Ovarian cancer is the second diagnosed gynecologic cancer in the USA; it is also a fatal condition because over 70% of patients are diagnosed with progressive-stage disease. In progressive-stage disease, the cure rates are only 20–30% 1. Current estimates authenticate that 1.4% (one in 72) of girls born today will be diagnosed with ovarian malignancy at some point in their lifetime 2. Nevertheless, discovery of the disease in the early stage has a perfect prognosis if treated. Given the impediment of treatment for advanced ovarian malignancy and the success of management for the early-stage disease, a screening test is ostensibly appealing 3. A lot of markers have been assigned to the management of ovarian malignancy in several different maneuvers. These include checking response to treatment, differentiating malignant from benign ovarian masses, judging prognosis, anticipating response to individual drugs, and detecting primary disease at an early stage. For epithelial ovarian cancer, carbohydrate antigen 125 (CA125) has received the most consideration to date among these markers 4. CA125 has very low sensitivity in distinguishing patients with early-stage malignancy. Thus, the need for a more sensitive and specific marker for ovarian cancer diagnosis, either alone or in combination with other markers, is demanding 5. The application of new technology to ovarian malignancy has resulted in the labeling of many genes that are overexpressed in the ovarian tissues 6. The gene for human epididymis protein 4 (HE4) is one of the most commonly overexpressed genes in ovarian malignancy compared with the normal tissues. HE4 is one of the most important novel markers found to be elevated in women with ovarian cancer. HE4 and CA125 provided the highest level of discrimination between benign and malignant ovarian tumors, and the combination of these two biomarkers dispensed a greater level of discriminatory potential than either marker alone 7. In a recent study, it was found that CA125 and HE4 may dispense evidence of ovarian malignancy 3 years before its clinical commencement, but the anticipated lead time affiliated with these markers seems to be less than 1 year. This means that the prognostic value of these two markers is still bounded but showed increasing precision with time approaching diagnosis 7. The aim of the current study was to evaluate the diagnostic value of serum HE4 in patients with ovarian masses in comparison with serum CA125.
Patients and methods
A diagnostic accuracy study was conducted in gynecological oncology unit at Ain-Shams University Maternity Hospital during the period from January 2010 to February 2012, after being approved by the local hospital ethics and research committee. A total of 110 participants were recruited from those patients attending the outpatient gynecological oncology clinic at Ain-Shams University Maternity Hospital. The study purpose and procedures were explained to all enrolled women and written informed consent was obtained from each participant. Patients aged 16–65 years were consecutively recruited on the basis of radiologic and clinical findings suspected for ovarian tumors. Patients who were candidates for exploratory laparotomy and for histopathological diagnosis were only included. The patients’ final diagnoses were those of the histopathological conclusions. All patients were subjected to full history-taking and thorough clinical examination and investigations. Routine pelvi-abdominal ultrasound was performed to assess the mass site, size, shape, ascitis, omental cake, and secondaries. Pelvi-abdominal computed tomography scan and MRI were performed to confirm the diagnosis when relevant and to evaluate any distant secondaries when malignancy is expected. Patients with renal impairment, which affects the level of HE4, were excluded from the study. None of the patient had been subjected to any preoperative chemotherapy. According to the histopathological finding, patients were categorized into two groups: group 1 included patients diagnosed as having benign masses and group 2 included patients diagnosed as having malignant masses. All malignant patients were properly staged according to the International Federation of Gynecology and Obstetrics criteria. The WHO criteria were used for histological grading of malignant patients 8.
Blood samples from patients were collected before surgical intervention. Within 4 h of collection, sera were obtained and frozen at −80°C until the time of HE4 and CA125 assays.
Serum HE4 and CA125 assays
Serum HE4 was measured using HE4 enzyme immunometric assay (Fujirebio Diagnostics Inc., Tokyo, Japan). HE4 assay is a solid-phase immunoassay that rests upon the direct sandwich technique using biotinylated anti-HE4 monoclonal antibody as described before, measuring range is 15–900 pmol/l 9. CA125 was assessed using Cobas e 411 analyzer (Roche, Tokyo, Japan) by electrochemiluminesence immunoassay based on sandwich principle utilizing two monoclonal antibodies as described before, measuring range is 0.6–5000 U/ml 9.
Sample size justification
Assuming an average level of HE4 of 45.5 pmol/l in patients with benign ovarian masses and an SD of 13.4 compared with a level of 1125.4 pmol/l in patients with malignant ovarian tumors and an SD of 2670.0, a sample size of 48 patients in each group is enough to detect such difference at 0.05 errors and 0.80 power of the test 10.
Statistical analysis was performed using statistical package for social sciences (SPSS) for Windows version 15.0 (SPSS Inc., Chicago, Illinois, USA). Normally distributed numerical data were presented as mean and SD and between-group differences were compared using unpaired Student’s t-test. Non-normally distributed numerical data were presented as median and interquartile range (IQR) and differences between the two groups were compared nonparametrically using the Mann–Whitney U-test. Categorical data were presented as number and percentage and intergroup differences were compared using the Pearson χ2-test (for nominal data) or the χ2-test for trends (for ordinal data). Associations between measured variables and ovarian malignancy were estimated using the receiver operator characteristic (ROC) curves. Validity of study parameters was evaluated in terms of sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Significance level was set at 0.05 (Fig. 1), participants' flow chart is shown in Table 1, after the statement hence, data of 96 patients were finally analyzed.
In all, 160 patients presented with pelvi-abdominal masses were assessed for eligibility. Fifty women were excluded, as they declined to participate (n=4) or they did not meet the inclusion criteria (n=46). Fourteen patients from the remaining 110 patients were also excluded, as their pelvi-abdominal masses were not ovarian in origin. Hence, data of 96 patients were finally analyzed. The mean age of included women was 42.39±12.84 years (range: 20–75 years). Of the included 96 patients, 62 (64.6%) were premenopausal, whereas 34 (35.4%) were postmenopausal. The median parity was 3 (range: 0–15; IQR: 2–4). The mean duration of marriage was 20.46±11.47 years (range: 2–44 years). The mean weight was 77.33±13.8 kg (range: 54–105 kg). The mean BMI was 28.29±4.47 kg/m2 (range: 21.09–37.34 kg/m2). Of the included women, six (6.3%) had chronic hypertension, five (5.2%) had diabetes mellitus, eight (8.3%) had combined hypertension and diabetes mellitus, and one (1%) had ischemic heart disease, whereas one (1%) had chronic liver disease. With respect to family history, two patients (2.1%) had a family history of breast cancer. The presenting complaint was abdominal pain/discomfort in 62 (46.6%) women, abnormal genital bleeding in 11 (11.5%) women, gastrointestinal (GI) symptoms in 10 (10.4%) women, abdominal distension in four (4.2%) women, and abdominal enlargement in two (2.1) women. Of the included 96 women, seven (7.3%) women were asymptomatic, and the mass was accidentally discovered during abdominal scan. The largest dimension of the mass in included women was 10.83±5.65 cm (range: 5–25 cm). The mass was bilateral in 24 (25%) women and unilateral in 72 (75%) women; of them, the mass was right-sided in 43 (44.8%) women and left-sided in 29 (30.2%) women. The mass was heterogeneous in 33 (34.4%) women and cystic in 63 (65.6%) women; of them, the cyst was unilocular in 49 (51%) women and multilocular in 14 (14.6%) women. Ascites was present in 14 (14.6%) women. The median serum level of CA125 was 58.06 mIU/ml (range: 3.21–3654; IQR: 25.06–243.53). The median serum level of HE4 was 186.2 pmol/l (range: 49–987; IQR: 112.2–502.6). Of the included 96 women, 58 (60.4%) had a benign ovarian lesion, whereas 38 (49.6%) had a malignant one. Histopathological results of the ovarian mass in included women are shown in Table 1. Table 2 shows the staging and grading of tumor in women with malignant ovarian mass. Differences between women with benign and malignant ovarian masses with respect to initial characteristics are shown in Table 3.
There was no significant difference between women who had benign ovarian lesions and those who had malignant ovarian lesions with respect to the past medical history. There were two (5.3%) patients with a family history of breast cancer among women with malignant ovarian lesions in contrast to none among women with benign ovarian lesions; this difference was not significant. There were no significant differences between women with benign ovarian lesions and those with malignant lesions with respect to the presenting complaint except for the GI symptoms, which were exclusively present in women with malignant ovarian lesions [10 (26.3%) vs. 0 (0%), P<0.001]. The median serum levels of CA125 were significantly higher in women with malignant ovarian lesions when compared with women with benign ovarian lesions [335.5 mIU/ml (range: 60–1127 mIU/ml) vs. 33.65 mIU/ml (range: 10.36–174 mIU/ml), respectively, P<0.001]. The median serum HE4 was significantly higher in women with malignant ovarian lesions when compared with women with benign ovarian lesions [558.3 pmol/l (range: 69.6–987 pmol/l) vs. 172 pmol/l (range: 49–410.3 pmol/l), respectively, P<0.001]. Table 4 shows the accuracy of measured serum tumor markers as predictors of ovarian malignancy in included women. According to the best cutoff values retrieved from the ROC curves, serum CA125 was the most sensitive, whereas serum HE4 was the most specific marker. Serum HE4 had a comparable sensitivity but was more specific than serum CA125. A serum CA125 concentration above the universally accepted cutoff value (35 mIU/ml) was associated with ovarian malignancy at a sensitivity of 94.7%, specificity of 56.9%, PPV of 59%, NPV of 94.3%, positive likelihood ratio (LR+) of 2.2, and a negative likelihood ratio (LR−) of 0.09. When serum CA125 cutoff level was raised to 58.1 mIU/ml, the test was more sensitive and more specific (sensitivity 94.7%, specificity 79.3%, PPV 75%, NPV 95.8%, LR+ 4.6, LR− 0.07). A serum HE4 concentration of at least 227 pmol/l was associated with ovarian malignancy at a sensitivity of 76.7%, specificity of 86.7%, PPV of 79.3%, NPV of 84.8%, LR+ of 5.8, and LR− of 0.3, being the most specific parameter through having the highest specificity, PPV, and LR+ (Table 5). A combination of CA125 and serum HE4 was associated with ovarian malignancy at a sensitivity of 76.7%, specificity of 100%, PPV of 95.8%, NPV of 86.3, LR+ of 34.5, and LR− of 0.2.
In the current study, women with malignant ovarian masses had a significantly higher mean age when compared with women with benign ovarian lesions. Ovarian cancer can occur at any age but Myers et al.11 mentioned that cancer incidence increases dramatically with age, being relatively rare before the age of 50 years. According to data reported by the Surveillance, Epidemiology, and End Results program, the median age at ovarian cancer diagnosis was 63 years, and ovarian cancer was diagnosed in 68.6% of patients after the age of 55 years 2. In our study, 64.6% of patients were premenopausal, whereas 35.4% were postmenopausal. The proportion of malignancy among postmenopausal women was significantly higher than that among premenopausal women (57.9%).
Givens et al.12 stated that, in postmenopausal women, 30% of adnexal masses are malignant, which is almost similar to our findings. The current study authenticated that there were no significant differences between patients with benign and malignant masses with respect to the presenting symptoms except for the GI symptoms, which were entirely present in patients with malignant ovarian masses (26.3%). A survey on 1725 patients with ovarian cancer from the USA and Canada was published, calculating the frequency of the presenting symptoms in patients with ovarian cancer. The most frequent symptoms were abdominal (77%), GI (70%), pain (58%), constitutional (50%), urinary (34%), and pelvic (26%) 13.
In this study, the median serum levels of CA125 and HE4 were significantly higher in women with malignant ovarian masses when compared with women with benign ovarian lesions. In our population, serum CA125 at a universally accepted cutoff value of 35 mIU/ml was associated with ovarian malignancy at a sensitivity of 94.7%, specificity of 56.9%, PPV of 59%, NPV of 94.3%, LR+ of 2.2, and LR− of 0.09. When serum CA125 cutoff level was raised to 58.1 mIU/ml, the test had the same sensitivity (94.7%) but more specificity (79.3%). We found that the median serum HE4 was significantly higher in women with malignant ovarian lesions when compared with women with benign ovarian lesions. Serum CA125 was found to be the most sensitive, whereas serum HE4 was the most specific marker. Serum HE4 had less sensitivity but was more specific than serum CA125. Combination of CA125 and HE4 was associated with ovarian malignancy at a sensitivity of 76.7%, specificity of 100%, PPV of 95.8%, NPV of 86.3%, LR+ of 34.5, and LR− of 0.2. HE4 may have a promising future in the diagnosis of ovarian malignancy. To improve its diagnostic accuracy, it is better to combine it with CA125 and other markers 14. Furthermore, Zheng et al.15 stated that evaluation of serum HE4 is superior to CA125 assay to differentiate epithelial ovarian cancer from pelvic inflammatory disease and ovarian endometriosis. In addition, in premenopausal patients with ovarian masses, the serum level of HE4 adds beneficial data to CA125 for differentiating patients with ovarian cancer from those with benign pelvic disease. Ninety-nine patients with different types of gynecological cancer (46 ovarian, 39 endometrial, and 14 cervical) and 40 patients with benign disease (22 endometriosis and 18 benign ovarian mass) together with 12 control participants were evaluated in a previous study. The results of this previous study authenticated that HE4 had a significantly higher area under the curve when compared with CA125 (0.99 vs. 0.91), with a sensitivity of 98% and specificity of 100%, respectively 16. Molina et al.17 found that there were statistically significant differences with respect to the serum levels of CA125 and HE4 (P=0.005) in patients with ovarian cancer compared with those with benign diseases. Both tumor markers expressed similar sensitivity in ovarian malignancy, slightly higher with CA125 (82.9%) than with HE4 (79.3%). Similarly, serum levels of HE4 were significantly higher in patients with ovarian malignancy than in patients with other gynecological malignancies. Kadija et al. 18 concluded that HE4 showed satisfactory competence of distinguishing endometriosis from ovarian cancer, which CA125 missed. Other investigators compared HE4 with CA125 individually, in combination, within the risk of malignancy algorithm and the risk of malignancy index. They chose a cohort of 160 patients that consisted of healthy patients and patients with benign diseases, and borderline tumors/adenocarcinomas of the endometrial, tubal, peritoneal, and ovarian origin were evaluated using HE4 and CA125. The major superiority of HE4 lies in its specificity and increased detection of borderline tumors and early-stage ovarian and tubal malignancies. HE4 is better than CA125 with or without risk of malignancy index and risk of malignancy algorithm indices. However, they concluded that there was no advantage from combining both markers in clinical practice 19.
Furthermore, a systematic review and meta-analysis conducted by Wu et al.20 authenticated that HE4 may be a useful marker in the diagnosis of ovarian malignancy. Serum HE4 detection is not only a valuable preoperative marker for anticipating the benign or malignant nature of ovarian masses, but also has the ability to be used as an initial step in ovarian cancer screening approach. In this study, ROC curves were constructed for measured markers as predictors of early stages of ovarian malignancy (stages I/II). Serum HE4 was the only significant predictor of early stages of ovarian malignancy (area under the curve=0.732). Serum HE4 of 533 pmol/l or less was associated with early stages of ovarian malignancy (stages I/II) at a sensitivity of 68.8%, specificity of 64.3%, PPV of 68.8%, NPV of 64.3%, LR+ of 1.9, and LR− of 0.5. Nolen et al.21 concluded that, in premenopausal patients, CA125 dispensed the highest level of discrimination of benign from malignant patients amid the early-stage tumors, whereas HE4 performed the best in the late-stage disease category. Havrilesky et al.22 appraised several biomarkers, including HE4 and CA125, in133 patients with stage I/II ovarian carcinoma (of whom 35 were of the serous subtype), 67 patients with stage III, and 396 healthy, age-matched controls. Using the ROC curve analysis, HE4 exhibited the highest sensitivity amid all single markers for the discovery of both early and late state ovarian cancer. Heterogeneity of data between our study and other studies and even between other studies are first because of the use of different kits from different companies and second because of the use of a different cutoff values in each study.
In ovarian cancer, HE4 seems to have higher specificity than CA125. Combination of both markers was a more accurate predictor of ovarian cancer than either marker alone.
Conflicts of interest
There are no conflicts of interest.
1. Jemal A, Tiwari RC, Murray T, Ghafoor A, Samuels A, Ward E, et al..Cancer statistics, 2004.CA Cancer J Clin2004;54:8–29.
2. Ries LAG, Harkins D, Krapcho M, Mariotto A, Miller BA, Feuer EJ, et al..SEER cancer statistics review, 1975–20032006.Bethesda, MD:National Cancer InstituteAvailable at: http://seer.cancer.gov/csr/1975_2003/
, based on November 2005 SEER data submission, posted to the SEER web site.
3. Schink JC.Current initial therapy of stage III and IV ovarian cancer
: challenges for managed care.Semin Oncol1999;26Suppl2–7.
4. Bast RC Jr, Badgwell D, Lu Z, Marquez R, Rosen D, Liu J, et al..New tumor markers: CA125 and beyond.Int J Gynecol Cancer2005;15Suppl 3274–281.
5. Moore RG, McMeekin DS, Brown AK, DiSilvestro P, Miller MC, Allard WJ, et al..A novel multiple marker bioassay utilizing HE4
and CA125 for the prediction of ovarian cancer
in patients with a pelvic mass.Gynecol Oncol2009;112:40–46.
6. Drapkin R, Von Horsten HH, Lin Y, Mok SC, Crum CP, Welch WR, Hecht JL.Human epididymis protein 4 (HE4
) is a secreted glycoprotein that is overexpressed by serous and endometrioid ovarian carcinomas.Cancer Res2005;65:2162–2169.
7. Andersen MR, Goff BA, Lowe KA, Scholler N, Bergan L, Drescher CW, et al..Use of a Symptom Index, CA125, and HE4
to predict ovarian cancer
8. El Sherbini MA, Sallam MM, Shaban EA, El-Shalakany AH.Diagnostic value of serum kallikrein-related peptidases 6 and 10 versus CA125 in ovarian cancer
.Int J Gynecol Cancer2011;21:625–632.
9. Abdel-Azeez HA, Labib HA, Sharaf SM, Refaie AN.HE4
and mesothelin: novel biomarkers of ovarian carcinoma in patients with pelvic masses.Asian Pac J Cancer Prev2010;11:111–116.
10. Huhtinen K, Suvitie P, Hiissa J, Junnila J, Huvila J, Kujari H, et al..Serum HE4
concentration differentiates malignant ovarian tumours from ovarian endometriotic cysts.Br J Cancer2009;100:1315–1319.
11. Myers ER, Bastian LA, Havrilesky LJ, Kulasingam SL, Terplan MS, Cline KE, et al..Management of adnexal mass.Evid Rep Technol Assess (Full Rep)2006;130:1–145.
12. Givens V, Mitchell G, Harraway-Smith C, Reddy A, Maness DL.Diagnosis and management of adnexal masses.Am Fam Physician2009;80:815–820821–822.
13. Goff BA, Mandel LS, Drescher CW, Urban N, Gough S, Schurman KM, et al..Development of an ovarian cancer
symptom index: possibilities for earlier detection.Cancer2007;109:221–227.
14. Yu S, Yang HJ, Xie SQ, Bao YX.Diagnostic value of HE4
for ovarian cancer
: a meta-analysis.Clin Chem Lab Med2012;50:1439–1446.
15. Zheng H, Gao Y.Serum HE4
as a useful biomarker in discriminating ovarian cancer
from benign pelvic disease.Int J Gynecol Cancer2012;22:1000–1005.
16. Montagnana M, Danese E, Giudici S, Franchi M, Guidi GC, Plebani M, Lippi G.HE4
in ovarian cancer
: from discovery to clinical application.Adv Clin Chem2011;55:1–20.
17. Molina R, Escudero JM, Augé JM, Filella X, Foj L, Torné A, et al..HE4
a novel tumour marker for ovarian cancer
: comparison with CA 125 and ROMA algorithm in patients with gynaecological diseases.Tumour Biol2011;32:1087–1095.
18. Kadija S, Stefanovic A, Jeremic K, Radojevic MM, Nikolic L, Markovic I, Atanackovic J.The utility of human epididymal protein 4, cancer antigen 125, and risk for malignancy algorithm in ovarian cancer
and endometriosis.Int J Gynecol Cancer2012;22:238–244.
19. Jacob F, Meier M, Caduff R, Goldstein D, Pochechueva T, Hacker N, et al..No benefit from combining HE4
and CA125 as ovarian tumor markers in a clinical setting.Gynecol Oncol2011;121:487–491.
20. Wu L, Dai ZY, Qian YH, Shi Y, Liu FJ, Yang C.Diagnostic value of serum human epididymis protein 4 (HE4
) in ovarian carcinoma: a systematic review and meta-analysis.Int J Gynecol Cancer2012;22:1106–1112.
21. Nolen B, Velikokhatnaya L, Marrangoni A, De Geest K, Lomakin A, Bast RC Jr, Lokshin A.Serum biomarker panels for the discrimination of benign from malignant cases in patients with an adnexal mass.Gynecol Oncol2010;117:440–445.
22. Havrilesky LJ, Whitehead CM, Rubatt JM, Cheek RL, Groelke J, He Q, et al..Evaluation of biomarker panels for early stage ovarian cancer
detection and monitoring for disease recurrence.Gynecol Oncol2008;110:374–382.