Key genes and associated mechanisms of PCOS and EC comorbidity: A bioinformatics analysis.

In this study, we utilized the RSF method to identify 5 key genes associated with the prognosis of the comorbidity between PCOS and EC, specifically SYTL1, PARVG, ID4, IL1RN, and S100A9. These key genes may regulate various related enrichment pathways, such as the TGF-β signaling pathway, as well as motif regulatory networks, miRNA regulatory networks, and immune cell micro-environments. Consequently, they can influence the occurrence, progression, and prognosis of the comorbidity between PCOS and EC. Future studies should focus on the role of these key genes in susceptibility, diagnosis, treatment, prevention, prognostic assessment, and disease management of PCOS and EC comorbidity.

Bioinformatics is defined as an interdisciplinary set of tools that integrates computer science, mathematics, and statistics for the classification and analysis of biological information. This toolset can be applied in various areas, including genome annotation and comparison, expression data analysis, protein structure prediction, and the study of biological networks. [ 1 , 2 ] Bioinformatics research offers substantial advantages in elucidating the occurrence and developmental pathways of these diseases, identifying biomarkers, and facilitating the development of new drugs. Polycystic ovary syndrome (PCOS) is one of the most prevalent female reproductive endocrine disorders, yet its etiology remains incompletely understood. Existing literature has identified various interactions among genes, as well as between genes and proteins, genes and environmental factors, highlighting the significance of genetic susceptibility in the onset and progression of PCOS. [ 3 ] Furthermore, numerous studies have sought to elucidate the pathogenesis of PCOS through different omics approaches, including genomics, transcriptomics, and proteomics, leading to the discovery of several candidate biomarkers that may be utilized for the diagnosis or prognostic evaluation of PCOS. [ 4 – 7 ] Endometrial carcinoma (EC) is a major reproductive system tumor and a common cancer in women. The mechanisms underlying their occurrence and development remain unclear. [ 8 – 11 ] Numerous multi-omics studies on EC have yielded extensive information regarding pathogenesis, molecular markers, therapeutic targets, treatment efficacy, and prognosis. [ 8 , 9 , 12 – 16 ] To date, while bioinformatics has made significant strides in the research and analysis of PCOS and EC, there has been a lack of thematic bioinformatics research focusing on the comorbidity between these conditions. Both PCOS and EC are characterized by high heterogeneity and complexity, and there exists a notable correlation between them. For instance, the incidence of EC is significantly higher in individuals with PCOS, being approximately 2 to 6 times greater than in those without PCOS. [ 17 – 23 ] This specialized study aims to conduct a bioinformatics comorbidity analysis of EC within the context of PCOS. This analysis seeks to elucidate the organic correlations and interaction mechanisms between the 2 diseases through series of exploration of relative key genes, thereby providing valuable insights for related prevention and control efforts.

Conceptualization: Yingsha Yao. Data curation: Yingsha Yao, Shulan Zhu. Software: Yingsha Yao, Shulan Zhu. Supervision: Xiaoming Zhu. Writing – original draft: Yingsha Yao, Shulan Zhu. Writing – review & editing: Xiaoming Zhu.

This study is a bioinformation analysis, the ethical review is not required. The cancer genome atlas (TCGA) database is currently the largest repository of cancer genetic information, encompassing a diverse array of data types. [ 24 ] The gene expression omnibus (GEO) database formally known as a comprehensive gene expression database. [ 25 ] This study utilized the package “Limma” to conduct differential expression analysis of comorbidity-related expression profiles, aiming to identify genes with significant differential expression between PCOS and EC samples, and to perform an accompanying molecular mechanism analysis. [ 25 ] In screening for differentially expressed genes, the criteria established were a P -value of 1. The results of the analysis were visualized through the generation of volcano plots and heat maps depicting the differential genes. This study utilizes the Metascape database to perform functional annotation and analysis of gene sets, including gene ontology pathway analysis, thereby comprehensively exploring the primary biological functions of these differential gene. [ 26 ] Results with a minimum overlap of ≥3 and a P -value of ≤.01 were deemed statistically significant. The random survival forest (RSF) is a variant of the survival forest algorithm that integrates concepts from both survival analysis and random forests. This method predicts survival time or survival probability by constructing multiple decision trees, thereby enhancing model performance through the introduction of randomness. [ 27 ] In this study, the package “randomForestSRC” is utilized to perform decision tree feature selection on the differential genes analyzed in the preceding step. Subsequently, the RSF algorithm is employed to rank the importance of prognosis-related genes. Genes with a relative importance >0.3 are identified and selected as the final marker genes. Gene Set Enrichment Analysis (GSEA) is a method employed to interpret high-throughput gene expression data, revealing underlying biological processes and pathways. [ 28 ] In this specific study, GSEA will be utilized to compare the differences in signaling pathways between the high gene expression group and the low-expression group, aiming to explore the core gene pathway molecular mechanisms related to the prognosis of the comorbidity between PCOS and EC. The number of permutations is set to 1000, and the replacement type is phenotypic. Gene Set Variation Analysis (GSVA) is a nonparametric, unsupervised method that evaluates the biological functions of a sample by comprehensively scoring the gene set of interest and converting gene-level changes into pathway-level alterations. [ 29 ] In this study, gene sets derived from PCOS and EC samples will be obtained from the Molecular Signatures Database (version 7.0). The GSVA algorithm will be employed to comprehensively score the core gene sets associated with the prognosis of the comorbidity between PCOS and EC, thereby evaluating potential biological function changes of core genes in samples originating from these 2 disease sources. This study utilizes the package “RcisTarget.hg19.motifDBs.cisbpOnly.500bp” to develop a Gene-motif rankings database pertinent to the comorbidity between PCOS and EC. [ 30 ] The analysis focuses on motifs associated with genes and transcription factors within the database, along with their respective normalized enrichment scores. Beyond the annotated motifs present in the source data, additional inferences were drawn based on motif similarities and gene sequences, leading to the creation of supplementary annotation files. MicroRNA (miRNA) is a type of small noncoding RNA that regulates gene expression by promoting the degradation of mRNA or inhibiting its translation. [ 31 ] In this study, we will obtain miRNA information related to key genes by querying the miRcode database. This information will be visualized using Cytoscape software to construct a hub gene-miRNA relationship network diagram, illustrating the interactions between genes and miRNAs. Furthermore, we will analyze the corresponding key genes and their miRNA regulatory networks, along with the relevant transcription and degradation processes, to elucidate the specific molecular mechanisms and interactive relationships associated with the comorbidity between PCOS and EC. [ 32 ] The CIBERSORT method is a widely utilized approach for assessing immune cell types within the microenvironment, capable of distinguishing 22 distinct human immune cell phenotypes, including T cells, B cells, plasma cells, and various myeloid cell subsets. [ 33 ] In this study, the CIBERSORT algorithm was employed to analyze data related to the comorbidity of PCOS and EC, aiming to infer the relative proportions of the 22 types of immune infiltrating cells. Additionally, a correlation analysis was performed to examine the relationship between the expression of key genes and the abundance of immune cells, thereby enhancing our understanding of the immune status in cases of the comorbidity between PCOS and EC. This special study will first utilize the package “Seurat” to read EC single-cell data sets obtained from public databases, filtering out low-expression genes. Next, the data will be standardized and homogenized, followed by the application of the t-SNE algorithm to visualize the positional relationships of various cell groups in 2-dimensional space. This approach will facilitate the subsequent screening of single-cell group clustering based on the expression of key genes. Subsequently, the cell groups will be annotated using the annotation files provided by the package “celldex” to ascertain the cell type, state, or characteristics of each cell group. Finally, the localization of key EC progression genes (TP53, MSH6, MLH1) within EC single-cell populations will be analyzed, and the location and co-expression of these key EC progression genes alongside key genes of the comorbidity between PCOS and EC within EC single-cell populations will be discussed. All statistical analyses were conducted using the R program (version 4.2), with a significance level set at P  <.05.

The PCOS sample data utilized in this study was downloaded from the NCBI GEO public database, specifically the GSE 34526 dataset. [ 34 ] This dataset comprises a total of 10 samples, including 3 cases from the normal control group and 7 cases from the PCOS disease group. Additionally, the EC sample data were retrieved from the TCGA public database, specifically the UCEC dataset, which contains a total of 589 samples, including 35 cases from the normal control group and 554 cases from the EC disease group. The EC single-cell dataset was also sourced from the NCBI GEO public database, specifically the GSE 203612 dataset, [ 35 ] which includes a total of 3 EC disease samples. A total of 1704 differential genes were identified in the PCOS-related GSE 34526 data set, comprising 1125 up-regulated genes and 579 down-regulated genes (Fig. 1 A and B). In the EC-related TCGA-UCEC data set, 2451 differential genes were identified, including 1071 up-regulated genes and 1380 down-regulated genes (Fig. 1 C and D). The intersections of both up-regulated and down-regulated genes from the PCOS and EC data sets were analyzed, resulting in 127 differential intersection genes. Specifically, there were 67 up-regulated gene intersections (Fig. 1 E) and 60 down-regulated gene intersections (Fig. 1 F). (A) Volcano plot illustrating differential genes associated with PCOS; (B) Heat map depicting differential genes in PCOS; (C) Volcano plot for differential genes related to EC; (D) Heat map of differential genes in EC; (E) Venn diagram showing up-regulated differential intersection genes between PCOS and EC; (F) Venn diagram depicting down-regulated differential intersection genes in PCOS and EC; (G) Schematic representation of functional enrichment for differential intersection genes in PCOS and EC; (H) Ranking of key intersection genes pertinent to the survival of patients with PCOS-EC comorbidity. EC = endometrial cancer, PCOS = polycystic ovary syndrome. Pathway enrichment analysis was conducted on the 127 differentially intersecting genes associated with the comorbidity between PCOS and EC. The results indicated that these differentially intersecting genes were predominantly enriched in pathways such as the regulation of transforming growth factor beta receptor signaling and the regulation of hormone biosynthetic processes (Fig. 1 G). RSF analysis identified a total of 7 candidate key genes with significant importance in comorbid survival. The order of importance is illustrated in Fig. 1 H. Survival analysis revealed that 5 genes were significantly associated with the survival of the comorbidity between PCOS and EC ( P  <.05). These 5 genes are: Synaptotagmin-like protein 1 gene (SYTL1) (Fig. 2 A), parvin gamma (actin-binding protein) gene (PARVG) (Fig. 2 A and B), inhibitor of DNA binding 4 gene (ID4) (Fig. 2 C), interleukin 1 receptor antagonist gene (IL1RN) (Fig. 2 D), and S100 calcium-binding protein A9 gene (S100A9) (Fig. 2 E). The survival curves related to key genes associated with PCOS-EC comorbidity are as follows: (A) SYTL1; (B) PARVG; (C) ID4; (D) IL1RN; (E) S100A9. The GSEA pathway enrichment of key genes is represented by: (F) SYTL1; (G) PARVG; (H) ID4; (I) IL1RN; (J) S100A9. The GSVA pathway enrichment for the key genes includes: (K) SYTL1; (L) PARVG; (M) ID4; (N) IL1RN; (O) S100A9. EC = endometrial cancer, GSEA = gene set enrichment analysis, GSVA = gene set variation analysis, ID4 = inhibitor of DNA binding 4, IL1RN = interleukin 1 receptor antagonist, PARVG = parvin gamma, PCOS = polycystic ovary syndrome, S100A9 = S100 calcium-binding protein A9, SYTL1 = synaptotagmin-like protein 1. The GSEA results indicate that the enriched pathways associated with the SYTL1 gene include the KEGG-ERBB-SIGNALING-PATHWAY and KEGG-TGF-BETA-SIGNALING-PATHWAY, among others (Fig. 2 F). For the PARVG gene, the enriched pathways encompass the KEGG-INSULIN-SIGNALING-PATHWAY and KEGG-TGF-BETA-SIGNALING-PATHWAY, among others (Fig. 2 G). The ID4 gene is linked to enriched pathways such as KEGG-FRUCTOSE-AND-MANNOSE-METABOLISM and KEGG-PENTOSE-PHOSPHATE-PATHWAY, among others (Fig. 2 H). The enriched pathways related to the IL1RN gene include the KEGG-GNRH-SIGNALING-PATHWAY and KEGG-WNT-SIGNALING-PATHWAY, among others (Fig. 2 I). Lastly, the S100A9 gene is associated with enriched pathways including the KEGG-GNRH-SIGNALING-PATHWAY and KEGG-OOCYTE-MEIOSIS, among others (Fig. 2 J). The GSVA results indicate that high expression of SYTL1 enriches the P53 pathway and the IL6-JAK-STAT3 signaling-pathway, among others, thereby triggering corresponding functional changes (Fig. 2 K). Similarly, high expression of PARVG enriches the P53 pathway and the IL2-STAT5 signaling-pathway, leading to analogous functional changes (Fig. 2 L). Furthermore, high expression of ID4 enriches the TGF-β signaling-pathway and the IL2-STAT5 signaling-pathway, resulting in corresponding functional changes (Fig. 2 M). Additionally, high expression of IL1RN integrates the IL6-JAK-STAT3 signaling-pathway and the IL2-STAT5 signaling-pathway, triggering related functional changes (Fig. 2 N). Lastly, high expression of S100A9 enriches the P53 pathway and the IL2-STAT5 signaling-pathway, thereby inducing corresponding functional changes (Fig. 2 O). Motif enrichment analysis revealed that these 5 key genes are jointly regulated by multiple transcription factors. We conducted Motif-TF annotation and cumulative recovery curve enrichment analysis on these transcription factors. The findings indicated that the Motif cisbp__M4556 exhibited the highest normalized enrichment score (normalized enrichment scores: 5.06). Figure 3 A illustrates the primary transcription factors and their corresponding relationships with the key genes regulated by these transcription factors. Figure 3 B presents the main transcription factors identified through motif enrichment analysis of the 5 key genes. (A) Diagram illustrating the key gene motifs enrichment; (B) Main transcription factors involved in key gene motifs enrichment; (C) Prediction of hub gene-miRNA pairs in reverse direction for PCOS-EC comorbidity; (D) Key regulatory genes associated with endometrial cancer; (E) Schematic representation of the correlation between the expression levels of key genes and the expression levels of disease regulatory genes. EC = endometrial cancer, PCOS = polycystic ovary syndrome. By querying the Mircode database, we obtained miRNA information related to 5 key genes associated with the comorbidity between PCOS and EC, which included 69 miRNAs and 135 hub gene-miRNA relationship pairs. The network relationship was constructed using Cytoscape software, as illustrated in Figure 3 C. Based on these 135 hub gene-miRNA relationship pairs, we further identified regulatory genes linked to the onset of the comorbidity between PCOS and EC through the GeneCards database. After analyzing the expression levels of the top 20 genes based on relevance scores, we found that the expression of these regulatory genes, such as ATM, BARD1, BRCA1, and BRCA2, exhibited significant changes in patients with PCOS and EC (Fig. 3 D). Furthermore, the expression levels of these disease-regulated genes were significantly correlated with the expression levels of 5 key genes. For instance, ID4 and PTEN demonstrated a significant positive correlation ( R  = 0.256), while SYTL1 and MSH6 showed a significant negative correlation ( r  = −0.431) (Fig. 3 E). For specific relevant information, please refer to Figure S1, Supplemental Digital Content, https://links.lww.com/MD/Q611 . The immune infiltration micro-environment primarily consists of immune cells, extracellular matrix, various growth factors, inflammatory factors, and distinct physical and chemical characteristics. These components can significantly influence the occurrence, progression, and clinical outcomes of the disease. Figure 4 A illustrates the proportion of 22 types of immune cells in PCOS-EC comorbid patients, as analyzed using the CIBERSORT method. Figure 4 B depicts the correlations among various immune cell categories in PCOS-EC comorbid patients. Figure 4 C presents the comparative analysis of the proportions and differences of these 22 immune cell types between PCOS-EC comorbid patients and the control group. Figure 4 D–H detail the specific relationships between 5 key genes associated with the comorbidity between PCOS and EC and the immune cell infiltration micro-environment. (A) Proportion of immune cell content in PCOS-EC comorbid patients; (B) correlation matrix depicting the relationships between immune cells in PCOS-EC comorbid patients; (C) proportion of immune cell categories in PCOS-EC comorbid patients compared to the corresponding control group, including ratio and comparative analysis; (D–H) Correlation analysis between key genes and immune cells, specifically: (D) SYTL1; (E) PARVG; (F) ID4; (G) IL1RN; (H) S100A9. EC = endometrial cancer, ID4 = inhibitor of DNA binding 4, IL1RN = interleukin 1 receptor antagonist, PARVG = parvin gamma, PCOS = polycystic ovary syndrome, S100A9 = S100 calcium-binding protein A9, SYTL1 = synaptotagmin-like protein 1. As shown in Figure 4 B, the immune cells in PCOS-EC comorbid patients, such as CD4 resting memory T cells, exhibit significant relationships with CD8 T cells, CD4 memory-activated T cells, and T cell follicular helper cells. Additionally, M0 macrophages show a strong correlation with CD8 T cells, while activated NK cells are significantly correlated with resting mast cells. As illustrated in Figure 4 C, the proportion of M0 macrophages, M1 macrophages, plasma cells, T follicular helper cells, and regulatory T cells (Tregs) in the immune cell profiles of PCOS-EC comorbid patients is significantly higher than that of the control group. Figure 4 D–H demonstrate that the 5 key genes show strong correlations with immune cell infiltration in the micro-environment. Notably, SYTL1 exhibits a significant positive correlation with Tregs and a significant negative correlation with resting mast cells. Similarly, PARVG shows a significant positive correlation with Tregs and a significant negative correlation with resting mast cells. ID4 is positively correlated with CD4 resting memory T cells and negatively correlated with M0 macrophages. IL1RN demonstrates a significant positive correlation with neutrophils while showing a significant negative correlation with resting mast cells. Lastly, S100A9 is significantly negatively correlated with neutrophils and significantly positively correlated with resting mast cells. Download the EC single-cell dataset GSE 203612 from the NCBI GEO public database. [ 35 ] The locations of 5 key genes (SYTL1, PARVG, ID4, IL1RN, S100A9) within this single-cell dataset are illustrated in Figure 5 A. (A) Cluster location of key genes in the single-cell dataset ( GSE203612 ); (B) expression of key genes in 6 associated cell groups. Figure 5 B illustrates the expression of key genes across 6 cell types – T cells, monocytes, neutrophils, epithelial cells, B cells, and tissue stem cells – that are associated with comorbidities. Gene data related to EC disease progression, specifically for TP53, MSH6, and MLH1, were obtained from the GeneCards database ( https://www.genecards.org/ ). These genes were co-expressed and visualized alongside 5 key genes. The analysis was conducted within the 6 associated cell groups mentioned above (see Fig. S2, Supplemental Digital Content, https://links.lww.com/MD/Q611 ). As illustrated in Figure S2, Supplemental Digital Content, https://links.lww.com/MD/Q611 , the 5 key genes, along with TP53, MSH6, and MLH1, exhibited significant co-expression.

This study utilized RSF analysis to identify 5 key genes associated with the prognosis of PCOS-EC comorbid patients, specifically SYTL1, PARVG, ID4, IL1RN, and S100A9. These findings were further validated through single-cell data analysis. Previous literature has established that single-disease EC is linked to these 5 key genes, while PCOS as a standalone condition is also associated with various key genes related to comorbidity. However, to date, no studies have specifically examined the relationship between these genes and the comorbidity between PCOS and EC. The SYTL1 gene encodes a protein that is situated in the cytoplasmic membrane and plays a crucial role in exocytosis. Research has demonstrated that its aberrantly elevated expression is associated with the onset and prognosis of EC. [ 36 ] The abnormal expression of the PARVG gene is associated with a poor prognosis in EC [ 37 ] and can serve as a predictive marker for EC prognosis. [ 38 ] The ID4 factor, encoded by the ID4 gene, regulates gene expression by interacting with specific transcription factors, which in turn influences the proliferation and tumorigenesis of various tissue cells. Research has indicated that the expression level of this gene is correlated with the abnormal proliferation (invasive growth) of ectopic endometrial glands and stromal cells. [ 39 , 40 ] Several studies have indicated that polymorphisms in the IL1RN gene are associated with susceptibility to PCOS. [ 41 ] Notably, allele V of the IL1RN gene has been significantly linked to the onset of PCOS. [ 42 ] Additional research has reported a correlation between abnormal IL1RN gene expression and the comorbidity between PCOS and nonalcoholic fatty liver disease. [ 43 ] In EC samples, elevated IL1RN expression suggests a potential role in the occurrence and progression of EC. [ 44 ] Similarly, in aggressive endometriosis tissue specimens, the IL1RN gene exhibited abnormally high expression specifically in translocated endometrial stromal cells. [ 45 ] Furthermore, the IL1RN gene is also found to be highly expressed in menstrual blood samples from patients with endometriosis, [ 46 ] implying that elevated IL1RN expression may be associated with the abnormal migration of endometrial cells. S100A9, encoded by the S100A9 gene, is an immunogenic protein that plays a crucial role in the body’s immune response and inflammation repair. Research has demonstrated that the overexpression of this gene is associated with the occurrence and susceptibility to PCOS. [ 47 ] The S100A9 protein often exists in a complex with the S100A8 protein, facilitating cell migration in the PCOS environment. [ 48 ] This mechanism may be linked to the exon-rich nature of S100A9, which significantly enhances the inflammatory state and increases steroidogenesis by activating the nuclear factor κB signaling-pathway. [ 49 ] Furthermore, studies have indicated that abnormal expression of the S100A9 gene is associated with the comorbidity between PCOS and nonalcoholic fatty liver disease. [ 43 ] Regarding the relationship between S100A9 and EC, research has shown that S100A9, functioning as an inflammatory mediator, exhibits significantly increased expression, suggesting an abnormal immune response in the endometrium that is related to the prognosis of EC. [ 50 ] To date, there have been no reports on the correlation between the SYTL1 gene, PARVG gene, and ID4 gene with PCOS. Abnormal expression of genes and their encoded proteins influence the occurrence and progression of diseases through various signaling-pathways. This study employed GSEA and GSVA methods to perform signaling-pathway enrichment analysis on 5 key genes. The results indicated that these key genes were predominantly enriched in signaling-pathways associated with hormone metabolism, cell proliferation, and tumor formation, specifically represented by the TGF-β signaling-pathway, the P53 signaling-pathway, and the IL-(JAK)-STAT signaling-pathway. Numerous research reports have examined the relationship between the TGF-β signaling-pathway and the onset and progression of PCOS [ 51 – 53 ] and EC. [ 54 – 59 ] For instance, overactivation of the receptor 1-mediated TGF-β signaling pathway enhances the migratory capacity of EC cells, while targeted inhibition of this pathway can reduce the proliferative activity of EC cells and impede the transition from an epithelial to a mesenchymal phenotype, thereby diminishing the invasive potential of these cells. [ 54 , 55 ] Tumor-related immune cell infiltration, also referred to as the immune tumor micro-environment, is closely associated with the occurrence and progression of tumors, as well as the efficacy of immunotherapy and prognosis. Tumors characterized by fewer somatic mutations are known to produce lower levels of immunogenic antigens, enabling them to evade detection by cytotoxic T cells. Consequently, patients with such tumors tend to have poorer prognoses and are more likely to progress to advanced stages. [ 60 , 61 ] Conversely, high levels of CD8 memory-activated T cell infiltration are linked to favorable prognoses in patients with EC. [ 60 , 62 – 64 ] Animal experiments have demonstrated that the introduction of exogenous CD8 T cells can enhance the prognosis of EC. [ 65 ] In the group of EC patients with a high mutation load, higher infiltration levels of CD4 memory-activated T cells, plasma cells, and CD8 memory-activated T cells were observed. In contrast, the low mutation load group exhibited higher infiltration levels of CD4 resting memory T cells and M0 macrophages. [ 66 ] The expression levels of key genes SYTL1 and PARVG, as revealed in this study, are positively correlated with the infiltration of T cell follicular helper cells, CD8 T cells, Tregs, and CD4 memory-activated T cells. Conversely, these gene expressions are inversely correlated with the infiltration levels of naive B cells, M2 macrophages, resting mast cells, and CD4 resting memory T cells. This indicates that high expression of SYTL1 and PARVG is associated with elevated infiltration of CD4 memory-activated T cells and CD8 T cells, which are linked to favorable prognosis, in the context of the PCOS-EC comorbidity state. Additionally, the association with CD4 resting memory T cells, which are correlated with poor prognosis, further underscores the complexity of immune cell infiltration in this condition. The observed low-level infiltration of certain cell types may significantly impact the onset and prognosis of PCOS-EC comorbidity. This study revealed that the expression levels of the key genes IL1RN and S100A9 are negatively correlated with the infiltration levels of CD4 resting memory T cells. This finding suggests that high expression of these 2 critical genes is more closely associated with the prognosis of PCOS-EC comorbidity in terms of immune cell infiltration. Conversely, the expression level of the key gene ID4 is positively correlated with the infiltration level of CD4 resting memory T cells, while being negatively correlated with the infiltration levels of CD4 memory-activated T cells and M0 macrophages. This indicates that elevated ID4 expression is related to PCOS-EC comorbidity from the perspective of immune cell infiltration, which are associated with poor prognosis. Furthermore, previous studies on immune cell infiltration in PCOS have primarily focused on its potential link to the adverse prognosis of PCOS-EC comorbidity. For instance, PCOS patients exhibit a significant decrease in CD4 resting memory T cells [ 67 ] and a notable increase in CD4 memory-activated T cells. [ 68 ] Additionally, the lipid metabolism abnormalities commonly observed in PCOS patients can influence the infiltration of CD8 T cells and enhance tumor cell invasiveness. [ 69 ] These findings regarding immune cell infiltration in PCOS suggest that patients may have a heightened susceptibility to EC, correlating with the poor prognosis observed in individuals with the comorbidity between PCOS and EC. This study is a bioinformatics investigation that relies solely on publicly available database data for analysis. This approach presents certain research flaws and shortcomings, including limited data sources and ethnic diversity, reliance on a single detection platform and a restricted sample size. These limitations may introduce risks of selection bias and reference bias in the findings. Therefore, there is an urgent need for further in-depth research involving prospective, multi-center studies with larger sample sizes and multiple detection platforms. Such efforts would help to verify and augment the existing research findings, thereby enhancing their reference value for research and development and improving the prospects for clinical application.