Dose-response relationship between ascites volume and survival in high-grade serous ovarian cancer: a prospective cohort study

Introduction

Ovarian cancer remains the most lethal gynecologic malignancy worldwide, with outcomes predominantly determined by stage at diagnosis and extent of cytoreduction (1,2). The stark survival disparity between early-stage disease (5-year survival exceeding 90%) and advanced-stage disease (approximately 30–40%) underscores the critical need for refined prognostic stratification (3,4). While current prognostic assessment relies on established clinicopathological parameters including International Federation of Gynecology and Obstetrics (FIGO) stage, residual disease status, TP53 mutation status, and platinum sensitivity, these factors inadequately explain the substantial heterogeneity observed in clinical outcomes among patients with similar disease characteristics (5). This prognostic uncertainty has motivated investigation into disease-specific manifestations, with ascites emerging as a parameter of particular interest given its high prevalence in advanced ovarian cancer and potential associations with tumor biology and peritoneal dissemination (6).

Ascites, the pathological accumulation of fluid in the peritoneal cavity, represents a hallmark manifestation of advanced ovarian cancer, occurring in 30–60% of patients at initial presentation (7,8). The relationship between ascites and ovarian cancer prognosis presents a complex landscape. Ascites is traditionally regarded as a marker of advanced disease burden and extensive peritoneal involvement, reflecting aggressive tumor biology characterized by increased vascular permeability, peritoneal carcinomatosis, and compromised lymphatic drainage (7,9,10). However, whether ascites independently predicts survival beyond established prognostic factors remains controversial, with studies yielding inconsistent findings. Iyoshi et al. examined 1,966 epithelial ovarian cancer patients, finding significant prognostic differences between low-volume (<100 mL) and high-volume (≥100 mL) groups (9). Szender et al. reported that patients with ascites >2,000 mL had significantly worse outcomes, and noted that massive ascites (>2,000 mL) was associated with reduced likelihood of optimal cytoreduction (7). A comprehensive analysis by the Arbeitsgemeinschaft Gynaekologische Onkologie (AGO)-OVAR group demonstrated that ascites >500 mL was independently associated with worse survival (11). While these investigations have associated the presence of ascites with inferior outcomes, others report no independent prognostic significance after adjusting for stage and surgical factors (12,13). These discordant results stem largely from a fundamental methodological limitation: the binary classification of ascites as merely present or absent, which fails to capture the quantitative relationship between ascites volume and clinical outcomes. Furthermore, most studies have not systematically examined whether a dose-response relationship exists between ascites volume and survival, nor have they identified clinically meaningful volume thresholds that could inform treatment decisions.

To address this critical gap, we conducted a prospective cohort study to systematically evaluate the prognostic impact of ascites volume in high-grade serous ovarian cancer (HGSOC) patients. The objectives of this study were to (I) characterize the dose-response relationship between ascites volume measured at primary surgery and progression-free survival (PFS) and overall survival (OS) using restricted cubic spline (RCS) analysis; (II) identify clinically relevant ascites volume thresholds based on nonlinear associations with outcomes; and (III) evaluate the independent prognostic significance of ascites volume categories after adjusting for established prognostic factors. This systematic exploration of ascites volume as a continuous prognostic variable seeks to inform evidence-based treatment planning strategies and guide clinical decision-making for risk stratification in patients with ovarian cancer. We present this article in accordance with the STROBE reporting checklist (available at https://gs.amegroups.com/article/view/10.21037/gs-2025-aw-481/rc).

Methods

Study cohort

The study cohort was derived from a prospective dataset of HGSOC patients at Xiangya Hospital, Central South University, covering cases consecutively enrolled between January 2017 and December 2020, with follow-up completed in June 2023 (14). After applying predefined exclusion criteria, 293 patients with confirmed HGSOC were included (Figure S1). Details of data preprocessing have been reported in our previous article. Missing values were addressed using the K-nearest neighbor (KNN) imputation method, and the imputed dataset served as the basis for the present analysis. All patients underwent cytoreductive surgery and standard platinum-based chemotherapy, either as primary treatment or following neoadjuvant chemotherapy with interval debulking surgery. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Central South University (No. 2017068222). Informed consent was taken from all the patients.

Ascites collection and measurement

Ascites volume measurement follows a standardized protocol that has been consistently implemented throughout the study period. At the time of primary cytoreductive surgery, before any manipulation of the peritoneal cavity or irrigation, free peritoneal fluid is systematically aspirated using suction. The surgical team begins by placing the patient in Trendelenburg position and aspirating fluid from the pelvis, followed by systematic aspiration from the paracolic gutters, Morrison’s pouch, and subphrenic spaces. All aspirated fluid is collected in calibrated suction canisters, and the volume is recorded directly from the canister measurements. Importantly, this measurement is performed before any surgical irrigation or lavage, thereby eliminating contamination from irrigation fluid. In cases where extensive adhesions or loculated fluid prevented complete aspiration, the operating surgeon documented this limitation in the operative report, and we conservatively recorded the measured volume as a minimum estimate.

To ensure consistency across surgeons, all gynecologic oncology surgeons in our department were trained in this standardized protocol. The protocol was reinforced through regular departmental meetings and quality assurance reviews. Additionally, the prospective nature of our study design, with pre-specified data collection forms that explicitly prompted documentation of ascites volume and measurement method, further enhanced measurement consistency.

Variables and outcome

Clinical and pathological variables were prospectively extracted from medical records and included age, body mass index (BMI), tumor characteristics [tumor, node, metastasis (TNM) stage, FIGO stage according to 2014 criteria, histologic subtype, tumor grade], treatment factors (neoadjuvant chemotherapy, surgical cytoreduction status, hyperthermic intraperitoneal chemotherapy), serum tumor markers [carbohydrate antigen 125 (CA125), human epididymal protein 4 (HE4), carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP)], and molecular features (TP53 mutation status by immunohistochemistry or sequencing). Ascites volume was measured intraoperatively at primary cytoreductive surgery and recorded from operative and pathology reports. Patients were categorized into three groups: no ascites (0 mL), low-volume ascites (<1,000 mL), and high-volume ascites (≥1,000 mL). Surgical cytoreduction status was classified as R0 (no visible residual disease), R1 (residual disease ≤1 cm), or R2 (residual disease >1 cm). Platinum sensitivity was defined according to Gynecologic Cancer InterGroup (GCIG) criteria based on platinum-free interval: platinum-sensitive (>12 months), partially platinum-sensitive (6–12 months), or platinum-resistant (<6 months). The primary outcomes were PFS, defined as the time from surgery to disease progression per RECIST 1.1 criteria or CA125 elevation per GCIG criteria or death from any cause, and OS, defined as the time from surgery to death from any cause. Patients without events were censored at the last follow-up.

Statistical analyses

RCS models with three knots positioned at the 10th, 50th, and 90th percentiles of the ascites volume distribution were employed to examine the potential nonlinear associations between ascites volume and survival outcomes (15). Cox proportional hazards models incorporating RCS terms estimated hazard ratios (HRs) with 95% confidence intervals (CIs) for the association between ascites and PFS and OS, while incidence rates (expressed as deaths per 100 person-years) were estimated using three-knot RCS Poisson regression models (16) with an offset term of 100 person-years. The nonlinearity of the association was assessed using a Wald test (17).

The prognostic impact of ascites groups (no ascites, <1,000 mL, ≥1,000 mL) on PFS and OS was evaluated using survival analysis. Unadjusted survival curves (18,19) were generated using the Kaplan-Meier method and compared using the log-rank test. Covariate-adjusted survival curves were derived from multivariable Cox proportional hazards models adjusting for age, FIGO stage, tumor grade, residual disease status, and treatment variables. The effects of ascites groups on survival were quantified using HRs with 95% CIs and 3-year survival probabilities.

Categorical variables were presented as frequencies (percentages) and compared using χ² tests. Continuous variables were expressed as medians [interquartile ranges (IQRs)] and compared using Mann-Whitney U tests. All analyses were performed using R version 4.4.3. Two-sided P values <0.05 were considered statistically significant. Significance levels are denoted as follows: ***P<0.001, **P<0.01, and *P<0.05.

Results

Patients characteristics

A total of 293 patients with ovarian cancer were included in the final analysis and stratified by ascites volume: 45 patients (15.4%) had no ascites, 164 patients (56.0%) had below 1,000 mL, and 84 patients (28.7%) had above 1,000 mL (Table 1). The median age was 54 years (IQR, 48–60 years) with no significant difference among groups (P=0.91). Patients with higher ascites volumes had significantly greater BMI (P=0.003) and more advanced disease characteristics. The majority (86.4%) presented with FIGO stage III or IV disease, and TP53 mutations were detected in 94.9% of patients. Complete cytoreduction (R0) rates significantly decreased with increasing ascites volume (77.8% in no ascites vs. 27.4% in above 1,000 mL group, P<0.001). Tumor markers HE4 and CA125 were significantly elevated in patients with higher ascites volumes (both P<0.001). During follow-up, 56.3% of patients experienced disease progression and 25.9% died. Median PFS was significantly shorter in patients with ascites above 1,000 mL compared to those without ascites (18.18 vs. 27.70 months, P=0.001), while OS showed no significant difference among groups (P=0.12).

Associations of ascites with PFS and OS

RCS analysis revealed a significant nonlinear association between ascites volume and PFS (Figure 1). In unadjusted analysis, ascites volume demonstrated a strong overall association with PFS (P<0.001) with significant nonlinearity (P for nonlinear =0.004). The HR increased steeply from 0 to approximately 1,000 mL, reaching a peak around 1,500–2,000 mL, then plateaued at higher volumes (Figure 1A). After adjusting for potential confounders, the overall association remained significant (P=0.009, P for nonlinear =0.02), although the effect magnitude was attenuated (Figure 1B). Similarly, death rates per 100 person-years showed a nonlinear pattern, with rates increasing sharply up to 1,000 mL and stabilizing thereafter in both unadjusted (P<0.001, P for nonlinear =0.02) and adjusted analyses (P<0.001, P for nonlinear =0.03) (Figure 1C,1D). The 1,000 mL threshold appeared to represent a critical inflection point for progression risk.

Figure 1 RCS analysis of the relationship between ascites volume and PFS in patients with high-grade serous ovarian cancer. RCS curves illustrate the nonlinear association between ascites volume and PFS. (A) Unadjusted HRs with 95% CIs. (B) The nonlinear association after multivariable adjustment for age, FIGO stage, tumor grade, residual disease status, and treatment variables. (C,D) Corresponding unadjusted and adjusted death rates per 100 person-years identify 1,000 mL as a critical threshold for progression risk. CI, confidence interval; FIGO, International Federation of Gynecology and Obstetrics; HRs, hazard ratios; PFS, progression-free survival; RCS, restricted cubic spline.

In contrast to PFS, ascites volume exhibited a predominantly linear dose-response relationship with OS (Figure 2). Unadjusted analysis showed a significant overall association (P<0.001) with no significant nonlinearity (P for nonlinear =0.21), indicating a consistent increase in mortality hazard with increasing ascites volume (Figure 2A). After multivariable adjustment, the linear relationship persisted (P<0.001, P for nonlinear =0.83), with HRs rising progressively across the entire ascites volume spectrum (Figure 2B). Death rates per 100 person-years demonstrated similar linear trends in both unadjusted (P=0.006, P for nonlinear =0.35) and adjusted models (P=0.02, P for nonlinear =0.86) (Figure 2C,2D). These findings suggest that ascites volume exerts a continuous detrimental effect on long-term survival, with each incremental increase in volume contributing to an elevated mortality risk.

Figure 2 Restricted cubic spline analysis of the relationship between ascites volume and OS. RCS curves show a predominantly linear dose-response association between ascites volume and OS. (A) Unadjusted HRs with 95% CIs. (B) The relationship after multivariable adjustment. (C,D) Corresponding unadjusted and adjusted death rates per 100 person-years suggest the relationship between incremental volume and survival outcomes. CI, confidence interval; HRs, hazard ratios; OS, overall survival; RCS, restricted cubic spline.

Effect of ascites groups on PFS and OS

Kaplan-Meier survival analysis demonstrated significant differences in PFS among the three ascites groups (Figure 3). In the unadjusted analysis, patients with ascites above 1,000 mL had substantially worse PFS compared to those without ascites (HR 2.16, 95% CI: 1.75–2.66, P<0.001), while the group below 1,000 mL showed moderately increased risk (HR 1.26, 95% CI: 1.03–1.53, P<0.05) (Figure 3A). The 3-year PFS rates were 57.9% (95% CI: 51.5–65.0%) for no ascites, 42.7% (95% CI: 39.1–46.7%) for below 1,000 mL, and 22.4% (95% CI: 18.2–27.6%) for above 1,000 mL groups. After adjusting for covariates, the prognostic impact of ascites was attenuated but remained significant for the above 1,000 mL group (HR 1.29, 95% CI: 1.02–1.64, P<0.05), while the below 1,000 mL group showed no significant difference from the reference (HR 1.12, 95% CI: 0.91–1.38, P>0.05) (Figure 3B). The adjusted 3-year PFS rates were 44.2%, 40.8%, and 36.5% for the three groups, respectively, indicating that high-volume ascites independently predicts disease progression.

Figure 3 Kaplan-Meier survival curves for PFS across ascites groups. Kaplan-Meier curves compare PFS among patients with no ascites (0 mL), low-volume ascites (<1,000 mL), and high-volume ascites (≥1,000 mL) in unadjusted analysis (A) and multivariable adjusted analysis (B). *, P<0.05; ***, P<0.001. CI, confidence interval; HR, hazard ratio; PFS, progression-free survival.

Similarly, OS differed significantly across ascites groups (Figure 4). Unadjusted analysis revealed markedly inferior OS in patients with ascites above 1,000 mL (HR 2.56, 95% CI: 1.84–3.54, P<0.001) and moderately reduced OS in the group below 1,000 mL (HR 1.37, 95% CI: 1.00–1.88, P<0.05) compared to patients without ascites (Figure 4A). The 3-year OS rates were 87.1% (95% CI: 82.5–91.8%), 80.1% (95% CI: 77.0–83.2%), and 74.8% (95% CI: 70.3–79.5%) for the no ascites, below 1,000 mL, and above 1,000 mL groups, respectively. Following multivariable adjustment, only the above 1,000 mL group retained independent prognostic significance (HR 1.54, 95% CI: 1.07–2.23, P<0.05), whereas the below 1,000 mL group showed no significant difference (HR 1.13, 95% CI: 0.81–1.56, P>0.05) (Figure 4B). The adjusted 3-year OS rates were 83.6%, 81.9%, and 76.5% for the respective groups.

Figure 4 Kaplan-Meier survival curves for OS across ascites groups. Kaplan-Meier curves illustrate OS differences among the three ascites groups in unadjusted analysis (A) and multivariable adjusted analysis (B). *, P<0.05; ***, P<0.001. CI, confidence interval; HR, hazard ratio; OS, overall survival.

Discussion

This prospective cohort study systematically evaluated the prognostic utility of ascites volume in HGSOC patients through continuous dose-response analysis. Our principal findings reveal that ascites volume exhibits distinct associations with survival outcomes: a nonlinear relationship with PFS characterized by a critical threshold at approximately 1,000 mL, and a predominantly linear dose-response relationship with OS. Importantly, high-volume ascites (≥1,000 mL) emerged as an independent predictor of both disease progression and mortality after adjusting for established prognostic factors. These findings provide quantitative evidence supporting ascites volume as a clinically meaningful prognostic biomarker that extends beyond its traditional role as a binary indicator of advanced disease.

The nonlinear association between ascites volume and PFS represents a key finding with important clinical implications. The steep increase in progression risk from 0 to 1,000 mL, followed by a plateau at higher volumes, suggests that the biological mechanisms underlying ascites formation exert their maximal detrimental effect within this range. This threshold phenomenon may reflect a critical tipping point in peritoneal tumor biology, beyond which additional fluid accumulation does not substantially worsen the already compromised peritoneal microenvironment (8,20). The 1,000 mL threshold aligns with clinical observations that massive ascites often indicates extensive peritoneal carcinomatosis with impaired lymphatic drainage and maximal vascular permeability (21-24). Our findings contrast with previous studies that treated ascites as a binary variable, which failed to capture this nuanced dose-response relationship and likely contributed to inconsistent prognostic associations reported in the literature (25). The identification of 1,000 mL as a clinically relevant cutpoint provides an objective, quantifiable criterion for risk stratification that can be readily implemented in clinical practice and incorporated into prognostic models (9).

In contrast to the nonlinear association observed between ascites volume and PFS, OS exhibited a predominantly linear and dose-dependent relationship with ascites burden. Specifically, patients with low-volume ascites experienced relatively modest survival impairment, whereas progressively increasing ascites volumes were associated with a stepwise escalation in mortality risk. This divergence suggests that limited ascites may primarily reflect localized peritoneal irritation or early tumor dissemination, while large-volume ascites represents a qualitatively distinct disease state characterized by extensive peritoneal involvement and systemic derangement (7). The linear relationship with OS implies that ascites volume functions as a continuous surrogate of cumulative tumor aggressiveness and host compromise, integrating peritoneal tumor load, mesothelial dysfunction, and inflammatory signaling. Importantly, the persistence of this association after comprehensive multivariable adjustment indicates that high-volume ascites conveys prognostic information beyond conventional staging metrics and surgical variables. Consistent with experimental and clinical evidence, massive ascites is increasingly recognized not merely as a passive consequence of advanced disease but as an active contributor to cancer progression, facilitating peritoneal metastasis, fostering hypoxic and immunosuppressive microenvironments, and promoting chemoresistance through soluble factors and tumor-stromal interactions (8,20,26).

Consistent with this biological gradient, our analyses demonstrated marked differences in clinicopathological characteristics between patients with minimal versus massive ascites, underscoring the prognostic heterogeneity conferred by ascites volume. High-volume ascites was strongly associated with substantially lower rates of complete cytoreduction (27.4% compared with 77.8% in patients without ascites), reflecting both extensive tumor dissemination and increased technical barriers to optimal debulking. Patients with larger ascites volumes also exhibited significantly elevated serum HE4 and CA125 levels, supporting the notion that ascites accumulation parallels heightened tumor metabolic activity and broader peritoneal disease extent. Notably, the higher BMI observed in patients with massive ascites likely represents fluid-related weight gain rather than preserved nutritional reserves, emphasizing the need to distinguish true body composition from ascites-driven volume overload (27). Collectively, these findings suggest that while low-volume ascites may exert limited prognostic impact, massive ascites integrates multiple adverse dimensions of tumor biology, surgical feasibility, and systemic physiology, thereby explaining its strong and independent association with inferior long-term survival (28).

The clinical implications of our findings are substantial for treatment planning and patient counseling. First, quantitative ascites volume measurement at initial surgery should be systematically documented and incorporated into prognostic assessment, moving beyond simple presence/absence documentation. The 1,000 mL threshold can guide risk stratification, with patients presenting with high-volume ascites potentially benefiting from more aggressive treatment approaches, including neoadjuvant chemotherapy to reduce disease burden, consideration of hyperthermic intraperitoneal chemotherapy, or enrollment in clinical trials evaluating novel therapeutic strategies. Second, ascites volume may serve as a quantifiable target for perioperative optimization, with aggressive ascites management potentially improving surgical outcomes and treatment tolerance (29). Third, the distinct dose-response patterns for PFS and OS suggest that ascites volume provides complementary prognostic information for both short-term and long-term outcomes, enabling more nuanced survival predictions and personalized treatment discussions (9). Finally, serial ascites monitoring during treatment could potentially serve as a dynamic biomarker for treatment response and recurrence, warranting further investigation (26,30).

Several limitations merit consideration. First, our study focused exclusively on HGSOC at a single institution, potentially limiting generalizability to other histological subtypes and broader populations. External validation in diverse cohorts is needed to confirm our findings. Second, although we made every effort to minimize heterogeneity in ascites collection and measurement, residual variability in ascites assessment remains inevitable. Third, the biological mechanisms linking ascites volume to survival outcomes were not directly investigated, and future studies incorporating molecular analyses of ascites fluid composition and tumor biology are needed to elucidate underlying pathways. Moreover, the lack of integrated analyses incorporating ascites burden alongside demographic, clinicopathological, treatment-related, and molecular variables limits our ability to fully characterize their combined prognostic impact (31-33). Finally, differences in treatment regimens—like the administration bevacizumab use or maintenance therapies—were not comprehensively incorporated into the analysis and may represent residual confounding affecting the observed associations. Despite these limitations, our prospective design, systematic ascites volume quantification, comprehensive covariate adjustment, and advanced statistical modeling using RCSs represent important methodological strengths that advance the field beyond prior binary classifications of ascites.

Conclusions

This study demonstrates that ascites volume represents a quantifiable, independent prognostic factor in ovarian cancer with distinct dose-response relationships for progression-free and OS. The identification of 1,000 mL as a clinically meaningful threshold for progression risk provides an objective criterion for risk stratification and treatment planning. Our findings support the routine documentation of ascites volume in clinical practice and its incorporation into prognostic models for personalized patient management. Future research should focus on validating these findings in external cohorts, investigating the molecular mechanisms underlying ascites-related poor prognosis, and evaluating whether targeted ascites management strategies can improve clinical outcomes.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-aw-481/rc

Data Sharing Statement: Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-aw-481/dss

Peer Review File: Available at https://gs.amegroups.com/article/view/10.21037/gs-2025-aw-481/prf

Funding: This work was supported by the National Natural Science Foundation of China (grant No. 82073323).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://gs.amegroups.com/article/view/10.21037/gs-2025-aw-481/coif). X.L. serves as an unpaid Associate Editor-in-Chief of Gland Surgery from July 2025 to June 2026. The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Institutional Review Board of Central South University (No. 2017068222). Informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Siegel RL, Giaquinto AN, Jemal A. Cancer statistics, 2024. CA Cancer J Clin 2024;74:12-49. [Crossref] [PubMed]
2. Bryant A, Hiu S, Kunonga PT, et al. Impact of residual disease as a prognostic factor for survival in women with advanced epithelial ovarian cancer after primary surgery. Cochrane Database Syst Rev 2022;9:CD015048. [Crossref] [PubMed]
3. Peres LC, Cushing-Haugen KL, Köbel M, et al. Invasive Epithelial Ovarian Cancer Survival by Histotype and Disease Stage. J Natl Cancer Inst 2019;111:60-8. [Crossref] [PubMed]
4. Qian L, Zhu J, Xue Z, et al. Proteomic landscape of epithelial ovarian cancer. Nat Commun 2024;15:6462. [Crossref] [PubMed]
5. Veneziani AC, Gonzalez-Ochoa E, Alqaisi H, et al. Heterogeneity and treatment landscape of ovarian carcinoma. Nat Rev Clin Oncol 2023;20:820-42. [Crossref] [PubMed]
6. Ford CE, Werner B, Hacker NF, et al. The untapped potential of ascites in ovarian cancer research and treatment. Br J Cancer 2020;123:9-16. [Crossref] [PubMed]
7. Szender JB, Emmons T, Belliotti S, et al. Impact of ascites volume on clinical outcomes in ovarian cancer: A cohort study. Gynecol Oncol 2017;146:491-7. [Crossref] [PubMed]
8. Dogra S, Adhikari L, Benbrook DM, et al. Harnessing ovarian cancer ascites for translational science: models, biomarkers, and therapeutics. Mol Cancer 2025;24:257. [Crossref] [PubMed]
9. Iyoshi S, Kimura M, Yoshihara M, et al. Detailed analysis of the histology-specific impact of ascites volume on the outcome of epithelial ovarian cancer: a multi-institutional retrospective cohort study. BMC Cancer 2024;24:1479. [Crossref] [PubMed]
10. Guigue PA, Brezinov Y, Yasmeen A, et al. Ascites and Serum Interleukin-10 Levels as a Prognostic Tool for Ovarian Cancer Outcomes. Cancers (Basel) 2024;16:2840. [Crossref] [PubMed]
11. Harter P, du Bois A, Hahmann M, et al. Surgery in recurrent ovarian cancer: the Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) DESKTOP OVAR trial. Ann Surg Oncol 2006;13:1702-10. [Crossref] [PubMed]
12. Chow LH, Ngai AMY, Tang CY, et al. Malignant ascites in ovarian cancer is compatible with long-term (10 year) survival with associations to clinicopathological features. J Ovarian Res 2025;18:98. [Crossref] [PubMed]
13. Iyoshi S, Sunohara M, Yoshihara M, et al. Long-term prognostic significance of ascites cytology in ovarian cancer cases in which R0 resection was achieved in the initial surgery: a multi-institutional retrospective cohort study. Jpn J Clin Oncol 2025;55:743-50. [Crossref] [PubMed]
14. Chen Z, Ouyang H, Sun B, et al. Utilizing explainable machine learning for progression-free survival prediction in high-grade serous ovarian cancer: insights from a prospective cohort study. Int J Surg 2025;111:3224-34. [Crossref] [PubMed]
15. Liu S, Lu C, He L, et al. Vitamin E Intake and Prevalence Rates of Thyroid Dysfunction and Autoimmune Thyroiditis: A Cross-Sectional Analysis of NHANES Data. Thyroid 2024;34:753-63. [Crossref] [PubMed]
16. Kondo T, Dewan P, Anand IS, et al. Clinical Characteristics and Outcomes in Patients With Heart Failure: Are There Thresholds and Inflection Points in Left Ventricular Ejection Fraction and Thresholds Justifying a Clinical Classification? Circulation 2023;148:732-49. [Crossref] [PubMed]
17. Harrell FE Jr. Regression Modeling Strategies With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Second Editon. Springer; 2015.
18. Nieto FJ, Coresh J. Adjusting survival curves for confounders: a review and a new method. Am J Epidemiol 1996;143:1059-68. [Crossref] [PubMed]
19. Tewari KS, Java JJ, Eskander RN, et al. Early initiation of chemotherapy following complete resection of advanced ovarian cancer associated with improved survival: NRG Oncology/Gynecologic Oncology Group study. Ann Oncol 2016;27:114-21. [Crossref] [PubMed]
20. Slattery K, Yao CH, Mylod E, et al. Uptake of lipids from ascites drives NK cell metabolic dysfunction in ovarian cancer. Sci Immunol 2025;10:eadr4795. [Crossref] [PubMed]
21. Asem M, Young A, Oyama C, et al. Ascites-induced compression alters the peritoneal microenvironment and promotes metastatic success in ovarian cancer. Sci Rep 2020;10:11913. [Crossref] [PubMed]
22. Kim S, Kim B, Song YS. Ascites modulates cancer cell behavior, contributing to tumor heterogeneity in ovarian cancer. Cancer Sci 2016;107:1173-8. [Crossref] [PubMed]
23. Almeida-Nunes DL, Mendes-Frias A, Silvestre R, et al. Immune Tumor Microenvironment in Ovarian Cancer Ascites. Int J Mol Sci 2022;23:10692. [Crossref] [PubMed]
24. Ghoneum A, Afify H, Salih Z, et al. Role of tumor microenvironment in ovarian cancer pathobiology. Oncotarget 2018;9:22832-49. [Crossref] [PubMed]
25. Penet MF, Krishnamachary B, Wildes FB, et al. Ascites Volumes and the Ovarian Cancer Microenvironment. Front Oncol 2018;8:595. [Crossref] [PubMed]
26. He T, Zhang J, Zeng L, et al. Composite score of PD-1 (+) CD8 (+) tumor-infiltrating lymphocytes and CD57 (+) CD8 (+) tumor ascites lymphocytes is associated with prognosis and tumor immune microenvironment of patients with advanced high-grade serous ovarian cancer. Chin J Cancer Res 2025;37:73-89. [Crossref] [PubMed]
27. Alves BC, Luchi-Cruz MM, Lopes AB, et al. Predicting dry weight in patients with cirrhotic ascites undergoing large-volume paracentesis. Clin Nutr ESPEN 2023;54:34-40. [Crossref] [PubMed]
28. Zheng X, Wang X, Cheng X, et al. Single-cell analyses implicate ascites in remodeling the ecosystems of primary and metastatic tumors in ovarian cancer. Nat Cancer 2023;4:1138-56. [Crossref] [PubMed]
29. Murphy B, Miyamoto T, Manning BS, et al. Myeloid activation clears ascites and reveals IL27-dependent regression of metastatic ovarian cancer. J Exp Med 2024;221:e20231967. [Crossref] [PubMed]
30. Kfoury M, Hazzaz RE, Sanson C, et al. Circulating Tumor DNA from Ascites as an alternative to tumor sampling for genomic profiling in ovarian cancer patients. Biomark Res 2023;11:93. [Crossref] [PubMed]
31. Scaglione GL, Pignata S, Pettinato A, et al. Homologous Recombination Deficiency (HRD) Scoring, by Means of Two Different Shallow Whole-Genome Sequencing Pipelines (sWGS), in Ovarian Cancer Patients: A Comparison with Myriad MyChoice Assay. Int J Mol Sci 2023;24:17095. [Crossref] [PubMed]
32. Caruso G, Weroha SJ, Cliby W. Ovarian Cancer: A Review. JAMA 2025;334:1278-91. [Crossref] [PubMed]
33. Fraser CC, Jia B, Hu G, et al. Ovarian Cancer Ascites Inhibits Transcriptional Activation of NK Cells Partly through CA125. J Immunol 2022;208:2227-38. [Crossref] [PubMed]