Phthalates unleashed: decoding ovarian carcinogenesis through multi-omics networks, single-cell insights, and molecular docking.

The molecular weights of three phthalates ranged from 194.19 to 390.56 g/mol, driven by structural variations in alkyl substituents. Dimethyl phthalate (194.19 g/mol) with short methyl groups exhibited the lowest weight, while the branched-chain derivative (390.56 g/mol) displayed the highest due to extended alkyl branches(Table  1 ). All compounds shared a core aromatic ring with dual ester groups, but side-chain complexity differed: linear chains in diethyl/dimethyl phthalates versus a tertiary carbon structure in the branched derivative. These structural distinctions—particularly chain length and branching—directly influence physicochemical properties (e.g., hydrophobicity and steric hindrance), which underpin bioinformatic parameter selection. Table 1 Chemical data of three phthalates Name SMILE Structure Molecular weight Diethyl phthalate CCOC(= O)c1ccccc1C(= O)OCC 222.24 Dimethyl phthalate COC(= O)c1ccccc1C(= O)OC 194.19 Dioctyl phthalate CCCCC(CC)COC(= O)c1ccccc1C(= O)OCC(CC)CCCC 390.56 Chemical data of three phthalates A thorough literature search was conducted across multiple databases to retrieve relevant information on diseases associated with phthalates, with special emphasis on diseases related to ovarian cancer. Intersection analysis of 324 phthalate-associated target genes revealed 8 conserved genes ( CPSF3, GART, GPI, IMPDH2, LDHD, RBKS, TPI1, XPOT ) common to DEP, DMP, and DOP (Fig.  2 A). Functional annotation linked these genes to metabolic reprogramming pathways—including glycolysis (e.g., GPI, TPI1 ; KEGG: hsa00010, FDR  = 0.0037), and aerobic glycolysis (WikiPathways: WP4629, FDR  = 0.00037). Notably, pathways like"Disorders of fructose metabolism"(WikiPathways: WP5178) and"Clear cell renal cell carcinoma pathways"(WP4018) suggest a conserved role in oncogenic metabolism, potentially explaining phthalate-driven ovarian cancer risk (Table S1). While DEP and DOP broadly targeted energy metabolism, DMP’s limited gene interactions (97 genes) may stem from its short alkyl chains reducing ligand-receptor affinity. These findings prioritize metabolic disruption as a key mechanism, supported by high-confidence pathways ( FDR  < 0.05), directing further exploration of phthalate-induced carcinogenesis. Fig. 2 Target analysis of ovarian cancer and phthalates. A Venn diagram of three phthalates targets from the database. B Venn diagram of phthalates and ovarian cancer targets. C , D PPI network of intersection targets and core genes of ovarian cancer and phthalates Target analysis of ovarian cancer and phthalates. A Venn diagram of three phthalates targets from the database. B Venn diagram of phthalates and ovarian cancer targets. C , D PPI network of intersection targets and core genes of ovarian cancer and phthalates Integrated analysis of OMIM and GeneCards databases identified 10,763 ovarian cancer-associated genes, with 234 overlapping phthalate targets (Fig.  2 B), among which 83 genes showed differential expression in the TCGA ovarian cancer cohort (Table S2). The PPI network of 234 targets (1,412 edges, node degree = 12.1; Fig.  2 C) was refined to 63 high-confidence core proteins, with GAPDH, CASP3, ESR1 , and CCND1 exhibiting maximal centrality (Fig.  2 D). These hub genes regulate hallmark cancer processes: GAPDH (glycolytic flux), ESR1 (estrogen signaling), and CCND1 (cell cycle progression), directly linking phthalate exposure to tumor initiation and progression. For example, GAPDH ’s dominance aligns with enriched glycolysis pathways, while ESR1 reinforces hormonal mechanisms implicated in ovarian cancer epidemiology. The network’s robustness (high node degree) and pathway coherence suggest a convergent mechanism where phthalates dysregulate energy metabolism and proliferative signaling—a hypothesis warranting experimental validation. A total of 692 significant GO terms were identified (adjusted p  < 0.05), including 574 biological processes (BP), 41 cellular components (CC), and 77 molecular functions (MF). The adjusted p value threshold was selected to minimize false positives while retaining biologically meaningful associations. Key enriched BP terms centered on nucleotide metabolism (e.g.,"nucleoside phosphate metabolic process," p  = 1.9e-10) and xenobiotic response (e.g.,"response to steroid hormone," p  = 1.3e-7), suggesting plasticizers may disrupt metabolic homeostasis and hormone signaling (Fig.  3 A-C, Table S3-5). Cellular component analysis highlighted mitochondrial matrix ( p  = 1.1e-6) and presynaptic membrane ( p  = 1.0e-4), implicating mitochondrial dysfunction and neuronal signaling cross-talk in ovarian carcinogenesis. Molecular functions were dominated by oxidoreductase activity (e.g.,"acting on CH-OH group of donors," p  = 3.4e-7) and kinase regulation (e.g.,"calcium-dependent protein kinase C activity," p  = 2.0e-6), aligning with known roles of redox imbalance and kinase signaling in tumor progression. Fig. 3 Functional enrichment analysis intersection targets ( A ) of ovarian cancer and phthalates. A Top 10 biological processes; B Top 10 cellular components; C molecular functions; D Top 10 KEGG pathway analysis. The size represents the number of genes, with larger dots indicating a greater number of genes. The color and the length of the bars represent the P -value, with lighter colors and longer bars indicating a smaller P -value Functional enrichment analysis intersection targets ( A ) of ovarian cancer and phthalates. A Top 10 biological processes; B Top 10 cellular components; C molecular functions; D Top 10 KEGG pathway analysis. The size represents the number of genes, with larger dots indicating a greater number of genes. The color and the length of the bars represent the P -value, with lighter colors and longer bars indicating a smaller P -value KEGG pathway analysis of the 234 targets revealed significant enrichment in cancer-related pathways, including"Chemical carcinogenesis – receptor activation"( p  = 2.5e-5) and"Non-small cell lung cancer"( p  = 9.8e-5), alongside metabolic pathways such as"Carbon metabolism"( p  = 6.8e-8) and"Glycolysis/Gluconeogenesis"( p  = 6.6e-6) (Fig.  3 D, Table S6). Notably,"Chemical carcinogenesis – receptor activation"(GeneRatio = 15/171) linked hormone receptors ( ESR2, AR ) to xenobiotic responses, suggesting a mechanism by which phthalates may mimic endocrine disruptors (Fig.  3 D, Table S5). The co-enrichment of metabolic and oncogenic pathways underscores a synergistic interplay between energy dysregulation and proliferative signaling in ovarian carcinogenesis. KEGG pathway analysis of the 63 core targets identified 99 significantly enriched pathways (adjusted p  < 0.05). A chord diagram (Fig.  4 , Table S7) revealed prominent gene-pathway interactions, with"Metabolic pathways"( p  = 2.1e-5) and"Pathways in cancer"( p  = 6.5e-5) displaying the highest gene overlap. Key metabolic sub-pathways included"Carbon metabolism"( p  = 1.2e-4) and"Glycolysis/Gluconeogenesis"( p  = 0.003), both linked to energy dysregulation. Notably," HIF-1 signaling pathway"( p  = 0.003) and"Glucagon signaling pathway"( p  = 0.0005) were recurrently associated with hub genes such as GAPDH (glycolysis) and CASP3 (apoptosis). These pathways implicate dual mechanisms: (1) metabolic reprogramming via ALDH6A1 and CAT in carbon metabolism, and (2) proliferative signaling through BIRC5 and AR in cancer pathways. The convergence of these processes suggests that plasticizers may drive ovarian carcinogenesis by disrupting redox balance, amplifying oncogenic signaling, and hijacking metabolic homeostasis. Fig. 4 Chord diagram showing the annotated core genes in top 10 enriched pathways. The darker the color, the stronger is the significance Chord diagram showing the annotated core genes in top 10 enriched pathways. The darker the color, the stronger is the significance Analysis of the TCGA ovarian cancer cohort (426 tumor [T] vs. 88 normal [N] samples, p  < 0.05) revealed significant dysregulation of six out of seven core hub genes (Fig.  5 A-F). CCND1 (cell cycle progression), CYCS (apoptosis regulation), ESR1 (estrogen signaling), and GAPDH (glycolytic flux) exhibited marked upregulation in tumors (fold change > 2, p  < 0.001), while PPARA (lipid metabolism) and SIRT1 (epigenetic regulation) were significantly downregulated (fold change < 0.5, p  < 0.005). Fig. 5 Expression of 6 core genes between ovarian cancer and healthy tissues in TCGA. A CCND1 ; B CYCS ; C ESR1 ; D GAPDH ; E PPARA ; F SIRT1 . * P  < 0.01 Expression of 6 core genes between ovarian cancer and healthy tissues in TCGA. A CCND1 ; B CYCS ; C ESR1 ; D GAPDH ; E PPARA ; F SIRT1 . * P  < 0.01 Notably, GAPDH ’s overexpression aligns with its role in metabolic reprogramming—a hallmark of ovarian cancer—while ESR1 ’s elevation underscores estrogen-driven proliferation mechanisms. Conversely, SIRT1 suppression may reflect epigenetic silencing commonly observed in advanced tumors. These expression patterns corroborate pathway analyses (Sect. 3.5), linking phthalate exposure to disrupted cell cycle control ( CCND1 ), apoptotic resistance ( CYCS ), and hormonal dysregulation ( ESR1 ). The absence of CASP3 expression differences ( p  = 0.12) suggests context-specific roles, warranting single-cell resolution studies. Using single-cell RNA sequencing (scRNA-seq) data, we analyzed in detail the expression levels of these genes across various cell types and states. The identification of cellular subpopulations revealed the heterogeneity of cells within ovarian cancer tissue (Fig.  6 A), and the distribution of cells in ovarian cancer tissue at different stages uncovered changes in cellular composition as the disease progresses (Fig.  6 B). Notably, the cellular heterogeneity analysis demonstrated that ovarian cancer tissue comprises diverse cell types, including epithelial cells, mesenchymal cells, endothelial cells, and immune cells, each contributing differently to tumor biology. UMAP plots (Fig.  6 C-H) and violin plots (Fig.  5 I-N) respectively illustrate the expression levels of the six key genes across different cellular subpopulations and cell types. We observed that GAPDH, CASP3, PPARG, ESR1, CYCS , and SIRT1 exhibited elevated expression levels in tumor cells, mesenchymal cells, endothelial cells, and immune cells compared to other cell types. This suggests that these genes may influence the occurrence and progression of ovarian cancer through multiple mechanisms. The elevated expression of metabolic genes such as GAPDH and PPARG in tumor and mesenchymal cells highlights their potential role in cancer metabolism and survival. Similarly, the heightened expression of ESR1 and SIRT1 in immune cells points to their involvement in immune modulation and tumor microenvironment interactions. These observations provide a foundation for further exploration of the biological pathways and molecular mechanisms underlying ovarian cancer progression, particularly in the context of cellular heterogeneity and intercellular communication. Fig. 6 Key gene expression levels in ovarian cancer at the single-cell level in GSE173682 . A Cellular subpopulation identification in ovarian cancer tissue, with each color representing a distinct subpopulation. B Cellular distribution across different stages of ovarian cancer, where different colors correspond to different cell types. C - H UMAP plots illustrate expression levels of six key genes ( GAPDH, CASP3, PPARG, ESR1, CYCS, SIRT1 ) across cell subpopulations, with a color scale from blue (low) to red (high). I-N Violin plots illustrate expression levels of six key genes ( GAPDH, CASP3, PPARG, ESR1, CYCS, SIRT1 ) in different cell types. The color coding from blue to red represents low to high expression levels, respectively Key gene expression levels in ovarian cancer at the single-cell level in GSE173682 . A Cellular subpopulation identification in ovarian cancer tissue, with each color representing a distinct subpopulation. B Cellular distribution across different stages of ovarian cancer, where different colors correspond to different cell types. C - H UMAP plots illustrate expression levels of six key genes ( GAPDH, CASP3, PPARG, ESR1, CYCS, SIRT1 ) across cell subpopulations, with a color scale from blue (low) to red (high). I-N Violin plots illustrate expression levels of six key genes ( GAPDH, CASP3, PPARG, ESR1, CYCS, SIRT1 ) in different cell types. The color coding from blue to red represents low to high expression levels, respectively Three types of phthalates exhibited docking scores ranging from −5.5 to −7.7 kcal/mol with proteins (PDB IDs: 8P6E ( CCND1 ), 1J3S ( CYCS ), 1SJ0 ( ESR1 ), 1ZNQ ( GAPDH ), 3ET1 ( PPARA ), and 4IF6 ( SIRT1 )), indicating stable binding affinities between the phthalates and the core targets. Based on the docking scores, we retained the results with the highest scores (Table  2 ). The interactions between phthalates and the core targets were analyzed based on their binding affinities, which are critical for understanding the potential molecular mechanisms underlying phthalate-induced biological effects. Table 2 Molecular docking results of three phthalates with 6 core targets Chemical Target Vina score Cavity volume (Å3) DEP CCND1 −5.5 759 CYCS −6 1208 ESR1 −6.2 1995 GAPDH −5.7 1945 PPARA −6.6 3755 SIRT1 −6.7 4634 DMP CCND1 −5.6 635 CYCS −6 1208 ESR1 −6.1 1995 GAPDH −5.5 9014 PPARA −6 3755 SIRT1 −6.4 4634 DOP CCND1 −5.6 635 CYCS −5.5 1208 ESR1 −7 1995 GAPDH −6 1945 PPARA −7.7 3755 SIRT1 −7.6 4634 Molecular docking results of three phthalates with 6 core targets Among the three phthalates, DEP demonstrated docking scores ranging from −5.5 to −6.7 kcal/mol with the core targets, with the highest binding affinity to 4IF6 ( SIRT1 ) at −6.7 kcal/mol (Fig.  7 A). DMP showed docking scores ranging from −5.5 to −6.4 kcal/mol with the core targets, with the highest binding affinity to 4IF6 ( SIRT1 ) at −6.4 kcal/mol (Fig.  7 B). DOP exhibited docking scores ranging from −5.5 to −7.7 kcal/mol with the core targets, with the highest binding affinity to 4IF6 ( SIRT1 ) at −7.7 kcal/mol, followed by 3ET1 ( PPARA ) at −7.6 kcal/mol (Fig.  7 C-D). The docking scores suggest that DOP has the strongest binding affinity among the three phthalates, particularly with SIRT1 and PPARA , which are implicated in cancer metabolism and progression. These bindings may disrupt the normal functions of these proteins, potentially influencing key biological processes such as cell cycle regulation ( CCND1 ), apoptosis ( CYCS ), hormone signaling ( ESR1 ), energy metabolism ( GAPDH ), lipid metabolism ( PPARA ), and epigenetic regulation ( SIRT1 ). These interactions provide avenues for further exploration of phthalates'roles in ovarian cancer tumorigenesis and progression. Fig. 7 Molecular docking results. A Docking results of DEP with SIRT1 ; B Docking results of DMP with SIRT1 ; C Docking results of DOP with SIRT1 ; D Docking results of DOP with PPARA Molecular docking results. A Docking results of DEP with SIRT1 ; B Docking results of DMP with SIRT1 ; C Docking results of DOP with SIRT1 ; D Docking results of DOP with PPARA

The chemical structures of DEP, DMP, and DOP were retrieved from the PubChem database (CID: 6781, 8554, 8343 respectively; accessed November 15, 2024). These structures were subsequently imported into three target prediction platforms: STITCH (Version 5.0), PharmMapper (Version 2017) and SwissTargetPrediction ( http://www.swisstargetprediction.ch/index.php ), with species restricted to Homo sapiens Targets predicted across all platforms were aggregated, duplicates removed, and nomenclature standardized using UniProt (Release 2024_06). This curated target library served as the foundation for analyzing molecular interactions and health implications of phthalates. To evaluate carcinogenic potential, we queried three toxicity databases: Chemical Toxicity Database ( https://www.drugfuture.com/ ), ProTox-3.0 ( https://tox.charite.de/protox3/ ), and ADMETlab 2.0 ( https://admetmesh.scbdd.com/service/evaluation/cal ). Data extraction focused on carcinogenicity endpoints, with prioritization of ovarian cancer-associated mechanisms. Ovarian cancer targets were identified using OMIM (Update 2024–11) and GeneCards (Version 5.22) with search terms "ovarian neoplasms", "ovarian cancer" and "malignant ovarian tumors". A median score threshold filtered genes, retaining those with scores ≥ median. Intersection analysis identified shared targets between phthalates and ovarian cancer, defining candidate phthalate-associated ovarian cancer targets. Intersecting targets were analyzed in STRING 12.0 ( https://cn.string-db.org/ ) under Homo sapiens-specific settings, full network mode, confidence score > 0.4, and all prediction methods enabled (parameters integrated for conciseness). Interaction networks were visualized in Cytoscape 3.9.0, and core genes were selected using three topological criteria: Closeness centrality > median; Radiality > median; Degree > 2 × median. Top-ranking genes underwent molecular docking to probe roles in phthalate-induced ovarian cancer. Metascape 3.5 facilitated Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses [ 32 ] Biological processes (BP), cellular components (CC), and molecular functions (MF) were evaluated for target roles. KEGG pathways with false discovery rate (FDR) < 0.05 were deemed significant. Core targets underwent focused KEGG analysis to identify critical signaling pathways. Results were visualized to enhance interpretability. Gene Expression Profiling Interactive Analysis (GEPIA; http://gepia.cancer-pku.cn ) compared gene expression between 9,736 tumor (TCGA) and 8,587 normal (GTEx) samples. Differentially expressed genes met |log2(fold change)|> 1 and p  < 0.05. Expression values were log2(TPM + 1)-transformed, with log2FC calculated as median(tumor) − median(normal). ScCancerExplorer 1.0 (Updated October 2024) was employed to investigate single-cell-level heterogeneity of phthalate-associated ovarian cancer targets. This cloud-based analytical platform enables integrated multi-omics analysis of > 6 million single-cell profiles across > 50 human cancer types, with prioritization of reproductive malignancies. Leveraging the GSE173682 dataset, we mapped the expression of six key genes across distinct cell clusters, tumor stages, and histopathological subtypes. The dataset comprises 12 samples—six ovarian carcinomas and six endometrial carcinomas—with ovarian cancer representation spanning stage IA endometrioid, stage IIIA serous, stage IIB/IIIC high-grade serous ovarian carcinoma (HGSOC), stage IV gastrointestinal stromal tumor (GIST), and stage IVB carcinosarcoma. Crystal structures of core targets were obtained from RCSB PDB ( https://www.rcsb.org ) and preprocessed in PyMOL 3.0.3 (water/ligand removal). CB-Dock 2 (Release 2022) performed multi-cavity docking ( n  = 5), followed by binding affinity (kcal/mol) calculations and binding mode evaluations. The overall workflow (Fig.  1 ) is as follows. Fig. 1 The workflow of the analysis The workflow of the analysis

This study demonstrates that phthalates (DEP, DMP, DOP) may contribute to ovarian cancer development by dysregulating metabolic pathways ( HIF-1 /glycolysis), strongly binding to key regulators ( SIRT1/PPARA ), and remodeling the tumor microenvironment, as evidenced through integrated computational analyses. While requiring experimental validation, these findings establish a framework for environmental carcinogen screening, identify potential exposure biomarkers, and provide scientific basis for reevaluating high-risk phthalates, bridging computational predictions with testable hypotheses for understanding environmental triggers in ovarian cancer.

Our integrated bioinformatic and structural biology approach reveals a convergent mechanism by which structurally diverse phthalates may promote ovarian carcinogenesis through metabolic reprogramming and endocrine disruption. The study provides three key advances: (1) identification of conserved phthalate-targeted pathways across molecular and cellular scales, (2) structural evidence for preferential binding to epigenetic/metabolic regulators, and (3) a novel hypothesis about phthalate-induced multi-cellular crosstalk in the tumor microenvironment. The consistent enrichment of glycolysis and HIF-1 signaling pathways among phthalate targets suggests these chemicals may induce a Warburg-like metabolic shift in ovarian cells. While previous studies linked phthalates to endocrine disruption [ 33 – 35 ], our network analysis reveals an underappreciated dimension—their potential to hijack energy metabolism through conserved targets like GAPDH and PPARA . Notably, the 8 shared target genes (e.g., GPI, TPI1 ) all participate in rate-limiting glycolytic steps, forming a"metabolic vulnerability axis"that could explain epidemiological associations between phthalate exposure and aggressive tumor phenotypes [ 36 ]. Unexpectedly, DOP's superior binding affinity to SIRT1 (−7.7 kcal/mol) and PPARA (−7.6 kcal/mol) suggests alkyl chain length may determine metabolic interference potency. This structure–activity relationship aligns with our chemical analysis, where branched-chain phthalates showed greater molecular interactions. We hypothesize that longer alkyl chains enhance hydrophobic interactions with metabolic enzymes'allosteric pockets—a mechanism warranting crystallographic validation. Single-cell data unveil a previously unrecognized dimension: phthalate targets show cell-type-specific expression patterns that may orchestrate tumor-stroma-immune crosstalk. For example: (1) ESR1 overexpression in tumor-associated macrophages could polarize them toward pro-tumorigenic phenotypes via estrogen receptor signaling [ 37 ]. (2) SIRT1 suppression in endothelial cells might impair vascular normalization, fostering hypoxic niches that drive metastasis [ 38 ]. These patterns suggest phthalates don't merely act on cancer cells, but may remodel the entire tumor ecosystem—a hypothesis supported by recent single-cell studies of endocrine disruptors [ 39 ]. The dysregulation patterns of core targets (e.g., CCND1, SIRT1) correlate strongly with poor prognosis in TCGA data, positioning them as potential biomarkers for phthalate-associated ovarian cancer subtypes. Particularly, GAPDH's dual role as a glycolytic enzyme and transcriptional regulator [ 40 ] may explain its central network position, making it a promising target for metabolic therapy. However, our study has important limitations: computational predictions of phthalate-protein interactions require experimental validation using techniques like surface plasmon resonance. The scRNA-seq data derive from bulk tumor analyses, necessitating future single-cell exposure studies. Pathway enrichments, while statistically robust, don't establish causality between phthalate exposure and pathway activation.

Recent advances in computational oncology, integrating computational intelligence and bioinformatics, have revolutionized cancer research by enabling systematic analysis of multi-omics data. Large-scale initiatives like The Cancer Genome Atlas (TCGA) ( https://tcga-data.nci.nih.gov/tcga/ ) and the International Cancer Genome Consortium (ICGC) [ 1 ] provide expansive genomic datasets that support mutation profiling, drug sensitivity prediction, and prognostic modeling. Tools such as cBioPortal [ 2 ] and Genomics of Drug Sensitivity in Cancer (GDSC) [ 3 ] accelerate the discovery of driver genes and therapeutic targets through data mining, while metabolomics platforms like XCMS [ 4 ] and MZmine [ 5 ] elucidate tumor metabolic regulation. Artificial intelligence further enhances precision medicine through applications in medical imaging, drug repositioning, and patient stratification [ 6 ], exemplified by deep learning models linking imaging phenotypes to genomic profiles [ 7 , 8 ]. These advancements underscore the transformative potential of computational tools in oncology, though challenges like data interoperability and algorithmic complexity persist, necessitating interdisciplinary collaboration for clinical translation. Ovarian cancer ranks as the seventh most prevalent malignancy and the eighth leading cause of cancer-related mortality among women globally, with approximately 239,000 new cases and 152,000 deaths annually [ 9 , 10 ]. Incidence rates exhibit significant regional disparities, notably higher in Western countries compared to other regions [ 11 ]. Epidemiological evidence indicates that 90% of ovarian cancer cases are correlated with environmental and lifestyle factors, while hereditary factors account for only 5%–10%, underscoring the critical role of environmental exposures in cancer etiology [ 12 ]. Key environmental risk factors include talc, pesticide derivatives (e.g., organophosphates, carbamates), and heavy metals (e.g., cadmium, lead, mercury), which elevate ovarian cancer risk through oxidative stress, DNA damage, and hormonal disruption [ 13 – 15 ]. Prolonged residential exposure to industrial pollutants or heavy traffic zones further increases mortality risk due to air and water contamination [ 16 , 17 ], while airborne particulate matter (PM2.5) and nitrogen oxides ( NO2 ) exacerbate carcinogenesis via systemic inflammation [ 18 , 19 ]. Gene-environment interactions amplify susceptibility, particularly in genetically predisposed individuals, highlighting the need for targeted prevention strategies [ 20 ]. Phthalates, a class of endocrine-disrupting plasticizers ubiquitous in consumer products (e.g., cosmetics, food packaging, medical devices), pose significant risks to women’s health [ 21 – 23 ]. Chronic exposure is associated with reproductive dysfunction, including diminished ovarian reserve, menstrual irregularities, and reduced IVF success rates [ 24 , 25 ], as well as pregnancy complications such as preterm birth and gestational diabetes, likely mediated by oxidative stress and inflammatory pathways [ 26 , 27 ]. These chemicals are also implicated in gynecological disorders (e.g., PCOS, endometriosis) and metabolic conditions (e.g., obesity, type 2 diabetes) through insulin resistance and glucose dysregulation [ 28 , 29 ]. Emerging evidence links phthalates to malignancies such as breast cancer and pediatric lymphoma [ 30 , 31 ]; however, their association with ovarian cancer remains unexplored. Given their endocrine-mimicking properties and widespread use, elucidating phthalate-induced ovarian carcinogenesis is imperative for refining preventive measures and public health policies. This study employs network toxicology—an interdisciplinary framework integrating bioinformatics, systems biology, and chemical informatics—to investigate interactions between common phthalates (diethyl phthalate [DEP], dimethyl phthalate [DMP], and dioctyl phthalate [DOP]) and ovarian cancer-associated proteins. Molecular docking complements this approach by simulating atomic-level binding patterns between phthalates and target proteins, revealing mechanisms through which these chemicals may drive carcinogenesis. By synergizing these methods, the research aims to decode how routine phthalate exposure influences ovarian cancer initiation and progression. The findings will advance phthalate safety assessments and inform prevention strategies, leveraging molecular insights to optimize therapeutic interventions.

Supplementary Material 1 Supplementary Material 1