WGCNA and LASSO Cox identify ACKR1 DAAM2 and PDE2A as prognostic genes and immune related biomarkers in cervical squamous cell carcinoma and endocervical adenocarcinoma

WGCNA and LASSO Cox identify ACKR1 DAAM2 and PDE2A as prognostic genes and immune related biomarkers in cervical squamous cell carcinoma and endocervical adenocarcinoma | 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 Research Article WGCNA and LASSO Cox identify ACKR1 DAAM2 and PDE2A as prognostic genes and immune related biomarkers in cervical squamous cell carcinoma and endocervical adenocarcinoma Heng Kong, XiaoXuan Kuang, XiaoPeng Wu, ChenQiong Gu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8656950/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 9 You are reading this latest preprint version Abstract Background CESC is a malignant tumor that seriously threatens women's health, and the prognosis of patients is poor, and there is an urgent clinical need to find molecular markers for the prognosis of CESC. Bioinformatics was used to screen molecular markers related to CESC prognosis to provide a basis for prognosis prediction of CESC. Methods Transcriptomic expression data of CESC and corresponding clinical data were downloaded from GEO and TCGA databases. CESC pivot genes were screened by WGCNA and LASSO-COX, and prognostic hub genes were verified using GEPIA. And to explore the correlation of prognostic pivot genes with immune cell infiltration and immune checkpoint gene expression.And compare the expression of prognosis-related genes in CESC and normal cervical tissues in the GEO dataset, TCGA combined with GTEx and HPA datasets. Results Through WGCNA analysis, we constructed a module based on gene synergy. The Darkgreen module was significantly negatively correlated with the CESC. GO analysis shows that most of the genes in the module are related to cell junction, and motility, etc. KEGG analysis showed that the genes in the light green module are more involved in CAMs and Proteoglycans in cancer, et al pathway.The prognostic model composed of three genes, namely: ACKR1, DAAM2 and PDE2A, and found a significant correlation between core prognosis model gene expression and immune infiltration and immune checkpoint genes, which provided guidance for the prognosis and immunotherapy of CESC. Conclusions The prognostic model consisting of three genes, ACKR1, DAAM2 and PDE2A, may be a prognostic and immunotherapy-related molecular marker for CESC. WGCNA LASSO-COX CESC Prognostic model immune prediction Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction CESC is a common malignancy and is the second leading cause of cancer death in women [ 1 ],CESC is a preventable, but still the leading cause of cancer death among women in 36 low-and middle-income countries[ 2 – 3 ].The early symptoms are more insidious, but the increasing volume of the tumor will compress and invade the surrounding organs and tissues, and induce the corresponding symptoms.When found, more has progressed to the middle and late stage, and even lost the opportunity of surgical treatment, the clinical principle of CESC to early diagnosis, early treatment. Early through surgical treatment, resection of focal tissue, preventing tumor cell spread or distant metastasis, improve overall survival. With advances in radiotherapy techniques such as intensity-modulated radiotherapy, treatment-related toxicity is reduced in women with locally advanced disease.But for women with metastatic or recurrent disease, the overall prognosis remains poor[ 4 ],The recurrence and metastasis of CESC is still a difficult problem in diagnosis and treatment. The development of sensitive diagnostic methods and finding new treatment strategies are of great significance for the prevention and treatment of CESC[ 5 ]. HPV is the main pathogenic cause of CESC, but HPV infection is associated not only with the CESC, but also with other anogenital cancers, including [ 6 ].For example, 25–40% of vulvar cancers, approximately 66% of vaginal cancers, and 70–90% of anal cancers are associated with HPV infection in [ 7 – 8 ]. With the deepening of research, it has been found that the interaction between various cytokines and chemokines forms a large and complex molecular biology network during the development of CESC. However, most studies address only one aspect of pathogenesis, ignoring the complex interactions between multiple factors in an organism. With the rapid development of high-throughput sequencing technologies, many sequencing technologies such as genomics, transcriptomics and proteomics have been widely used in the study of multifactor complex diseases such as cancer and cardiovascular disease[ 9 – 10 ].In order to obtain critical information from a large amount of histological data and to study the interaction between critical information, various biological network-based integrated analysis techniques [ 11 ] were developed. Common ones are gene regulatory networks, protein interaction networks, signaling, etc. The co-expression network [ 12 ] was constructed via WGCNA for finding modules of highly correlated genes using module eigengenes and associated with external sample traits using the eigengene network approach. Related networks facilitate network-based genetic screening approaches for identifying candidate biomarkers or therapeutic targets. Applications in complex disease research are increasingly widespread in [ 13 – 15 ].Our study identified 14 modules and 1 key gene as important components of CESC etiology, which may improve our understanding of the molecular mechanisms, such as: pathway networks, metabolic networks, and gene co-expression networks. In this study, we constructed a weighted gene expression network based on gene expression data from the GEO database [ 16 – 17 ], which contains gene expression matrices of CESC cancer tissues and adjacent tissues.We used co-expression network and other related bioinformatics methods, using the expression matrix in TCGA database and clinical survival information to explore the module and hub genes related to CESC prognosis, in order to obtain potential markers closely related to CESC prognosis. Finally, we obtained a prognostic regression model composed of three genes to study the role of core model genes in gene set enrichment analysis (GSEA), tumorigenesis, progression and immune response. And predicted the correlation between core prognostic model gene expression and immune infiltration and immune checkpoints. 2. Results 2.1.Sample collection According to the inclusion criteria[ 18 ]:(a) complete gene expression information (b) complete survival information, exclusion criteria: (a) incomplete gene expression (b) no survival time and survival status in the clinical information. Transcriptomic data and clinical data from 306 CESC tissues and three adjacent tissues were obtained from the TCGA database, including overall survival time, survival status, age, and sex. Transcriptomic data for 18 tumors and 17 normal samples were obtained from GSE63678 (Table S2) 2.2. Identification of DGEs in the GSE63678 dataset Figure 1 shows the identification of the DGEs in the GSE63678 data set.When the threshold was | log2 (fold-change) |> 1 and FDR < 0.01,504 genes were identified, including 306 downregulated genes and 198 up-regulated genes.The top 30 genes with the most significant expression differences were selected for the P-value and differential expression heatmaps and the top 10 significantly differential expression volcano plots, respectively. 2.3. Construction of the co-expression network between CESC and normal samples in the GSE63678 dataset. Figure 2 shows the construction of the CESC and normal samples coexpression network in the GSE63678 dataset.The analysis performed by WGCNA was extended to include the gene dataset GSE63678, excluding the top 50% of genes with minimal MAD and removing outlying genes and samples using the goodSamplesGenes method. All samples were checked and 100 was selected as the height cutoff to draw the sample cluster heatmap (Fig. 2 A), “sft $ powerEstimate” was utilized in order to ascertain the soft-power threshold, and the value of 16(scale independence of 0.86) was selected for the purpose of carrying out additional research on CESC as well as normal.(Fig. 2 B- 2 C)We use the one-step network building function of the WGCNA R package to build the gene networks and find the modules within them. For cluster segmentation, the soft threshold power was set to 16, and the minimum module gene was set to 30, and the sensitivity was set to: 4 (which means high sensitivity). In addition, we merged the distance of less than. 25, the genes finally obtained 13 co-expression modules (Fig. 2 -D).According to the results of this analysis, the high independence between the modules in terms of gene expression. Further analysis of the feature vector (2-E) showed that the Darkgreen module (including 165 genes) had the highest correlation with CESC (r = -0.80, P = 1.3e-8) and was confirmed as a significant module (Fig. 3 A). The Darkgreen module mean and gene significance (GS) have a highly significant correlation (Fig. 3 B).The Darkgreen module genes (46 genes) were selected by MM > 0.80 and GS > 0.65, and crossed with the differentially expressed 504 genes, yielding 38 key genes (Fig. 3 C). 2.4. Functional enrichment analysis Figure 4 shows the functional analysis of the Darkgreen module.Genes included in the Darkgreen module were analysed using GO and KEGG (Fig. 4 A-B). The GO enrichment results showed that Darkgreen module genes were enriched in the functions of “ locomotion”, “negative regulation of cellular process”, “movement of cell or subcellular component”, “cell motility、localization of cell”, “cell migration”, “cell adhesion”, “biological adhesion”, “negative regulation of dendritic cell apoptotic process”, “dendritic cell apoptotic process”, “elastic fiber”, “extracellular matrix”, “cell junction”, “Wnt signalosome”, “cell-cell junction”, “extracellular matrix structural constituent”, “Roundabout binding”, “structural molecule activity”, “ion channel binding”, “cytokine binding”, “Wnt-protein binding”, “phosphatidylinositol 3-kinase binding”, “phosphatidylserine binding”, “protein binding, bridging”, “phosphoric ester hydrolase activity” and so on. KEGG enrichment results show that the main signal pathways of darkgreen module genes are located at “Axon guidance”, “Leukocyte transendothelial migration”, “Malaria”, “Cell adhesion molecules (CAMs)”, “EGFR tyrosine kinase inhibitor resistance”, “Proteoglycans in cancer”, “Aldosterone synthesis and secretion”, and “Viral protein interaction with cytokine and cytokine receptor”. (A)The relationship of two traits and 32 modules; (B) The scatterplot describing the relationship between MM and GS in lightgreen module;(C)Venn diagram of significant darkgreen module genes and GSE63678-DGEs. 2.5. Identification of key genes For the 38 core network genes, the differential expression boxplot (Figure.5A) was drawn using expression profile data and clinical information from the TCGA database, where the differential expression of ATF6B and EXOSC7 in tumor and normal tissues was not statistically significant.Next, we performed univariate Cox prognosis regression analysis (Figure.5B) and multivariate COX prognosis regression analysis (Figure.5C) on the genes with significant differences to obtain three prognostic genes, The relationship between the expression level of the three core genes and the clinical survival rate was subsequently analyzed by integrating the survival time, survival status and gene expression data by the Lasso-Cox prognostic regression model (Figure.5D-E). In addition, we also set a 10-fold cross-validation to obtain the optimal model. Further identify genes associated with CESC prognosis. When the Lambda value was 0.0651047237417135, three genes were finally obtained:PDE2A, DAAM2 and ACKR1,Coef (PDE2A) = 0.0862991933466534,Coef(DAAM2) = 0.0166165369047536, Coef (ACKR1)= -0.0213422619729048,The risk scoring formula for this sample is as follows:RiskScore = 0.0862991933466534*PDE2A + 0.0166165369047536*DAAM2-0.0213422619729048*ACKR1.Risk scores for all samples according to the risk score formula, Risk scores were calculated by combining the survival status and survival time of the samples, The best cut-off value for RiskScore was calculated using the R package maxstat, Set the minimum number of grouped samples greater than 25%, Maximum sample number was grouped by less than 75%, The best cutoff value is: 0.010479882817014, Based on this, the patients were divided into high and low two groups, (Fig. 5 F), The differences between the two groups were further analyzed using the survfit function of the R package survival, The significance of prognostic differences between different groups was assessed using the logrank test method, Eventually we observed a significant prognostic difference (P < 0.001), Patients in the high-risk group may die earlier than low patients, The survival status is also worse, The prognosis differences between the two groups were statistically significant.We drew the ROC curves (Fig. 5 G) to verify the accuracy and reliability of the predictability of the Lasso-Cox prognostic regression model. We performed the ROC analysis to obtain AUC using the R package pROC (version 1.17.0.1). Specifically, 1,3 and 5 years of ROC using the RUC function of pROC and the AUC value of final AUC results, demonstrating that the Lasso-Cox prognostic regression curve was reliable.We generated risk curve, risk, survival scatter plot (Fig. 5 H) and heat map of model gene expression (Fig. 5 I), we analyzed the relationship between different risk scores and follow-up time, survival status and expression of each gene, observed significant decrease in survival with increasing risk scores (Figure.5H), as expected, the ACKR1 gene was a protective factor, decreased expression with increasing risk scores. PDE2A and DAAM2 genes are risk factors and showed a trend of up-regulated expression with increasing risk scores. 2.6. Enrichment analysis of the GSEA cellular pathway model genes To investigate the significantly associated pathways in the Lasso-Cox prognostic regression model, we performed a GSEA pathway-related enrichment analysis for the three prognostic genes (Fig. 4 (A) -(C)).ACKR1 enriched out of a significant correlation of 35 pathways;There were 58 pathways significantly associated with DAAM2; PED2A has 41 significantly associated pathways. 2.7. Effect of model genes on tumor immunity. Next, we determined whether the genes in the Lasso-Cox prognostic regression model affected the immune response.We drew the histograms of the various immune cell contents in the samples (Fig. 5 (A-B)) and the correlation between the levels of the various immune cells (Fig. 5 (C-D)). Correlation Figure C shows that, in the GEO dataset, B_cells_naive and T_cells_CD4_memory_activated had the highest correlation (r = 0.75), Mast_cells_resting and Mast_cells_activated present a significant negative correlation(r=-0.56). Correlation Figure D shows that, in the TCGA dataset, B_cells_naive and Plasma_cellshad the highest correlation (r = 0.54), T_cells_CD8 and T_cells_CD4_memory_restingpresent a significant negative correlation (r=-0.49), Secondly, we used the analysis of immune cell expression immune checkpoint gene in the online tool sangerbox and immune infiltration analysis to study the correlation between the three prognostic genes and immune cell expression, immune infiltration and immune checkpoint, and draw the correlation map (Fig. 5 (H) -5 (Q)).as shown in the figure, ACKR1 had a significant correlation with Bcells, CAFs, Endothelial, and Macrophages and otherCells (P < 0.05); DAAM2 was significantly associated with Bcells, CAFs, and Endothelial and otherCells (P < 0.05); PDE2A was significantly associated with Bcells, CAFs, CD4_Tcells, Endothelial, Macrophages and otherCells (P < 0.05): pearson correlation was calculated for marker genes of ACKR1, DAAM2 and PDE2A, and five classes of immune pathways.The expression of prognostic genes is closely related to immune checkpoint gene expression (Fig. 5 (R) -5 (V)). The expression of the prognostic genes ACKR1 and DAAM2 and PDE2A were significantly and positively correlated with the immune infiltration scores of the samples, with all differences being statistically significant (P < 0.05).Thus, we concluded that the three prognostic genes in the Lasso-Cox prognostic regression model were associated with immune cells and significantly positively associated with immune checkpoints. 2.8. Expression of prognostic model genes in various databases In this study, the expression of three prognostic model genes was selected in TCGA-CESC, GTEx and HPA database. As shown in the figure, the three prognostic model genes were significantly low expressed in tumor tissue, consistent with the expression in GSE63678 dataset, which suggests that agonists of these three genes may be potential targets for the treatment of CESC. 3. Discussion The etiology of CESC is complex, and the clinical diagnosis is still not completely clear. The traditional treatment methods are mainly surgical resection, radiotherapy and chemotherapy. The sensitivity of different patients to radiotherapy and chemotherapy varies greatly, and the possible [ 19 , 20 ], recurrence and metastasis will still occur even after patients receive the above treatment. The recurrence and metastasis of CESC are currently the main causes of CESC death. Although cervical biopsy is the gold standard for the diagnosis of CESC, it is somewhat traumatic and is not suitable for early screening of [ 21 , 22 ]. There is an urgent need to find molecular markers of CESC. WGCNA is a method to identify gene modules and key hub genes associated with phenotypic traits [ 12 ] and classifying them into modules according to their different expression patterns [ 23 ].Since genes with the same expression pattern are largely regulated by the same or similar factors, they are likely to perform the same or similar biological functions [ 11 ].Compared with the traditional co-expression network analysis, WGCNA overcomes the previous co-expression network dividing the associations between continuous variables such as gene expression data as related or unrelated simple binary relationships, thus avoiding the loss of important information [ 24 ].The co-expression network constructed by WGCNA method needs to meet the scale-free network distribution standard. By setting the soft threshold, it meets the objective fact that a few molecules play an important role in various biological life activities [ 15 ]. WGCNA found that ACKR1 [ 25 ] and RRM2 [ 26 ] gene may be related to lymph node metastasis and prognosis in CESC.However, the use of a single bioinformatics analysis method may cause excessive interference data and affect the accuracy of the results. LASSO Regression is to use the compression estimation method to construct the compression function and shrink the partial regression coefficient to zero, which is one of the main methods for dealing with multicolynity data[ 27 , 28 ],Therefore, the joint analysis of WGCNA and LASSO-COX model increases the accuracy and authenticity of the research results. In this study, we screened three genes highly negatively associated with CESC: ACKR1, DAAM2 and PED2A, Atypical chemokine receptoR1 (ACKR1) as key regulators of chemokines involved in combining inflammatory responses and cancer proliferation, angiogenesis and metastasis. Chemokine patterns and chemokine receptor signaling are an integral part of tumor cell proliferation and spread [ 29 ],It has been shown that ACKR1 prevents tumor angiogenesis and subsequent metastasis when it is expressed on malignant cells.It is possible that ACKR1 promotes cell cycle regulation through other interactions, including tumor suppressor CD82 / KAI 1, direct interaction of CD82 with ACKR1 leading to p21 cyclin-dependent kinase inhibition and prevention of metastasis escape [ 30 ], and studies also predict that ACKR1 is associated with cervical lymph node metastasis and prognosis [ 25 ] DAAM2 is involved in the tumorigenesis and progression of human cancers and seems to play different roles in different types of cancers, such as DAAM2 that accelerates the progression of glioma and hepatocellular carcinoma [ 31 , 32 ]. DAAM2 promotes the invasion of CRC cells and plays an important role in CRC invasion [ 33 ].Overexpression of DAAM2 was found in BRCA tissues, and the knockdown of DAAM2 delayed the proliferation, invasion, and migration of BRCA cells. However, the high expression level of DAAM2 has a higher survival rate in LGG and LIHC [ 34 ], and phosphodiesterase 2 (PDE2A) regulates the level of cAMP / cGMP, which is closely associated with various types of tumor progression.Tumors with low PDE2A expression show decreased immune function, PDE2A closely involved in HCC proliferation and metastasis [ 35 ] PDE2A is significantly downregulated in glioma, and overexpression slows the progression of glioma [ 36 ].It plays a key role in the progression of cancers such as colorectal cancer [ 37 ] and melanoma [ 38 ]. PDE2A may be a biomarker for early diagnosis and prognostic evaluation in patients with CESC [ 39 ] The development of tumors is affected by many biological and behavioral characteristics, including the influence of the tumor immune microenvironment. TME is a key factor in tumor growth, metastasis and regulation of tumor immune response [ 40 ].Various components of the tumor microenvironment not only play important roles in tumor progression, immune escape and metastasis, but also have profound effects on the therapeutic effect of patients in [ 41 , 42 ]. For example, immunosuppressive cells within the tumor microenvironment play a key role in promoting tumor immune escape and promoting local suppression of antitumor immune responses by the release of immunosuppressive cytokines [ 43 ]. Similarly, the expression level of tumor-infiltrating lymphocytes is usually correlated with survival in CESC as well as in patients with other solid tumors [ 44 ] In this study, we used WGCNA to select Darkgreen module genes with significant correlation with CESC, including 46 significantly associated genes, performed limma differential analysis of the data set in GSE63678 in GEO database, and 506 differentially expressed genes were obtained, and the intersection of the two genes to obtain 38 key genes,Using LASSO-COX for 38 key genes, three hub genes, ACKR1 and DAAM2 and PDE2A, were obtained to explore their expression in CESC and normal cervical tissue, and preliminarily analyzed the relationship with the immune infiltration level of TME of CESC, which may provide some reference for immunotherapy of CESC. 4. Materials and Methods 4.1. Data sources and searches In this study, the transcriptomic data and clinical data were obtained from the TCGA-CESC [ 45 ] and GEO databases. The TCGA clinical data included clinically relevant information on overall survival time, survival status, age, sex, and disease locus. On the GEO website, we obtained the raw data for the GSE63678[ 17 ] dataset, and the limma package [ 46 ] was used for the statistical analysis of the data, with a p-value < 0.01, FDR < 0.01, | log2FC |≥ 1. 4.2. WGCNA network construction and module identification First, We used the gene expression profiling, We calculated the Median Absolute Deviation (MAD) for each gene separately, Excluding the top 50% of genes with minimal MAD, Outlier genes and samples were removed using the goodSamplesGenes method of the R package WGCNA, Further construction of scale-free co-expression network using WGCNA, Soft thresholds of the co-expression network were calculated using the 'pickSoftThreshold()' function.When the soft threshold is equal to 4, the co-expression network is more close to the scale-free network. The weighted adjacency matrix is constructed to conduct hierarchical clustering based on the phase difference (1-TOM) of topological overlap matrix (TOM) to construct related modules. After linking the module with the clinical characteristics data, the scatter plot of the module membership (MM) and the gene significance (GS) are drawn to clarify the significance of the genes within the module. 4.3. Functional enrichment analysis R package “cluster Profiler”[ 47 ] was used to obtain results for gene set enrichment. Set a minimum gene set to 5 and a maximum gene set to 5000, P value of < 0.05 and a FDR of < 0.25 were considered statistically significant. We performed KEGG and GO enrichment analysis on the genes extracted from the modules with the most significant correlation with CESC, in order to explore the function of the module genes, the involved biological pathways and the localization in the cells. 4.4. Identification of key genes The prognostic value of the core genes was determined by univariate Cox regression analysis, and a P < 0.05 was considered statistically significant.In this study, we used the R package glmnet[ 48 ], integrating survival time, survival status and gene expression data for regression analysis using the Lasso-cox method. The survival time, survival status and gene expression data were integrated for regression analysis using the lasso-cox method. Kaplan-Meier survival curves were generated to evaluate the predictive performance of associated risk genes. We performed ROC analysis to obtain AUC using our R software package pROC. Specifically, we obtained patient survival time, survival status, and risk score, performed ROC analysis of 365,1095,1825 days ROC analysis using the ROC function of pROC and evaluated AUC and confidence interval using the ci function of pROC to obtain final AUC results to verify the performance of model prediction. 4.5.Correlation of prognostic models with tumor immunity. IBOR [ 49 ] is a computational tool for immune tumor biology research. In order to explore the influence and correlation of the expression of prognostic genes in Lasso regression prediction model, we selected the CIBERSORT [ 50 ] algorithm with the R package IBOR based on the expression profile of CESC obtained in TCGA and the expression profile of GSE63678, and plotted the bar chart and correlation chart of immune cell content. In addition, we used the online analysis tool SangerBox[ 51 ] to map the expression of prognostic genes and various immune cells, explore the relationship between the expression of prognostic genes and immunotherapy, and map the correlation between prognostic genes and immune checkpoint.The EPIC [ 52 ] selected Bcells, CAFs, CD4_Tcells, CD8_Tcells, Endothelial, Macrophages, NKcells, otherCells infiltration scores using the R package IOBR, and the immune infiltrating cells scored for each sample using ESTIMATE[ 53 ]. 4.6 Enrichment analysis of the GSEA cellular pathway model genes To investigate the association of tumor-related and immune-related pathway genes in the Lasso-Cox prognostic regression model, we performed GSEA[ 54 ] pathway-related enrichment analysis for prognostic genes, and all pathways were screened with P < 0.01 significant difference and FDR < 0.25 to identify pathways significantly associated with prognostic genes (Fig. 5 (A) -5 (C)). 5. Conclusions In conclusion, in this study, ACKR1, a hub gene related with DAAM2 and PDE2A prognosis, three prognostic genes were significantly low expression in CESC, explored the correlation between three genes with immune cells and immune checkpoint genes, and three prognostic related hub genes were positively correlated with the expression of some immune checkpoint genes, and showed significant correlation with immune infiltration.These results suggest that the three prognostic molecular markers screened are associated with the immune infiltration level of CESC TME, which provides guidance for the prognosis and immunotherapy of CESC, this model has a useful for predicting the effect of CESC clinical immunotherapy This study is a retrospective bioinformatics analysis based on public transcriptomic datasets. Furthermore, this study also has certain limitations. The TCGA-CESC cohort included 306 tumor samples and 3 adjacent samples, and the limited number of adjacent samples may affect the robustness of differential expression interpretation. The external validation dataset GSE63678 included 18 tumor and 17 normal samples, and the relatively small sample size and cross platform differences may introduce bias. In addition, the prognostic model and immune related findings were primarily derived from computational inference and correlation analyses and were not validated in an independent prospective cohort or experimentally confirmed in cellular, animal, or clinical samples. Therefore, our results should be interpreted as hypothesis generating, and further independent cohort validation and mechanistic experiments are warranted. Declarations Supplementary Materials: The following supporting information can be downloaded at: , Figure S1; Table S1 Author Contributions: The following statements should be used “Conceptualization, H.K. and C.G.; methodology, H.K.; software, C.G.; validation, H.K., C.G. and X.K.; formal analysis, C.G.; investigation, H.K.; resources, C.G.; data curation, H.K.; writing—original draft preparation, H.K. and C.G.; writing—review and editing, H.K.; visualization, C.G.; supervision, X.K. and X.W.; project administration, H.K.; funding acquisition, C.G. All authors have read and agreed to the published version of the manuscript.” Please turn to the CRediT taxonomy for the term explanation. Authorship must be limited to those who have contributed substantially to the work reported. Funding: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Informed Consent Statement: This study used information in public databases, which excluded informed consent forms. Data Availability Statement: The transcriptomic and clinical data analyzed in this study were obtained from the following public resources with explicit identifiers and direct URLs to ensure transparency and reproducibility: (1) the TCGA Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma cohort (project ID: TCGA-CESC, https://portal.gdc.cancer.gov/projects/TCGA-CESC ) from the NCI Genomic Data Commons (GDC); (2) the GEO external validation dataset GSE63678 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE63678 ); (3) GTEx datasets (https://www.gtexportal.org/home/datasets ) used as normal tissue references; (4) the Human Protein Atlas (HPA, https://www.proteinatlas.org/ ) used for expression cross validation; (5) GEPIA2 (https://gepia2.cancer-pku.cn/ ) used for survival verification (integrating TCGA and GTEx); and (6) SangerBox (https://sangerbox.com/ ) used for immune related visualization analyses. Acknowledgments: In this section, Thanks to my dear mentor for my great support. Ethics declaration: Not applicable. The authors declare that all bioinformatics analyses in this study comply with the ethical standards of the relevant institutional research committees, the 1964 Declaration of Helsinki and its subsequent amendments, as well as the data usage policies of the public databases employed. Conflicts of Interest: The authors declare no conflict of interest. Consent to publish: Not applicable. This study is a bioinformatics data analysis. All data are from public or de-identified datasets without personally identifiable information, and no individual participants are involved, so consent to publish is not applicable. References Paskeh, M.D.A.; Mirzaei, S.; Gholami, M.H.; Zarrabi, A.; Zabolian, A.; Hashemi, M.; Hushmandi, K.; Ashrafizadeh, M.; Aref, A.R.; Samarghandian, S. Cervical cancer progression is regulated by SOX transcription factors: Revealing signaling networks and therapeutic strategies. Biomed Pharmacother 2021, 144, 112335, doi:10.1016/j.biopha.2021.112335. Arbyn M, Weiderpass E, Bruni L, de Sanjosé S, Saraiya M, Ferlay J, et al. Estimates of incidence and mortality of cervical cancer in 2018: a worldwide analysis. Lancet Glob Health. 2020; 8: e191-e203. IARC Scientific Publications. In: Vaccarella S, Lortet-Tieulent J, Saracci R, Conway DI, Straif K, Wild CP, editors. Reducing social inequalities in cancer: evidence and priorities for research. Lyon (FR): International Agency for Research on Cancer.© International Agency for Research on Cancer, 2019. For more information contact [email protected] .; 2019. Cohen PA, Jhingran A, Oaknin A, Denny L. Cervical cancer. Lancet. 2019; 393: 169-82. Matsuo K, Blake EA, Machida H, Mandelbaum RS, Roman LD, Wright JD. Incidences and risk factors of metachronous vulvar, vaginal, and anal cancers after cervical cancer diagnosis. Gynecologic Oncology. 2018; 150: 501-8. Schiffman M, Castle PE, Jeronimo J, Rodriguez AC, Wacholder S. Human papillomavirus and cervical cancer. Lancet. 2007; 370: 890-907. Martin D, Rödel F, Balermpas P, Rödel C, Fokas E. The immune microenvironment and HPV in anal cancer: Rationale to complement chemoradiation with immunotherapy. Biochim Biophys Acta Rev Cancer. 2017; 1868: 221-30. Nelson, E.L.; Stockdale, C.K. Vulvar and vaginal HPV disease. Obstet Gynecol Clin North Am 2013, 40, 359-376, doi:10.1016/j.ogc.2013.03.003. Slatko BE, Gardner AF, Ausubel FM. Overview of Next-Generation Sequencing Technologies. Curr Protoc Mol Biol. 2018; 122: e59. Kumar KR, Cowley MJ, Davis RL. Next-Generation Sequencing and Emerging Technologies. Semin Thromb Hemost. 2019; 45: 661-73. Al-Jumayli M, Brown SL, Chetty IJ, Extermann M, Movsas B. The Biological Process of Aging and the Impact of Ionizing Radiation. Semin Radiat Oncol. 2022; 32: 172-8. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008; 9: 559. Tian Z, He W, Tang J, Liao X, Yang Q, Wu Y, et al. Identification of Important Modules and Biomarkers in Breast Cancer Based on WGCNA. Onco Targets Ther. 2020; 13: 6805-17. Chen S, Yang D, Lei C, Li Y, Sun X, Chen M, et al. Identification of crucial genes in abdominal aortic aneurysm by WGCNA. PeerJ. 2019; 7: e7873. Wu C, Huang ZH, Meng ZQ, Fan XT, Lu S, Tan YY, et al. A network pharmacology approach to reveal the pharmacological targets and biological mechanism of compound kushen injection for treating pancreatic cancer based on WGCNA and in vitro experiment validation. Chin Med. 2021; 16: 121. Pyeon D, Newton MA, Lambert PF, den Boon JA, Sengupta S, Marsit CJ, et al. Fundamental differences in cell cycle deregulation in human papillomavirus-positive and human papillomavirus-negative head/neck and cervical cancers. Cancer Res. 2007; 67: 4605-19. Pappa KI, Polyzos A, Jacob-Hirsch J, Amariglio N, Vlachos GD, Loutradis D, et al. Profiling of Discrete Gynecological Cancers Reveals Novel Transcriptional Modules and Common Features Shared by Other Cancer Types and Embryonic Stem Cells. PLoS One. 2015; 10: e0142229. Xiao, Y.; Zhang, G.; Wang, L.; Liang, M. Exploration and validation of a combined immune and metabolism gene signature for prognosis prediction of colorectal cancer. Front Endocrinol (Lausanne) 2022, 13, 1069528, doi:10.3389/fendo.2022.1069528. Liontos M, Kyriazoglou A, Dimitriadis I, Dimopoulos MA, Bamias A. Systemic therapy in cervical cancer: 30 years in review. Crit Rev Oncol Hematol. 2019; 137: 9-17. Li H, Wu X, Cheng X. Advances in diagnosis and treatment of metastatic cervical cancer. J Gynecol Oncol. 2016; 27: e43. Singh U, Qureshi S, Negi N, Singh N, Goel M, Srivastava K. Comparative study between liquid-based cytology & conventional Pap smear for cytological follow up of treated patients of cancer cervix. Indian J Med Res. 2018; 147: 263-7. Wei C, Wang H, Xu F, Liu Z, Jiang R. LncRNA SOX21-AS1 is associated with progression of hepatocellular carcinoma and predicts prognosis through epigenetically silencing p21. Biomedicine & Pharmacotherapy. 2018; 104: 137-44. Nangraj AS, Selvaraj G, Kaliamurthi S, Kaushik AC, Cho WC, Wei DQ. Integrated PPI- and WGCNA-Retrieval of Hub Gene Signatures Shared Between Barrett's Esophagus and Esophageal Adenocarcinoma. Front Pharmacol. 2020; 11: 881. Wan Q, Tang J, Han Y, Wang D. Co-expression modules construction by WGCNA and identify potential prognostic markers of uveal melanoma. Exp Eye Res. 2018; 166: 13-20. Liu J, Li S, Lin L, Jiang Y, Wan Y, Zhou S, et al. Co-expression network analysis identified atypical chemokine receptor 1 (ACKR1) association with lymph node metastasis and prognosis in cervical cancer. Cancer Biomark. 2020; 27: 213-23. Wang J, Yi Y, Chen Y, Xiong Y, Zhang W. Potential mechanism of RRM2 for promoting Cervical Cancer based on weighted gene co-expression network analysis. Int J Med Sci. 2020; 17: 2362-72. The spike-and-slab lasso Cox model for survival prediction and associated genes detection, Bioinformatic Tang Z, Lei S, Zhang X, Yi Z, Guo B, Chen JY, et al. Gsslasso Cox: a Bayesian hierarchical model for predicting survival and detecting associated genes by incorporating pathway information. BMC Bioinformatics. 2019; 20: 94. Balkwill, F. Cancer and the chemokine network. Nat Rev Cancer 2004, 4, 540-550, doi:10.1038/nrc1388. Bandyopadhyay, S.; Zhan, R.; Chaudhuri, A.; Watabe, M.; Pai, S.K.; Hirota, S.; Hosobe, S.; Tsukada, T.; Miura, K.; Takano, Y.; et al. Interaction of KAI1 on tumor cells with DARC on vascular endothelium leads to metastasis suppression. Nat Med 2006, 12, 933-938, doi:10.1038/nm1444. Zhu W, Krishna S, Garcia C, Lin CJ, Mitchell BD, Scott KL, et al. Daam2 driven degradation of VHL promotes gliomagenesis. Elife. 2017; 6. Fang X, Zhang D, Zhao W, Gao L, Wang L. Dishevelled Associated Activator Of Morphogenesis (DAAM) Facilitates Invasion of Hepatocellular Carcinoma by Upregulating Hypoxia-Inducible Factor 1α (HIF-1α) Expression. Med Sci Monit. 2020; 26: e924670. Chen A, Liu Z, Wu Q, Li H. Disheveled-associated activator of morphogenesis 2 promotes invasion of colorectal cancer by activating PAK1 and promoting MMP7 expression. Genes Genomics. 2021; 43: 923-35. Li Z, Wei X, Zhu Y. The prognostic value of DAAM2 in lower grade glioma, liver cancer, and breast cancer. Clin Transl Oncol. 2023; 25: 2224-38. Chen L, Zhou J, Zhao Z, Zhu Y, Xing J, An J, et al. Low Expression of Phosphodiesterase 2 (PDE2A) Promotes the Progression by Regulating Mitochondrial Morphology and ATP Content and Predicts Poor Prognosis in Hepatocellular Carcinoma. Cells. 2022; 12. Li, S.Z.; Ren, K.X.; Zhao, J.; Wu, S.; Li, J.; Zang, J.; Fei, Z.; Zhao, J.L. miR-139/PDE2A-Notch1 feedback circuit represses stemness of gliomas by inhibiting Wnt/β-catenin signaling. Int J Biol Sci 2021, 17, 3508-3521, doi:10.7150/ijbs.62858. Zhao, Y.; Wang, Y.; Zhao, J.; Zhang, Z.; Jin, M.; Zhou, F.; Jin, C.; Zhang, J.; Xing, J.; Wang, N.; et al. PDE2 Inhibits PKA-Mediated Phosphorylation of TFAM to Promote Mitochondrial Ca(2+)-Induced Colorectal Cancer Growth. Front Oncol 2021, 11, 663778, doi:10.3389/fonc.2021.663778. Hiramoto K, Murata T, Shimizu K, Morita H, Inui M, Manganiello VC, et al. Role of phosphodiesterase 2 in growth and invasion of human malignant melanoma cells. Cell Signal. 2014; 26: 1807-17. Ding, H.; Xiong, X.X.; Fan, G.L.; Yi, Y.X.; Chen, Y.R.; Wang, J.T.; Zhang, W. The New Biomarker for Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma (CESC) Based on Public Database Mining. Biomed Res Int 2020, 2020, 5478574, doi:10.1155/2020/5478574. Ngwa, V.M.; Edwards, D.N.; Philip, M.; Chen, J. Microenvironmental Metabolism Regulates Antitumor Immunity. Cancer Res 2019, 79, 4003-4008, doi:10.1158/0008-5472.Can-19-0617. Binnewies M, Roberts EW, Kersten K, Chan V, Fearon DF, Merad M, et al. Understanding the tumor immune microenvironment (TIME) for effective therapy. Nat Med. 2018; 24: 541-50. De Jaeghere EA, Denys HG, De Wever O. Fibroblasts Fuel Immune Escape in the Tumor Microenvironment. Trends Cancer. 2019; 5: 704-23. Fortes C, Mastroeni S, Mannooranparampil TJ, Passarelli F, Zappalà A, Annessi G, et al. Tumor-infiltrating lymphocytes predict cutaneous melanoma survival. Melanoma Res. 2015; 25: 306-11. Mao X, Xu J, Wang W, Liang C, Hua J, Liu J, et al. Crosstalk between cancer-associated fibroblasts and immune cells in the tumor microenvironment: new findings and future perspectives. Mol Cancer. 2021; 20: 131. Burk RD, Chen Z, Saller C, Tarvin K, Carvalho AL, Scapulatempo-Neto C, et al. Integrated genomic and molecular characterization of cervical cancer. Nature. 2017; 543: 378-84. Ritchie, M.E.; Phipson, B.; Wu, D.; Hu, Y.; Law, C.W.; Shi, W.; Smyth, G.K. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015, 43, e47, doi:10.1093/nar/gkv007. Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021; 2: 100141. Engebretsen S, Bohlin J. Statistical predictions with glmnet. Clin Epigenetics. 2019; 11: 123. Zeng D, Ye Z, Shen R, Yu G, Wu J, Xiong Y, et al. IOBR: Multi-Omics Immuno-Oncology Biological Research to Decode Tumor Microenvironment and Signatures. Front Immunol. 2021; 12: 687975. Newman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015; 12: 453-7. Shen W, Song Z, Zhong X, Huang M, Shen D, Gao P, et al. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform. iMeta. 2022; 1: e36. Racle J, de Jonge K, Baumgaertner P, Speiser DE, Gfeller D. Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data. Elife. 2017; 6. Yoshihara, K.; Shahmoradgoli, M.; Martínez, E.; Vegesna, R.; Kim, H.; Torres-Garcia, W.; Treviño, V.; Shen, H.; Laird, P.W.; Levine, D.A.; et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 2013, 4, 2612, doi:10.1038/ncomms3612. Yang T, Hui R, Nouws J, Sauler M, Zeng T, Wu Q. Untargeted metabolomics analysis of esophageal squamous cell cancer progression. J Transl Med. 2022; 20: 127. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 05 May, 2026 Reviews received at journal 04 May, 2026 Reviewers agreed at journal 04 May, 2026 Reviewers agreed at journal 01 May, 2026 Reviewers invited by journal 16 Mar, 2026 Editor invited by journal 05 Mar, 2026 Editor assigned by journal 03 Mar, 2026 Submission checks completed at journal 28 Feb, 2026 First submitted to journal 28 Feb, 2026 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-8656950","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":607021821,"identity":"32e023ed-b55c-4972-beef-d1fe82257c16","order_by":0,"name":"Heng Kong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYBACCSBmhjCZDxz48IM0LWyJB2f2kKaFx/gwBxsRWiTbew+/Lqi4Y7fh+JkPhxl4GOT5xQ7g1yLNcy7NesaZZ8kbzuRuOFxgwWA4c3YCfi1yEjlmxrxth5MNDgC1zOBhSDC4TZSWf0At5988OMzDRoQWaYkc48e8DYftDG7kMBCnRbLnjBkzz7HDCZI3nhkAA1mCsF8kjvcYf+apOWzPdz758YcPP2zk+aUJaAECNlDcJC44ADGCoHIQYP4AJOzlG4hSPApGwSgYBSMRAABj90o2anD7ngAAAABJRU5ErkJggg==","orcid":"","institution":"Shenzhen Bao'an District Songgang People's Hospital, 518105, Guangdong, P.R. China.","correspondingAuthor":true,"prefix":"","firstName":"Heng","middleName":"","lastName":"Kong","suffix":""},{"id":607021822,"identity":"a406e884-c69f-4308-a2eb-513835b8b38e","order_by":1,"name":"XiaoXuan Kuang","email":"","orcid":"","institution":"Shenzhen Bao'an District Songgang People's Hospital, 518105, Guangdong, P.R. China.","correspondingAuthor":false,"prefix":"","firstName":"XiaoXuan","middleName":"","lastName":"Kuang","suffix":""},{"id":607021823,"identity":"1bf69897-7750-4664-b627-1693046c61ca","order_by":2,"name":"XiaoPeng Wu","email":"","orcid":"","institution":"Shenzhen Bao'an District Songgang People's Hospital, 518105, Guangdong, P.R. China.","correspondingAuthor":false,"prefix":"","firstName":"XiaoPeng","middleName":"","lastName":"Wu","suffix":""},{"id":607021824,"identity":"8d6bf78e-64a9-4ec6-9d8c-9391da6457f2","order_by":3,"name":"ChenQiong Gu","email":"","orcid":"","institution":"South China Normal University, 510630, Guangdong ,P. R. China.","correspondingAuthor":false,"prefix":"","firstName":"ChenQiong","middleName":"","lastName":"Gu","suffix":""}],"badges":[],"createdAt":"2026-01-21 08:15:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8656950/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8656950/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104887809,"identity":"fdb5465b-8653-45f2-a35a-2f0b4858920f","added_by":"auto","created_at":"2026-03-18 10:12:40","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":374142,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of DGEs in the GSE63678. \u003c/strong\u003e(A) Heatmap of 60 DEGsin GSE63678.(B) The down-regulated and up-gulated genes with | log2 (FoldChange) |\u0026gt; 1 with FDR \u0026lt;0.01.Green bar represents the number of downregulated genes. Red dots represents the number of up-regulated genes.\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/611b4c980b253ed1f1d1a3a8.jpg"},{"id":104887934,"identity":"a01dff9b-e9af-4589-a168-3820496baa03","added_by":"auto","created_at":"2026-03-18 10:12:57","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":272112,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConstruction of the co-expression network between CESC and normal samples in the GSE63678 dataset.\u003c/strong\u003e (A )Sample clustering was conducted to detect outliers. All samples are located in the clusters and pass the cutoff thresholds. (B) and (C) Soft-thresholding power analysis was used to obtain the scale-free fit index of network topology. (D) Hierarchical cluster analysis was conducted to detect co-expression clusters with corresponding color assignments. Each color represents a module in the constructed gene co-expression network by WGCNA. (E) Heatmap depicts the Topological Overlap Matrix (TOM) of genes selected for weighted co-expression network analysis.\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/5eb9e5b986c0f802a341c310.jpg"},{"id":104887937,"identity":"e7cf37a9-adf1-4e87-b459-5bbf17480025","added_by":"auto","created_at":"2026-03-18 10:12:58","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":152583,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe CESC important modules analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e(A)The relationship of two traits and 32 modules; (B) The scatterplot describing the relationship between MM and GS in lightgreen module;(C)Venn diagram of significant darkgreen module genes and GSE63678-DGEs.\u003c/p\u003e","description":"","filename":"Picture3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/dfe72c01d523570ade45afb7.jpg"},{"id":104888046,"identity":"64d55f0e-2edf-40d9-8abf-0833f7c6a721","added_by":"auto","created_at":"2026-03-18 10:13:14","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":195183,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFunctional Analysis of the Darkgreen Module genes.\u003c/strong\u003e (A) GO analysis of the genes involved in the Darkgreen module regarding biological process, cellular component, and molecular function. (B) KEGG analysis of genes involved in Darkgreenn module. The node size reflects the gene count, and the node color reflects the − log10 (FDR).\u003c/p\u003e","description":"","filename":"Picture4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/281494022477bc34540e95e4.jpg"},{"id":104887807,"identity":"6063b580-3877-429f-92dc-d0962604d962","added_by":"auto","created_at":"2026-03-18 10:12:40","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":123474,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIdentification of the prognostic genes. \u003c/strong\u003e(A) Boxplot of darkgreen module core genes expression in the TCGA database. (B) Unigenic Cox prognostic regression analysis of darkgreen module core genes. (C) Multigenic Cox prognostic regression analysis of darkgreen module core genes. (D, E) Lasso prognostic regression model. (F) Survival curves in the TCGA-CESC. (G) ROC curve to verify the accuracy of risk. (H) Risk curve, risk, and survival scatter plot. (I) Heatmap of darkgreen model gene expression in the high- and low-risk groups.\u003c/p\u003e","description":"","filename":"Picture5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/6edac569cdfb8792416995b3.jpg"},{"id":104887901,"identity":"46856759-ce6e-43f8-b4c7-ad334b274ba9","added_by":"auto","created_at":"2026-03-18 10:12:54","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":353719,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGSEA enrichment analysis. \u003c/strong\u003eThe top ten pathways with the most significant GSEA enrichment results for ACKR1 (A), DAAM2 (B), and PDE2A (C).\u003c/p\u003e","description":"","filename":"Picture6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/a11e92f7ae888343187e18e9.jpg"},{"id":104887958,"identity":"44403562-7b06-414f-a4f5-86e24a908265","added_by":"auto","created_at":"2026-03-18 10:13:04","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":447247,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig.6. Immune relevance of the prognostic genes. \u003c/strong\u003eHistogram of the multifarious immune cell contents in the GEO (A) and TCGA (B) database. Graph of correlations between the levels of multifarious immune cells of GEO (C) and TCGA (D) data-set. (E, F, G)The map of the linear regression between expression levels of multifarious immune checkpoints and prognostic gene expression. (H, I, J) Linear regression plots of immune infiltration scores and prognostic gene expression relationships in different samples. (K) Correlation plot of three prognostic genes and immune cells (calculated using EPIC method).\u003c/p\u003e","description":"","filename":"Picture7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/c5dc96cc324c1c6587c1a8d4.jpg"},{"id":104887895,"identity":"66c0dbe2-178b-41e4-b1e9-c568ee16f81e","added_by":"auto","created_at":"2026-03-18 10:12:53","extension":"jpg","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":247346,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eFig.7.Differential expression of core prognostic genes in survival curves in GEPIA and in different data. \u003c/strong\u003e(A) KM survival curves for ACKR1 in CESC in GEPIA. (B)KM survival curves for DAAM2 in CESC in GEPIA. (C) KM survival curves for PDE2A in CESC in GEPIA.(D)Expression of the three prognostic genes in TCGA, HPA, and GTEx.\u003c/p\u003e","description":"","filename":"Picture8.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/22fb820fe84f21b4a81ef4ea.jpg"},{"id":105034474,"identity":"8de06133-3423-4e7f-b8a9-befcdef3b0a8","added_by":"auto","created_at":"2026-03-20 07:23:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3186204,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8656950/v1/3a62eab9-3f4c-488f-8a15-16a816d08836.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"WGCNA and LASSO Cox identify ACKR1 DAAM2 and PDE2A as prognostic genes and immune related biomarkers in cervical squamous cell carcinoma and endocervical adenocarcinoma","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eCESC is a common malignancy and is the second leading cause of cancer death in women [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e],CESC is a preventable, but still the leading cause of cancer death among women in 36 low-and middle-income countries[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].The early symptoms are more insidious, but the increasing volume of the tumor will compress and invade the surrounding organs and tissues, and induce the corresponding symptoms.When found, more has progressed to the middle and late stage, and even lost the opportunity of surgical treatment, the clinical principle of CESC to early diagnosis, early treatment. Early through surgical treatment, resection of focal tissue, preventing tumor cell spread or distant metastasis, improve overall survival. With advances in radiotherapy techniques such as intensity-modulated radiotherapy, treatment-related toxicity is reduced in women with locally advanced disease.But for women with metastatic or recurrent disease, the overall prognosis remains poor[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e],The recurrence and metastasis of CESC is still a difficult problem in diagnosis and treatment. The development of sensitive diagnostic methods and finding new treatment strategies are of great significance for the prevention and treatment of CESC[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHPV is the main pathogenic cause of CESC, but HPV infection is associated not only with the CESC, but also with other anogenital cancers, including [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].For example, 25\u0026ndash;40% of vulvar cancers, approximately 66% of vaginal cancers, and 70\u0026ndash;90% of anal cancers are associated with HPV infection in [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWith the deepening of research, it has been found that the interaction between various cytokines and chemokines forms a large and complex molecular biology network during the development of CESC. However, most studies address only one aspect of pathogenesis, ignoring the complex interactions between multiple factors in an organism. With the rapid development of high-throughput sequencing technologies, many sequencing technologies such as genomics, transcriptomics and proteomics have been widely used in the study of multifactor complex diseases such as cancer and cardiovascular disease[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e].In order to obtain critical information from a large amount of histological data and to study the interaction between critical information, various biological network-based integrated analysis techniques [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] were developed. Common ones are gene regulatory networks, protein interaction networks, signaling, etc.\u003c/p\u003e \u003cp\u003eThe co-expression network [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] was constructed via WGCNA for finding modules of highly correlated genes using module eigengenes and associated with external sample traits using the eigengene network approach. Related networks facilitate network-based genetic screening approaches for identifying candidate biomarkers or therapeutic targets. Applications in complex disease research are increasingly widespread in [\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].Our study identified 14 modules and 1 key gene as important components of CESC etiology, which may improve our understanding of the molecular mechanisms, such as: pathway networks, metabolic networks, and gene co-expression networks.\u003c/p\u003e \u003cp\u003eIn this study, we constructed a weighted gene expression network based on gene expression data from the GEO database [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], which contains gene expression matrices of CESC cancer tissues and adjacent tissues.We used co-expression network and other related bioinformatics methods, using the expression matrix in TCGA database and clinical survival information to explore the module and hub genes related to CESC prognosis, in order to obtain potential markers closely related to CESC prognosis. Finally, we obtained a prognostic regression model composed of three genes to study the role of core model genes in gene set enrichment analysis (GSEA), tumorigenesis, progression and immune response. And predicted the correlation between core prognostic model gene expression and immune infiltration and immune checkpoints.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"2. Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1.Sample collection\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eAccording to the inclusion criteria[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]:(a) complete gene expression information (b) complete survival information, exclusion criteria: (a) incomplete gene expression (b) no survival time and survival status in the clinical information. Transcriptomic data and clinical data from 306 CESC tissues and three adjacent tissues were obtained from the TCGA database, including overall survival time, survival status, age, and sex. Transcriptomic data for 18 tumors and 17 normal samples were obtained from GSE63678 (Table S2)\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Identification of DGEs in the GSE63678 dataset\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the identification of the DGEs in the GSE63678 data set.When the threshold was | log2 (fold-change) |\u0026gt; 1 and FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.01,504 genes were identified, including 306 downregulated genes and 198 up-regulated genes.The top 30 genes with the most significant expression differences were selected for the P-value and differential expression heatmaps and the top 10 significantly differential expression volcano plots, respectively.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3. Construction of the co-expression network between CESC and normal samples in the GSE63678 dataset.\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the construction of the CESC and normal samples coexpression network in the GSE63678 dataset.The analysis performed by WGCNA was extended to include the gene dataset GSE63678, excluding the top 50% of genes with minimal MAD and removing outlying genes and samples using the goodSamplesGenes method. All samples were checked and 100 was selected as the height cutoff to draw the sample cluster heatmap (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), \u0026ldquo;sft\u003cspan\u003e$\u003c/span\u003epowerEstimate\u0026rdquo; was utilized in order to ascertain the soft-power threshold, and the value of 16(scale independence of 0.86) was selected for the purpose of carrying out additional research on CESC as well as normal.(Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB-\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC)We use the one-step network building function of the WGCNA R package to build the gene networks and find the modules within them. For cluster segmentation, the soft threshold power was set to 16, and the minimum module gene was set to 30, and the sensitivity was set to: 4 (which means high sensitivity). In addition, we merged the distance of less than. 25, the genes finally obtained 13 co-expression modules (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e-D).According to the results of this analysis, the high independence between the modules in terms of gene expression. Further analysis of the feature vector (2-E) showed that the Darkgreen module (including 165 genes) had the highest correlation with CESC (r = -0.80, P\u0026thinsp;=\u0026thinsp;1.3e-8) and was confirmed as a significant module (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). The Darkgreen module mean and gene significance (GS) have a highly significant correlation (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB).The Darkgreen module genes (46 genes) were selected by MM\u0026thinsp;\u0026gt;\u0026thinsp;0.80 and GS\u0026thinsp;\u0026gt;\u0026thinsp;0.65, and crossed with the differentially expressed 504 genes, yielding 38 key genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC).\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4. Functional enrichment analysis\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e shows the functional analysis of the Darkgreen module.Genes included in the Darkgreen module were analysed using GO and KEGG (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA-B). The GO enrichment results showed that Darkgreen module genes were enriched in the functions of \u0026ldquo; locomotion\u0026rdquo;, \u0026ldquo;negative regulation of cellular process\u0026rdquo;, \u0026ldquo;movement of cell or subcellular component\u0026rdquo;, \u0026ldquo;cell motility、localization of cell\u0026rdquo;, \u0026ldquo;cell migration\u0026rdquo;, \u0026ldquo;cell adhesion\u0026rdquo;, \u0026ldquo;biological adhesion\u0026rdquo;, \u0026ldquo;negative regulation of dendritic cell apoptotic process\u0026rdquo;, \u0026ldquo;dendritic cell apoptotic process\u0026rdquo;, \u0026ldquo;elastic fiber\u0026rdquo;, \u0026ldquo;extracellular matrix\u0026rdquo;, \u0026ldquo;cell junction\u0026rdquo;, \u0026ldquo;Wnt signalosome\u0026rdquo;, \u0026ldquo;cell-cell junction\u0026rdquo;, \u0026ldquo;extracellular matrix structural constituent\u0026rdquo;, \u0026ldquo;Roundabout binding\u0026rdquo;, \u0026ldquo;structural molecule activity\u0026rdquo;, \u0026ldquo;ion channel binding\u0026rdquo;, \u0026ldquo;cytokine binding\u0026rdquo;, \u0026ldquo;Wnt-protein binding\u0026rdquo;, \u0026ldquo;phosphatidylinositol 3-kinase binding\u0026rdquo;, \u0026ldquo;phosphatidylserine binding\u0026rdquo;, \u0026ldquo;protein binding, bridging\u0026rdquo;, \u0026ldquo;phosphoric ester hydrolase activity\u0026rdquo; and so on. KEGG enrichment results show that the main signal pathways of darkgreen module genes are located at \u0026ldquo;Axon guidance\u0026rdquo;, \u0026ldquo;Leukocyte transendothelial migration\u0026rdquo;, \u0026ldquo;Malaria\u0026rdquo;, \u0026ldquo;Cell adhesion molecules (CAMs)\u0026rdquo;, \u0026ldquo;EGFR tyrosine kinase inhibitor resistance\u0026rdquo;, \u0026ldquo;Proteoglycans in cancer\u0026rdquo;, \u0026ldquo;Aldosterone synthesis and secretion\u0026rdquo;, and \u0026ldquo;Viral protein interaction with cytokine and cytokine receptor\u0026rdquo;.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003e(A)The relationship of two traits and 32 modules; (B) The scatterplot describing the relationship between MM and GS in lightgreen module;(C)Venn diagram of significant darkgreen module genes and GSE63678-DGEs.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5. Identification of key genes\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eFor the 38 core network genes, the differential expression boxplot (Figure.5A) was drawn using expression profile data and clinical information from the TCGA database, where the differential expression of ATF6B and EXOSC7 in tumor and normal tissues was not statistically significant.Next, we performed univariate Cox prognosis regression analysis (Figure.5B) and multivariate COX prognosis regression analysis (Figure.5C) on the genes with significant differences to obtain three prognostic genes, The relationship between the expression level of the three core genes and the clinical survival rate was subsequently analyzed by integrating the survival time, survival status and gene expression data by the Lasso-Cox prognostic regression model (Figure.5D-E).\u003c/p\u003e \u003cp\u003eIn addition, we also set a 10-fold cross-validation to obtain the optimal model. Further identify genes associated with CESC prognosis. When the Lambda value was 0.0651047237417135, three genes were finally obtained:PDE2A, DAAM2 and ACKR1,Coef (PDE2A)\u0026thinsp;=\u0026thinsp;0.0862991933466534,Coef(DAAM2)\u0026thinsp;=\u0026thinsp;0.0166165369047536, Coef (ACKR1)= -0.0213422619729048,The risk scoring formula for this sample is as follows:RiskScore\u0026thinsp;=\u0026thinsp;0.0862991933466534*PDE2A\u0026thinsp;+\u0026thinsp;0.0166165369047536*DAAM2-0.0213422619729048*ACKR1.Risk scores for all samples according to the risk score formula, Risk scores were calculated by combining the survival status and survival time of the samples, The best cut-off value for RiskScore was calculated using the R package maxstat, Set the minimum number of grouped samples greater than 25%, Maximum sample number was\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003egrouped by less than 75%, The best cutoff value is: 0.010479882817014, Based on this, the patients were divided into high and low two groups, (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF), The differences between the two groups were further analyzed using the survfit function of the R package survival, The significance of prognostic differences between different groups was assessed using the logrank test method, Eventually we observed a significant prognostic difference (P\u0026thinsp;\u0026lt;\u0026thinsp;0.001), Patients in the high-risk group may die earlier than low patients, The survival status is also worse, The prognosis differences between the two groups were statistically significant.We drew the ROC curves (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG) to verify the accuracy and reliability of the predictability of the Lasso-Cox prognostic regression model. We performed the ROC analysis to obtain AUC using the R package pROC (version 1.17.0.1). Specifically, 1,3 and 5 years of ROC using the RUC function of pROC and the AUC value of final AUC results, demonstrating that the Lasso-Cox prognostic regression curve was reliable.We generated risk curve, risk, survival scatter plot (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH) and heat map of model gene expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eI), we analyzed the relationship between different risk scores and follow-up time, survival status and expression of each gene, observed significant decrease in survival with increasing risk scores (Figure.5H), as expected, the ACKR1 gene was a protective factor, decreased expression with increasing risk scores. PDE2A and DAAM2 genes are risk factors and showed a trend of up-regulated expression with increasing risk scores.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e2.6. Enrichment analysis of the GSEA cellular pathway model genes\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eTo investigate the significantly associated pathways in the Lasso-Cox prognostic regression model, we performed a GSEA pathway-related enrichment analysis for the three prognostic genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e (A) -(C)).ACKR1 enriched out of a significant correlation of 35 pathways;There were 58 pathways significantly associated with DAAM2; PED2A has 41 significantly associated pathways.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e2.7. Effect of model genes on tumor immunity.\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eNext, we determined whether the genes in the Lasso-Cox prognostic regression model affected the immune response.We drew the histograms of the various immune cell contents in the samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (A-B)) and the correlation between the levels of the various immune cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (C-D)). Correlation Figure C shows that, in the GEO dataset, B_cells_naive and T_cells_CD4_memory_activated had the highest correlation (r\u0026thinsp;=\u0026thinsp;0.75), Mast_cells_resting and Mast_cells_activated present a significant negative correlation(r=-0.56). Correlation Figure D shows that, in the TCGA dataset,\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv description=\"F6-2\" class=\"Drawing\" id=\"2\" name=\"图片 2\"\u003e\u003c/div\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eB_cells_naive and Plasma_cellshad the highest correlation (r\u0026thinsp;=\u0026thinsp;0.54), T_cells_CD8 and T_cells_CD4_memory_restingpresent a significant negative correlation (r=-0.49), Secondly, we used the analysis of immune cell expression immune checkpoint gene in the online tool sangerbox and immune infiltration analysis to study the correlation between the three prognostic genes and immune cell expression, immune infiltration and immune checkpoint, and draw the correlation map (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (H) -5 (Q)).as shown in\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003ethe figure, ACKR1 had a significant correlation with Bcells, CAFs, Endothelial, and Macrophages and otherCells (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05); DAAM2 was significantly associated with Bcells,\u003c/p\u003e \u003cp\u003eCAFs, and Endothelial and otherCells (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05); PDE2A was significantly associated with Bcells, CAFs, CD4_Tcells, Endothelial, Macrophages and otherCells (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05): pearson correlation was calculated for marker genes of ACKR1, DAAM2 and PDE2A, and five classes of immune pathways.The expression of prognostic genes is closely related to immune checkpoint gene expression (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (R) -5 (V)). The expression of the prognostic genes ACKR1 and DAAM2 and PDE2A were significantly and positively correlated with the immune infiltration scores of the samples, with all differences being statistically significant (P\u0026thinsp;\u0026lt;\u0026thinsp;0.05).Thus, we concluded that the three prognostic genes in the Lasso-Cox prognostic regression model were associated with immune cells and significantly positively associated with immune checkpoints.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.8. Expression of prognostic model genes in various databases\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn this study, the expression of three prognostic model genes was selected in TCGA-CESC, GTEx and HPA database. As shown in the figure, the three prognostic model genes were significantly low expressed in tumor tissue, consistent with the expression in GSE63678 dataset, which suggests that agonists of these three genes may be potential targets for the treatment of CESC.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"3. Discussion","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eThe etiology of CESC is complex, and the clinical diagnosis is still not completely clear. The traditional treatment methods are mainly surgical resection, radiotherapy and chemotherapy. The sensitivity of different patients to radiotherapy and chemotherapy varies greatly, and the possible [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], recurrence and metastasis will still occur even after patients receive the above treatment. The recurrence and metastasis of CESC are currently the main causes of CESC death. Although cervical biopsy is the gold standard for the diagnosis of CESC, it is somewhat traumatic and is not suitable for early screening of [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. There is an urgent need to find molecular markers of CESC.\u003c/p\u003e \u003cp\u003eWGCNA is a method to identify gene modules and key hub genes associated with phenotypic traits [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] and classifying them into modules according to their different expression patterns [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].Since genes with the same expression pattern are largely regulated by the same or similar factors, they are likely to perform the same or similar biological functions [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e].Compared with the traditional co-expression network analysis, WGCNA overcomes the previous co-expression network dividing the associations between continuous variables such as gene expression data as related or unrelated simple binary relationships, thus avoiding the loss of important information [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e].The co-expression network constructed by WGCNA method needs to meet the scale-free network distribution standard. By setting the soft threshold, it meets the objective fact that a few molecules play an important role in various biological life activities [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. WGCNA found that ACKR1 [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] and RRM2 [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] gene may be related to lymph node metastasis and prognosis in CESC.However, the use of a single bioinformatics analysis method may cause excessive interference data and affect the accuracy of the results. LASSO Regression is to use the compression estimation method to construct the compression function and shrink the partial regression coefficient to zero, which is one of the main methods for dealing with multicolynity data[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e],Therefore, the joint analysis of WGCNA and LASSO-COX model increases the accuracy and authenticity of the research results.\u003c/p\u003e \u003cp\u003eIn this study, we screened three genes highly negatively associated with CESC: ACKR1, DAAM2 and PED2A, Atypical chemokine receptoR1 (ACKR1) as key regulators of chemokines involved in combining inflammatory responses and cancer proliferation, angiogenesis and metastasis. Chemokine patterns and chemokine receptor signaling are an integral part of tumor cell proliferation and spread [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e],It has been shown that ACKR1 prevents tumor angiogenesis and subsequent metastasis when it is expressed on malignant cells.It is possible that ACKR1 promotes cell cycle regulation through other interactions, including tumor suppressor CD82 / KAI 1, direct interaction of CD82 with ACKR1 leading to p21 cyclin-dependent kinase inhibition and prevention of metastasis escape [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e], and studies also predict that ACKR1 is associated with cervical lymph node metastasis and prognosis [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eDAAM2 is involved in the tumorigenesis and progression of human cancers and seems to play different roles in different types of cancers, such as DAAM2 that accelerates the progression of glioma and hepatocellular carcinoma [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. DAAM2 promotes the invasion of CRC cells and plays an important role in CRC invasion [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].Overexpression of DAAM2 was found in BRCA tissues, and the knockdown of DAAM2 delayed the proliferation, invasion, and migration of BRCA cells. However, the high expression level of DAAM2 has a higher survival rate in LGG and LIHC [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], and phosphodiesterase 2 (PDE2A) regulates the level of cAMP / cGMP, which is closely associated with various types of tumor progression.Tumors with low PDE2A expression show decreased immune function, PDE2A closely involved in HCC proliferation and metastasis [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] PDE2A is significantly downregulated in glioma, and overexpression slows the progression of glioma [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].It plays a key role in the progression of cancers such as colorectal cancer [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e] and melanoma [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. PDE2A may be a biomarker for early diagnosis and prognostic evaluation in patients with CESC [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eThe development of tumors is affected by many biological and behavioral characteristics, including the influence of the tumor immune microenvironment. TME is a key factor in tumor growth, metastasis and regulation of tumor immune response [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].Various components of the tumor microenvironment not only play important roles in tumor progression, immune escape and metastasis, but also have profound effects on the therapeutic effect of patients in [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. For example, immunosuppressive cells within the tumor microenvironment play a key role in promoting tumor immune escape and promoting local suppression of antitumor immune responses by the release of immunosuppressive cytokines [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Similarly, the expression level of tumor-infiltrating lymphocytes is usually correlated with survival in CESC as well as in patients with other solid tumors [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/p\u003e \u003cp\u003eIn this study, we used WGCNA to select Darkgreen module genes with significant correlation with CESC, including 46 significantly associated genes, performed limma differential analysis of the data set in GSE63678 in GEO database, and 506 differentially expressed genes were obtained, and the intersection of the two genes to obtain 38 key genes,Using LASSO-COX for 38 key genes, three hub genes, ACKR1 and DAAM2 and PDE2A, were obtained to explore their expression in CESC and normal cervical tissue, and preliminarily analyzed the relationship with the immune infiltration level of TME of CESC, which may provide some reference for immunotherapy of CESC.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"4. Materials and Methods","content":"\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e4.1. Data sources and searches\u003c/h2\u003e \u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn this study, the transcriptomic data and clinical data were obtained from the TCGA-CESC [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e] and GEO databases. The TCGA clinical data included clinically relevant information on overall survival time, survival status, age, sex, and disease locus. On the GEO website, we obtained the raw data for the GSE63678[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e] dataset, and the limma package [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e] was used for the statistical analysis of the data, with a p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.01, FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.01, | log2FC |\u0026ge; 1.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e4.2. WGCNA network construction and module identification\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eFirst, We used the gene expression profiling, We calculated the Median Absolute Deviation (MAD) for each gene separately, Excluding the top 50% of genes with minimal MAD, Outlier genes and samples were removed using the goodSamplesGenes method of the R package WGCNA, Further construction of scale-free co-expression network using WGCNA, Soft thresholds of the co-expression network were calculated using the 'pickSoftThreshold()' function.When the soft threshold is equal to 4, the co-expression network is more close to the scale-free network. The weighted adjacency matrix is constructed to conduct hierarchical clustering based on the phase difference (1-TOM) of topological overlap matrix (TOM) to construct related modules. After linking the module with the clinical characteristics data, the scatter plot of the module membership (MM) and the gene significance (GS) are drawn to clarify the significance of the genes within the module.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e4.3. Functional enrichment analysis\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eR package \u0026ldquo;cluster Profiler\u0026rdquo;[\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e] was used to obtain results for gene set enrichment. Set a minimum gene set to 5 and a maximum gene set to 5000, P value of \u0026lt;\u0026thinsp;0.05 and a FDR of \u0026lt;\u0026thinsp;0.25 were considered statistically significant. We performed KEGG and GO enrichment analysis on the genes extracted from the modules with the most significant correlation with CESC, in order to explore the function of the module genes, the involved biological pathways and the localization in the cells.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e4.4. Identification of key genes\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe prognostic value of the core genes was determined by univariate Cox regression analysis, and a P\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.In this study, we used the R package glmnet[\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e], integrating survival time, survival status and gene expression data for regression analysis using the Lasso-cox method. The survival time, survival status and gene expression data were integrated for regression analysis using the lasso-cox method. Kaplan-Meier survival curves were generated to evaluate the predictive performance of associated risk genes. We performed ROC analysis to obtain AUC using our R software package pROC. Specifically, we obtained patient survival time, survival status, and risk score, performed ROC analysis of 365,1095,1825 days ROC analysis using the ROC function of pROC and evaluated AUC and confidence interval using the ci function of pROC to obtain final AUC results to verify the performance of model prediction.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.5.Correlation of prognostic models with tumor immunity.\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eIBOR [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e] is a computational tool for immune tumor biology research. In order to explore the influence and correlation of the expression of prognostic genes in Lasso regression prediction model, we selected the CIBERSORT [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e] algorithm with the R package IBOR based on the expression profile of CESC obtained in TCGA and the expression profile of GSE63678, and plotted the bar chart and correlation chart of immune cell content. In addition, we used the online analysis tool SangerBox[\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e] to map the expression of prognostic genes and various immune cells, explore the relationship between the expression of prognostic genes and immunotherapy, and map the correlation between prognostic genes and immune checkpoint.The EPIC [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e] selected Bcells, CAFs, CD4_Tcells, CD8_Tcells, Endothelial, Macrophages, NKcells, otherCells infiltration scores using the R package IOBR, and the immune infiltrating cells scored for each sample using ESTIMATE[\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e].\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e4.6 Enrichment analysis of the GSEA cellular pathway model genes\u003c/h2\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eTo investigate the association of tumor-related and immune-related pathway genes in the Lasso-Cox prognostic regression model, we performed GSEA[\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e] pathway-related enrichment analysis for prognostic genes, and all pathways were screened with P\u0026thinsp;\u0026lt;\u0026thinsp;0.01 significant difference and FDR\u0026thinsp;\u0026lt;\u0026thinsp;0.25 to identify pathways significantly associated with prognostic genes (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e (A) -5 (C)).\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003e \u003cdiv class=\"BlockQuote\"\u003e \u003cp\u003eIn conclusion, in this study, ACKR1, a hub gene related with DAAM2 and PDE2A prognosis, three prognostic genes were significantly low expression in CESC, explored the correlation between three genes with immune cells and immune checkpoint genes, and three prognostic related hub genes were positively correlated with the expression of some immune checkpoint genes, and showed significant correlation with immune infiltration.These results suggest that the three prognostic molecular markers screened are associated with the immune infiltration level of CESC TME, which provides guidance for the prognosis and immunotherapy of CESC, this model has a useful for predicting the effect of CESC clinical immunotherapy This study is a retrospective bioinformatics analysis based on public transcriptomic datasets.\u003c/p\u003e \u003cp\u003eFurthermore, this study also has certain limitations. The TCGA-CESC cohort included 306 tumor samples and 3 adjacent samples, and the limited number of adjacent samples may affect the robustness of differential expression interpretation. The external validation dataset GSE63678 included 18 tumor and 17 normal samples, and the relatively small sample size and cross platform differences may introduce bias. In addition, the prognostic model and immune related findings were primarily derived from computational inference and correlation analyses and were not validated in an independent prospective cohort or experimentally confirmed in cellular, animal, or clinical samples. Therefore, our results should be interpreted as hypothesis generating, and further independent cohort validation and mechanistic experiments are warranted.\u003c/p\u003e \u003c/div\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eSupplementary Materials: \u003c/strong\u003eThe following supporting information can be downloaded at: , Figure S1; Table S1 \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions:\u003c/strong\u003e The following statements should be used \u0026ldquo;Conceptualization, H.K. and C.G.; methodology, H.K.; software, C.G.; validation, H.K., C.G. and X.K.; formal analysis, C.G.; investigation, H.K.; resources, C.G.; data curation, H.K.; writing\u0026mdash;original draft preparation, H.K. and C.G.; writing\u0026mdash;review and editing, H.K.; visualization, C.G.; supervision, X.K. and X.W.; project administration, H.K.; funding acquisition, C.G. All authors have read and agreed to the published version of the manuscript.\u0026rdquo; Please turn to the CRediT taxonomy for the term explanation. Authorship must be limited to those who have contributed substantially to the work reported.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003eThis research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eInformed Consent Statement: \u003c/strong\u003eThis study used information in public databases, which excluded informed consent forms.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability Statement:\u003c/strong\u003e The transcriptomic and clinical data analyzed in this study were obtained from the following public resources with explicit identifiers and direct URLs to ensure transparency and reproducibility: (1) the TCGA Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma cohort (project ID: TCGA-CESC, https://portal.gdc.cancer.gov/projects/TCGA-CESC ) from the NCI Genomic Data Commons (GDC); (2) the GEO external validation dataset GSE63678 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE63678 ); (3) GTEx datasets (https://www.gtexportal.org/home/datasets ) used as normal tissue references; (4) the Human Protein Atlas (HPA, https://www.proteinatlas.org/ ) used for expression cross validation; (5) GEPIA2 (https://gepia2.cancer-pku.cn/ ) used for survival verification (integrating TCGA and GTEx); and (6) SangerBox (https://sangerbox.com/ ) used for immune related visualization analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e In this section, Thanks to my dear mentor for my great support.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declaration:\u003c/strong\u003e Not applicable. The authors declare that all bioinformatics analyses in this study comply with the ethical standards of the relevant institutional research committees, the 1964 Declaration of Helsinki and its subsequent amendments, as well as the data usage policies of the public databases employed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest:\u003c/strong\u003eThe authors declare no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to publish:\u003c/strong\u003eNot applicable. This study is a bioinformatics data analysis. All data are from public or de-identified datasets without personally identifiable information, and no individual participants are involved, so consent to publish is not applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003ePaskeh, M.D.A.; Mirzaei, S.; Gholami, M.H.; Zarrabi, A.; Zabolian, A.; Hashemi, M.; Hushmandi, K.; Ashrafizadeh, M.; Aref, A.R.; Samarghandian, S. Cervical cancer progression is regulated by SOX transcription factors: Revealing signaling networks and therapeutic strategies. Biomed Pharmacother 2021, 144, 112335, doi:10.1016/j.biopha.2021.112335.\u003c/li\u003e\n\u003cli\u003eArbyn M, Weiderpass E, Bruni L, de Sanjos\u0026eacute; S, Saraiya M, Ferlay J, et al. Estimates of incidence and mortality of cervical cancer in 2018: a worldwide analysis. Lancet Glob Health. 2020; 8: e191-e203.\u003c/li\u003e\n\u003cli\u003eIARC Scientific Publications. In: Vaccarella S, Lortet-Tieulent J, Saracci R, Conway DI, Straif K, Wild CP, editors. Reducing social inequalities in cancer: evidence and priorities for research. Lyon (FR): International Agency for Research on Cancer.\u0026copy; International Agency for Research on Cancer, 2019. For more information contact [email protected].; 2019.\u003c/li\u003e\n\u003cli\u003eCohen PA, Jhingran A, Oaknin A, Denny L. Cervical cancer. Lancet. 2019; 393: 169-82.\u003c/li\u003e\n\u003cli\u003eMatsuo K, Blake EA, Machida H, Mandelbaum RS, Roman LD, Wright JD. Incidences and risk factors of metachronous vulvar, vaginal, and anal cancers after cervical cancer diagnosis. Gynecologic Oncology. 2018; 150: 501-8.\u003c/li\u003e\n\u003cli\u003eSchiffman M, Castle PE, Jeronimo J, Rodriguez AC, Wacholder S. Human papillomavirus and cervical cancer. Lancet. 2007; 370: 890-907.\u003c/li\u003e\n\u003cli\u003eMartin D, R\u0026ouml;del F, Balermpas P, R\u0026ouml;del C, Fokas E. The immune microenvironment and HPV in anal cancer: Rationale to complement chemoradiation with immunotherapy. Biochim Biophys Acta Rev Cancer. 2017; 1868: 221-30.\u003c/li\u003e\n\u003cli\u003eNelson, E.L.; Stockdale, C.K. Vulvar and vaginal HPV disease. Obstet Gynecol Clin North Am 2013, 40, 359-376, doi:10.1016/j.ogc.2013.03.003.\u003c/li\u003e\n\u003cli\u003eSlatko BE, Gardner AF, Ausubel FM. Overview of Next-Generation Sequencing Technologies. Curr Protoc Mol Biol. 2018; 122: e59.\u003c/li\u003e\n\u003cli\u003eKumar KR, Cowley MJ, Davis RL. Next-Generation Sequencing and Emerging Technologies. Semin Thromb Hemost. 2019; 45: 661-73.\u003c/li\u003e\n\u003cli\u003eAl-Jumayli M, Brown SL, Chetty IJ, Extermann M, Movsas B. The Biological Process of Aging and the Impact of Ionizing Radiation. Semin Radiat Oncol. 2022; 32: 172-8.\u003c/li\u003e\n\u003cli\u003eLangfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008; 9: 559.\u003c/li\u003e\n\u003cli\u003eTian Z, He W, Tang J, Liao X, Yang Q, Wu Y, et al. Identification of Important Modules and Biomarkers in Breast Cancer Based on WGCNA. Onco Targets Ther. 2020; 13: 6805-17.\u003c/li\u003e\n\u003cli\u003eChen S, Yang D, Lei C, Li Y, Sun X, Chen M, et al. Identification of crucial genes in abdominal aortic aneurysm by WGCNA. PeerJ. 2019; 7: e7873.\u003c/li\u003e\n\u003cli\u003eWu C, Huang ZH, Meng ZQ, Fan XT, Lu S, Tan YY, et al. A network pharmacology approach to reveal the pharmacological targets and biological mechanism of compound kushen injection for treating pancreatic cancer based on WGCNA and in vitro experiment validation. Chin Med. 2021; 16: 121.\u003c/li\u003e\n\u003cli\u003ePyeon D, Newton MA, Lambert PF, den Boon JA, Sengupta S, Marsit CJ, et al. Fundamental differences in cell cycle deregulation in human papillomavirus-positive and human papillomavirus-negative head/neck and cervical cancers. Cancer Res. 2007; 67: 4605-19.\u003c/li\u003e\n\u003cli\u003ePappa KI, Polyzos A, Jacob-Hirsch J, Amariglio N, Vlachos GD, Loutradis D, et al. Profiling of Discrete Gynecological Cancers Reveals Novel Transcriptional Modules and Common Features Shared by Other Cancer Types and Embryonic Stem Cells. PLoS One. 2015; 10: e0142229.\u003c/li\u003e\n\u003cli\u003eXiao, Y.; Zhang, G.; Wang, L.; Liang, M. Exploration and validation of a combined immune and metabolism gene signature for prognosis prediction of colorectal cancer. Front Endocrinol (Lausanne) 2022, 13, 1069528, doi:10.3389/fendo.2022.1069528.\u003c/li\u003e\n\u003cli\u003eLiontos M, Kyriazoglou A, Dimitriadis I, Dimopoulos MA, Bamias A. Systemic therapy in cervical cancer: 30 years in review. Crit Rev Oncol Hematol. 2019; 137: 9-17.\u003c/li\u003e\n\u003cli\u003eLi H, Wu X, Cheng X. Advances in diagnosis and treatment of metastatic cervical cancer. J Gynecol Oncol. 2016; 27: e43.\u003c/li\u003e\n\u003cli\u003eSingh U, Qureshi S, Negi N, Singh N, Goel M, Srivastava K. Comparative study between liquid-based cytology \u0026amp; conventional Pap smear for cytological follow up of treated patients of cancer cervix. Indian J Med Res. 2018; 147: 263-7.\u003c/li\u003e\n\u003cli\u003eWei C, Wang H, Xu F, Liu Z, Jiang R. LncRNA SOX21-AS1 is associated with progression of hepatocellular carcinoma and predicts prognosis through epigenetically silencing p21. Biomedicine \u0026amp; Pharmacotherapy. 2018; 104: 137-44.\u003c/li\u003e\n\u003cli\u003eNangraj AS, Selvaraj G, Kaliamurthi S, Kaushik AC, Cho WC, Wei DQ. Integrated PPI- and WGCNA-Retrieval of Hub Gene Signatures Shared Between Barrett\u0026apos;s Esophagus and Esophageal Adenocarcinoma. Front Pharmacol. 2020; 11: 881.\u003c/li\u003e\n\u003cli\u003eWan Q, Tang J, Han Y, Wang D. Co-expression modules construction by WGCNA and identify potential prognostic markers of uveal melanoma. Exp Eye Res. 2018; 166: 13-20.\u003c/li\u003e\n\u003cli\u003eLiu J, Li S, Lin L, Jiang Y, Wan Y, Zhou S, et al. Co-expression network analysis identified atypical chemokine receptor 1 (ACKR1) association with lymph node metastasis and prognosis in cervical cancer. Cancer Biomark. 2020; 27: 213-23.\u003c/li\u003e\n\u003cli\u003eWang J, Yi Y, Chen Y, Xiong Y, Zhang W. Potential mechanism of RRM2 for promoting Cervical Cancer based on weighted gene co-expression network analysis. Int J Med Sci. 2020; 17: 2362-72.\u003c/li\u003e\n\u003cli\u003eThe spike-and-slab lasso Cox model for survival prediction and associated genes detection, Bioinformatic\u003c/li\u003e\n\u003cli\u003eTang Z, Lei S, Zhang X, Yi Z, Guo B, Chen JY, et al. Gsslasso Cox: a Bayesian hierarchical model for predicting survival and detecting associated genes by incorporating pathway information. BMC Bioinformatics. 2019; 20: 94.\u003c/li\u003e\n\u003cli\u003eBalkwill, F. Cancer and the chemokine network. Nat Rev Cancer 2004, 4, 540-550, doi:10.1038/nrc1388.\u003c/li\u003e\n\u003cli\u003eBandyopadhyay, S.; Zhan, R.; Chaudhuri, A.; Watabe, M.; Pai, S.K.; Hirota, S.; Hosobe, S.; Tsukada, T.; Miura, K.; Takano, Y.; et al. Interaction of KAI1 on tumor cells with DARC on vascular endothelium leads to metastasis suppression. Nat Med 2006, 12, 933-938, doi:10.1038/nm1444.\u003c/li\u003e\n\u003cli\u003eZhu W, Krishna S, Garcia C, Lin CJ, Mitchell BD, Scott KL, et al. Daam2 driven degradation of VHL promotes gliomagenesis. Elife. 2017; 6.\u003c/li\u003e\n\u003cli\u003eFang X, Zhang D, Zhao W, Gao L, Wang L. Dishevelled Associated Activator Of Morphogenesis (DAAM) Facilitates Invasion of Hepatocellular Carcinoma by Upregulating Hypoxia-Inducible Factor 1\u0026alpha; (HIF-1\u0026alpha;) Expression. Med Sci Monit. 2020; 26: e924670.\u003c/li\u003e\n\u003cli\u003eChen A, Liu Z, Wu Q, Li H. Disheveled-associated activator of morphogenesis 2 promotes invasion of colorectal cancer by activating PAK1 and promoting MMP7 expression. Genes Genomics. 2021; 43: 923-35.\u003c/li\u003e\n\u003cli\u003eLi Z, Wei X, Zhu Y. The prognostic value of DAAM2 in lower grade glioma, liver cancer, and breast cancer. Clin Transl Oncol. 2023; 25: 2224-38.\u003c/li\u003e\n\u003cli\u003eChen L, Zhou J, Zhao Z, Zhu Y, Xing J, An J, et al. Low Expression of Phosphodiesterase 2 (PDE2A) Promotes the Progression by Regulating Mitochondrial Morphology and ATP Content and Predicts Poor Prognosis in Hepatocellular Carcinoma. Cells. 2022; 12.\u003c/li\u003e\n\u003cli\u003eLi, S.Z.; Ren, K.X.; Zhao, J.; Wu, S.; Li, J.; Zang, J.; Fei, Z.; Zhao, J.L. miR-139/PDE2A-Notch1 feedback circuit represses stemness of gliomas by inhibiting Wnt/\u0026beta;-catenin signaling. Int J Biol Sci 2021, 17, 3508-3521, doi:10.7150/ijbs.62858.\u003c/li\u003e\n\u003cli\u003eZhao, Y.; Wang, Y.; Zhao, J.; Zhang, Z.; Jin, M.; Zhou, F.; Jin, C.; Zhang, J.; Xing, J.; Wang, N.; et al. PDE2 Inhibits PKA-Mediated Phosphorylation of TFAM to Promote Mitochondrial Ca(2+)-Induced Colorectal Cancer Growth. Front Oncol 2021, 11, 663778, doi:10.3389/fonc.2021.663778.\u003c/li\u003e\n\u003cli\u003eHiramoto K, Murata T, Shimizu K, Morita H, Inui M, Manganiello VC, et al. Role of phosphodiesterase 2 in growth and invasion of human malignant melanoma cells. Cell Signal. 2014; 26: 1807-17.\u003c/li\u003e\n\u003cli\u003eDing, H.; Xiong, X.X.; Fan, G.L.; Yi, Y.X.; Chen, Y.R.; Wang, J.T.; Zhang, W. The New Biomarker for Cervical Squamous Cell Carcinoma and Endocervical Adenocarcinoma (CESC) Based on Public Database Mining. Biomed Res Int 2020, 2020, 5478574, doi:10.1155/2020/5478574.\u003c/li\u003e\n\u003cli\u003eNgwa, V.M.; Edwards, D.N.; Philip, M.; Chen, J. Microenvironmental Metabolism Regulates Antitumor Immunity. Cancer Res 2019, 79, 4003-4008, doi:10.1158/0008-5472.Can-19-0617.\u003c/li\u003e\n\u003cli\u003eBinnewies M, Roberts EW, Kersten K, Chan V, Fearon DF, Merad M, et al. Understanding the tumor immune microenvironment (TIME) for effective therapy. Nat Med. 2018; 24: 541-50.\u003c/li\u003e\n\u003cli\u003eDe Jaeghere EA, Denys HG, De Wever O. Fibroblasts Fuel Immune Escape in the Tumor Microenvironment. Trends Cancer. 2019; 5: 704-23.\u003c/li\u003e\n\u003cli\u003eFortes C, Mastroeni S, Mannooranparampil TJ, Passarelli F, Zappal\u0026agrave; A, Annessi G, et al. Tumor-infiltrating lymphocytes predict cutaneous melanoma survival. Melanoma Res. 2015; 25: 306-11.\u003c/li\u003e\n\u003cli\u003eMao X, Xu J, Wang W, Liang C, Hua J, Liu J, et al. Crosstalk between cancer-associated fibroblasts and immune cells in the tumor microenvironment: new findings and future perspectives. Mol Cancer. 2021; 20: 131.\u003c/li\u003e\n\u003cli\u003eBurk RD, Chen Z, Saller C, Tarvin K, Carvalho AL, Scapulatempo-Neto C, et al. Integrated genomic and molecular characterization of cervical cancer. Nature. 2017; 543: 378-84.\u003c/li\u003e\n\u003cli\u003eRitchie, M.E.; Phipson, B.; Wu, D.; Hu, Y.; Law, C.W.; Shi, W.; Smyth, G.K. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res 2015, 43, e47, doi:10.1093/nar/gkv007.\u003c/li\u003e\n\u003cli\u003eWu T, Hu E, Xu S, Chen M, Guo P, Dai Z, et al. clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2021; 2: 100141.\u003c/li\u003e\n\u003cli\u003eEngebretsen S, Bohlin J. Statistical predictions with glmnet. Clin Epigenetics. 2019; 11: 123.\u003c/li\u003e\n\u003cli\u003eZeng D, Ye Z, Shen R, Yu G, Wu J, Xiong Y, et al. IOBR: Multi-Omics Immuno-Oncology Biological Research to Decode Tumor Microenvironment and Signatures. Front Immunol. 2021; 12: 687975.\u003c/li\u003e\n\u003cli\u003eNewman AM, Liu CL, Green MR, Gentles AJ, Feng W, Xu Y, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015; 12: 453-7.\u003c/li\u003e\n\u003cli\u003eShen W, Song Z, Zhong X, Huang M, Shen D, Gao P, et al. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform. iMeta. 2022; 1: e36.\u003c/li\u003e\n\u003cli\u003eRacle J, de Jonge K, Baumgaertner P, Speiser DE, Gfeller D. Simultaneous enumeration of cancer and immune cell types from bulk tumor gene expression data. Elife. 2017; 6.\u003c/li\u003e\n\u003cli\u003eYoshihara, K.; Shahmoradgoli, M.; Mart\u0026iacute;nez, E.; Vegesna, R.; Kim, H.; Torres-Garcia, W.; Trevi\u0026ntilde;o, V.; Shen, H.; Laird, P.W.; Levine, D.A.; et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun 2013, 4, 2612, doi:10.1038/ncomms3612.\u003c/li\u003e\n\u003cli\u003eYang T, Hui R, Nouws J, Sauler M, Zeng T, Wu Q. Untargeted metabolomics analysis of esophageal squamous cell cancer progression. J Transl Med. 2022; 20: 127.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"discover-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"dion","sideBox":"Learn more about [Discover Oncology](https://www.springer.com/12672)","snPcode":"","submissionUrl":"","title":"Discover Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"WGCNA, LASSO-COX, CESC, Prognostic model, immune prediction","lastPublishedDoi":"10.21203/rs.3.rs-8656950/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8656950/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eCESC is a malignant tumor that seriously threatens women's health, and the prognosis of patients is poor, and there is an urgent clinical need to find molecular markers for the prognosis of CESC. Bioinformatics was used to screen molecular markers related to CESC prognosis to provide a basis for prognosis prediction of CESC.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eTranscriptomic expression data of CESC and corresponding clinical data were downloaded from GEO and TCGA databases. CESC pivot genes were screened by WGCNA and LASSO-COX, and prognostic hub genes were verified using GEPIA. And to explore the correlation of prognostic pivot genes with immune cell infiltration and immune checkpoint gene expression.And compare the expression of prognosis-related genes in CESC and normal cervical tissues in the GEO dataset, TCGA combined with GTEx and HPA datasets.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eThrough WGCNA analysis, we constructed a module based on gene synergy. The Darkgreen module was significantly negatively correlated with the CESC. GO analysis shows that most of the genes in the module are related to cell junction, and motility, etc. KEGG analysis showed that the genes in the light green module are more involved in CAMs and Proteoglycans in cancer, et al pathway.The prognostic model composed of three genes, namely: ACKR1, DAAM2 and PDE2A, and found a significant correlation between core prognosis model gene expression and immune infiltration and immune checkpoint genes, which provided guidance for the prognosis and immunotherapy of CESC.\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe prognostic model consisting of three genes, ACKR1, DAAM2 and PDE2A, may be a prognostic and immunotherapy-related molecular marker for CESC.\u003c/p\u003e","manuscriptTitle":"WGCNA and LASSO Cox identify ACKR1 DAAM2 and PDE2A as prognostic genes and immune related biomarkers in cervical squamous cell carcinoma and endocervical adenocarcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-18 10:11:04","doi":"10.21203/rs.3.rs-8656950/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-05T14:21:11+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-05T02:43:36+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"129466915626151958798893708704064558784","date":"2026-05-05T02:34:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210328813875657720924434310229483741929","date":"2026-05-01T09:46:27+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-16T14:58:15+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-05T09:25:43+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-03T06:54:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-28T09:19:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Discover Oncology","date":"2026-02-28T08:50:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"discover-oncology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"dion","sideBox":"Learn more about [Discover Oncology](https://www.springer.com/12672)","snPcode":"","submissionUrl":"","title":"Discover Oncology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Discover Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"e5f84513-a992-44ea-8fff-d262962fa898","owner":[],"postedDate":"March 18th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-05T14:21:11+00:00","index":80,"fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-05T02:43:36+00:00","index":79,"fulltext":""},{"type":"reviewerAgreed","content":"129466915626151958798893708704064558784","date":"2026-05-05T02:34:19+00:00","index":78,"fulltext":""},{"type":"reviewerAgreed","content":"210328813875657720924434310229483741929","date":"2026-05-01T09:46:27+00:00","index":65,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-03-18T10:11:08+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-18 10:11:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8656950","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8656950","identity":"rs-8656950","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}