From 1 January 1998 to 31 December 2005, all patients treated for advanced epithelial ovarian cancer at the Department of Obstetrics and Gynecology of Chonnam National University Hospital (CNUH) were identified from the tumor registry database. All patients were screened retrospectively for their initial serum CA-125 level, age at diagnosis, tumor histology and grade, stage of disease based on the International Federation of Gynecology and Obstetrics (FIGO) staging system, availability of serial serum CA-125 level determinations and timing of recurrence and/or disease status at last follow-up. Patients with epithelial tumors of borderline malignancy, germ cell tumor and gonadal stromal tumor were excluded from this investigation.

Following initial eligibility screening, the following inclusion criteria were applied to determine the final study population: 1) elevated (≥ 35 U/mL) serum CA-125 level at the time of diagnosis, 2) histologic documentation of epithelial ovarian cancer, 3) stage of disease III (A, B, and C) and IV treated with primary cytoreductive surgery and postoperative chemotherapy, 4) complete clinical and radiographic response to initial treatment with normalization of CA-125 (<35 U/mL), 5) follow-up serum CA-125 determinations and/or radiographic scans at 1- to 3-month intervals, and 6) clinical and radiographic determination of disease status at the time of last follow-up or recurrence.

After primary surgical staging procedure, all patients received combination chemotherapy with paclitaxel (175 mg/m² by 3 hour infusion) and carboplatin (area under the curve of 5) every 3 weeks for a total of six cycles. The treatment response was assessed by a complete history, physical examination, appropriate imaging studies (chest X-ray, computed tomography, magnetic resonance imaging and positron emission tomography) and serum CA-125 level. Serum CA-125 levels were measured preoperatively and within 1 week before each cycle of primary adjuvant chemotherapy. All serum samples were analyzed in the same laboratory at CNUH. Optimal cytoreductive surgery was defined as a following review of the surgery and pathology report <1 cm of residual disease was present (following the recommendations of the Gynecology Oncology Group Protocol 17). Complete response was defined as the absence of all measurable disease for at least 4 weeks and the normalization of the serum CA-125 level.

Overall differences in the serum CA-125 level according to the nadir serum CA-125 level were assessed by Student's t-test and the Wilcoxon rank-sum test. A receiver operating characteristic (ROC) curve was used to determine the most clinically useful serum CA-125 threshold value to predict a recurrence. To estimate linear correlation between nadir CA-125 levels and clinical variables, simple lineal regression analysis was used. A multivariate Cox proportional hazards model for OS and PFS was used to assess differences in outcome on the basis of nadir serum CA-125 levels, including other variables, such as age (>65 years vs. ≤65 years), stage (IV vs. III), tumor grade (3 vs. ≤2), histology (serous vs. non-serous) and residual tumor size. Step-wise regression techniques were used to build multivariate models using a significance level of 0.10 to remain in the model. The differences in OS and PFS by nadir serum CA-125 levels and normalization timing were evaluated using the Kaplan-Meier method with log-rank tests. Statistical analyses were performed using SPSS ver.17.0 (SPSS Inc., Chicago, IL, USA). All P-values reported are two-sided, and a P-value <0.05 was considered statistically significant.