RESEARCH Impact of Preoperative Serum Levels of CA 125 on Epithelial Ovarian Cancer Survival

Background: CA125 is very helpful in treatment monitoring and detection of epithelial ovarian cancer (EOC) recurrence. However there is controversy as to its accuracy and optimal usage. What is the impact of the CA125 levels before primary surgery treatment to the survival of patients? This study aimed to detect any association of preoperative serum levels with prognosis and survival in EOC patients. Materials and Methods: Our cohort comprised EOC patients in Dr. Sardjito Hospital, Yogyakarta, Indonesia, who complied with follow up. To explore the effect of preoperative CA125 levels and other variables on survival Cox’s regression models were applied. Results: A total of 90 cases of EOC who had surgery were available for follow up. The level of CA125 proved to be a prognostic factor for overall survival of EOC patients, with an adjusted HR of 4.10 (p = 0.03). Adjuvant chemotherapy was another prognostic factor, 1 - 2 cycles having an adjusted HR of 0.17 (p = 0.04) and 3 - 8 cycles


Introduction
Ovarian cancer is gynecologic cancer as the fourth highest cause of death in women. Ovarian cancer in the deaths (Siegel et al. 2015). Despite maximum efforts have been made to develop a new way of screening, diagnosis, and treatment strategies, but the incidence and such as cancer of the uterus, stomach, colon and rectum (Jemal et al., 2008). Many factors affect the prognosis of patients with ovarian cancer, such as age, performance status, disease stage, histopathology type, the degree of differentiation, ascites more than 500 ml and residual tumor (du Bois et al., 2009). Ovarian cancer is gynecologic malignancies gave overall the highest risk of death (Colombo et al., 2006;Siegel et al., 2015).
Cancer antigen (CA)125 is a protein in the blood obtained in various conditions including ovarian cancer. A review of 15 studies showed that rising CA125 levels in 50% of patients with stage I disease, 90% in stage II, 92% stage III, 94% in stage IV disease (Jacobs and Bast, Heru Pradjatmo 1989). Although the C125 is a useful tumor marker for monitoring disease progression of ovarian cancer, CA125 alone is not helpful for a test screening of ovarian cancer. This test is not sensitive enough to diagnose the early stages of the disease. More than 85% of women with advanced disease occur rising levels of CA125 (> 35 U/ ml). Nearly 6% of women without ovarian cancer CA125 levels gained more than 35 U/ml (Urban, 2003).

Impact of Preoperative Serum Levels of CA 125 on Epithelial Ovarian Cancer Survival
CA125 is very useful in treatment monitoring and detection of ovarian cancer recurrence. CA125 measurements can be helpful only in women where ovarian cancer has been diagnosed, CA125 as a prognostic marker, not a diagnostic marker for detecting ovarian cancer. It is agreed that CA125 potential as a useful marker for ovarian cancer, although controversy about how accurately these markers to determine the status and how the optimal way to use it. Some studies have suggested that the levels of serum CA125 may be a prognostic factor for survival in ovarian cancer patients. CA125 levels were mostly decreased in patients with ovarian cancer who respond to chemotherapy (Rustin et al., 2009). CA125 levels were very high before the operation was associated with a poor prognosis, and patients with CA125 levels of more than 259 U/ml before chemotherapy begins has a poor prognosis (Parker et al., 1988). But information or third chemotherapy is still high indicating the patient has a poor prognosis, so it will affect the consideration of the clinician to stop or change treatment. Several research second or third chemotherapy is a prognosis variable for survival (Redman et al., 1990;Rustin et al., 2009). How do the CA125 levels before primary operations treatment to the survival of patients? Several studies are still obtained controversial results so that still need to be investigated to get more research then a meta-analysis of research will be able to perform and will found the best evidence.

Materials and Methods
The design was a prospective cohort study, a population study of all patients with ovarian cancer who were treated at the Dr. Sardjito Hospital, Yogyakarta, Indonesia. The research sample was epithelial ovarian cancer (EOC) patients who can be performed following up. Some of the variables examined and recorded, such as CA125 serum levels before surgery, age, weight and the degree of differentiation, number of chemotherapy, disease stage, and residual tumor during surgery. The Each of these variables was grouped into two or three groups. Distribution of variables recorded was presented in Table 1. To explore the effect of preoperative CA125 levels and other variables with the survival of the EOC patients for bivariable and multivariable analysis with Cox's regression models.

Results
The study obtained 90 cases of epithelial ovarian cancer who had surgery and patients was able to follow up. The shortest followed-up of patients was 18 months, and the longest followed-up was 54 months. None of the patients received chemotherapy before surgery. The distribution of observed variables is presented in Table  1. All patients examined serum CA125 levels before surgery and in the analysis were grouped into two groups, were 61/90 (67.8%). Most patients aged >50 years there were 59/90 (65.6%). BMI of the patients in the analysis there were 65/90 (65.5%) and overweight to obese there were 25/90 (34.5%), histological type were divided into three groups: serous type 25/90 (27.8%) mucinous type 44 (48.9%) other types 21/90 (23.4%). The degree of and poor differentiated. The number of chemotherapy

Discussion
Ovarian cancer is rarely found at an early stage as it develops in silent and almost no symptoms. When clinical symptoms are present generally as the result of the growth, development and complications that often arise at the level of an advanced stage. The late stage caused by the absence time symptoms arise ovarian tumor may have grown large enough so that there has been a release of cancer cells into the abdominal cavity. In Pakistan Sarwar et al (2006) reported that 21.2% EOC patients found in early stage. A previous report in Indonesia ovarian cancer in RSCM Jakarta was obtained early stage ovarian cancer 31.2% (Sihombing and Sirait, 2007). In the absence of an it is more than 80% of ovarian cancers are diagnosed at the late stage (Havrilesky et al., 2008). The time when in stage I become 20-25% in the late stage despite the treatment that should be (Rapkiewicz et al., 2004). High risk of mortality mainly caused by cancer has spread to peritoneum or abdominal organs so that treatment may not be optimal.
Some studies suggest that serum CA125 levels may be an important prognostic factor for survival in ovarian cancer patients. CA125 serum levels associated with the tumor, but not all of the increase means the progression of the disease. CA125 is not synthesized only in ovarian cancer cells or ovarian tissue. CA125 may be increased in the state instead of malignancy. Illnesses such as adnexitis, hepatitis, pancreatitis, liver cirrhosis, autoimmune diseases or early pregnancy there is also a rise in CA125. Increased levels of CA125 may also be derived from other epithelial tissues, including pleural coelom, pericardial, as well as epithelial cells of the lung, breast, and even the conjunctiva (Whitehouse and Solomon, 2003).  Biomarkers for the diagnosis, prognosis, predicts survival and response to therapy are increasingly gaining attention, especially in epithelial ovarian cancer. CA125 is the best-known biomarker for epithelial ovarian cancer and the only biomarker that is still used in everyday practice before other biomarkers acceptable to replace it. Many gynecologist oncologists emphasize the necessity of this biomarker in determining a patient's individual treatment basis (Tingulstad et al., 2003;Hodgall, 2008). It has been agreed that CA125 can not be used as a marker for early diagnosis of ovarian cancer. Only 50% early stage of ovarian cancers there are increased levels of CA125, and 10% advanced ovarian cancer levels CA125 is normal (Nustad et al., 1996;Urban, 2003), so that CA125 for ovarian cancer screening the result of positive predictive value is low. It was said that CA 125 can be used for early diagnosis of ovarian cancer  (1987) and Sevelda et al (1989) using a cutoff point CA125 35 U/ml. Others researchers reported that CA125 was a prognostic factor for survival in ovarian cancer patients. Nagele et al (1995), Moebus et al (1988), Petri et al (2006) using a cut-off CA125 65 U/ml, Tang et al (2012) using a cut-off CA125 50 U/ml. In which patients higher survival compared to patients who had CA125 levels above the cutoff. Knowledgeable preoperative state of ovarian cancer patients with groups that have a good or poor prognosis that the predictions are accurate and crucial for determining treatment strategies.
In this study used a cut-off CA125 70 U/ml based on the results of research which state that the cut-off 70 U/ (Rustin et al., 2009). They also said that to choose cut-off between 35 U/ml to 90 U/ml. This study used a cut-off CA125 70 U/ml found that preoperative serum CA125 levels of patients with epithelial ovarian cancer are prognostic factor of survival in EOC patients. As seen from the results of the bivariable and multivariable analysis with Cox's regression model. The Crude Hazard Ratio HR 5.13 (95% CI: 1.54-17.10) with p <0.01 and adjusted HR 4.10 (95% CI: 1.07-15.76) with p=0:03 (Table 2). In this study also attempted to be analyzed by cut off 65 U/ ml and cut off 75 U/ml. The result found that for cut-off 65 U/ml crude HR 4,16 (95% CI: 1.32-13.92) with p=0.02 and adjusted HR 2.81 (95% CI: 0.73-10.79) with p=0.13. 5.13 (95% CI 1.54-17.10) with p <0:01 and adjusted HR 1.72 (95% CI 0.36-8.07) with p=0.48. So with a cut-off 70 U/ml, showed bigger survival differences as well as and three respectively. These results support the statement of Rustin et al (2009) that the cut-off 70 U/ml provides serum levels of CA125 is a prognostic factor of survival in epithelial ovarian cancer.
The analysis results of the age of the patient obtained in bivariable as well as multivariable analysis showed with p=0.22 and adjusted HR 1.54 with p=0.32 means likelihood of mortality more than 1.5 times compared to the age group <50 years. The result due to the power of this studied is small or the sample size of this studied is not enough to show the differences of age as a prognostic that ages was not as a prognostic survival in EOC patients. Other investigators have reported that age was a prognostic factor for survival in ovarian cancer patients (Osman et al., 2008;du Bois et al., 2009;Tang et al., 2012;Liu et al., 2014;Nassar et al., 2015).
The nutritional status of patients did not appear to be seems nutritional status or a higher BMI had a protective effect where the crude and adjusted HR <1, meaning a better nutritional status has an effect of increasing the survival of patients. Results of a meta-analysis of previous studies concluded that BMI at diagnosis can not be a prognostic factor of survival in ovarian cancer patients either (Bae et al. 2014). The National Cancer Institutes said that the 5-year survival rate for ovarian cancer patients also depends on histopathological type and the degree of differentiation of tumor cells. This study found that mucinous type is a poor prognostic factor for survival of EOC patients with HR 2.18 compared to the serous type, it was clinically researchers reported that histological types were a prognostic factor of survival in ovarian cancer (du Bois et al., 2009;Tang et al., 2012;Chen et al., 2013;Liu et al., 2014).
The degree of differentiation appears affected to the survival of patients in which the degree of differentiation is getting worse with the differentiation is getting worse either. Well differentiated as the reference then the moderate differentiated and poor differentiated had the crude HR 1.56 (p=0.35) adjusted HR 1.61 (p=0.51) and crude HR 2.22 (p=0.07) adjusted HR 3.41 (p=0.15) the degree of differentiation is also a prognostic factor 2009; Liu et al., 2014;Nassar et al., 2015).
Chemotherapy seems to be the only factor that of EOC in this studied besides CA125 as the variable of interest of the study. The administration of chemotherapy to the patients provide protection where the survival longer than the patient who did not receive chemotherapy as seen in the multivariable analysis the adjusted HR of the patients who got chemotherapy the two groups adjusted HR 0.17 (p=0.04) and HR 0.39 (p=0.06). Liu et al (2014) reported in their study found that adjuvant chemotherapy either.
Previous researchers have reported that stage of the disease is a prognostic factor of survival in ovarian cancer patients (du Bois et al., 2009;Tang et al., 2012;Liu et al., 2014;Chen et al., 2015;Nassar et al., 2015;Pradjatmo, 2015). Becomes more evident that the early diagnosis of ovarian cancer will reduce the mortality. This study found factor of survival in EOC patients with crude HR 3.56 (p=0.01) and adjusted HR 1.98 (p=0.27).
This study found that the residual tumor when surgery was not a prognostic factor for patient survival, the analysis obtained crude HR=1.27 (p=0.57) adjusted HR=1.03 (p=0.95), even though with nonoptimal surgery the risk of death of EOC patients increase. Other investigators have reported that the residual tumor is a prognostic factor for survival in ovarian cancer patients et al., 1995;Tingulstad et al., 2003;Osman et al., 2008;du Bois et al., 2009;Pongsanon et al., 2011;Liu et al., 2014).
Summary: The study obtained 90 cases of EOC, who had surgery and patients was able to follow up. The variable of interest was the preoperative level of CA125, and the found that level of CA125 is a prognostic factor for overall survival of EOC patients, the adjusted HR 4.10 (p=0.03). The role of CA125 levels for the prediction of prognosis survival of ovarian cancer has been reported. However, several researchers obtained different results. Adjuvant chemotherapy was a prognostic factor either, patients who got chemotherapy 1-2 cycles adjusted HR 0.17 (p=0.04) and who got 3-8 cycles HR 0.39 (p=0.06). Other factors such as; age of patients adjusted HR 1.54 (p=0.32), grade of differentiation for moderate differentiated adjusted HR 1.61 (p=0.51) for poor differentiated adjusted HR 3.41 (p=0.15), stage of disease adjusted HR 1.98 (p=0.27).
is low or the sample size was not enough to show the