"\u003cp\u003eThe rising occurrence of prostate cancer presents an increasingly formidable obstacle due to factors like the aging population, improved standards of living, and heightened healthcare awareness\u003csup\u003e[13]\u003c/sup\u003e. Amidst the array of treatment options for advanced prostate cancer, androgen deprivation therapy (ADT) emerges as a cornerstone approach. Despite achieving short-term relief from symptoms, the disease often persists and evolves, culminating in recurrence and the development of Castration-Resistant Prostate Cancer (CRPC)\u003csup\u003e[14]\u003c/sup\u003e. Consequently, there is a pressing need to identify biomarkers indicative of prostate cancer initiation and progression to CRPC, enabling early intervention to inform clinical diagnosis and treatment strategies. The objective of this study was to pinpoint shared fundamental genes linked to both prostate cancer (PCa) and CRPC, enabling early detection of high-risk patients and development of a prognostic model rooted in these core genes, shedding light on their potential impact on tumor prognosis.\u003c/p\u003e\n\u003cp\u003eInitially, GSE35988 was selected as the training dataset to conduct differential expression analysis between the normal versus prostate cancer group and the prostate cancer versus Castration-Resistant Prostate Cancer (CRPC) group, with the aim of pinpointing commonly differentially expressed genes. Following this, a prognostic risk model was established through Cox proportional hazard modeling and Lasso Cox regression analysis, integrating seven genes (KIF4A, UBE2C, FAM72D, CCDC78, HOXD9, LIX1, and SLC5A8).\u003c/p\u003e\n\u003cp\u003eNotably, Kinesin family member 4A (KIF4A), belonging to the kinesin 4 subfamily, plays a pivotal role in regulating chromosome cohesion and segregation during mitosis\u003csup\u003e[15]\u003c/sup\u003e. Its overexpression has been linked to adverse outcomes in lung, breast, and colon cancers, highlighting its significance in cancer biology. Understanding KIF4A\u0026apos;s role in mitosis and its correlation with cancer prognosis could inform potential therapeutic approaches for these malignancies.\u0026nbsp;Our findings revealed a connection between elevated KIF4A expression and advanced pathological stage and higher Gleason score in prostate cancer patients. This association suggests potential implications for disease progression and clinical outcomes in these individuals. Consistent with prior studies, heightened KIF4A expression was linked to enhanced proliferation and migration abilities in prostate cancer cells. Moreover, downregulation of KIF4A was demonstrated to counteract the progression of endocrine therapy-resistant CRPC through modulation of the androgen receptor (AR)\u003csup\u003e[16]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eUbiquitin-binding enzyme E2 C (UBE2C) serves as a critical regulator in eukaryotic protein degradation pathways, contributing to the disruption of mitotic cycle proteins and affecting cell cycle progression. Elevated UBE2C expression has been linked to the development and advancement of several cancers, such as lung, esophageal adenocarcinoma, hepatocellular carcinoma, nasopharyngeal carcinoma, and breast cancer. Moreover, its overexpression is frequently correlated with poor prognoses in breast, thyroid, cervical, bile duct, and gastric cancers\u003csup\u003e[17]\u003c/sup\u003e. In our study, we observed significant up-regulation of UBE2C during prostate cancer formation, with increased expression observed in castration-resistant Prostate Cancer (CRPC). Elevated UBE2C expression was found to correlate with higher pathological T and N stages, clinical T stage, patient age, PSA level, Gleason score, and poorer Progression-Free Interval (PFI) in CRPC patients.\u003c/p\u003e\n\u003cp\u003eOn the other hand, FAM72D, located on Chromosome 1q21.1, has been reported to be up-regulated in high-risk multiple myeloma and is associated with enhanced MM cell proliferation, indicative of a poor prognosis\u003csup\u003e[18]\u003c/sup\u003e. However, little is known about its function in prostate cancer and further exploration is needed. Limb Expression 1 (LIX1) is localized in mitochondria, where it regulates mitochondrial shape and redox signaling. It is predominantly expressed in gastrointestinal mesenchymal tumors, often indicating an unfavorable prognosis. Knockdown of LIX1 has been shown to inhibit the MAPK pathway in GIST cells and enhance the anti-tumor effect of imatinib\u003csup\u003e[19]\u003c/sup\u003e. Our findings reveal a significant downregulation of LIX1 expression in both prostate cancer and Castration-Resistant Prostate Cancer (CRPC), indicating a potential correlation with poorer prognoses. Conversely, Homeobox D9 (HOXD9), known for its pivotal role in governing cellular processes, exhibits a typical pattern of overexpression in various cancers including gastric, cervical, pancreatic, colorectal, and hepatocellular carcinomas, often associated with unfavorable clinical outcomes\u003csup\u003e[20]\u003c/sup\u003e. Surprisingly, our investigation unveils a decreased expression of HOXD9 in prostate cancer, which further diminishes in CRPC compared to normal prostate tissues. Furthermore, this downregulation aligns with higher Gleason scores and diminished disease-free survival in patients. SLC5A8, a protein-coding gene, is known to exert tumor-suppressive effects, particularly in colon and thyroid cancer cells. This gene promotes apoptosis, facilitating the programmed cell death of tumor cells, while simultaneously inhibiting their proliferation, thereby impeding tumor progression\u003csup\u003e[21]\u003c/sup\u003e. Our study findings align with this role of SLC5A8 in prostate cancer.\u003c/p\u003e\n\u003cp\u003eThis study also has some limitations. The mechanism of action of these genes on prostate cancer cell proliferation, invasion, and apoptosis remains unclear and requires further exploration. Additionally, the sample size in this study is small, highlighting the need for subsequent multicenter, large-sample, prospective studies to validate the findings.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis study established a prognostic model utilizing KIF4A, UBE2C, FAM72D, CCDC78, HOXD9, LIX1, and SLC5A8, accurately forecasting the outcomes of prostate adenocarcinoma (PRAD) patients. We systematically investigated how these genes correlate with clinical characteristics in prostate cancer patients, indicating their promise as diagnostic and prognostic markers. Furthermore, these genes may offer valuable therapeutic targets for individualized treatment strategies.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eZ.H.H. contributed to the study design, data analysis, result interpretation and manuscript drafting. H.X., T.H.Z., H.Y.W. and C.H.F. were involved in the reviewing the manuscript. Z.H.H., H.X., J.F.G., C.B.X., and C.H.F. were involved in the manuscript revision. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eTCGA gene expression profile (FPKM value) and corresponding clinical information during the current study are publicly available in the TCGA-PRAD repository (https:// portal.gdc.cancer.gov/repository/). The data on microarray expression profiling and corresponding survival information from the GEO database are publicly available in GSE35988, GSE66187 and GSE116918 repository (https://www.ncbi.nlm.nih.gov/geo/). Further inquiries can be directed to the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eGLOBAL BURDEN OF DISEASE CANCER COLLABORATION, FITZMAURICE C, AKINYEMIJU T F, et al. Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic Analysis for the Global Burden of Disease Study[J/OL]. JAMA oncology, 2018, 4(11): 1553-1568. https://doi.org/10.1001/jamaoncol.2018.2706.\u003c/li\u003e\n\u003cli\u003eHOFFMAN K E, PENSON D F, ZHAO Z, et al. Patient-Reported Outcomes Through 5 Years for Active Surveillance, Surgery, Brachytherapy, or External Beam Radiation With or Without Androgen Deprivation Therapy for Localized Prostate Cancer[J/OL]. JAMA, 2020, 323(2): 149-163. https://doi.org/10.1001/jama.2019.20675.\u003c/li\u003e\n\u003cli\u003eWATSON P A, ARORA V K, SAWYERS C L. Emerging mechanisms of resistance to androgen receptor inhibitors in prostate cancer[J/OL]. Nature Reviews. Cancer, 2015, 15(12): 701-711. https://doi.org/10.1038/nrc4016.\u003c/li\u003e\n\u003cli\u003eEL FAKIRI M, GEIS N M, AYADA N, et al. PSMA-Targeting Radiopharmaceuticals for Prostate Cancer Therapy: Recent Developments and Future Perspectives[J/OL]. Cancers, 2021, 13(16): 3967. https://doi.org/10.3390/cancers13163967.\u003c/li\u003e\n\u003cli\u003eGRASSO C S, WU Y M, ROBINSON D R, et al. The mutational landscape of lethal castration-resistant prostate cancer[J/OL]. Nature, 2012, 487(7406): 239-243. https://doi.org/10.1038/nature11125.\u003c/li\u003e\n\u003cli\u003eZHANG X, COLEMAN I M, BROWN L G, et al. SRRM4 Expression and the Loss of REST Activity May Promote the Emergence of the Neuroendocrine Phenotype in Castration-Resistant Prostate Cancer[J/OL]. Clinical Cancer Research: An Official Journal of the American Association for Cancer Research, 2015, 21(20): 4698-4708. https://doi.org/10.1158/1078-0432.CCR-15-0157.\u003c/li\u003e\n\u003cli\u003elimma powers differential expression analyses for RNA-sequencing and microarray studies - PubMed[EB/OL]. [2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/25605792/.\u003c/li\u003e\n\u003cli\u003eYU G, WANG L G, HAN Y, et al. clusterProfiler: an R package for comparing biological themes among gene clusters[J/OL]. Omics: A Journal of Integrative Biology, 2012, 16(5): 284-287. https://doi.org/10.1089/omi.2011.0118.\u003c/li\u003e\n\u003cli\u003eGOEMAN J J. L1 penalized estimation in the Cox proportional hazards model[J/OL]. Biometrical Journal. Biometrische Zeitschrift, 2010, 52(1): 70-84. https://doi.org/10.1002/bimj.200900028.\u003c/li\u003e\n\u003cli\u003eComprehensive investigation of a novel differentially expressed lncRNA expression profile signature to assess the survival of patients with colorectal adenocarcinoma - PubMed[EB/OL]. [2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/28187432/.\u003c/li\u003e\n\u003cli\u003eA nonparametric test for the association between longitudinal covariates and censored survival data - PubMed[EB/OL]. [2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/30796830/.\u003c/li\u003e\n\u003cli\u003eJAIN S, LYONS C A, WALKER S M, et al. Validation of a Metastatic Assay using biopsies to improve risk stratification in patients with prostate cancer treated with radical radiation therapy[J/OL]. Annals of Oncology: Official Journal of the European Society for Medical Oncology, 2018, 29(1): 215-222. https://doi.org/10.1093/annonc/mdx637.\u003c/li\u003e\n\u003cli\u003eSHOWALTER T N. Commentary: In search of answers regarding the benefits and harms of short term ADT for intermediate-risk prostate cancer[J]. The Canadian Journal of Urology, 2017, 24(1): 8663.\u003c/li\u003e\n\u003cli\u003eCHOWDHURY S, BJARTELL A, LUMEN N, et al. Real-World Outcomes in First-Line Treatment of Metastatic Castration-Resistant Prostate Cancer: The Prostate Cancer Registry[J/OL]. Targeted Oncology, 2020, 15(3): 301-315. https://doi.org/10.1007/s11523-020-00720-2.\u003c/li\u003e\n\u003cli\u003eKAHM Y J, KIM I G, JUNG U, et al. Impact of KIF4A on Cancer Stem Cells and EMT in Lung Cancer and Glioma[J/OL]. Cancers, 2023, 15(23): 5523. https://doi.org/10.3390/cancers15235523.\u003c/li\u003e\n\u003cli\u003eCHEN J, LI M, FANG S, et al. KIF4A: A potential biomarker for prediction and prognostic of prostate cancer[J/OL]. Clinical and Investigative Medicine. Medecine Clinique Et Experimentale, 2020, 43(3): E49-59. https://doi.org/10.25011/cim.v43i3.34393.\u003c/li\u003e\n\u003cli\u003eONG K H, LAI H Y, SUN D P, et al. Ubiquitin-conjugating enzyme E2C (UBE2C) is a prognostic indicator for cholangiocarcinoma[J/OL]. European Journal of Medical Research, 2023, 28(1): 593. https://doi.org/10.1186/s40001-023-01575-9.\u003c/li\u003e\n\u003cli\u003eCHATONNET F, PIGNARRE A, S\u0026Eacute;RANDOUR A A, et al. The hydroxymethylome of multiple myeloma identifies FAM72D as a 1q21 marker linked to proliferation[J/OL]. Haematologica, 2020, 105(3): 774-783. https://doi.org/10.3324/haematol.2019.222133.\u003c/li\u003e\n\u003cli\u003eS R D, E T, S D, et al. LIX1 Controls MAPK Signaling Reactivation and Contributes to GIST-T1 Cell Resistance to Imatinib[J/OL]. International journal of molecular sciences, 2023, 24(8)[2024-02-25]. https://pubmed.ncbi.nlm.nih.gov/37108337/.\u003c/li\u003e\n\u003cli\u003eLI J, YANG P, HONG L, et al. BST2 promotes gastric cancer metastasis under the regulation of HOXD9 and PABPC1[J/OL]. Molecular Carcinogenesis, 2024. https://doi.org/10.1002/mc.23679.\u003c/li\u003e\n\u003cli\u003eYANG Y, LIAO C, YANG Q, et al. Role of hypermethylated SLC5A8 in follicular thyroid cancer diagnosis and prognosis prediction[J/OL]. World Journal of Surgical Oncology, 2023, 21(1): 367. https://doi.org/10.1186/s12957-023-03240-1.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"prostate cancer, castration resistance, TCGA, differentially expressed genes, prognostic modeling","lastPublishedDoi":"10.21203/rs.3.rs-4203768/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4203768/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eObjective\u003c/h2\u003e \u003cp\u003eThe objective of this study was to employ machine learning to identify shared differentially expressed genes (DEGs) in prostate cancer (PCa) initiation and castration resistance, aiming to establish a robust prognostic model and enhance understanding of patient prognosis for personalized treatment strategies.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003emRNA transcriptome data associated with Castration-Resistant Prostate Cancer (CRPC) were obtained from the GEO database. Differential expression analysis was conducted using the limma R package to compare normal prostate samples with PCa samples, and PCa samples with CRPC samples. Next, we applied LASSO regression, univariate, and multivariate COX regression analyses to pinpoint genes linked to prognosis and build prognostic models. Validation was performed using the TCGA_PRAD dataset to confirm expression differences of hub genes and explore their correlation with clinical variables and prognostic significance.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eWe successfully established a prostate cancer risk prediction model containing seven genes (KIF4A, UBE2C, FAM72D, CCDC78, HOXD9, LIX1 and SLC5A8) and verified its accuracy on an independent data set. The results of calibration curve and decision curve show that the model has potential clinical application value. The nomogram can accurately predict the prognosis of patients. Additionally, elevated expression of KIF4A, UBE2C, and FAM72D, or reduced expression of LIX1, correlated with advanced pathological T and N stages, clinical T stage, prostate-specific antigen (PSA) level, age at diagnosis, Gleason score, and shorter progression-free interval (PFI) (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eBy integrating bioinformatics analysis and clinical data, we not only established a reliable prognostic model for prostate cancer but also identified key genes pivotal in disease progression and treatment resistance. These findings provide novel insights and methodologies for assessing prognosis and tailoring treatment strategies for prostate cancer patients.\u003c/p\u003e","manuscriptTitle":"Machine Learning-Based Identification of Co-expressed Genes in Prostate Cancer and CRPC and Construction of Prognostic Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-04-17 20:50:36","doi":"10.21203/rs.3.rs-4203768/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-12-26T17:26:09+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-25T07:07:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"330339856972996514994608849718842037782","date":"2024-12-23T06:57:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-04-22T09:47:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"028baded-a372-44d8-bf8a-2e85110c0659","date":"2024-04-12T01:22:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"9a67a1c9-e402-4b7c-8023-1971391595c2","date":"2024-04-11T23:56:32+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-04-09T23:40:37+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-04-09T23:34:30+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-04-09T16:10:52+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-04-09T16:09:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-04-02T04:31:04+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"2ec9c56f-a4d8-44e1-9acc-9eef03de7ea2","owner":[],"postedDate":"April 17th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":30578173,"name":"Health sciences/Urology/Prostate"},{"id":30578174,"name":"Health sciences/Oncology/Cancer"},{"id":30578175,"name":"Biological sciences/Molecular biology"}],"tags":[],"updatedAt":"2025-02-17T16:10:37+00:00","versionOfRecord":{"articleIdentity":"rs-4203768","link":"https://doi.org/10.1038/s41598-025-90444-y","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-02-16 15:58:13","publishedOnDateReadable":"February 16th, 2025"},"versionCreatedAt":"2024-04-17 20:50:36","video":"","vorDoi":"10.1038/s41598-025-90444-y","vorDoiUrl":"https://doi.org/10.1038/s41598-025-90444-y","workflowStages":[]},"version":"v1","identity":"rs-4203768","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4203768","identity":"rs-4203768","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}