Accuracy of preoperative prediction of malignancy in ovarian mass by ultrasound examination and CA 125 serum level

Background: Ovarian cancer is the second most common malignancy of the female reproductive system and one of the leading lethal gynecologic malignancies. Screening of ovarian cancer in certain high risk groups is very important due to unspecificity and late appearance of symptoms. Its risk factors include positive family history, older age of menopause and low parity as pregnancy protects against ovarian cancer. Objectives: to compare the accuracy of preoperative prediction of malignancy in ovarian mass by morphological ultrasound (US) examination, Doppler indices and CA 125 serum level with the result of histopathological examination mass after laparotomy. Methods: One hundred and four cases of ovarian masses predicted to be malignant by US examination and CA 125 serum level were subjected to laparotomy and histopathological examination. The main outcome measures in the ovarian masses were: athe US signs of malignancy [ such as solid mass, multiple septation in cystic mass, mixed solid and cystic components, thick cyst wall (> 3 mm), nodule in a cyst wall ] , bDoppler indices(resistance index and pulsatility index) , cCA125 serum level, and dhistopathological examination findings after laparotomy. Results: The histopathology identified 20 benign (B) and 84 malignant (M) ovarian masses. The benign tumors were 9(45%) endometroitic cyst, 6(30%) pseudomucinous cyst adenoma and 5(25%) serous cyst adenoma. The malignant ones included 43(51.2%) papillary serous cyst adenocarcinoma 18(21.4%) endometrioid adenocarcinoma , 10(11.9%) pseudomucinous cyst adenocarcinoma, 5(5.9%) clear cell adenocarcinoma, 2(2.4%) papillary serous borderline cyst adenocarcinoma, 2(2.4%) borderline serous adenocarcinoma, 1(1.2%) serous adenocarcinoma, 1(1.2%) borderline endometroid adenocarcinoma, 1(1.2%) dysgerminoma and 1(1.2%) Pseudomucinous borderline cyst adenocarcinoma]. The US showed no morphological signs of malignancy in 10 [9.6% (9 M vs. 1 B)] masses, thick cyst wall and mixed solid & cystic components 1(1%) M; thick cyst wall1 and nodule in the cyst wall 1(1%) M, mixed solid and cystic components 15[14.4% (14 M vs. 1 B)], solid components 17(16.3%) M, thick cyst wall (> 3 mm) 27[26% (10 M vs. 17 B)] and nodules in the cyst wall in 33[31.7% (32 M vs. 1 B)] masses. Doppler studies of ovarian mass vasculature showed that < 0.4 resistance index and < 1 pulsatility index prevailed significantly in 83 and 82 malignant masses respectively (P< 0.001) while CA125 serum cutoff level 30 IU/ ml alone failed to differentiate between the benign and malignant masses Conclusion: using CA125 serum cutoff level 30 IU/ ml combined with US grey scale or color Doppler examination can discriminate between benign and malignant adnexal masses especially in positive Doppler indices. INTRODUCTION Ovarian cancer (OC) is the second most common malignancy of the female reproductive system and one of the leading causes of death among gynaecologic malignancies.(1) The disease is more common in industrialized nations, with the exception of Japan. In the United States, females have 1 % to 2.5% (1 out of 40-60 women) lifetime 13 Bull High Inst Public Health Vol.42 No.1 [2012] chance of developing OC. Older women are at highest risk. More than half of the deaths from OC occur in women between 55 and 74 years of age and approximately one quarter of OC deaths occur in women between 35 and 54 years of age.(2) There are no statistics that describe disease incidence in Egypt. Signs and symptoms of OC are frequently absent early and when they exist they may be subtle. In most cases, the symptoms persist for several months before being recognized and diagnosed.(3) The five-year survival rate for all stages of OC is 47%.(4) For cases where a diagnosis is made early in the disease, when the cancer is still confined to the primary site, the five-year survival rate is 92.7 %.(5) So prognosis is good for women diagnosed at an early stage, whereas the majority, diagnosed at later stages, is likely to survive less than 5 years.(6) Symptoms as bloating, fullness, and pressure in the abdomen are the most prominent symptoms. Pain and fatigue are also important, followed by problems in urination and constipation.(7) Ovarian cancer is neither an asymptomatic disease nor a socalled ‘silent killer’. Recent studies have demonstrated that patients at all stages of the disease have symptoms. Examination can reveal abdominal or pelvi-abdominal mass only in the late stages and bimanual pelvic examination can reveal adenexal mass or fullness.(9,20) Studies exploring the value of screening those women for OC are lacking and urgently required. Even though population-based screening for OC is not recommended, and although there is no level of evidence that this group of women should undergo screening, it seems prudent that, until evidence is available, measurement of CA 125 levels and transvaginal ultrasound should be undertaken at least on yearly basis.(21) CA125 is still the most extensively studied biomarker for possible use in the early detection of OC, and has proved valuable in both detection and disease monitoring.(23,24) CA125 is elevated in the serum of most Hossam Hassan Aly Hassan El Sokkary 14 women with OC, but pre-operative serum levels of CA125 are below the conventional cutoff level of 35 U/ml in roughly 50% of clinically detected stage I cases (25) and in the majority of women with occult cancers identified at prophylactic surgery.(26) Using vaginal ultrasound examination can add to the predictive value of CA125. US morphological signs of malignancy include large ovarian volume more than 18 ml before menopause and 8 ml after, thick cyst wall, solid component in ovarian mass, mixed solid and cystic component, nodule in the cyst wall and abnormal vascular pattern proved by Doppler study (27,28,29). Objectives: To evaluate the accuracy of preoperative prediction of malignancy in ovarian mass by the morphological ultrasound examination, Doppler indices and CA 125 serum level. Methods: Following approval by Alexandria Faculty of Medicine Institutional Ethics Committee, 104 patients with ovarian masses that fulfilled the inclusion criteria attending the outpatient clinic of Oncology Department of El-Shatby University Hospital were included in the study after taking their consents. The study was the "One-shot prospective case study" without control group. The main inclusion criteria were ovarian mass with one or more of the followings aUltrasonographic (US) signs of malignancy [ such as solid component, mixed solid and cystic component, nodule in the cyst wall, thick cyst wall (more than 3mm) ], bDoppler studies of ovarian mass vessels (including resistance index (RI) and pulsatilty index (PI) with a cut level values of less than 0.4 for RI and less than 1 for PI.(31,32) For enrolment in the study there should be: 1at least one positive US sign whether morphological appearance or Doppler indices. 2This positive US sign must be combined with CA 125 serum level more than 30 u/ml.(30) Patients with these criteria were admitted and subjected to laparotomy and histopathological examination of the ovarian masses. The women were 15 Bull High Inst Public Health Vol.42 No.1 [2012] examined with both real-time 3.5-5 MHz transabdominal transducer and 5.5-7 MHZ vaginal transducer. Laparatomy was done to all cases at ElShatby University Hospital and histopathological examination was done to all ovarian masses at the Clinical Pathology Department of the Main University Hospital. Table (1) Sample size of one group according to disease prevalence Formula n= t2 x p(1-pr) m2 Description: n = required sample size t = confidence level at 95% (standard value of 1.96) p = estimated prevalence of ovarian carcinoma (estimated as 1 % ) pr = probability(0.4) m = margin of error at 5% (standard value of 0.5) Calculation of sample size (N): n= 1.962 x 10.0(1-.4) .52 n = 3.8416 x 6 .25 n = 23.0496 .25 n = 92.1 92 This table showed that the sample size should be more than 92 cases. RESULTS Histopathology reports of the ovarian masses showed that 43 masses were papillary serous cystadenocarcinoma, 2 papillary serous borderlines cystadenocarcinoma, 1 serous adenocarcinoma, 2 borderlines serous adenocarcinoma, 18 endometrioid adenocarcinoma, 1 borderline endometroid adenocarcinoma, 10 pseudomucinous cyst adenocarcinoma, 1 pseudomucinous borderline cyst adenocarcinoma, 1 dysgerminoma, 5 clear cell adenocarcinoma, 5 serous, 6 pseudomucinous cyst adenoma, and 9 endometriotic cysts. Malignant masses were 84 and benign ones were 20. Regarding the ultrasonographic signs, 10 cases showed any morphological sign of malignancy, 17 showed solid components, 15 Hossam Hassan Aly Hassan El Sokkary 16 mixed solid and cystic components, 33 nodule in the cyst wall, 27 thick cyst wall (> 3 mm), 1 thick cyst wall and mixed solid and cystic component and 1 showed thick cyst wall and nodule in the cyst wall. Doppler studies of tumor vasculature showed that the resistance index was less than 0.4 in 83 cases and the pulsatility index was less than 1 in 82 cases. Table (2): Distribution of the studied ovarian masses according to the histopathological diagnosis. Diagnosis Ovarian mass Total Malignant Benign Borderline endometroid adenocarcinoma no. 1 0 1 % 1.2% .0% 1.0% Clear cell adenocarcinoma no. 5 0 5 % 5.9% .0% 4.8% Dysgerminoma no. 1 0 1 % 1.2% .0% 1.0% Endometriod adenocarcinoma no. 18 0 18 % 21.4% .0% 17.3% Endometroitic cyst no. 0 9 9 % .0% 45% 8.7% Pseudomucinous border line cyst adenocarcinoma no. 1 0 1 % 1.2% .0% 1.0% Pseudomucinous cyst adenocarcinoma no. 10 0 10 % 11.9% .0% 9.6% Papillary serous border line cyst adenocarcinoma no. 2 0 2 % 2.4% .0% 2.0% Papillary serous cyst adenocarcinoma no. 43 0 43 % 51.2% .0% 41.3% Pseudomucinous cyst adenoma no. 0 6 6 % .0% 30% 5.7% Serous cyst adenocarcinoma no. 1 0 1 % 1.2% .0% 1.0% Serous cyst adenocarcinoma border line no. 2 0 2 % 2.4% .0% 1.9% Serous cyst adenoma no. 0 5 5 % .0% 25.0


INTRODUCTION
Ovarian cancer (OC) is the second most common malignancy of the female reproductive system and one of the leading causes of death among gynaecologic malignancies. (1)The disease is more common in industrialized nations, with the exception of Japan.In the United States, females have 1 % to 2.5% (1 out of 40-60 women) lifetime Bull High Inst Public Health Vol.42 No. 1 [2012]   chance of developing OC.Older women are at highest risk.More than half of the deaths from OC occur in women between 55 and 74 years of age and approximately one quarter of OC deaths occur in women between 35 and 54 years of age. (2)There are no statistics that describe disease incidence in Egypt.Signs and symptoms of OC are frequently absent early and when they exist they may be subtle.
In most cases, the symptoms persist for several months before being recognized and diagnosed. (3)The five-year survival rate for all stages of OC is 47%. (4)For cases where a diagnosis is made early in the disease, when the cancer is still confined to the primary site, the five-year survival rate is 92.7 %. (5) So prognosis is good for women diagnosed at an early stage, whereas the majority, diagnosed at later stages, is likely to survive less than 5 years. (6)mptoms as bloating, fullness, and pressure in the abdomen are the most prominent symptoms.Pain and fatigue are also important, followed by problems in urination and constipation. (7)Ovarian cancer is neither an asymptomatic disease nor a socalled 'silent killer'.Recent studies have demonstrated that patients at all stages of the disease have symptoms. (8,18,19)Examination can reveal abdominal or pelvi-abdominal mass only in the late stages and bimanual pelvic examination can reveal adenexal mass or fullness. (9,20)udies exploring the value of screening those women for OC are lacking and urgently required.Even though population-based screening for OC is not recommended, and although there is no level of evidence that this group of women should undergo screening, it seems prudent that, until evidence is available, measurement of CA 125 levels and transvaginal ultrasound should be undertaken at least on yearly basis. (21)125 is still the most extensively studied biomarker for possible use in the early detection of OC, and has proved valuable in both detection and disease monitoring. (23,24)omen with OC, but pre-operative serum levels of CA125 are below the conventional cutoff level of 35 U/ml in roughly 50% of clinically detected stage I cases (25) and in the majority of women with occult cancers identified at prophylactic surgery.(26) Using vaginal ultrasound examination can add to the predictive value of CA125.US morphological signs of malignancy include large ovarian volume more than 18 ml before menopause and 8 ml after, thick cyst wall, solid component in ovarian mass, mixed solid and cystic component, nodule in the cyst wall and abnormal vascular pattern proved by Doppler study (27,28,29) .

Objectives:
To evaluate the accuracy of preoperative prediction of malignancy in ovarian mass by  than 1 for PI. (31,32)For enrolment in the study there should be: 1-at least one positive US sign whether morphological appearance or Doppler indices.2-This positive US sign must be combined with CA 125 serum level more than 30 u/ml. (30)Patients with these

ROC Curve
Di ago na l se gm e nts are prod uced by ti es.
1 -Specificity  (33) Differentiating benign from malignant tumors might be achieved by several methods such as clinical signs and symptoms, serum CA 125 and ultrasound. (34)netheless, using one item alone to differentiate between benign and malignant cases shows low positive predictive value.
For example, in predicting malignancy in ovarian tumors, abdominal ultrasonography had a positive predictive value of 39% and a negative predictive value of 94%.If a negative sonogram had been relied upon, 6% of malignant ovarian tumors in postmenopausal women might have been missed. (35)so, serum levels of CA125 have cancer but result in many false positives in patients with benign conditions. (36) the study of van Nagell et al, the transvaginal grey scale US had a sensitivity of 85.0%, specificity 98.7%, a positive predictive value of 14.01%, and a negative predictive value of 99.9%. (37)Tailor et al using CA125 serum level and morphological vaginal ultrasonographic examination showed that sensitivity of ultrasound screening was 92% and the specificity was 97.8%. (38)Varras concluded that the combination of physical examination with serum CA-125 levels and pelvic ultrasound scan seemed to improve the sensitivity and specificity of predicting the adnexal malignancies in postmenopausal women.In contrast, in premenopausal women, the consideration of CA-125 levels with Doppler ultrasonographic findings might confuse the differential diagnosis of ovarian masses. (39) the current study, we attempted to use the morphological ultrasound examination, Doppler indices and CA 125 serum level.
Ethics Committee, 104 patients with ovarian masses that fulfilled the inclusion criteria attending the outpatient clinic of Oncology Department of El-Shatby University Hospital were included in the study after taking their consents.The study was the "One-shot prospective case study" without control group.The main inclusion criteria were ovarian mass with one or more of the followings a-Ultrasonographic (US) signs of malignancy [ such as solid component, mixed solid and cystic component, nodule in the cyst wall, thick cyst wall (more than 3mm) ], b-Doppler studies of ovarian mass vessels (including resistance index (RI) and pulsatilty index (PI) with a cut level values of less than 0.4 for RI and less criteria were admitted and subjected to laparotomy and histopathological examination of the ovarian masses.The women were examined with both real-time 3.5-5 MHz transabdominal transducer and 5.5-7 MHZ vaginal transducer.Laparatomy was done to all cases at El-Shatby University Hospital and histopathological examination was done to all ovarian masses at the Clinical Pathology Department of the Main University Hospital.
been used widely for distinguishing benign from malignant pelvic masses.However, CA125 is elevated in only about half of stage I/II ovarian cancer patients.Lowering the cutoff of CA125 less than 30 IU/ML would increase its sensitivity in detecting Bull High Inst Public Health Vol.42 No.1 [2012] the combination of the ultrasonographic morphological appearance, Doppler indices and CA125 serum level to differentiate between benign and malignant ovarian masses .Our results revealed that using the ultrasound grey scale examination together with Doppler indices of mass vascularity (at least one of them is positive) combined with CA125 serum cutoff level more than 30 IU/ML succeeded significantly to differentiate between the benign (n=20) and malignant(n=84) ovarian masses (p=0.0001).Using the ultrasound morphological picture alone failed to differentiate between benign and malignant ovarian masses (p=0.58).The malignant morphological ultrasound signs have been seen in 75 out of 84 malignant patients and all benign (20) cases except for one.Solid component ultrasonographic sign was the most accurate sign in the differentiation between benign and malignant ovarian masses as it was present in 17 malignant cases while it was seen in either benign case.As regards the Doppler indices, both pulsatility index and resistance index were sensitive and specific in the differentiation between both benign and malignant groups In the same context, CA125 serum cutoff level more than 30 IU / ml level was not significantly either sensitive or specific enough to discriminate between the two groups.CONCLUSION From the current study, it is concluded that using CA125 with serum cutoff level > 30 IU/ml combined with ultrasonographic grey scale or color Doppler examination can effectively discriminate between benign and malignant adnexal masses especially with positive Doppler indices.