Clinical Characteristics and Prognosis of Ovarian Clear Cell Carcinoma: a Retrospective Study of 112 Patients in Beijing

Objectives This retrospective study aimed to evaluate the clinical characteristics and prognosis of ovarian clear cell carcinoma (OCCC) and to further explore the monitoring value of cancer antigen 125 (CA-125). The medical records of 112 OCCC patients who were treated in Hospital (PUMCH) between 2014 and 2019 were collected and reviewed, and data such as age, Federation of Gynecology and Obstetrics (FIGO) stage, CA-125 level, treatment, recurrence, and death were extracted.

Conclusions FIGO stage and chemotherapy resistance are independent risk factors for prognosis. CA-125 levels following treatment are a valid indicator for treatment monitoring. Regardless of chemosensitivity to CA-125, CA-125 normalization before chemotherapy cycle 2 may not be a distinct in ection point for PFS and OS.

Background
As a form of epithelial ovarian cancer (EOC), ovarian clear cell carcinoma (OCCC) shows distinctive epidemiological and clinical characteristics. 1 In Asia, OCCC accounts for 11.6% (Korea) 2 and 30% (Japan) 3 of all EOCs, surpassing 6% in the United States 4 . In addition, its probability is on the rise, with an observed increase from 3.56% in 1979-1984 to 18.13% in 2005-2008 in Taiwan 5 and from 23.4% in 2010 to 29.1% as of 2010 in Japan 3 . The clinical management of OCCC consists of cytoreductive surgery followed by platinum-and taxane-based chemotherapy 6 . Compared to other EOCs, due to the relative resistance of OCCC to platinum drugs, the e cacy of platinum-based chemotherapy is less than 40% for patients in stage III/IV. 7 The clinical outcomes of patients with advanced OCCC are dismal. Regarding   stage I/II disease, the one-year survival rate of patients with OCCC and other histotypes is no less than   80%, while for stage III/IV disease, the one-year and ve-year survival rates of patients with OCCC are 60% and 22%, respectively, which are lower than those of patients with other histotypes (80% and 35%, respectively). 8 For patients who experience recurrence, the 5-year post-recurrence survival rate is only 13.2%. 9 Given the prevalence and poor prognosis of OCCC in Asia, we retrospectively analysed OCCC patients who visited Peking Union Medical College Hospital (PUMCH) between 2015 and 2019, aimed to describe the treatment methods, recurrence rate and survival rate of OCCC patients and to determine risk factors for the recurrence and survival of OCCC patients.

Patients and data collection
This retrospective single-centre study used essentially the same method as the two-academic-Institute study published in Oncotarget in 2016 10 . We reviewed patients with OCCC (International Federation of Gynecology and Obstetrics [FIGO] stage I − IV) who received treatment at PUMCH (Beijing, China) between 2015 and 2019. We excluded OCCC patients who chose fertility-preserving surgery.
We extracted and evaluated the following information from medical records: demographic and pathological characteristics, immunohistochemistry indicators (Napsin A, HNF1B), stage at diagnosis, surgery, subsequent systemic chemotherapy, recurrence and disease status at last contact. Serum cancer antigen 125 (CA-125) levels were measured using a radioimmunoassay kit (Roche F170 Modular system). The commonly accepted normal upper limit for CA-125 is 35 U/ml. Serum CA-125 levels were measured at the time of the rst attack, preoperation and post-operation, at each cycle of chemotherapy, and at each contact during the follow-up period. The need for informed consent was waived because of the retrospective nature of the study. The study protocol was approved by the Ethics Committee of PUMCH.

Main outcome indicators
Progression-free survival 1 (PFS1) and overall survival (OS) were de ned as the time intervals from the date of the primary surgery to the date of rst recurrence and death or last contact, respectively 11 . Progression-free survival 2 (PFS2) was de ned as the time interval from the date of the second surgery or chemotherapy to the date of second recurrence and death or last contact. Patients were considered to have platinum-resistant disease if the interval time was less than 6 months from the completion of the last platinum-based chemotherapy to disease recurrence, and the patients were considered to have platinum-sensitive disease if the time exceeded 6 months.

Statistical analysis
Statistical analysis was performed using SAS V9.3. The distributions of clinicopathologic events were evaluated using the chi-square test or Fisher's exact test. The Wilcoxon rank-sum test or the Kruskal-Wallis H test was used to determine the distribution of continuous variables between groups. The survival curves were compared by employing the log-rank test. A Cox proportional hazards model was used to evaluate the independent factors affecting survival. A P value < 0.05 was considered signi cant.

Discussion
This study is one of the largest retrospective series of OCCC patients in which data were collected from patients at diagnosis to second recurrence. We mainly described the stage composition, recurrence status, and PFS of OCCC patients and found that FIGO stage and chemotherapy resistance are independent risk factors for prognosis. We also aimed to explain the limitations associated with CA-125 in the staging and recurrence monitoring of OCCC. The limitations of CA-125 in patients with FIGO stage I and those who experience recurrence are highlighted.
Among the 112 patients included, 53.57% were in stage I and 28.57% were in stage III, which is similar to the results of a national study performed in South Korea 2 but quite different from other single-centre studies 6 7 . All patients underwent staging or cytoreductive surgery. Except for one patient who refused chemotherapy after surgery, all the other patients received platinum chemotherapy drugs. Approximately 1/4 of patients in stage III/IV receive advanced chemotherapy, although we found that advanced chemotherapy is not bene cial to prognosis. It is worth noting that the one-year PFS rate of patients with stage I, II and III disease was approximately 90%, and the three-year PFS rate varied greatly, with 86.44% for patients in stage I, 66.67% for patients in stage II, and 66.67% for patients in stage III 56.67%. These three indicators were slightly higher than the national research data of South Korea. Staging has been con rmed as a risk factor for the prognosis of OCCC in many studies, so I will not repeat it here. OCCC patients with platinum-resistant recurrence had a shorter PFS duration than those with platinum-sensitive recurrence (8 months vs 19 months). The median survival duration with single-agent chemotherapy and bevacizumab does not exceed 16 months 12 . Most other chemotherapy regimens are almost ineffective. 11 A recent study summarized the biological rationale and available clinical data on immunotherapy in platinum-resistant ovarian cancer and discussed the challenges and future areas of research in the eld 13 .
In contrast to other studies 10 14 , we found that the monitoring of CA-125 is of limited signi cance and has no predictive signi cance for prognosis. 10 14 We previously published a study in which OCCC patients who visited PUMCH between 1993 and 2013 were analysed and found that CA-125 levels following treatment were a valid indicator for treatment monitoring 10 . Although the same method was used, in the current study, we obtained a negative result, that is, the normalization of CA-125 levels before 2 cycles of chemotherapy does not affect PFS or OS. This may be because the samples we collected are different. In addition, we found that 61.11% of stage I patients had CA-125 levels in the normal range when they rst became ill. During treatment, there was no difference in the number of courses of chemotherapy or prognosis between these patients and other stage I patients, and 72.73% of these patients had normal CA-125 levels after surgery or 1 course of chemotherapy. Therefore, we believe that for patients who are not sensitive to the CA-125 test the rst time, ultrasound, CT or MRI should be considered rst. Importantly, an examination can reduce the disease burden on patients. In contrast to stage I patients, 76.67% of stage III patients had CA-125 levels greater than 35 IU/ml. We believe these patients are more suitable for CA-125 monitoring. Moreover, 50% of patients who experienced recurrence had CA-125 levels in the normal range, which further showed that monitoring CA-125 levels is of limited signi cance and may delay the discovery of recurrence.
We veri ed the value of Napsin A and HNF1B in the diagnosis of OCCC by calculating the false positive rate. In 2013, Skirnisdottir I 15 proposed that the positivity of Napsin A is helpful for the diagnosis of OCCC. In 2020, Fatemeh Nili 16 reported that the sensitivity of Napsin A in the diagnosis of OCCC was 85%. However, we found that 39.62% of patients were still negative for Napsin A. Only 15% of patients were negative for HNF1B, which is consistent with Wenbin Huang's ndings 17 . We combined the two indicators and found that the false positive rate was 9%. It is also worth noting that patients who were in remission had a shorter follow-up time than patients who relapsed (22.50 vs 33.00), (not shown in the table due to space reasons). There is no doubt that disease monitoring has been gradually relaxed in these patients. We believe that clinicians should provide patients with continuous follow-up to detect disease recurrence in advance and intervene as soon as possible.
The patients included in the current study were from economically developed areas in China, and they had satisfactory economic and medical conditions, which suggests that in poor areas in China, the accessibility of treatment and prognosis may be worse. Due to the habit of case writing, clinicians are

Ethics Approval and Consent to Participate
This study was approved by the Institutional Review Board of PUMCH.

Author Contributions
Huimei Zhou designed the study, performed the data analyses and wrote the manuscript. Qian Liu performed the data analyses and revised the manuscript. Jiaxin Yang designed the study, performed the data analyses, revised the manuscript and approved the nal version of the manuscript. Xiaohua Shi, Dongyan Cao, Tao Wang and Keng Shen collected and reviewed the data and approved the nal version of the manuscript.