1. INTRODUCTION

The global prevalence of endometrial cancer, which has become one of the leading gynecological malignancies, has steadily increased over the last 2 decades, with the age-standardized incidence rate increasing significantly, by 0.69% per year. This trend has been observed in most countries and territories worldwide, regardless of sociodemographic status,1 with >75% of endometrial cancers diagnosed at an early stage. The primary treatment for endometrial cancer is a combined staging surgery, which is comprised of a total hysterectomy, bilateral salpingo-oophorectomy, and retroperitoneal lymphadenectomy. Adjuvant treatment decisions are based on several clinical and pathologic prognostic parameters, such as tumor grade, histologic type, age, depth of myometrial invasion, and lymphovascular space involvement (LVSI).2–4 Although the majority of women diagnosed with early stage endometrial cancer have a favorable prognosis, approximately 10% to 15% will experience a recurrence.5 In patients who experience a cancer recurrence, particularly as distant metastasis or peritoneal carcinomatosis, the outcome is typically abysmal, with a median survival of under a year.6

Carbohydrate antigen 125 (CA-125) is a tandem repeating epitope of mucin-16 (MUC16), which is overexpressed in multiple malignancies and is associated with tumorigenesis, cell proliferation, and metastasis.7–9 Abnormal serum CA-125 levels are found in a variety of malignancies10; therefore, CA-125 is widely used as a biomarker for numerous clinical manifestations, such as the following: (1) as a diagnostic factor to distinguish between benign and malignant tumors; (2) to predict disease severity before treatment; (3) as a prognostic factor of disease outcome; and (4) for monitoring patient response to treatment.

Unlike ovarian cancer, the potential applications of serum CA-125 in endometrial cancer have not been well studied. Most of the available studies included all stages of endometrial cancer, with inconclusive results. Preoperative serum CA-125 levels in patients with endometrial cancer are associated with disease severity, with higher CA-125 levels frequently found in patients with advanced stage cancer, higher grading, cervical involvement, lymph node metastasis, or extrauterine spreading.11,12 Based on the results of a 2-year prospective nonrandomized study, a preoperative CA-125 level >20 U/mL and/or a grade 3 tumor correctly identified 75% to 87% of patients requiring lymphadenectomy.13 Preoperative CA-125 levels, therefore, have been suggested as an adjunct, to be incorporated into preoperative risk stratification models for treatment planning in women diagnosed with endometrial cancer.14,15

The roles of CA-125 levels in predicting cancer recurrence have not yet been elucidated; therefore, the present study explored the roles of serum CA-125 levels on predicting recurrence in patients with stage I endometrial cancer, focusing particularly on postoperative and other clinicopathological factors.

2. METHODS 2.1. Patients

A total of 733 patients with endometrial cancer limited to the uterus underwent staging surgery at our hospital between January 2010 and March 2019, for whom we reviewed the electronic medical records for clinicopathological data. Patients were staged using the 2009 International Federation of Gynecology and Obstetrics (FIGO) staging system, and disease recurrence was confirmed based on radiological and/or histological findings. Histological grading of the surgical specimens was based on the 2004 World Health Organization pathological classification, which is dependent on the percentage of solid components and nuclear atypia of a given specimen. The preoperative CA-125 level was defined as that obtained closest to the operation, while the postoperative level was defined as that obtained within 6 to 12 months after the surgery. Serum CA-125 levels were determined via enzyme immunoassay, using commercially available kits according to the manufacturer’s standard instructions. Body mass index (BMI) was obtained by dividing each patient’s weight in kilograms by the square of their height in meters. The Charlson Comorbidity Index (CCI) was calculated for each patient, along with the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) and International Classification of Diseases and Related Health Problems, 10th Revision, Canada (ICD-10-CA) diagnosis codes from our hospital’s abstract data and the associated weights for each category.

Adjuvant treatment strategies for endometrial cancer include external beam pelvic irradiation, vaginal vault brachytherapy, chemotherapy, and chemoradiotherapy. Indications for postoperative treatment were based on certain clinical and pathological risk factors, based on the recommendations of the Society of Gynecologic Oncology (SGO) or European Society of Gynaecological Oncology (ESGO)/European SocieTy for Radiotherapy and Oncology (ESTRO)/European Society of Pathology (ESP) guidelines.16,17

All patients received periodic follow-up at our institution, as the postoperative monitoring of endometrial cancer includes physical examinations and systemic reviews performed every 3 months for the first 2 years, every 6 months for the next 3 years, and annually thereafter. Additionally, serum CA-125 levels and vaginal cytology were checked every 6 months for 3 years, and annually thereafter. In cases of suspected recurrence, patients underwent imaging via computed tomography (CT), magnetic resonance imaging (MRI), or positron emission tomography (PET).

2.2. Inclusion criteria

The inclusion criteria for the present study were as follows: (1) patients with pathologically diagnosed stage I endometrial cancer with the presence of endometrioid, serous, clear cell, and/or other cell types, such as squamous, mucinous, and undifferentiated/dedifferentiated cells and 2) patients underwent complete staging surgery, including total hysterectomy, bilateral salpingo-oophorectomy, and retroperitoneal lymphadenectomy at our hospital. Patients with uterine carcinosarcoma were excluded from the present study, due to its poor prognosis and earlier recurrence rate compared with other cell types,18 as were those with synchronous endometrial and ovarian carcinomas, or other primary malignancies. Based on the inclusion criteria, 215 patients were excluded from the present study due to incomplete staging (n = 133), recurrence or loss of follow-up within 6 months (n = 33), carcinosarcoma histology (n = 27), or a second or synchronous primary malignancy developing within 5 years (n = 22). After applying the exclusion criteria, a total of 518 patients were included in the present study.

The protocol for the present study was approved by our hospital’s institutional review board (no. 2022-07-007C).

2.3. Statistical analysis

The prognostic variables analyzed in the present study were as follows: patient age, BMI, CCI, myometrial invasion depth, cell type, histological grading, tumor size, LVSI, preoperative hysteroscopy, peritoneal cytology, adjuvant treatments, surgical method, lymphadenectomy field, and preoperative and postoperative CA-125 levels. Baseline characteristics were compared using the t, chi-square, or Fisher exact test, as appropriate, while univariate and multivariate Cox proportional hazards analyses were used to assess the prognostic significance of various characteristics. Variables with a p < 0.2 on the chi-square or univariate Cox proportional hazard analysis were selected for further multivariate Cox regression analysis.

Recurrence-free survival (RFS) was defined as the length of time between the primary surgery and the disease recurrence or the final oncological visit. Kaplan-Meier analysis was used to determine RFS, and patient survival was compared among subgroups using the log-rank test.

All statistical analyses were performed using SPSS statistical software v26.0 (IBM, Chicago, IL, USA), with a two-tailed p < 0.05 was considered to be statistically significant.

3. RESULTS 3.1. Patient characteristics

A total of 518 patients who met the inclusion criteria were included in the present study, with a mean patient age of 57 years (22–87 years) and a mean BMI of 25.5 kg/m2 (14.4–47.5 kg/m2). The median preoperative CA-125 level was 18.5 U/mL (5–4239 U/mL), while the median postoperative CA-125 level was 9.3 U/mL (2–143 U/mL). Based on the 2009 FIGO stage classification, 423 (81.7%) patients were diagnosed with stage IA disease, while 95 (18.3%) were diagnosed with stage IB disease. No adjuvant treatment was necessary in 301 (60%) patients, and of the remaining 217 patients, 147 (28.4%) received radiotherapy, 25 (4.8%) received chemotherapy, and 45 (6.8%) received combined chemoradiotherapy. The median follow-up duration was 49 months (7–136 months), during which time 41 (7.9%) patients experienced recurrence. The detailed numbers and frequencies (percentages) of patients with each clinicopathological characteristic are shown in Table 1.

Table 1 - Patient characteristics Characteristics Patient number (n) Percentage (%) Age group  20-40 40 7.7  41-50 100 19.3  51-60 208 40.4  60-70 129 24.9  >70 41 7.9 Body mass index  <27 319 61.6  ≥27 199 38.4 Charlson Comorbidity Index  <5 435 84.0  ≥5 83 16.0 Histologic group  Endometrioid 465 89.8  Serous 9 1.7  Clear cell 4 0.8  Others 40 7.7 Endometrioid cell grading  1 81 15.6  2 358 69.1  3 68 13.1 Myometrial invasion, %  <50 423 81.7  ≥50 95 18.3 Lymphovascular invasion  Negative 453 87.5  Positive 65 12.5 Peritoneal cytology  Negative 479 96.0  Positive 17 3.4 Tumor size, cm  <2 231 44.6  ≥2 287 55.4 Preoperative hysteroscopy  No 405 78.2  Yes 11 21.8 Surgical approach method  Laparotomy 359 69.3  Microinvasive 159 30.7 Lymphadenectomy field  Pelvic 243 46.9  Pelvic + para-aortic 275 53.1 Adjuvant therapy  No adjuvant therapy 311 60.0  IVRT only 113 21.8  EBRT only 29 5.6  IVRT + EBRT 5 1.0  Chemotherapy only 25 4.8  Chemoradiotherapy 45 6.8 Preoperative CA-125, U/mL  <35 393 75.9  ≥35 107 20.7 Postoperative CA-125, U/mL  <35 481 92.9  ≥35 10 1.9  <13.75 393 75.8  ≥13.75 98 18.9 Recurrence  No 477 92.1  Yes 41 7.9

CA-125 = carbohydrate antigen 125; EBRT = external beam radiotherapy; IVRT = intravaginal radiotherapy.

3.2. Relationship between clinicopathological factors and recurrence

In the present study, we explored whether the following factors were associated with tumor recurrence: age (<65 vs ≥65 years), BMI (<27 vs ≥27), CCI (<5 vs ≥5), endometrioid cell grade (1 + 2 vs 3), cell type (endometrioid vs nonendometrioid), myometrial invasion (<50 vs ≥50%), tumor size (<2 vs ≥2 cm), LVSI (no vs yes), peritoneal cytology (negative vs positive), preoperative hysteroscopy (no vs yes), surgical method (laparotomy vs laparoscopy), lymphadenectomy field (pelvis only vs pelvis + para-aortic), adjuvant treatment (none vs radiotherapy vs chemotherapy vs chemoradiotherapy), preoperative CA-125 level (<35 vs ≥35 U/mL), and postoperative CA-125 levels at different cutoff values (<35 vs ≥35 U/mL and <13.75 vs ≥13.75 U/mL). None of the aforementioned factors was significantly associated with recurrence of stage I endometrial cancer based on the chi-square analysis, except for postoperative CA-125 levels (Table 2).

Table 2 - The relationship between clinicopathological factors and recurrence Clinicopathological factors Recurrence (n = 41) No recurrence (n = 477) p a Age  <65 30 389 0.190  ≥65 11 88 BMI  BMI < 27 29 290 0.209  BMI ≥ 27 12 187 Charlson Comorbidity Index  Score < 5 30 405  Score ≥ 5 11 72 0.050 Endometrioid cell grade  Low (1 + 2) 32 411  High (3) 9 59 0.090 Cell type  Endometrioid 38 427 0.520  Nonendometrioid 3 50 Myometrial invasion, %  <50 30 393 0.143  ≥50 11 84 Tumor size, cm  <2 18 213 0.920  ≥2 23 264 Lymphovascular space involvement  No 34 419 0.360  Yes 7 58 Peritoneal cytology  Negative 36 442 0.127  Positive 3 17 Preoperative hysteroscopy  No 33 372 0.710  Yes 8 105 Surgical method  Laparotomic 32 372 0.206  Laparoscopic 9 150 Lymphadenectomy field  Pelvis only 21 222 0.339  Pelvis + para-aortic 20 255 Adjuvant treatments  No 20 291 0.543  Radiotherapy only 14 128  Chemotherapy only 2 23  Chemoradiotherapy 4 31 Preoperative CA-125, U/mL  <35 30 369 0.560  ≥35 10 97 Postoperative CA-125, U/mL  <35 33 449 0.002b  ≥35 5 4  <13.75 22 371 0.002b  ≥13.75 15 83

BMI = body mass index; CA-125 = carbohydrate antigen 125.

ap value was calculated by chi-square method.

bp < 0.05 means statistically significant.

3.3. Univariate and multivariate analysis of clinicopathological factors as predictors of recurrence in stage I endometrial cancer

Based on prior results, the following factors were evaluated via Cox hazard regression analysis to determine which factors are predictors of tumor recurrence: age, endometrioid cell grade, CCI, myometrial invasion depth, peritoneal cytology, preoperative CA-125 level, and postoperative CA-125 level. As shown in Table 3, the postoperative CA-125 level was the only statistically significant predictor of recurrence in stage I endometrial cancer, based on the univariate Cox regression analysis. Furthermore, we analyzed factors with a p < 0.2 via multivariate Cox regression analysis, including endometrial cell grade, CCI, myometrial invasion depth, peritoneal cytology, and postoperative CA-125 levels, and found that postoperative CA-125 levels were independently associated with cancer recurrence.

Table 3 - Univariable and multivariable analysis of clinicopathological factors as predictors of recurrence in stage 1 endometrial cancer Clinicopathological factors Univariate analysis Multivariate analysis HR (95% CI) p HR (95% CI) p Age (≥65 vs <65) 0.473 (0.139-1.604) 0.229 Endometrioid cell grade (grade 3 vs grade 1 + 2) 1.959 (0.891-4.309) 0.094 1.673 (0.703-3.98) 0.245 Charlson Comorbidity Index (≥5 vs <5) 2.026 (0.98-4.30) 0.054 1.939 (0.859-4.378) 0.111 Myometrial invasion (≥50 vs <50) 1.993 (0.993-4.069) 0.058 1.60 (0.715-3.60) 0.252 Peritoneal cytology (positive vs negative) 2.67 (0.724-9.60) 0.141 2.5.5 (0.606-10.43) 0.204 Preoperative CA-125, U/mL (≥35 vs <35) 1.247 (0.589-2.641) 0.563 Postoperative CA-125, U/mL (≥35 vs <35) 10.88 (2.79-42.4) 0.001a 2.339 (1.094-5.00) 0.038a Postoperative CA-125, U/mL (≥13.75 vs <13.75) 3.04 (1.51-6.12) 0.002a

CA-125 = carbohydrate antigen 125; HR = hazard ratio.

ap < 0.05 means statistically significant.

3.4. Relationship of postoperative CA-125 to RFS

In the present study, the median RFS of patients with postoperative CA-125 levels <35 and ≥35 U/mL was 48.0 months (1–136 months) and 13.0 months (8–66 months), respectively. Based on the log-rank test, a higher postoperative CA-125 level (cutoff value, 35 U/mL) was significantly associated with worse RFS (Fig. 1A); however, using 35 U/mL as the cutoff value, the sensitivity and specificity were 13.2% and 99.1%, respectively. We therefore used the receiver operating characteristic (ROC) curve to determine the optimal postoperative CA-125 cutoff value for predicting the recurrence of stage I endometrial cancer. Using the Youden index, if the cutoff value was set at 13.75 U/mL, the sensitivity and specificity were 50.5% and 81.7%, respectively. The number of patients with postoperative CA-125 levels >13.75 U/mL were as follows: 14 of 35 (40%) in the recurrent subgroup vs 83 of 454 (18.2%) in the nonrecurrent subgroup. The recurrent hazard ratio for stage I endometrial cancer patients with a CA-125 level ≥13.75 U/mL to those <13.75 U/mL was 3.04 (95% CI, 1.51-6.12; p = 0.002), based on the univariate Cox regression. Based on the multivariate Cox regression analysis results, a recurrence hazard ratio of 2.339 showed statistical significance (95% CI, 1.09-5.00; p = 0.038).

Fig. 1:

Recurrence-free survival stratified by different postoperative CA-125 cutoff values in surgical stage I endometrial cancer. A, Cutoff value of 35 U/mL. B, Cutoff value of 13.75 U/mL. CA-125 = carbohydrate antigen 125.

A Kaplan-Meier survival curve was constructed to demonstrate the time to recurrence. By adopting a cutoff value of 13.75 U/mL, the time to recurrence in patients with stage I endometrial cancer patients with a postoperative CA-125 level ≥13.75 U/mL was significantly shorter than those with CA-125 levels <13.75 U/mL (median [range], 35.5 months [4–109 months] vs 50.5 months [4–136 months]), with a p = 0.001, according to the log-rank test (Fig. 1B). By using 13.75 U/mL as the cutoff value, the 5-year RFS in patients with elevated CA-125 levels was 84.7%, compared with that in patients with lower CA-125 levels, which was 95.4%.

Furthermore, we used the Pearson method to evaluate the correlation between the preoperative and postoperative CA-125 levels. The results of this analysis indicated that these two values were not linearly related. The Pearson correlation coefficient was 0.062, with a two-tailed p = 0.18.

4. DISCUSSION

From the Surveillance, Epidemiology, and End Results (SEER) database of people diagnosed with endometrial cancer between 2011 and 2017, the 5-year relative survival rates of localized, regional, and distant endometrial cancers were 96, 71, and 20%, respectively. According to the European Society for Medical Oncology-European Society of Gynaecological Oncology-European Society for Radiotherapy & Oncology (ESMO-ESGO-ESTRO) classification, the recurrence rates in low-, intermediate-, and high-risk endometrial cancers were 6% to 9%, 9% to 16%, and 21% to 35%, respectively.19,20 The results of the present study indicated a similar recurrence rate in patients with stage I endometrial cancer, at 7.9%.

In the present study, the common prognostic factors for early-stage endometrial cancers, such as tumor grade, histologic type, LVSI, and myometrial invasion depth, were not found to predict cancer recurrence. There are two possible reasons for this—first, the patients in the present study received adequate adjuvant treatment based on known risk factors; and second, the power of the sample size was not sufficiently large enough to achieve statistical significance.

Whether preoperative CA-125 levels can serve as a useful prognostic factor for patients with stage I endometrial cancer remains uncertain, although the results of the present study showed that preoperative CA-125 levels were not useful predictive markers for recurrence. This may be because the patients in the present study did not have extrauterine metastases, and received adjuvant treatment based on their clinicopathological risk factors. This result is similar to that of a retrospective Korean multicenter study, in which they reported that preoperative serum CA-125 levels may not be useful for predicting most prognostic factors, and may not contribute to the preoperative selection of patients with intermediate- or high-risk disease who need adjuvant radiotherapy in early-stage endometrial cancer.21 Conversely, Jiang et al,14 in an analysis of 995 patients with endometrial cancer, found that elevated CA-125 was an independent prognostic factor; however, their study included stage I to IV patients, and almost 20% of the patients with endometrial cancer were excluded due to a lack of preoperative serum CA-125 levels. Recently, a prospective multicenter cohort PIpelle Prospective ENDOmetrial carcinoma (PIPENDO) study, which included women with endometrial cancer, demonstrated that elevated preoperative CA-125 levels were correlated with poor prognosis and independently associated with survival, particularly in low-grade subgroups. Unlike the present study, most women with low-grade endometrial cancer in the aforementioned study did not undergo lymphadenectomy or lymph node sampling, which could have resulted in an inability to detect occult nodal metastasis and were, therefore, understaged.

In the present study, postoperative CA-125 levels were obtained between 6 and 12 months after the completion of primary staging surgery and adjuvant therapy to avoid transient elevation of serum CA-125 levels. Serum CA-125 levels are often falsely elevated in patients with disease-free endometrial cancer, and decrease to normal levels 3 months after treatment. Some possible mechanisms for this elevation include peritoneal irritation, mediators of inflammation that induce CA-125 production in the mesothelium, tumor cell shedding from tumor tissues during surgery, or cardiopulmonary function changes after treatment.22,23

Among patients with stage I endometrial cancer, postoperative CA-125 levels were found to be significantly higher in cases of recurrence, and were shown to be independent predictors of recurrence via a multivariate analysis. Patients with increased serum CA-125 levels had higher recurrence rates and shorter RFS periods than those with lower serum CA-125 levels, regardless of the cutoff value selected. The results of the present study also showed that postoperative CA-125 levels had no correlation with preoperative CA-125 levels. Similarly, Patsner et al24 reported that preoperative serum CA-125 levels were normal in 123 of 125 (98.4%) patients with early clinical and surgical stage endometrial adenocarcinoma, and remained normal in all patients, without evidence of either isolated vaginal recurrence or postoperative radiation enteritis. Marked elevations in CA-125 levels were noted in patients with pelvic, abdominal, or pulmonary metastases; therefore, they suggested that serum CA-125 levels may be beneficial in the posttreatment monitoring of patients with early-stage endometrial carcinoma.24 In contrast to the present study, postoperative CA-125 levels were obtained more frequently in the aforementioned study (every 3–4 months postoperatively), with a shorter follow-up time (median, 18 months) and a smaller study population. Furthermore, we reported that obtaining the postoperative CA-125 level within 6 to 12 months after surgery is not only an indicator, but is also a predictor of, recurrence in patients with stage I endometrial cancer.

Compared with epithelial ovarian cancer, the role of CA-125 in the surveillance of endometrial cancer is inconclusive, as the CA-125 blood test is not routinely performed as part of treatment follow-up in some countries. The monitoring of serum CA-125 is suggested in select patients, for example, those with advanced disease, serous cell type, or elevated preoperative CA-125 levels.25,26 Based on the findings of the present study, and the fact that the CA-125 assay is inexpensive, reproducible, and objective, obtaining CA-125 levels within 6 to 12 months after staging surgery may play a role in stage I endometrial cancer follow-up. If the patient has postoperative CA-125 levels >13.75 U/mL, a minimal follow-up schedule may not be recommended. More studies, therefore, are needed to validate an optimal cutoff value of postoperative CA-125 in predicting cancer recurrence.

To the best of the authors’ k