Diagnostic Accuracy and Appropriate Cut Off Value of Risk of Malignancy Index in Preoperative Discrimination Between Malignant and Benign Ovarian Tumors: Prospective Cross-sectional Study


 BackgroundRisk of malignancy index (RMI) is scoring system which was introduced to differentiate between benign and malignant ovarian tumor. It incorporates CA-125, ultrasound score and menopausal status for prediction of ovarian malignancies in preoperative period. There is no universal screening method to discriminate between benign and malignant adnexal masses yet. So, this study was conducted to determine the diagnostic accuracy of RMI and determine best cut off value for RMI.MethodsProspective cross-sectional study was carried out among women with ovarian mass admitted to Gynecology ward and operated from September 1, 2019 to June 30, 2020.Data analysis was carried out using SPSS version 26. CA-125 level, menopausal status and ultrasound score were used to calculate RMI. Finally, RMI score was compared to histopathology result used as gold standard.ResultsNinity nine patients were enrolled in this study. Prevalence of benign ovarian tumors were 61.6% (61/99) and that of malignant ovarian tumors were 38.4% (38/99). The mean age for benign tumors was 30±9yrs and the mean age for malignant tumors was 50.6±10.8yrs. Among benign tumors, serous cystadenoma was the most common (36%), followed by dermoid cyst (32.9%), mucinous cyst adenoma (14.8%). The most common malignant ovarian tumor was serous cyst adenocarcinoma (63.2%), followed by mucinous cystadenocarcinoma (23.8%) and dysgerminoma (5.3%). Overall, using RMI score cut off value 220 has good sensitivity (84.2%), specificity (77%), PPV (69.5%), NPV (88.7%) and diagnostic accuracy (79.8%) for discriminating between benign and malignant ovarian tumors.ConclusionFrom this study there were high proportion of women with RMI>=220 in malignant ovarian tumors group. The study shows that there is significant role of RMI in prediction of ovarian malignancy thus helping in deciding which patients need referral to a center where gynecologic oncologists are available. It is good practice to use it in developing countries including our country because of its simplicity, safety and applicability in initial evaluations of patients with adnexal mass.


Introduction
Ovarian cancer is the second most common type of female reproductive cancer, and more women die from ovarian cancer than cervical cancer and uterine cancer combined (1). Woman's risk at birth of ovarian cancer at some time in her life is 1 % to 1.5% and that of dying from cancer is almost 0.5% (1).In 2018 4.4% of all cancer related mortality is attributed to ovarian cancer (2). It is the seventh most common cancer in women, and incidence rates are highest in developed countries (2,3). The incidence of ovarian cancer increases with age. Interest in early detection as a method of reducing mortality developed with the discovery of serum tumor markers associated with ovarian malignancies particularly CA 125 and with the improved diagnostic accuracy of pelvic ultrasonography (3).
The high mortality rate in ovarian malignancy is mainly due to late detection of disease (4). If it can be detected at an early stage then disease can be treated with optimal primary cytoreduction and achievement of optimal cytoreduction (single most important prognostic criteria) becomes possible (5).Pelvic examination is not speci c and sensitive in detecting ovarian malignancy. Cancers detected by pelvic examination are often far advanced, so pelvic examination for screening is not recommended (5). Tumor markers when used alone are not speci c to be used as screening test. They are raised in a number of benign conditions and are not raised in poorly differentiated cancer, borderline tumors and mucinous tumors (6). In a prospective study conducted to evaluate sensitivity and speci city of CA-125 as a marker for ovarian malignancy concluded that measurement of serum CA-125 levels, particularly at a reference value of 35 IU/mL, is not su ciently sensitive to be used alone as a screening test for the detection of ovarian cancer (6). Ultrasound can differentiate between solid, cystic and multilocular masses, although malignancy cannot be diagnosed, unequivocally (6).Risk of malignancy index is a scoring system which can be introduced to differentiate between malignant and benign ovarian tumor (5). RMI in ovarian malignancy incorporates CA-125, USG and Menopausal status for Predicting if an ovarian mass is malignant or benign, screening for suspected pelvic mass, deciding appropriate management protocol and triage of patients (5). This score has given signi cantly superior results than the use of a single parameter (7). RMI Score (RMI) -Total Score = USG Score X Menopausal Score X CA -125(U/ml). USG score: 1 -No risk factor or one risk factor, 3 -Two -Five risk factors. High risk factors in USG: Multiloculated cysts, solid areas, bilateral lesions, ascites and evidence of metastasis.
Menopausal status: 1-Pre-menopausal, 3 -Post-menopausal CA125-Absolute value (IU/ml). The anatomic site of ovaries makes it inaccessible to simple diagnostic procedure like smear and biopsy like cancers of uterus and cervix (4). Absence of effective screening method for ovarian cancer yet is another contemporary challenge. Patients with malignant tumors should be evaluated by gynecologic oncologist as the quality of cytoreduction surgery and surgical staging/ lymph node dissection are critical prognostic parameters in ovarian malignancy (5,7). The nding of an ovarian mass raises questions about the most suitable management and the place where this management is to be implemented (7). The nding of the study will rebound to bene t the society considering that ovarian tumor is one of the major health problems of women in all age groups, as there is no effective screening strategy yet in the world. It will also put the basis for health policy makers and resource allocators to reconsider strengthen further study to decrease mortality associated with ovarian tumor. The nding of the study will also help us to triage patients preoperatively whether to be operated by Gynecologic oncologist or General Gynecologist thus decreasing health costs.
Local data on cancer epidemiology in Ethiopia are lacking (8). Studies from the Global Burden of Disease Cancer Collaboration and the Cancer Incidence in Five Continents Collaboration have estimated cancer incidence by cause for countries globally, and both studies used evidence from neighboring African countries to estimate cancer incidence in Ethiopia (8). In Ethiopia in females 15 years and older, the most common cancer was breast cancer, followed by cervical cancer , ovarian cancer, colorectal cancer, and leukemia (8).Ovarian cancer is 3 rd most common cancer next to breast and cervical cancer among female older than 15yrs old with crude incidence rate(CIR) of 4.9 per 100,000 population and age standardized incidence rate(ASIR) of 8.1 per100,000 population (8). Based on 2013 data from the Addis Ababa Cancer Registry, breast cancer accounted for 31.4%, cervical cancer for 14.3% and ovarian cancer for 6.3% of all cancer cases (9) As the symptoms of the ovarian cancer are very vague like bloating, pelvic or abdominal pain, poor appetite, feeling full quickly, and urinary urgency it is also known as "silent killer". Thus, silent occurrence and slow progression, added to the fact that few effective methods for early diagnosis and no universal screening method for diagnosis of malignant ovarian tumor exists, made its mortality rate highest among gynecologic malignancies (10)(11)(12). Of all gynecologic malignancies ovarian cancer has the worst prognosis since is detected at advanced stage (13).The main challenge is to identify patients with highrisk adnexal masses preoperatively and this is compounded by the lack of de nitive noninvasive diagnostic test (14). The discrimination between benign and malignant adnexal mass is central to decision regarding clinical management and surgical planning in such patients (15).Jacob et.al originally developed a risk of malignancy index based on ultrasound ndings, menopausal status and CA-125 (15).The main advantage of this method compared to other approaches such as color Doppler ultrasound (15) or use of different tumor markers (16,17) is that RMI can be used easily in less specialized unit (18). United states preventive service task force found adequate evidence that screening for ovarian cancer doesn't reduce ovarian cancer mortality and thus recommend against screening for ovarian cancer in asymptomatic women who are not known to have high risk hereditary cancer syndrome (19,20). Prompt identi cation of ovarian malignancies and referral to Gynecologic oncologist can enhance patient survival, but a single method which can accurately predict ovarian malignancies are unavailable (21).RMI is widely studied for prediction of ovarian malignancies in western populations. However, little is known about its implication in developing countries (21)(22)(23). Subsequent studies have shown RMI is reliable tool in differentiating benign and malignant adnexal mass (24,25). Because of this Authors wish to see if RMI score, can be applied in present low resource setting of our population for setting up a better diagnosis, referral and management system

Study area and design
An institutional based Prospective cross-sectional study was employed in Jimma University medical center, Jimma zone, from September 1, 2019 to June 30, 2020. Jimma University teaching Hospital (JUTH) is one of the oldest public hospitals in the country. It was established in 1922 G.C. Geographically, it is located in Jimma city 352 km to southwest of the capital Addis Ababa, Ethiopia. It has been governed under the Ethiopian government by the name of "Ras Desta Damtew Hospital" and later "Jimma Hospital" during Dergue regime and currently Jimma University Specialized Teaching Hospital and recently Jimma Medical center. Currently it is the only teaching and referral hospital in the south western part of the country providing service for approximately 115,000 in patient ,16000 outpaient,11000 emergency case and 4500 deliveries per year.it gives services to about 15 million populations coming from Oromia, Gambella and SNNP (26) .

Study participants
All patients with adnexal mass who visited Gynecologic OPD of Jimma Medical Center and admitted to gynecology ward and operated during study period were included in the study. Patients with previous history of ovarian cancer, abdominal mass other than ovarian mass was excluded. For patients who had undergone hysterectomy age cut off point greater than or equals to 51.3yrs was considered as postmenopausal.

Sample size and sampling technique
Purposive sampling technique was employed. All patients with adnexal mass who were admitted to gynecology ward of Jimma medical center and underwent laparotomy were included in the study. A total of 99 patients with adnexal mass were registered for surgery and provided informed consent to participate in the study.

Data collection methods and tools
Structured questionnaire which was prepared in English by reviewing related relevant literature (5,7,10,15,17,21,25) and translated to regional language Afaan Oromoo by language experts was used to collect data. Data was collected by interviewing the patients and reviewing their charts. The questionnaire has seven parts. The 1 st part assesses sociodemographic characteristics; the 2 nd part assesses menstrual status; the 3 rd part evaluate patients signs and symptoms; the 4 th part has absolute value of CA-125; the 5 th part assess ultrasound ndings of adnexal mass; the 6 th part has RMI score absolute value and the last part has histopathology result of the adnexal mass. USG done by radiologist or Gynecologist were used for USG score. Two year-II Gynecology residents were selected to collect data after getting proper training by principal investigator. At the end of each data collection day the questionnaire was checked for consistency and completeness and close supervision was carried out during data collection by principal investigator

Data analysis
The collected data were entered into epidata version 4.6.0.2 and then exported to SPSS (statistical packages for social sciences) version 26 computer software for analysis. Descriptive statistics and analytic statistics methods were used for analysis. To determine the best cut off value of RMI, Receiver operating characteristic curve (ROC) was plotted and odds ratio with 95% CI was calculated. Best cut off value was chosen according to highest sensitivity and lowest false positive rate. Histopathology result was used as gold standard to differentiate malignant from benign adnexal mass. P-value less than 0.05 considered statistically signi cant. Finally obtained results were presented using tables and gures.

Results
A total of 99 patients were enrolled in this study. The incidence of benign ovarian tumor in all patients presented with adnexal mass and operated for suspected ovarian malignancy at JMC is 61.6% and that of malignant ovarian tumor is 38.4%. From this study the youngest patient was 15 years old and the oldest patient is 73 years old. All benign tumors were in age group 15- and the rest were from urban 37(37.4%). Fifty (50.5%) patients were Muslims in religion 30(30.3%) were Orthodox,17(17.2%) were protestants and 2(2%) were Catholic. Majority of patients were Oromo in ethnicity (58.6%), followed by Amhara (12.1%), Kaffa (12.1%), Waliyita (4%) and others (1%). Seventy-six (76.8%) patients were married,12(12.1%) were single,4(4%) were divorced and 7(7%) were widowed. Majority of patients can't read and write (58.6%),11.1% learned up to grade 8 and 12.1% were in grade 9-10. The mean income of patients was 2809±1928 birr. There was moderate association between income level and histopathology result(eta=0.541)

Discussion
Among all gynecologic malignancies ovarian cancer has the worst prognosis since it is diagnosed at advance stage (10)(11)(12)(13).De nitive diagnosis of ovarian cancer can be made only after laparotomy (13).About 10% women undergo exploratory laparotomy for ovarian tumor during their life time (27).To detect ovarian cancer at early stage several approaches have been tried including, single cut off CA-125,USG score, Doppler USG parameters but none of them found to be effective (11,20,22).Clinical impression and USG examinations are still major preoperative diagnostic tools for adnexal mass. However due to their limitation, Gynecologists are often faced with unexpected nding intraoperatively and has to perform unplanned procedure. RMI is the most widely used method for preoperative discrimination between benign and malignant adnexal mass (12,13).
In this study the incidence of benign ovarian tumor in all patients presented with adnexal mass and operated for suspected ovarian malignancy is 61.6% and that of malignant ovarian tumor is 38.4%.
among malignant cases 14.3% occurs in premenopausal patients and 80.6% cases occur in postmenopausal patients and it is comparable to previous study report (28,29). The mean age for benign tumor is 30±9yrs and the mean age for malignant tumor is 50.6±10.8yrs. The mean age is comparable to previous studies (30) Most studies reported an increased diagnostic accuracy and performance of RMI with cut off 200 (4,5,7,13,17,19,28). Both FA and AK were involved in study proposal commenting, data analysis and write up. All authors read and approved the nal manuscript.

Funding
Data collection process of this study was funded by Jimma University. The funding body only followed the process to con rm whether the allocated fund was used for proposed research.

Availability of data and materials
The data sets used during current study are available from corresponding author on reasonable request at email: Bezzakedida2@gmail.com Ethics approval and consent to participate Ethical clearance was obtained from institutional review board (IRB) of Jimma University institute of health with reference number of IRB000218/20.Permission letter was obtained from department of Gynecology and Obstetrics and submitted to Jimma University medical center. Informed consent was taken from each study participants before data collection. All information obtained from study participants was kept with con dentiality. Participants were also told they have right to withdraw from the study at any time. All experimental protocol of this study was done in accordance with declaration of Helsinki and approved by Institutional review board (IRB) of Jimma University.

Consent for Publication
Not applicable.

Competing interests
The authors declare that there is no competing interest Comparison of ROC curves of RMI and CA-125 for discriminating between benign and malignant adnexal mass in JMC,2020