Prognosis and adjuvant chemotherapy for patients with malignant peritoneal cytology in early-stage non-endometrioid endometrial cancer

Objective To investigate the in�uence of malignant peritoneal cytology (MPC) on the prognosis of early-stage patients with endometrial clear cell carcinoma(CCC) and serous carcinoma(SC), and the value of chemotherapy in their treatment.


Introduction
Endometrial cancer (EC) is one of most common gynecological cancers, with 66,200 estimated new cases in the United States in 2023 (1).Traditionally, EC is classi ed as endometrioid and non-endometrioid cancer, and the latter account for about 20% of EC characterized by a higher rate of extra-uterine spread and worse prognosis (2).Of note, clear cell cancer and serous cancer are the major non-endometrioid endometrial cancer, together accounting for about 75% of non-endometrioid cancer (2)(3)(4).
Endometrial cancer is commonly diagnosed at an early stage.The standard surgical treatment approach for patients with apparent early-stage disease is hysterectomy, salpingo-oophorectomy, and additional lymphadenectomy, which allows primary tumor removal and identi cation of major prognostic factors (e.g., myometrial invasion, lymph vascular space invasion) for tailoring postoperative adjuvant therapy (5).The presence of tumor cells seen in the peritoneal cytologic specimen at hysterectomy, known as malignant peritoneal cytology, is another possible risk factor considered as an indication of adjuvant therapy.
In the National Comprehensive Cancer Network® (NCCN®) guidelines, noninvasive stage IA CCC and SC are strati ed according to peritoneal cytology results for tailored adjuvant treatment strategies(6).
The objective of this study was to examine the association between malignant peritoneal cytology and survival in women with early-stage endometrial clear cell carcinoma and serous carcinoma, and to explore whether malignant peritoneal cytology may be used as an indication to guide the use of chemotherapy in early-stage CCC and SC.

Data Source
The Surveillance, Epidemiology, and End Results (SEER) database, a population-based cancer registry established by the National Cancer Institute (NCI), was searched for the analysis (14).Patient identi cation, data accumulation and entry, and rigorous quality control for the database are managed by registered trained personnel (15).We obtained the information of patients diagnosed between 2010 and 2019 from the SEER * Stat version 8.4.0.1 (https://seer.cancer.gov/seerstat/).
Women with stage I-II (T1/N0/M0, T2/N0/M0) endometrial cancer with histologic con rmation who had available data on survival time and peritoneal cytology results from 2010 to 2019 were included.Exclusion criteria included non-CCC and non-SC, and no hysterectomy or unknown surgery type.

Outcome Measure
Overall survival (OS) and cancer-speci c survival (CSS) were assessed for outcome analysis.OS was considered as the time from diagnosis to death for any cause or to the last contact.CSS was de ned as the time from diagnosis to death due to endometrial carcinoma.Women who had no survival event at the last follow-up were censored.

Statistical analyses
The rst step of the analysis was to identify the independent characteristics associated with malignant peritoneal cytology.Differences in the baseline characteristics were assessed with chi-square test for univariable analysis.All the signi cant covariates with a p-value < 0.05 on univariable analysis were entered in the binary logistic regression model, and effect size was expressed with an adjusted odds ratio (OR) and corresponding 95% con dence interval (CI).Hosmer-Lemeshow test was used to assess the goodness-of-t in the nal model, and a P > 0.05 was interpreted as a good-t model.
The second step of the analysis was to evaluate the survival effect of malignant peritoneal cytology in stage I-II endometrial CCC and SC.The propensity score was calculated based on the covariates age at diagnosis, race, marital status, tumor differentiation, stage, tumor size, hysterectomy type, para-aortic lymphadenectomy, pelvic lymphadenectomy, chemotherapy, radiotherapy to reduce the imbalance between the malignant and negative peritoneal cytology groups.Malignant peritoneal cytology cases were matched to negative peritoneal cytology group in a 1:4 ratio and a caliper width of 0.05 standard deviation (SD) was speci ed.Survival analysis was assessed with the Kaplan-Meier method, and the results were compared with the log-rank test.The Kaplan-Meier method was used to plot survival curves, and a Cox proportional hazard regression model was tted to estimate hazard ratio (HR) with 95% con dence interval (CI) for malignant peritoneal cytology.
Lastly, cox regression analyses were employed to identify whether adjuvant chemotherapy improved the outcome of endometrial CCC/SC patients with malignant peritoneal cytology by reporting the hazard ratios (HR) and 95% CI.
In addition, we conducted multiple subgroup analyses to enhance the validity of this study.First, the effect of peritoneal cytology results on survival outcome was examined, strati ed by histology (clear cell versus serous), and stage (stage IA, stage IB, stage II).Second, the effect of chemotherapy on the prognosis of women with stage IA was examined based on peritoneal cytology results (malignant versus negative).Propensity score inverse probability of treatment weighting (IPTW) was used to balance the measured covariates in each subcohort(16).The Kaplan-Meier method was used to plot survival curves, and the survival analysis results were compared with the log-rank test.

Factors associated with malignant peritoneal cytology
In the univariable and multivariable analysis (Table 1), race, marital status, and postoperative adjuvant chemotherapy were the factors associated with peritoneal cytology results (all p < 0.05).Blacks were more likely to have negative peritoneal cytology than whites (OR 0.565; 95%CI 0.406-0.787).Women in the malignant peritoneal cytology were more likely to receive postoperative adjuvant chemotherapy (OR 2.033; 95%CI 1.589-2.602).

Survival effect of malignant peritoneal cytology
All of the 12 measured covariates were well-balanced after propensity score matching (PSM) (Table S1).In the PSM model, 365 women with malignant peritoneal cytology were compared with 1406 women with negative peritoneal cytology for survival outcome analysis.The cox proportional hazard regression model showed that malignant peritoneal cytology was an independent risk prognostic factor for overall survival (OS) (Univariate: HR 2.028; 95%CI  S2).The 5-year OS rates were 56.5% for women with malignant peritoneal cytology and 74.4% for those with negative peritoneal cytology (p < 0.0001) (Fig. 2A), and the 5-year CSS rates were 60.8% versus 80.0% (p < 0.0001) (Fig. 2B).

Signi cance of adjuvant chemotherapy in patients based on peritoneal cytology results
The association between adjuvant chemotherapy and survival was examined among women with malignant peritoneal cytology (Table 2).According the multivariate cox regression analysis, after controlling for confounding factors, adjuvant chemotherapy as an independent prognostic factor could improve the survival of patients with malignant peritoneal cytology (OS: HR 0.557, 95%CI 0.370-0.839,p = 0.005; CSS: HR 0.553, 95%CI 0.354-0.866,p = 0.010).Additionally, in stage IA patients with malignant peritoneal cytology, adjuvant chemotherapy could improve the survival outcomes (OS: P = 0.025; CSS: P = 0.038) (Fig. 4A;B).Of note, in stage IA patients with negative peritoneal cytology, chemotherapy was a good prognostic factor for OS (P = 0.001) but not for CSS (P = 0.300) (Fig. 4C;D).

Discussion
Our study found that malignant peritoneal cytology was signi cantly correlated with shorter OS and CSS in early-stage endometrial clear cell carcinoma and serous carcinoma.Moreover, the survival effect of adjuvant chemotherapy may differ depending on peritoneal cytology status for early-stage endometrial clear cell carcinoma and serous carcinoma.
In our study, malignant peritoneal cytology was seen in 10.2% of stage I-II ECCC and SC, compared to 6.0-10.9%which has been reported in prior studies (7,(17)(18)(19)(20)(21).This prevalence is far higher than in early-stage endometrioid endometrial cancer, where malignant peritoneal cytology is reported to be around 5%(8, 10, 11).One possible mechanism of malignant peritoneal cytology is the retrograde spread of tumor cells via the fallopian tubes.A study of 4489 EC patients, including 686 SC patients and 172 CCC patients, revealed that tubal ligation is associated with decreased risk of peritoneal tumor metastasis (GOG-210) (22).The risk reduction is much more signi cant in SC group than endometroid histology (SC: OR 0.28, 95% CI 0.11-0.68;Low-grade endometroid: OR 1.27, 95% CI 0.41-3.89).In the CCC group, although no statistically signi cant conclusions can be drawn due to the small number of CCC patients, CCC patients with tubal ligation tend to have a lower risk of peritoneal tumor metastasis (OR 0.27, 95% CI 0.00-2.24).Non-endometrioid EC is more likely to have trans-tubal spread, which may explain the higher incidence of malignant peritoneal cytology in SC and CCC than endometrioid tumors.Based on a SEER study (7), the increased risk of malignant peritoneal cytology is associated with diagnosis at an older age (> 78 years old), serous histology, and larger tumor size (4.1-6.0cm vs. <2cm) in stage I non-endometrioid EC.Our study, however, failed to nd an association between the above factors and malignant peritoneal cytology, which is consistent with several studies targeted SC/CCC histology (21,23).
Our study demonstrated that malignant peritoneal cytology was a poor prognostic factor for stage I -II ECCC and SC histology, which was in line with the results of previous studies, but our study speci cally targeted SC and CCC groups and the sample size was larger (2,7,12,21,23).A big-data SEER analysis reported by Matsuo K, et al. revealed that malignant peritoneal cytology was associated with a nearly two-fold increase in all-cause mortality risk compared to negative peritoneal cytology (5-year rates, 63.4% versus 80.2%, HR 2.18, 95% CI 1.78-2.66) in stage I non-endometrioid EC (7).In an analysis of a large cohort of stage I SC/CCC patients with long-term follow-up, Jeans EB, et al. demonstrated any recurrence was worse for tumors of at least 3.5 cm (HR 3.8, 95%CI 1.3-11.7,p = 0.02) and patients with positive/suspicious cytology (HR 4.4, 95%CI 1.5 to 12.4, p ≤ 0.01), and multivariate modeling showed worse any recurrence with positive/suspicious cytology after adjusting for the effect of age (2).A recent study of 101 patients with stage I SC and CCC showed that patients with malignant cytology had a higher likelihood of peritoneal recurrence and a shorter time to relapse (13 vs. 38 months, p = 0.022), as compared to patients with negative cytology (21).A retrospective study of 148 patients diagnosed with uterine serous carcinoma reported malignant peritoneal cytology had a statistically signi cant effect on OS (HR: 2.09, 95% CI 1.19-3.68)and on CSS (HR: 2.02, 95% CI 1.06-3.82),but the study included stage III SC patients (23).A study enrolling patients with stage I-II uterine serous carcinoma without adjuvant therapy demonstrated malignant peritoneal cytology was signi cant prognosticators (DFS: HR 4.24; 95% CI 1.56-11.50;p = 0.005; OS: HR 4.96; 95% CI 1.34-18.30;p = 0.017) (12).
Our study explored the prognostic value of malignant peritoneal cytology in not only early-stage non-endometrioid EC patients but also in various subgroups strati ed based on stage and histological type.In the subgroup analyses, malignant peritoneal cytology was associated with decreased OS and CSS in serous, clear cell histology group, and stage IA disease, but not for stage IB or stage II disease, which was consistent with the big-data results reported by Matsuo K, et al (7).The previous retrospective study on 68 cases with USC without adjuvant therapy also showed that malignant peritoneal cytology had signi cant worse prognostic impact on OS (p = 0.0012) and DFS (p = 0.0001) in stage IA disease (n = 42), but not in stage IB (n = 7) (OS: p = 0.722; DFS: p = 0.911) and II (n = 19) (OS: p = 0.48; DFS : p = 0.85) (12).Yang J, et al, however, reported that the detrimental effects of malignant cytology on survival were more prominent in patients with stage IB disease.Notably, the limited number of patients with malignant peritoneal cytology may result in a few underpowered analyses in the study reported by Yang J, et al (21).
Adjuvant therapy for early-stage endometrial clear cell carcinoma and serous carcinoma is highly individualized, and the role of chemotherapy remains controversial.Previous studies have shown that chemotherapy could improve the prognosis of patients with stage I SC and CCC (19,21).These study, however, did not consider peritoneal cytology as a strati cation factor to explore the role of chemotherapy.In our study, patients with malignant peritoneal cytology tended to receive adjuvant chemotherapy, and subgroup analysis found that malignant peritoneal cytology was an adverse prognostic factor in SC/CCC patients with stage IA, so we further explored whether adjuvant chemotherapy could improve the prognosis of patients with malignant peritoneal cytology in stage IA.We found that in stage IA patients with malignant peritoneal cytology, those who received chemotherapy had longer OS (p = 0.025) and CSS (p = 0.038), and for those with negative peritoneal cytology, chemotherapy improved OS (p = 0.001) but not CSS (p = 0.300).Therefore, we believed that SC and CCC patients with stage IA and malignant peritoneal cytology could be considered for chemotherapy.Our nding was in accordance with the NCCN recommendation.According to the 2023 NCCN guideline, noninvasive stage IA CCC and SC are strati ed according to peritoneal cytology results for tailored adjuvant therapy strategies, and systemic therapy and vaginal brachytherapy is recommended if positive washings(6).Additionally, for patients with invasive stage IA, systemic therapy could be considered regardless of peritoneal cytology(6).
A recent retrospective study revealed that chemotherapy did not improve RFS or OS in type II ECs (SC: n = 22, CCC: n = 12) with stage IA and malignant peritoneal cytology.Although the results were not statistically signi cant, it could be found that the 5-year relapse-free survival rate and 5-year overall survival rate in the chemotherapy group were higher than those in the non-chemotherapy group (RFS:85.2% vs. 66.7%, p = 0.11; OS: 91.9% vs. 77.8%,p = 0.15), so it may be better to treat them with adjuvant chemotherapy (24).An observational retrospective study of 16,851 stage I-II ECs including 973 serous carcinomas and 190 clear cell carcinomas revealed that adjuvant chemotherapy was associated with increased overall survival in early-stage EC patients with malignant peritoneal cytology (HR 0.62, 95%CI 0.40-0.95,P = 0.03).However, studies have not fully discussed which histologic subtype of early-stage patients with malignant peritoneal cytology may be able to experience increased survival with adjuvant chemotherapy (9).A study, involving stage I SC and CCC patients who had received adjuvant vaginal cuff brachytherapy and/or chemotherapy, reported that positive/suspicious cytology were at increased risk for relapse and worse survival, and should be considered for additional upfront adjuvant treatments, such as platinum-based chemotherapy(2).However, the role of chemotherapy has not been individually investigated in stage IA patients in this analysis(2).For early-stage non-endometrioid endometrial cancer with negative peritoneal cytology, Matsuo K, et al. stated that women who received chemotherapy-based postoperative therapy had higher 5-year OS rates compared to those received whole pelvic irradiation (79.6% for chemotherapy ± brachytherapy, 77.4% for chemotherapy with whole pelvic irradiation, and 56.9% for whole pelvic irradiation, P < 0.001) (7).However, a study reported by Tate K, et al. revealed that patients with stage IA or IB USC showing negative peritoneal cytology might have an extremely favorable prognosis (5-year DFS rate:88.8%;5-year OS rate:93.6%) and did not need any adjuvant therapies (12).More clinical studies are needed to investigate the signi cance of adjuvant chemotherapy in patients with stage IA non-endometrioid endometrial cancer with negative peritoneal cytology.
Strengths of the current study include: this was a population-based analysis with the large sample size for endometrial clear cell carcinoma and serous carcinoma; the long follow-up period; several subgroup analyses with stage and histologic type as strati cation factors enhanced the interpretation of the study results; the analytic approach with propensity score matching and propensity score inverse probability of treatment weighting enriched the statistical rigor.
Additionally, our study does have several limitations.Predominantly, it was a retrospective study with inherent unmeasured bias.For example, details of surgical-pathological factors such as lympho-vascular space invasion, and omental biopsy were not available in the SEER Program, but may be associated with the incidence of malignant peritoneal cytology as well as the oncologic outcomes.Second, the accuracy of malignant cytology results was unknown due to the lack of central pathology review in the study.Third, as this study only examined the U.S. population, whether our ndings are generalizable to other study populations needs to be further examined.Fourth, per the SEER Program de nition, cases with no and unknown status are grouped together in chemotherapy, and details of chemotherapy (regimen and cycle) were not available.

Conclusion
conclusion, the current study clearly showed that malignant peritoneal cytology was identi ed as an adverse prognostic factor in analyses of all early-stage endometrial clear cell carcinoma and serous carcinoma as well as in CCC histology type, SC histology type, and stage IA subgroup analyses.Based on the results of the present study, stage IA SC/CCC patients with malignant peritoneal cytology could further bene t from chemotherapy.Of note, peritoneal cytology results should not be taken in isolation to guide adjuvant therapy, and it should also be considered in combination with other risk factors for survival outcomes.Further prospective clinical trials are necessary to evaluate the e cacy of adjuvant therapy in stage IA SC/CCC groups.

Declarations
Ethics approval and consent to participate: Not applicable Consent for publication: applicable

Supplementary Files
This is a list of supplementary les associated with this preprint.Click to download.

Figure 1 Flow
Figure 1 Flow chart of patient selection from the Surveillance, Epidemiology, and End Results (SEER) database.

Figure 2
Figure 2Overall survival (A) and cancer-speci c survival (B) based on the peritoneal cytology results (negative or malignant) for the entire cohort.

Figure 3 Based
Figure 3 Based on the peritoneal cytology results (negative or malignant), overall survival (A) and cancer-speci c survival (B) for CCC histology; overall survival (C) and cancer-speci c survival (D) for SC histology; overall survival (E) and cancer-speci c survival (F) for stage IA patients.

Figure 4 Based
Figure 4Based on the peritoneal cytology results (negative or malignant), overall survival (A) and cancer-speci c survival (B) for stage IA patients with malignant peritoneal cytology; overall survival (C) and cancer-speci c survival (D) for stage IA patients with negative peritoneal cytology. FigureS1.tifTableS1andS2.docx
Abbreviations: NOS, not otherwise signi cant.Bold values indicate signi cance at p < 0.05.

Table 2
Univariate and Multivariate cox regression analysis of patients with malignant peritoneal cytology.(n= 368) Abbreviations: NOS, not otherwise signi cant.OS,overall survival.CSS,speci c cancer survival.Bold values indicate signi cance at p < 0.05.