Prognostic analysis of epithelial ovarian cancer patients with normal CA125 level: a population-based comparative study Authors: Chuanzhong Liu, Liying Zhang, Yibao Huang, Haijuan Shi, Zhijun Yang Published: 2026 ABSTRACT The epithelial ovarian cancer (EOC) with normal CA125 level is relatively rare. This study aims to develop a nomogram model for predicting the overall survival (OS) of EOC patients with normal CA125 level. A total of 13,068 patients with epithelial ovarian cancer (EOC) were extracted from the SEER database, and an additional 138 patients were recruited from our hospital as an external validation cohort. We found that the OEC patients with normal CA125 levels tend to have a better prognosis than those with elevated CA125 levels. Multivariate Cox regression analysis showed that age, summary stage, histology, and lymphadenectomy status were important factors affecting the prognosis of EOC patients with normal CA125 level. Introduction Ovarian cancer, as a prevalent malignant tumour in the female reproductive system, poses a significant threat to the lives and wellbeing of women globally. The latest report “Global Cancer Statistics 2020” from International Agency for Research on Cancer (IARC) indicated that approximately 9.2 million new cancer cases emerged among women worldwide in 2020, where ovarian cancer accounted for 3.4%. Furthermore, ovarian cancer was responsible for 4.7% of the 4.4 million female cancer-related deaths during the same period1. Due to the minimal discernible symptoms during the early stage, it is frequently undetected until progressed to the advanced stage with metastasis, resulting in a poor prognosis with the 5-year survival rate of less than 50%2. Ovarian cancer has a diverse array of histological subtypes, including epithelial tumour, germ cell tumour, and sex cord stromal tumour, among others. Epithelial ovarian cancer (EOC) is the most common subtype, and its main histology includes serous, mucinous, endometrioid and clear cell. EOC is the primary cause for gynecological cancer-related fatalities in USA and its mortality rate ranks fifth among American women3. Currently, clinical recommendations for ovarian cancer primarily rely on the research findings derived from high-grade serous tumour and moderately to poorly differentiated endometrioid tumour4. With the continuous advancement in medical technology, early screening techniques for various malignant tumors have been significantly improved. However, early screening methods for ovarian cancer are still inadequately effective in clinical practice. For example, as a pivotal biomarker for EOC, CA125 has been widely used in the diagnosis, prognosis evaluation, and recurrence monitoring of advanced EOC5–8. Nevertheless, a considerable proportion of EOC patients with advanced stage and approximately 50% of those with early stage still exhibit negative results in CA125 testing9. This observation underscores the challenge in early diagnosis, prognostic evaluation and recurrence surveillance for ovarian cancer, prompting the reliance on more expensive and potentially invasive methods such as computed tomography (CT), biopsy and other examinations10. In a small-scale clinical study, the scholars found that the EOC patients with normal CA125 level tend to exhibit better clinical prognosis compared to those with elevated CA125 level11. However, the correlation between the prognosis of EOC patients and CA125 status is unclear. Therefore, this study aimed to investigate the independent determinants of survival benefits for the EOC patients with normal CA125 level. Patients and methods Patients from SEER database We used the Surveillance, Epidemiology, and End Results (SEER) software (version 8.4.3) to obtain the clinical data of EOC patients. The data were mainly extracted from the Incidence-SEER Research Data, 17 Registries, Nov 2023 Sub (2000–2021). The inclusion criteria encompassed: (1) patients diagnosed with ovarian cancer (“Site and Morphology. Site recode ICD-O-3/WHO 2008″, Ovary); (2) patients with a positive pathology confirmation (“Diagnostic Confirmation”, positive histology). The exclusion criteria were as follows: (1) missing essential data, such as patient ID, SEER stage, survival months, and survival status; (2) no record of CA125 level; (3) not EOC. Demographic and clinical characteristics The demographic variables were primarily included in this study: age (years) (10–39, 40–59, 60–79, over 80), race (Black, White, other), diagnosis (histological diagnosis, clinical diagnosis), first malignant (yes, no), summary stage (localized, regional, distant, unknown), histology (serous, mucinous, clear cell, endometrioid, other, unknown), grade (grade I, grade II, grade III, grade IV, unknown), stage (stage I, stage II, stage III, stage IV, unknown), T stage (T1, T2, T3, unknown), lymphadenectomy (yes, no), chemotherapy (yes, no), chemotherapy type (neoadjuvant, postoperative, no/unknown), radiotherapy (yes, no), cause of death (alived, this cancer, other cause), vital status and survival months. Patients from institutional registry The external validation cohort consisted of 138 patients recruited from the Cancer Hospital of Guangxi Medical University between 2012 and 2015. Among them, 22 patients had normal CA125 levels, whereas 116 patients had elevated CA125 levels. Statistical analysis Statistical analysis was performed using R software (version 4.3.0). Variable comparisons were conducted by the two-tailed t test or χ 2 test, as appropriate. Kaplan-Meier method was used to construct overall survival (OS) curve, and log-rank test was employed to analyze the statistical significance of survival differences. To identify potential predictive factors, the hazard ratio (HR) and their 95% confidence interval (CI) were calculated. Univariate and multivariate Cox regression analysis were performed to evaluate the independent factors influencing OS. For the EOC patients with normal CA125 level, a nomogram was developed based on the findings from multivariate Cox regression analysis. The predictive accuracy was assessed by using receiver operating characteristic (ROC) curve, calibration curve and decision curve analysis. A P-value of less than 0.05 indicates statistical significance of the difference. Result Patient characteristics The flowchart outlined in Fig. 1 encapsulates the process of EOC patient selection and nomogram development. Initially, a total of 65,535 patients diagnosed with ovarian cancer were included from the SEER database. After excluding cases with unknown CA125 level (52,329 cases) and not EOC (138 cases), a final dataset of 13,068 EOC patients were included in the study. In addition, 138 patients were recruited from our hospital as an external validation cohort. The clinical characteristics of these EOC patients were shown in Tables 1 and 2. Fig. 1Flow diagram of patient selection and nomogram development. Flow diagram of patient selection and nomogram development. Table 1Baseline characteristics of patients with EOC from SEER database.VariablesOverall (N = 13068)CA125 levelP-valueNormal (N = 1,550)Elevated (N = 1,1518)Age < 0.001 10–39585 (4.5)150 (9.7)435 (3.8)40–594512 (34.5)649 (41.9)3863 (33.5)60–796321 (48.4)644 (41.5)5677 (49.3)>801650 (12.6)107 (6.9)1543 (13.4)Race 0.002 White10,412 (79.7)1212 (78.2)9200 (79.9)Black929 (7.1)93 (6.0)836 (7.2)Other1727 (13.2)245 (15.8)1482 (12.9)Diagnosis 0.003 Histological12,903 (98.7)1543 (99.5)11,360 (98.6)Clinical165 (1.3)7 (0.5)158 (1.4)First malignantYes11,379 (87.1)1345 (86.8)10,034 (87.1)0.737No1689 (12.9)205 (13.2)1484 (12.9)Summary stage < 0.001 Localized1109 (8.5)406 (26.2)703 (6.0)Regional1873 (14.3)380 (24.5)1493 (13.0)Distant5710 (43.7)220 (14.2)5490 (47.7)Unknown4376 (33.5)544 (35.1)3832 (33.3)Histology < 0.001 Serous9080 (69.5)619 (39.9)8461 (73.5)Mucinous752 (5.8)282 (18.2)470 (4.1)Clear cell1075 (8.2)312 (20.1)763 (6.6)Endometrioid1443 (11.0)270 (17.4)1173 (10.2)Other72 (0.6)19 (1.3)53 (0.6)Unknown646 (4.9)48 (3.1)598 (5.2)Grade < 0.001 Grade I559 (4.3)164 (10.6)395 (3.4)Grade II858 (6.6)195 (12.6)663 (5.8)Grade III2156 (16.5)222 (14.3)1934 (16.8)Grade IV2281 (17.5)209 (13.5)2072 (18.0)Unknown7214 (55.1)760 (49.0)6454 (56.0)Stage < 0.001 Stage I1002 (7.7)325 (21.0)677 (5.9)Stage II327 (2.5)57 (3.7)270 (2.3)Stage III1530 (11.7)92 (5.9)1438 (12.5)Stage IV1266 (9.7)41 (2.6)1225 (10.6)Unknown8943 (68.4)1035 (66.8)7908 (68.7)T stage < 0.001 T11159 (8.9)337 (21.8)822 (7.2)T2441 (3.4)71 (4.6)370 (3.2)T32332 (17.8)103 (6.6)2229 (19.4)Unknown9136 (69.9)1039 (67.0)8097 (70.2)Lymphadenectomy < 0.001 Yes6234 (47.7)1043 (67.3)5191 (45.0)No6793 (52.0)507 (32.7)6286 (54.6)Unknown41 (0.3)0 (0.0)41 (0.4)Chemotherapy < 0.001 Yes10,256 (78.5)981 (63.3)9275 (80.5)No2812 (21.5)569 (36.7)2243 (19.5)Chemotherapy type < 0.001 Neoadjuvant2793 (21.4)60 (3.9)2733 (23.7)Postoperative6023 (46.1)892 (57.5)5131 (44.5)No/Unknown4252 (32.5)598 (38.6)3654 (31.8)Radiotherapy0.802Yes294 (2.2)33 (2.1)261 (2.3)No12,774 (97.8)1517 (97.9)11,257 (97.7)Cause of death < 0.001 Alived6369 (48.7)1189 (76.8)5180 (45.0)This cancer5892 (45.1)275 (17.7)5617 (48.7)Other cause807 (6.2)86 (5.5)721 (6.3) Table 1Baseline characteristics of patients with EOC from SEER database.VariablesOverall (N = 13068)CA125 levelP-valueNormal (N = 1,550)Elevated (N = 1,1518)Age < 0.001 10–39585 (4.5)150 (9.7)435 (3.8)40–594512 (34.5)649 (41.9)3863 (33.5)60–796321 (48.4)644 (41.5)5677 (49.3)>801650 (12.6)107 (6.9)1543 (13.4)Race 0.002 White10,412 (79.7)1212 (78.2)9200 (79.9)Black929 (7.1)93 (6.0)836 (7.2)Other1727 (13.2)245 (15.8)1482 (12.9)Diagnosis 0.003 Histological12,903 (98.7)1543 (99.5)11,360 (98.6)Clinical165 (1.3)7 (0.5)158 (1.4)First malignantYes11,379 (87.1)1345 (86.8)10,034 (87.1)0.737No1689 (12.9)205 (13.2)1484 (12.9)Summary stage < 0.001 Localized1109 (8.5)406 (26.2)703 (6.0)Regional1873 (14.3)380 (24.5)1493 (13.0)Distant5710 (43.7)220 (14.2)5490 (47.7)Unknown4376 (33.5)544 (35.1)3832 (33.3)Histology < 0.001 Serous9080 (69.5)619 (39.9)8461 (73.5)Mucinous752 (5.8)282 (18.2)470 (4.1)Clear cell1075 (8.2)312 (20.1)763 (6.6)Endometrioid1443 (11.0)270 (17.4)1173 (10.2)Other72 (0.6)19 (1.3)53 (0.6)Unknown646 (4.9)48 (3.1)598 (5.2)Grade < 0.001 Grade I559 (4.3)164 (10.6)395 (3.4)Grade II858 (6.6)195 (12.6)663 (5.8)Grade III2156 (16.5)222 (14.3)1934 (16.8)Grade IV2281 (17.5)209 (13.5)2072 (18.0)Unknown7214 (55.1)760 (49.0)6454 (56.0)Stage < 0.001 Stage I1002 (7.7)325 (21.0)677 (5.9)Stage II327 (2.5)57 (3.7)270 (2.3)Stage III1530 (11.7)92 (5.9)1438 (12.5)Stage IV1266 (9.7)41 (2.6)1225 (10.6)Unknown8943 (68.4)1035 (66.8)7908 (68.7)T stage < 0.001 T11159 (8.9)337 (21.8)822 (7.2)T2441 (3.4)71 (4.6)370 (3.2)T32332 (17.8)103 (6.6)2229 (19.4)Unknown9136 (69.9)1039 (67.0)8097 (70.2)Lymphadenectomy < 0.001 Yes6234 (47.7)1043 (67.3)5191 (45.0)No6793 (52.0)507 (32.7)6286 (54.6)Unknown41 (0.3)0 (0.0)41 (0.4)Chemotherapy < 0.001 Yes10,256 (78.5)981 (63.3)9275 (80.5)No2812 (21.5)569 (36.7)2243 (19.5)Chemotherapy type < 0.001 Neoadjuvant2793 (21.4)60 (3.9)2733 (23.7)Postoperative6023 (46.1)892 (57.5)5131 (44.5)No/Unknown4252 (32.5)598 (38.6)3654 (31.8)Radiotherapy0.802Yes294 (2.2)33 (2.1)261 (2.3)No12,774 (97.8)1517 (97.9)11,257 (97.7)Cause of death < 0.001 Alived6369 (48.7)1189 (76.8)5180 (45.0)This cancer5892 (45.1)275 (17.7)5617 (48.7)Other cause807 (6.2)86 (5.5)721 (6.3) Baseline characteristics of patients with EOC from SEER database. Table 2Baseline characteristics of patients with EOC from our registry.VariablesOverall (N = 138)CA125 levelP-valueNormal (N = 22)Elevated (N = 1,16)Age0.14710–391831540–5993187560–7927126Summary stage 0.005 Localized1367Regional19118Distant1061591Histology 0.021 Serous1081395Mucinous1468Clear cell624Endometrioid413Other606Grade0.247Grade I1129Grade II26620Grade III73766Grade IV101Unknown27720Stage 0.012 Stage I1367Stage II19118Stage III801268Stage IV26323T stage 0.005 T11367T219118T31061591Lymphadenectomy0.460Yes12519106No13310Chemotherapy0.377Yes13422112No404Chemotherapy type 0.044 Neoadjuvant671552Postoperative71764Cause of death0.283Alived731558This cancer58652Other cause716 Table 2Baseline characteristics of patients with EOC from our registry.VariablesOverall (N = 138)CA125 levelP-valueNormal (N = 22)Elevated (N = 1,16)Age0.14710–391831540–5993187560–7927126Summary stage 0.005 Localized1367Regional19118Distant1061591Histology 0.021 Serous1081395Mucinous1468Clear cell624Endometrioid413Other606Grade0.247Grade I1129Grade II26620Grade III73766Grade IV101Unknown27720Stage 0.012 Stage I1367Stage II19118Stage III801268Stage IV26323T stage 0.005 T11367T219118T31061591Lymphadenectomy0.460Yes12519106No13310Chemotherapy0.377Yes13422112No404Chemotherapy type 0.044 Neoadjuvant671552Postoperative71764Cause of death0.283Alived731558This cancer58652Other cause716 Baseline characteristics of patients with EOC from our registry. Overall survival analysis for patients with EOC between elevated and normal CA125 level OS was calculated from the date of treatment until the patient’s death or the date of the final follow-up consultation. For the patients with normal CA125 level, the 2-year, 4-year, 6-year, 8-year, and 10-year survival rates were 88.9%, 80.8%, 72.9%, 67.4%, and 64.8%, respectively, and the median OS has not yet been reached. For patients with elevated CA125 level, the 2-year, 4-year, 6-year, 8-year, and 10-year survival rates were 65.7%, 47.2%, 36.6%, 31.0%, and 27.6%, respectively, with a median OS of 44 months. Therefore, the OS of EOC patients with normal CA125 level was better than those with elevated CA125 level (P < 0.0001) Fig. 2A). Consistent with the SEER database findings, our external validation cohort demonstrated that normal CA125 level is a strong predictor of favorable overall survival in EOC patients (P = 0.02) (Fig. 2B). Fig. 2Kaplan-Meier survival curves for EOC patients with elevated or normal CA125 level from SEER database (A) and our registry (B). Kaplan-Meier survival curves for EOC patients with elevated or normal CA125 level from SEER database (A) and our registry (B). Development and validation of nomogram For internal validation, the 1,550 EOC patients with normal CA125 level were randomly allocated into a training set (N = 1,085, 70%) and a test set (N = 465, 30%). The findings from Cox regression analysis of OS in the training set were presented in Table 3. The multivariate Cox regression analysis showed that age (4.016, 95% CI: 2.082–7.748), summary stage (0.205, 95% CI: 0.078–0.544), histology (0.308, 95% CI: 0.138–0.688), and lymphadenectomy status (0.679, 95% CI: 0.516–0.894) emerged as independent predictors of OS. Then, a nomogram was developed based on findings from the multivariate Cox regression analysis (Fig. 3A). The 5-year AUC of training set and test set were 0.763 and 0.816, respectively (Fig. 3B,C), reflecting satisfactory accuracy of the nomogram. The 1-year, 3-year and 5-year calibration curves for both the training set and test set closely adhered to the ideal diagonal line, indicating good agreement (Fig. 3D and I). In addition, decision curve analysis revealed a substantial net benefit associated with the nomogram in both the training set and test set (Fig. 3J and O). Table 3Cox regression analysis of overall survival (OS) in the training set (N = 1,085).VariablesOS-UnivariateOS-MultivariateHR (95% CI)P-valueHR.95% CIP-valueAge7.735 (4.240-14.112)< 0.0014.016 (2.082–7.748) < 0.001 Race0.804 (0.567–1.138)0.218Diagnosis0.172 (0.064–0.462)< 0.0011.7 (0.408–7.090)0.466First malignant0.601 (0.436–0.827)0.0020.796 (0.565–1.122)0.193Summary stage0.369 (0.235–0.581)< 0.0010.205 (0.078–0.544) 0.001 Histology0.094 (0.047–0.185)< 0.0010.308 (0.138–0.688) 0.004 Grade0.297 (0.162–0.546)< 0.0010.532 (0.267–1.058)0.072Stage4.143 (2.156–7.961)< 0.0011.647 (0.479–5.660)0.428T stage1.834 (1.031–3.262)0.0390.767 (0.207–2.838)0.691Lymphadenectomy0.446 (0.348–0.572)< 0.0010.679 (0.516–0.894) 0.006 Chemotherapy1.14 (0.877–1.481)0.327Chemotherapy type2.606 (1.591–4.270)< 0.0011.171 (0.671–2.042)0.578Radiotherapy3.071 (1.72–5.484)< 0.0011.555 (0.698–3.465)0.280 Table 3Cox regression analysis of overall survival (OS) in the training set (N = 1,085).VariablesOS-UnivariateOS-MultivariateHR (95% CI)P-valueHR.95% CIP-valueAge7.735 (4.240-14.112)< 0.0014.016 (2.082–7.748) < 0.001 Race0.804 (0.567–1.138)0.218Diagnosis0.172 (0.064–0.462)< 0.0011.7 (0.408–7.090)0.466First malignant0.601 (0.436–0.827)0.0020.796 (0.565–1.122)0.193Summary stage0.369 (0.235–0.581)< 0.0010.205 (0.078–0.544) 0.001 Histology0.094 (0.047–0.185)< 0.0010.308 (0.138–0.688) 0.004 Grade0.297 (0.162–0.546)< 0.0010.532 (0.267–1.058)0.072Stage4.143 (2.156–7.961)< 0.0011.647 (0.479–5.660)0.428T stage1.834 (1.031–3.262)0.0390.767 (0.207–2.838)0.691Lymphadenectomy0.446 (0.348–0.572)< 0.0010.679 (0.516–0.894) 0.006 Chemotherapy1.14 (0.877–1.481)0.327Chemotherapy type2.606 (1.591–4.270)< 0.0011.171 (0.671–2.042)0.578Radiotherapy3.071 (1.72–5.484)< 0.0011.555 (0.698–3.465)0.280 Cox regression analysis of overall survival (OS) in the training set (N = 1,085). Fig. 3Development and validation of nomogram for EOC patients with normal CA125 level from SEER database. (A) A nomogram for predicting OS; The AUC curve of training set (B) and test set (C); (D-F) The 1-year, 3-year and 5-year calibration curves for the training set; (G-I) The 1-year, 3-year and 5-year calibration curves for the test set; (J-L) The 1-year, 3-year and 5-year decision curves for the training set; (M-O) The 1-year, 3-year and 5-year decision curves for the test set. Development and validation of nomogram for EOC patients with normal CA125 level from SEER database. (A) A nomogram for predicting OS; The AUC curve of training set (B) and test set (C); (D-F) The 1-year, 3-year and 5-year calibration curves for the training set; (G-I) The 1-year, 3-year and 5-year calibration curves for the test set; (J-L) The 1-year, 3-year and 5-year decision curves for the training set; (M-O) The 1-year, 3-year and 5-year decision curves for the test set. External validation of nomogram for EOC patients with normal CA125 level The AUC values for the 1-year, 3-year and 5-year predictions in the external validation cohort were 1.00, 0.80, and 0.91, respectively (Fig. 4A-C), demonstrating the satisfactory accuracy of the nomogram. Furthermore, the calibration curves for the external validation cohort closely aligned with the ideal diagonal line (Fig. 4D-F), indicating excellent agreement between predicted and observed outcomes. In addition, decision curve analysis confirmed a substantial net clinical benefit of the nomogram in the external validation cohort (Fig. 4G-I). Fig. 4External validation of nomogram for EOC patients with normal CA125 level. The 1-year (A), 3-year (B) and 5-year (C) AUC curve of external validation cohort; (D-F) The 1-year, 3-year and 5-year calibration curves for the external validation cohort; (G-I) The 1-year, 3-year and 5-year decision curves for the external validation cohort. External validation of nomogram for EOC patients with normal CA125 level. The 1-year (A), 3-year (B) and 5-year (C) AUC curve of external validation cohort; (D-F) The 1-year, 3-year and 5-year calibration curves for the external validation cohort; (G-I) The 1-year, 3-year and 5-year decision curves for the external validation cohort. Discussion Clinical research focusing on the EOC patients with normal CA125 level is relatively scarce. This study selected 13,068 eligible EOC patients from the SEER database and 138 patients from our hospital to date examining correlation between the prognosis and CA125 status in EOC patients. The research indicates that the EOC patients with normal CA125 level have a better prognosis than those with elevated CA125 level. The prognosis of EOC is influenced by multiple factors, including age, tumor stage, postoperative residual tumor size, number of adjuvant chemotherapy cycles, and the performance of lymph node dissection. Notably, accumulating evidence indicates that age plays a significant role in determining the treatment outcomes of ovarian cancer12. Elderly patients are often constrained by their physiological status and immune function, which limits treatment options and results in generally lower survival rates. In contrast, younger EOC patients showed significantly improved OS, particularly when the postoperative residual disease is less than 1 cm. Nevertheless, residual disease remained a key predictor of OS. Other research confirmed that SEER stage, FIGO stage, and lymph node metastasis all strongly influence survival13,14. Notably, systematic lymph node dissection was linked to better outcomes in early-stage ovarian cancer15. In addition, histological subtype is a major prognostic factor. Zhou et al. found that among EOC, endometrioid carcinoma had better survival outcomes than serous, mucinous, clear cell carcinomas16,17. In addition to the above-mentioned factors, the results of this study suggest that a normal CA125 level may be another independent prognostic factor for EOC. CA125 is regarded as a glycoprotein complex characterized by an extensive epitope map, which is the extracellular segment of mucin 16 (MUC16) that is secreted into serum18. The CA125, a highly glycosylated protein, can protect the extracellular domain from endogenous and exogenous damage19. In addition, CA125 plays a pivotal role in the occurrence and metastasis of tumors20. Downregulation of MUC16 on the cell surface can reduce DNA repair capacity, exacerbate DNA damage, and sensitize tumor cells to genotoxic drugs by suppressing FOXO3a expression and nuclear localization21. In addition, Okła et al.22 revealed a linkage between the mesothelin concentrations in plasma and peritoneal fluid with its expression levels in tumor tissues, and mesothelin levels were positively correlated with CA125 levels. Gubbels et al.23 discovered that the combination between CA125 and mesothelin can enhance cancer cell adhesion and promote metastasis by increasing the level of metalloproteinases-7 and promoting the phosphorylation of p38 mitogen activated protein kinase. This CA125-mesothelin complex plays an important role in the multi-step process of ovarian cancer peritoneal metastasis24. Current therapeutic endeavors targeting CA125 have exhibited promising prospects. For example, NAV-001, a high-efficacy antibody-drug conjugate targeting mesothelin, can effectively inhibit the growth of tumor cells by recognizing the glycosylation site of CA125 binding domain and then blocking the interaction between mesothelin and CA12525. However, the current studies are confined to preclinical stages. In this study, the EOC patients with normal CA125 level exhibit favorable prognosis. The patient’s age and FIGO stage were critical prognostic factors. These findings fit within the broader prognostic landscape of ovarian cancer. In clinical practice, recurrent ovarian cancer typically presents with a substantial tumor burden, which poses significant challenges to achieving satisfactory secondary cytoreduction. Given the lack of unequivocal evidence regarding the survival benefit of reoperation, major clinical guidelines, such as those from the NCCN, maintain a conservative stance toward secondary surgery for recurrent EOC. Moreover, a larger tumor burden is frequently associated with elevated CA125 levels. This observation underscores the clinical correlation between CA125 and tumor metastasis, suggesting that EOC patients with normal CA125 levels may harbor a lower tumor burden. For example, Plotti F et al.11 conducted a small-scale clinical study involving 8 recurrent ovarian cancer patients with normal CA125 level. Their findings showed that all patients achieved satisfactory surgery for the tumor reduction. This observation raises the question of whether normal CA125 levels should serve as an independent predictive factor in surgical decision-making for EOC. However, the underlying mechanism remains unclear, and further research is of great significance. In addition to CA125, HE4 is the most widely utilized serum biomarker for EOC. HE4 is overexpressed in both early-stage and recurrent EOC, with high preoperative levels strongly associated with poor prognosis26. Several diagnostic tests based on CA125 or HE4 have been clinically established and have demonstrated favorable performance in the diagnosis and management of EOC. However, studies have shown that HE4 primarily provides diagnostic value in patients with elevated CA125 levels27. Consequently, these observations underscore the clinical significance of investigating EOC cases presenting with normal CA125 levels. However, this study has several limitations. First, it does not define the threshold criteria for “normal CA125 level” in SEER database. Without a standardized cutoff, the normal level may differ across settings, reducing comparability and generalizability. Second, due to the inherent limitations of the SEER database, detailed treatment information was not available for analysis, such as chemotherapy regimens, cycles, and response to therapy. Third, the nomogram was externally validated with a small sample size and was not enough to confirm its reliability, larger independent external cohorts are needed. Conclusion Patients with EOC who have normal CA125 levels tend to have a better prognosis than those with elevated CA125 levels. The patient’s age, summary stage, histology, and lymphadenectomy status were important factors affecting their prognosis. Larger studies are required to validate the above findings.