A risk model of gene signatures for predicting platinum response and survival in ovarian cancer

Integrating bulk RNA-seq and scRNA-seq analyses with machine learning to predict platinum response and prognosis in ovarian cancer

Platinum-based therapy is an integral part of the standard treatment for ovarian cancer. However, despite extensive research spanning several decades, the identification of dependable predictive biomarkers for platinum response in clinical practice has proven to be a formidable challenge. Recently, the development of single-cell technology has enabled more precise investigations into the heterogeneity of cancer. In this study, we isolated cancer cells from the single-cell transcriptomic data of platinum-sensitive and platinum-resistant patients with ovarian cancer. Differential gene analysis of platinum-sensitive and platinum-resistant cancer cells revealed that several of the differentially expressed genes had previously been reported in other studies to be associated with platinum resistant. Gene set enrichment analysis revealed the up-regulation of pathways involved in processes such as autophagy, cell cycle regulation, and DNA damage repair, which are known to promote platinum resistance in ovarian cancer. Based on these findings, we hypothesized that these differentially expressed genes could be used to predict the response of ovarian cancer patients to platinum-based chemotherapy. To validate this hypothesis, we explored 7 different machine learning models for predicting platinum chemotherapy response at varying feature gene counts. Ultimately, the random forest model performed the best, with 5 genes (PAX2, TFPI2, APOA1, ADIRF and CRISP3) and achieve an AUC of 0.993 in test cohort and 0.989 in GSE63885 independent validation cohorts. We named this model GPPS (Genes to Predict Platinum response Signature). Furthermore, we discovered that the GPPS model can also predict patient prognosis.

Prognosis and immune infiltration prediction in neuroblastoma based on neutrophil extracellular traps-related gene signature

Neuroblastoma (NB) is a malignant tumor originating from the peripheral sympathetic nervous system and high-risk NB patients have a dismal prognosis. Recent studies have underscored the pivotal role of neutrophil extracellular traps (NETs) in the proliferation, metastasis and immune evasion of cancer. To explore the effect of NETs on NB, we have carried out a systematic analysis and showed several findings in the present work. First, expression profiles along with clinical data were analyzed using the training dataset GSE62564 and 36 NETs-related genes were identified to be significantly associated with overall survival. Following LASSO regression analysis, 11 genes were enrolled to construct the NETs signature, which exhibited a robust predictive capability for overall survival with exhibiting high AUC values within the training set. Validation cohorts confirmed a similar predictive efficacy. Next, NB patients were classified into subgroups based on median risk scores and differentially expressed genes were analyzed. Furthermore, the study performed comprehensive analyses encompassing functional enrichment, immune infiltration and drug sensitivity. Enrichment analysis revealed that the high-risk NBs with high-risk score displayed characteristics of oncogenic malignancy, poor prognosis and immunosuppression. Notably, the risk score exhibited a strong correlation with infiltration levels of various immune cells and the sensitivity to anti-cancer drugs, and was further recognized as an independent prognostic factor for NB patients. In summary, our study elucidates a novel NETs-related gene signature comprising 11 genes, which serves a reliable predictor for NB prognosis.

Hesperidin inhibits colon cancer progression by downregulating SLC5A1 to suppress EGFR phosphorylation

Objective: Hesperidin, an active constituent of traditional Chinese medicine, Chenpi, exhibits anticancer properties across different cancers. This study aimed to clarify the efficacy of Hesperidin against tumors and its mechanisms of action in colon cancer. Method: We assessed the efficacy of Hesperidin on human colon cancer cells (HCT-116 and DLD-1) and normal colonic epithelial cells (NCM460). We quantified cell viability at various Hesperidin concentrations using the CCK8 assay in a series of experiments. We employed clone formation, EdU incorporation, Transwell, and wound healing assays to clarify Hesperidin efficacy on cancer cell proliferation, invasion, and migration. Western blot analyses revealed modulations in epithelial-mesenchymal transition-related proteins, SLC5A1, EGFR, and phosphorylated EGFR levels following Hesperidin exposure. Co-IP assays further validated the interaction between SLC5A1 and EGFR. Our findings were significantly restored following SLC5A1 overexpression in colon cancer cells, highlighting its pivotal role in Hesperidin-induced responses. Results: Hesperidin selectively impaired the viability of HCT-116 and DLD-1 colon cancer cells at specific concentrations while preserving normal NCM460 cells. This flavonoid dose-dependently reduced cancer cell proliferation, migration, and invasion. It significantly suppressed SLC5A1 and phosphorylated EGFR expression. We identified a direct SLC5A1-EGFR interaction essential for regulating EGFR activity in colon cancer. Overexpressing SLC5A1 significantly reduced the inhibitory effects of Hesperidin, highlighting its crucial role in this context. Conclusion: Hesperidin exerts its anticancer effects on colon cancer by inhibiting SLC5A1 expression and consequently downregulating EGFR phosphorylation.

Development and Validation of a Predictive Model for Resistance to Platinum-Based Chemotherapy in Patients with Ovarian Cancer through Proteomic Analysis

Platinum resistance in ovarian cancer poses a significant challenge, substantially impacting patient outcomes. Developing an accurate predictive model is crucial for improving clinical decision-making and guiding treatment strategies. Proteomic data from 217 high-grade serous ovarian cancer (HGSOC) biospecimens obtained from JHU, PNNL, and PTRC were used to construct a prediction model for identifying individuals who are resistant to platinum-based chemotherapy. A total of 6437 common proteins were detected across all data sets, with 26 proteins overlapping between the development cohorts JHU and PNNL. Using LASSO and logistic regression analysis, a six-protein model (P31323_PRKAR2B, Q13309_SKP2, Q14997_PSME4, Q6ZRP7_QSOX2, Q7LGA3_HS2ST1, and Q7Z2Z2_EFL1) was developed, which accurately predicted platinum resistance, with an AUC of 0.964 (95% CI, 0.929-0.999). Internal validation by resampling resulted in a C-index of 0.972 (95% CI 0.894-0.988). External validation performed on the PTRC cohort achieved an AUC of 0.855 (95% CI 0.748-0.963). Calibration curves showed good consistency, and DCA indicated superior clinical utility. The model also performed well in predicting PFS and OS at various time points. Based on these proteins, our predictive model can precisely predict platinum response and survival outcomes in HGSOC patients, which can assist clinicians in promptly identifying potentially platinum-resistant individuals.

Senescence-Related LncRNAs: Pioneering Indicators for Ovarian Cancer Outcomes

In gynecological oncology, ovarian cancer (OC) remains the most lethal, highlighting its significance in public health. Our research focused on the role of long non-coding RNA (lncRNA) in OC, particularly senescence-related lncRNAs (SnRlncRNAs), crucial for OC prognosis. Utilizing data from the genotype-tissue expression (GTEx) and cancer genome Atlas (TCGA), SnRlncRNAs were discerned and subsequently, a risk signature was sculpted using co-expression and differential expression analyses, Cox regression, and least absolute shrinkage and selection operator (LASSO). This signature's robustness was validated through time-dependent receiver operating characteristics (ROC), and multivariate Cox regression, with further validation in the international cancer genome consortium (ICGC). Gene set enrichment analyses (GSEA) unveiled pathways intertwined with risk groups. The ROC, alongside the nomogram and calibration outcomes, attested to the model's robust predictive accuracy. Of particular significance, our model has demonstrated superiority over several commonly utilized clinical indicators, such as stage and grade. Patients in the low-risk group demonstrated greater immune infiltration and varied drug sensitivities compared to other groups. Moreover, consensus clustering classified OC patients into four distinct groups based on the expression of 17 SnRlncRNAs, showing diverse survival rates. In conclusion, these findings underscored the robustness and reliability of our model and highlighted its potential for facilitating improved decision-making in the context of risk assessment, and demonstrated that these markers potentially served as robust, efficacious biomarkers and prognostic tools, offering insights into predicting OC response to anticancer therapeutics.

Supplementary Information

The online version contains supplementary material available at 10.1007/s43657-024-00163-z.

Integrated analysis of single-cell and bulk RNA-sequencing identifies a signature based on drug response genes to predict prognosis and therapeutic response in ovarian cancer

Ovarian cancer represents a severe gynecological malignancy with a dire prognosis, underscoring the imperative need for dependable biomarkers that can accurately predict drug response and guide therapeutic choices. In this study, we harnessed online single-cell RNA sequencing (scRNAseq) and bulk RNA sequencing (RNAseq) datasets, applying the Scissor algorithm to identify cells responsive to paclitaxel. From these cells, we derived a gene signature, subsequently used to construct a prognostic model that demonstrated high sensitivity and specificity in predicting patient outcomes. Moreover, we conducted pathway and functional enrichment analyses to uncover potential molecular mechanisms driving the prognostic gene signature. This study illustrates the critical role of scRNAseq and bulk RNAseq in developing precise prognostic models for ovarian cancer, potentially transforming clinical decision-making.

Establishment of an ovarian cancer exhausted CD8+T cells-related genes model by integrated analysis of scRNA-seq and bulk RNA-seq

Ovarian cancer (OC) was the fifth leading cause of cancer death and the deadliest gynecological cancer in women. This was largely attributed to its late diagnosis, high therapeutic resistance, and a dearth of effective treatments. Clinical and preclinical studies have revealed that tumor-infiltrating CD8+T cells often lost their effector function, the dysfunctional state of CD8+T cells was known as exhaustion. Our objective was to identify genes associated with exhausted CD8+T cells (CD8TEXGs) and their prognostic significance in OC. We downloaded the RNA-seq and clinical data from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. CD8TEXGs were initially identified from single-cell RNA-seq (scRNA-seq) datasets, then univariate Cox regression, the least absolute shrinkage and selection operator (LASSO), and multivariate Cox regression were utilized to calculate risk score and to develop the CD8TEXGs risk signature. Kaplan-Meier analysis, univariate Cox regression, multivariate Cox regression, time-dependent receiver operating characteristics (ROC), nomogram, and calibration were conducted to verify and evaluate the risk signature. Gene set enrichment analyses (GSEA) in the risk groups were used to figure out the closely correlated pathways with the risk group. The role of risk score has been further explored in the homologous recombination repair deficiency (HRD), BRAC1/2 gene mutations and tumor mutation burden (TMB). A risk signature with 4 CD8TEXGs in OC was finally built in the TCGA database and further validated in large GEO cohorts. The signature also demonstrated broad applicability across various types of cancer in the pan-cancer analysis. The high-risk score was significantly associated with a worse prognosis and the risk score was proven to be an independent prognostic biomarker. The 1-, 3-, and 5-years ROC values, nomogram, calibration, and comparison with the previously published models confirmed the excellent prediction power of this model. The low-risk group patients tended to exhibit a higher HRD score, BRCA1/2 gene mutation ratio and TMB. The low-risk group patients were more sensitive to Poly-ADP-ribose polymerase inhibitors (PARPi). Our findings of the prognostic value of CD8TEXGs in prognosis and drug response provided valuable insights into the molecular mechanisms and clinical management of OC.

Prediction of Response to Chemoradiotherapy by Dynamic Changes of Circulating Exosome Levels in Patients with Esophageal Squamous Cell Carcinoma

Background

The exosomes-based liquid biopsy represents a prospective biomarker for tumor screening, prognosis prediction, and tumor regression. This study aimed to isolate circulating exosomes (CEs) from plasma of the esophageal squamous cell carcinoma (ESCC) patients who received chemoradiotherapy through exosome detection method via the ultrafast-isolation system (EXODUS) and investigated the association between the dynamic changes of CE levels and therapeutic effect.

Methods

We isolated and quantitatively analyzed CEs from plasma of locally advanced ESCC patients received chemoradiotherapy at 2 time points: baseline (pre-chemoradiotherapy) and 2 months after the chemoradiotherapy (post-chemoradiotherapy). We isolated exosomes from plasma by EXODUS platform and confirmed them through nanoparticle tracking analysis (NTA), transmission electron microscope (TEM), and Western blot. The associations of CE level with clinicopathological characteristics, tumor regression, and progression-free survival (PFS) were analyzed.

Results

The average diameter of CEs was 107.4±14.3 nm at baseline and 101.7±17.1 nm at post-chemoradiotherapy. The mean exosome concentration significantly decreased after chemoradiotherapy (7.3×1011 particles/mL vs 5.4×1011 particles/mL, P < 0.001). The patients with stage III-IVA and tumor length ≥5cm had obviously higher baseline CE levels. Dynamic changes in CE levels were successfully applied for evaluation of chemoradiotherapy response and PFS. Furthermore, through multivariate Cox regression analysis, it was revealed that dynamic changes of CE levels were an independent predictor of PFS in locally advanced ESCC patients who received chemoradiotherapy.

Conclusion

Here, we demonstrated EXODUS platform isolated and enriched CEs from plasma of ESCC patients with high-purity and high-yield. The EXODUS platform can facilitate liquid biopsy based on exosomes translation to the clinic. Baseline CE levels can reflect ESCC tumor burden. The dynamic changes of CE levels during chemoradiotherapy allow the prediction of treatment effect and PFS of ESCC patients, requiring further investigations in larger patient cohorts.

Heterogeneity of computational pathomic signature predicts drug resistance and intra-tumor heterogeneity of ovarian cancer

BACKGROUND

Chemotherapy resistance is the main cause of ovarian cancer progression and even death. However, there are no clear indicators for predicting the risk of drug resistance in patients. Intra-tumor heterogeneity (ITH) is one of the characteristics of malignant tumors, which is associated with the treatment and prognosis of tumors. Accordingly, our study aims to investigate the correlation between the image features of intra-tumor heterogeneity and drug resistance of ovarian cancer based on artificial intelligence.

METHODS

We obtained hematoxylin and eosin staining frozen histopathological images of ovarian cancer and paracarcinoma tissues from the Cancer Genome Atlas. We extracted quantitative image features of whole-slide images based on the automatic image nuclear segmentation processing technology. After that, we used bioinformatics analysis to find the relationship between image features of intra-tumor heterogeneity and drug resistance.

RESULTS

Our results show that our automatic image processing process based on computer artificial intelligence can extract image features effectively, and the key image features extracted are closely related to ITH. Among them, the Perimeter.sd image feature with the most prominent ITH feature can accurately predict the risk of platinum-based chemotherapy drug resistance in ovarian cancer patients.

CONCLUSION

Automatic image processing and feature extraction based on artificial intelligence have excellent results. Perimeter.sd can be used as a useful image feature indicator for evaluating ITH. ITH is associated with drug resistance of ovarian cancer, so ITH characteristics can be used as an effective indicator to evaluate drug resistance in patients with ovarian cancer.

Predictive Value of Machine Learning for Platinum Chemotherapy Responses in Ovarian Cancer: Systematic Review and Meta-Analysis

BACKGROUND

Machine learning is a potentially effective method for predicting the response to platinum-based treatment for ovarian cancer. However, the predictive performance of various machine learning methods and variables is still a matter of controversy and debate.

OBJECTIVE

This study aims to systematically review relevant literature on the predictive value of machine learning for platinum-based chemotherapy responses in patients with ovarian cancer.

METHODS

Following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, we systematically searched the PubMed, Embase, Web of Science, and Cochrane databases for relevant studies on predictive models for platinum-based therapies for the treatment of ovarian cancer published before April 26, 2023. The Prediction Model Risk of Bias Assessment tool was used to evaluate the risk of bias in the included articles. Concordance index (C-index), sensitivity, and specificity were used to evaluate the performance of the prediction models to investigate the predictive value of machine learning for platinum chemotherapy responses in patients with ovarian cancer.

RESULTS

A total of 1749 articles were examined, and 19 of them involving 39 models were eligible for this study. The most commonly used modeling methods were logistic regression (16/39, 41%), Extreme Gradient Boosting (4/39, 10%), and support vector machine (4/39, 10%). The training cohort reported C-index in 39 predictive models, with a pooled value of 0.806; the validation cohort reported C-index in 12 predictive models, with a pooled value of 0.831. Support vector machine performed well in both the training and validation cohorts, with a C-index of 0.942 and 0.879, respectively. The pooled sensitivity was 0.890, and the pooled specificity was 0.790 in the training cohort.

CONCLUSIONS

Machine learning can effectively predict how patients with ovarian cancer respond to platinum-based chemotherapy and may provide a reference for the development or updating of subsequent scoring systems.

Personalization of Therapy in High-Grade Serous Tubo-Ovarian Cancer-The Possibility or the Necessity?

High-grade serous tubo-ovarian cancer (HGSTOC) is the most lethal tumor of the female genital tract. The foregoing therapy consists of cytoreduction followed by standard platinum/taxane chemotherapy; alternatively, for primary unresectable tumors, neo-adjuvant platinum/taxane chemotherapy followed by delayed interval cytoreduction. In patients with suboptimal surgery or advanced disease, different forms of targeted therapy have been accepted or tested in clinical trials. Studies on HGSTOC discovered its genetic and proteomic heterogeneity, epigenetic regulation, and the role of the tumor microenvironment. These findings turned attention to the fact that there are several distinct primary tumor subtypes of HGSTOC and the unique biology of primary, metastatic, and recurrent tumors may result in a differential drug response. This results in both chemo-refractoriness of some primary tumors and, what is significantly more frequent and destructive, secondary chemo-resistance of metastatic and recurrent HGSTOC tumors. Treatment possibilities for platinum-resistant disease include several chemotherapeutics with moderate activity and different targeted drugs with difficult tolerable effects. Therefore, the question appears as to why different subtypes of ovarian cancer are predominantly treated based on the same therapeutic schemes and not in an individualized way, adjusted to the biology of a specific tumor subtype and temporal moment of the disease. The paper reviews the genomic, mutational, and epigenetic signatures of HGSTOC subtypes and the tumor microenvironment. The clinical trials on personalized therapy and the overall results of a new, comprehensive approach to personalized therapy for ovarian cancer have been presented and discussed.

Endoplasmic Reticulum Stress-Related Ten-Biomarker Risk Classifier for Survival Evaluation in Epithelial Ovarian Cancer and TRPM2: A Potential Therapeutic Target of Ovarian Cancer

Epithelial ovarian cancer (EOC) is the most lethal gynecological malignant tumor. Endoplasmic reticulum (ER) stress plays an important role in the malignant behaviors of several tumors. In this study, we established a risk classifier based on 10 differentially expressed genes related to ER stress to evaluate the prognosis of patients and help to develop novel medical decision-making for EOC cases. A total of 378 EOC cases with transcriptome data from the TCGA-OV public dataset were included. Cox regression analysis was used to establish a risk classifier based on 10 ER stress-related genes (ERGs). Then, through a variety of statistical methods, including survival analysis and receiver operating characteristic (ROC) methods, the prediction ability of the proposed classifier was tested and verified. Similar results were confirmed in the GEO cohort. In the immunoassay, the different subgroups showed different penetration levels of immune cells. Finally, we conducted loss-of-function experiments to silence TRPM2 in the human EOC cell line. We created a 10-ERG risk classifier that displays a powerful capability of survival evaluation for EOC cases, and TRPM2 could be a potential therapeutic target of ovarian cancer cells.

Gynecological cancer prognosis using machine learning techniques: A systematic review of last three decades (1990–2022)

OBJECTIVE

Many Computer Aided Prognostic (CAP) systems based on machine learning techniques have been proposed in the field of oncology. The objective of this systematic review was to assess and critically appraise the methodologies and approaches used in predicting the prognosis of gynecological cancers using CAPs.

METHODS

Electronic databases were used to systematically search for studies utilizing machine learning methods in gynecological cancers. Study risk of bias (ROB) and applicability were assessed using the PROBAST tool. 139 studies met the inclusion criteria, of which 71 predicted outcomes for ovarian cancer patients, 41 predicted outcomes for cervical cancer patients, 28 predicted outcomes for uterine cancer patients, and 2 predicted outcomes for gynecological malignancies broadly.

RESULTS

Random forest (22.30 %) and support vector machine (21.58 %) classifiers were used most commonly. Use of clinicopathological, genomic and radiomic data as predictors was observed in 48.20 %, 51.08 % and 17.27 % of studies, respectively, with some studies using multiple modalities. 21.58 % of studies were externally validated. Twenty-three individual studies compared ML and non-ML methods. Study quality was highly variable and methodologies, statistical reporting and outcome measures were inconsistent, preventing generalized commentary or meta-analysis of performance outcomes.

CONCLUSION

There is significant variability in model development when prognosticating gynecological malignancies with respect to variable selection, machine learning (ML) methods and endpoint selection. This heterogeneity prevents meta-analysis and conclusions regarding the superiority of ML methods. Furthermore, PROBAST-mediated ROB and applicability analysis demonstrates concern for the translatability of existing models. This review identifies ways that this can be improved upon in future works to develop robust, clinically translatable models within this promising field.

Increased response to TPF chemotherapy promotes immune escape in hypopharyngeal squamous cell carcinoma

Background: There is an urgent need to identify which patients would benefit from TPF chemotherapy in hypopharyngeal squamous cell carcinoma (HPSCC) and to explore new combinations to improve the treatment effect. Materials and methods: Gene-expression profiles in 15 TPF-sensitive patients were compared to 13 resistant patients. Immunohistochemistry (IHC) was performed to detect CD8⁺ T cells in 28 samples. Patient-Derived Tumor Xenograft (PDX) model and IHC were used to verify markers that optimize treatment for HPSCC. Results: Through RNA sequencing 188 genes were up-regulated in TPF chemotherapy-resistant (CR) tissues were involved in T cell activation, while 60 down-regulated genes were involved in glycolysis. Gene set enrichment analysis (GSEA) showed that chemotherapy-sensitive (CS) group upregulation of the pathways of glycolysis, while immune response was downregulated. CIBERSORT, MCP-counter, and IHC proved that most immune cells including CD8⁺ T cells in the CR significantly higher than that in CS group. Among the 16 up-regulated genes in CS had close associations, the most significant negative correlation between the gene level and CD8⁺ T cells existed in SEC61G. SEC61G was related to glycolysis, which was transcriptionally regulated by E2F1, and participated in antigen degradation through ubiquitin-dependent protein catabolic process. Palbociclib, combined with Cetuximab decreased the tumor burden and significantly suppressed the expression of E2F1 and SEC61G while activating MHC-I in PDX model. Conclusion: Enhanced glycolysis promoted immune escape, but increased response to TPF chemotherapy. SEC61G was the center of the molecular network and targeting the E2F1/SEC61G pathway increased the expression level of MHC-I.

Advances in small molecule maintenance therapies for high-grade serous ovarian cancer

INTRODUCTION

Ovarian cancer is one of the most lethal gynecological tumors with a lack of effective treatment modalities especially in advanced/recurrent disease. Nevertheless, recently, new small molecules have emerged as an effective approach for the management of ovarian cancer patients, especially in the maintenance setting.

AREAS COVERED

This review summarizes the role of small molecules used in the management of high-grade serous ovarian cancer. The authors performed a critical review of current evidence and ongoing studies. Of note, tyrosine kinase inhibitors (TKIs) and poly(ADP-ribose) polymerase (PARP) inhibitors are the most intriguing medications in this setting.

EXPERT OPINION

Protein-targeted therapies against tumor tissues have progressed significantly in the last years due to an enhanced knowledge of the biological and molecular processes of carcinogenesis. Treatment with small molecules allows the targeting of specific proteins involved in cancer biology. TKIs seem promising but further data are necessary to assess the pros and cons of adopting this treatment modality. PARP inhibitors represent the new standard of care for ovarian cancer patients harboring either a BRCA mutation or with homologous recombination deficiency (HRD). Interestingly, the accumulation of data has highlighted that PARP inhibitors provide benefits even in patients with HR proficient tumors.

Development of Nomogram Models Based on Peripheral Blood Score and Clinicopathological Parameters to Predict Preoperative Advanced Stage and Prognosis for Epithelial Ovarian Cancer Patients

Purpose

Nutritional and inflammatory states are crucial in cancer development. The purpose of this study is to construct a scoring system grounded on peripheral blood parameters associated with nutrition and inflammation and explore its value in stage, overall survival (OS), and progression-free survival (PFS) prediction for epithelial ovarian cancer (EOC) patients.

Patients and Methods

Four hundred and fifty-three EOC patients were retrospectively identified and their clinical data and relevant peripheral blood parameters were collected. The ratio of neutrophil to lymphocyte, lymphocyte to monocyte, fibrinogen to lymphocyte, total cholesterol to lymphocyte and albumin level were calculated and dichotomized. A scoring system named peripheral blood score (PBS) was constructed. Univariate and multivariate Logistic or Cox regression analyses were used to select independent factors; these factors were then used to develop nomogram models of advanced stage and OS, PFS, respectively. The internal validation and DCA analysis were performed to evaluate models.

Results

Lower PBS indicated a better prognosis and higher PBS indicated inferior. High PBS is associated with advanced stage, high CA125, serous histological type, poor differentiation, and accompanied ascites. The logistic regression showed age, CA125, and PBS were independent factors for the FIGO III-IV stage. The nomogram models for advanced FIGO stage based on these factors showed good efficiency. FIGO stage, residual disease, and PBS were independent factors affecting OS and PFS, the nomogram models composed of these factors had good performance. DCA curves revealed the models augmented net benefits.

Conclusion

PBS can be a noninvasive biomarker for EOC patients' prognosis. The related nomogram models could be powerful, cost-effective tools to provide information of advanced stage, OS, and PFS for EOC patients.