Evaluation of adnexal masses is usually performed with. rezime ...

Transcription

1 /STRU^NI RAD UDK DOI: /ACI L Introduction of new Tumor Marker Age Score in clinical practice: Validity evaluation for differentiation benign from malignant adnexal masses... Ivana Liki} Ladjevi} 1,2, Milan Terzi} 1,2, Neboj{a Ladjevi} 2,3, Jelena Dotli} 1, Igor Pili} 1, Jovan Bila 2, Neboj{a Kne evi} 4, Svetlana Milenkovi} 5, Sanja Mari~i} 6, Neboj{a Arsenovi} 7, Nata{a Brndu{i} 1, Branka Nikoli} 2,8 1 Clinical Center of Serbia, Clinic of Obstetrics and Gynecology, Belgrade, Serbia 2 Faculty of Medicine, University of Belgrade, Serbia 3 Clinical Center of Serbia, Center for anesthesiology and resuscitation, Belgrade, Serbia 4 Department of Anesthesiology, Advocate Illinois Masonic Medical Center, Chicago, IL 5 Department of Pathology, Clinical Center of Serbia, Belgrade, Serbia 6 Occupational Health Department, General Health Center "Savski Venac", Belgrade, Serbia 7 Department of Cellular Pathology, PathLinks Pathology Services, Lincoln County Hosp, UK 8 Obst/Gyn Clinic "Narodni front", Belgrade, Serbia OBJECTIVE: The aim of the study was to examine several tumor markers and their correlation with pathohistological findings in patients with adnexal masses. METHODS: Study involved 139 patients, 84 of them with benign, 47 with malignant and 8 with borderline adnexal tumor. Levels of CA 125, CA 19-9, CEA and CA 15-3 were obtained preoperatively and assessed regarding the specific pathohistological diagnose and the patient s age. Obtaining these results led us to divide the patient s CA 125 levels with age and by doing that we have attained a new Tumor Marker Age score (TMA score). Results: Patients with malignant adnexal tumors had significantly higher levels of CEA (p<0.05), CA 125, CA 19-9 and CA 15-3 tumor markers (p<0.01), in comparison with patients with benign tumors. TMA score highly statistically correlate with the tumor type (benignant/malignant). CONCLUSIONS: With the increase of tumor marker levels and the patient s age the malignant nature of adnexal tumors is more often. Results of our study highlight the importance of the use of combined tumor markers (at least CA-125 and CA 19-9) in women with adnexal masses. Those levels along with the patient s age and new TMA score could preoperatively predict malignant nature of the tumor. Keywords: tumor markers; adnexal masses; borderline; ovary rezime INTRODUCTION: Evaluation of adnexal masses is usually performed with measurement of one or more serum tumor markers, such as CA 125, CA 19-9, CA 15-3, and/or carcino embrio antigen-cea 1. Moderately elevated levels of CA 125 can be detected in various nonmalignant gynecological diseases and some physiological conditions, such as ovarian endometriosis, uterine adenomyosis, leiomyoma, pelvic inflammatory disease and menstruation 2,3. Nevertheless, very high levels of serum CA 125 can usually be detected in patients with malignant adnexal masses, and rarely in patients with benign tumors, and therefore, it can be used in their differentiation4. In 90% of cases with advanced stage of epithelial ovarian carcinoma levels of CA 125 are higher than normal 5. However, serum CA 125 levels for ovarian cancer screening, according to the available literature, have not been proven to be efficient enough when assessed alone. On the other hand, combination of serum CA 125 assay and transvaginal ultrasound have shown better results 6,7. Moreover, the most accurate assessment can be established using other tumor markers such as CEA, CA 19-9 and CA 15-3 combined with CA 125 levels8,9. OBJECTIVE The aim of our study was to assess the correlation between serum tumor markers CA 125, CEA, CA 19-9, and CA 15-3 and pathohistological changes in patients with adnexal masses. METHODS The study was performed on patients with adnexal masses, hospitalized at the Clinic for Gynecology and Obstetrics, Clinical Center of Serbia during the year 2011, who gave informed consent to participation in the study. Levels of CA 125, CA 19-9, CEA and CA 15-3 were obtained from each patient preoperatively, and analyzed on IMMULITE 2000 Immunoassay System (Siemens AG, Munich, Germany). Referral levels used in this study were: 0-35 IU/L for CA-125, 0-33 IU/L for CA 19-9, 0-38 IU/L for CA 15-3 and IU/L for Carcino Embrio Antigen (CEA).

2 50 I. Liki} Ladjevi} et al. ACI Vol. LIX TABLE 1 DESCRIPTIVE STATISTICAL PARAMETERS FOR TUMOR MARKER VALUES Tumor Malignant ηm=47 Borderline ηbl=8 Benignant ηb=84 Statistical parameters Ca 125 Ca 19-9 CEA Ca 15.3 Min Max Mean Median SD SEx Min Max Mean Median SD SEx Min Max Mean Median SD SEx After surgery, pathohistological findings of tumors were analyzed in order to determine the final diagnosis and stage of disease. Obtained tumor marker levels were assessed regarding the specific pathohistological diagnose and the patient s age. Achieving these results led us to divide the patient s Ca 125 levels with age and by doing that we have attained a new Tumor Marker Age score (TMA score). We have also validated new Tumor Marker Age score (TMA) score: TMA score = CA 125 level (IU/L)/patients age (years). For statistical analysis standard methods of data description (the mean value and the measures of variability) as well as tests for evaluation of the data correlation and differences significance were used. Sensitivity ((true positive/true positive+false negative) x 100), specificity ((true negative/true negative+false positive) x100), positive ((true positive/true positive+false positive) x100) and negative ((true negative/true negative+false negative) x100) predictive values were calculated for each of the examined markers. The Receiver Operator Characteristic (ROC) curves were drawn and the trash hold analysis was performed. For data analysis SPSS software (SPSS 15.0, Chicago, IL) was used and p < 0.05 was considered statistically significant. RESULTS Study involved 139 patients, 84 of them (60.43%) with benign, 47 (33.81%) with malignant and eight (5.76%) with borderline adnexal tumors. The average age of investigated women who had malignant tumors was 58 years ( ), while in the group of women who had benign tumors average age was 31 years ( ). Patients with malignant adnexal tumors were highly significantly older than those who had benign tumors (t=13.258; df=129; p=0.000; p<0.01). Correlation of age and CEA levels was not significant (t=0.046; p=0.600; p>0.05). On the other hand, correlation of patients age in total and levels of CA 125, CA 19-9 and CA 15-3 was highly significant (p<0.01). Moreover, age of the patients with benign tumors was positively and highly significantly correlated with CA 125 levels (t=0.352; p=0.001; p<0.01). The average age for patients with border line tumors was 36 years. Three of those eight patients had an atypically proliferating mucinous borderline tumor and five had the atypically proliferating serous borderline tumor. However, all investigated tumor markers were within a normal range in all eight borderline tumor patients and therefore, no further analysis was performed in these patients.

3 Br. 3 New Tumor Marker Age Score 51 TABLE 2 DESCRIPTIVE STATISTICAL PARAMETERS FOR TUMOR MARKER VALUES AND PATIENTS AGE REGARDING THE BENIGNANT TUMOR DIAGNOSES Dg Simple ovarian cyst Dermoidal ovarian cyst Hemorrhagic ovarian cyst Endometriotic ovarian cyst Cysis corporis lutei Statistical parameters Ca 125 Ca 19-9 CEA Ca 15.3 Age Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD On Table 1 descriptive statistical parameters for tumor marker levels in patients with malignant, borderline and benign tumors are shown. Patients with malignant adnexal tumors had significantly higher levels CEA (p<0.05), and highly significantly (p<0.01) higher levels of tumor markers CA 125 (W=3667.0; Z=9.008; p=0.000), CA 19-9 (t=8.146; df=129; p=0.000) and CA 15-3 (W=4721.5; Z=3.956; p=0.000), in comparison with patients with benign tumors. Detailed review of pathohistological findings revealed that among the investigated patients, there were six types of benign diagnoses: simple ovarian cyst (36 pts), dermoidal ovarian cyst (19 pts), hemorrhagic ovarian cyst (5 pts), endometriotic ovarian cyst (17 pts), cystis corporis lutei (6 pts) and ovarian fibrotecoma (1 pts). This single case of fibrotecoma had to be excluded from further analyses. Descriptive statistical parameters for tumor marker levels and age of patients with benign tumors are shown on the Table 2. The frequency of patients with benign tumors was significantly different (p<0.01). We had statistically significantly more patients with simple ovarian cyst, dermoidal ovarian cyst and endometriotic ovarian cyst (χ2=9.084; df=2; p<0.01), while we had significantly less patients with hemorrhagic ovarian cyst and cystis corporis lutei (χ2=37.904; df=4; p=0.000). Levels of CA 125 were highly significantly different concerning the benign tumor diagnoses (KW χ2=50.112; df=4; p=0.000; F=40.486; df1=4; df2=78; p=0.000). CA 125 values were significantly higher in patients who had endometriotic ovarian cyst than in patients with any other diagnosis. Levels of CA 19-9 were highly significantly different concerning the benign tumor diagnoses (KW χ2=14.510; df=4; p=0.006; F=4.509; df1=4; df2=78; p=0.002). The lowest levels of CA 19-9 were in patients with simple ovarian cyst. Concentrations of CA 15-3 were highly significantly different regarding the diagnoses of benign tumors (KW χ2=17.737; df=4; p=0.001; F=6.378; df1=4; df2=78; p=0.000). Levels of CA 15-3 were significantly lower in patients who had simple ovarian cyst in comparison with patients with all other diagnoses.

4 52 I. Liki} Ladjevi} et al. ACI Vol. LIX 1, Malignant Se ns itivity,8,5 R e fe re nc e Lin e C A Benignant G O D 2 500,3 C A C E A L inear (M alignant ) 0,0 0,0,3,5,8 1,0 C A C A y = 8,5005x - 60,704 R2 = 0,2234 Linear (Benignant) 1 - S p e c i f ic ity AGE y = 1,278x -2,1848 R2 = 0,0816 FIGURE 1: ROC CURVES FOR EXAMINED TUMOR MARKERS FIGURE 3: PATIENT S AGE AND CA 125 LEVEL RELATIONSHIP REGARDING THE TUMOR S NATURE C A y = 11,342x -282,38 R = 0, FIGURE 2: PATIENT S AGE AND CA 125 LEVEL RELATIONSHIP Levels of CEA did not differ regarding the diagnoses of benign tumors (KW χ2=2.129; df=4; p=0.712; F=0.683; df1=4; df2=78; p=0.606) (Table 3). There were 11 different malignant diagnoses: serous (11 pts), mucinous (5 pts), endometriotic (5 pts), serous papillary carcinoma (8 pts), mixed malignant Muller s tumor (6 pts), dysgerminoma (2 pts), granulosa cell tumor (3 pts), clear cell (1 pts), yolk sac tumor (1 pts), atypically proliferating serous tumor (1 pts) and Krukenberg tumor (4 pts). All diagnoses, that were present in less than five patients, were analyzed together, as a new group called "other diagnoses". Descriptive statistical parameters for a quarter of our patients with malignant diagnoses had serous ovarian. The frequency of patients did not differ significantly concerning the malignant tumor diagnoses (χ2=5.979; df=5; p=0.308; p>0.05). There were no significant (p>0.05) δifferences in levels of neither one of the investigated tumor markers regarding the different diagnoses of malignant tumors (CA- 125: F=0.470; p=0.796; KW χ2=8.082; p=0.152; CA19- AGE Sensitivity ROC Curve Specificity FIGURE 4: TMA SCORE ROC CURVE (AREA 95.9%) 9: F=1.225; p=0.315; KW χ2=7.417; p=0.191; CEA: F=1.186; p=0.333; KW χ2=4.352; p=0.500; CA 15-3: F=0.695; p=0.631; KW χ2=4.068; p=0.540). Descriptive statistical parameters (mean + SD) for tumor markers regarding the stages of malignancy tumors are presented on the Table 4. Serous was mainly diagnosed in the FIGO stage III, as well as serous papillary carcinoma. In contrast, mucinous was mostly detected in the stage I, while the tumors from the group called "other diagnoses" (dysgerminoma, granulosa cell tumor, clear cell, yolk sac tumor, atypically proliferating serous tumor) were, fortunately, all in the FIGO stage I in the time of their discovery. On the other hand, endometriotic and mixed malignant Muller s tumor, were diagnosed, almost equally in stages I, II and III. Except in patients with Krukenberg tumor, there were no cases of stage IV at the time of tumor diagnosis and treatment. Due to a small number of patients in different groups and stages no fur-

5 Br. 3 New Tumor Marker Age Score 53 TABLE 3 DESCRIPTIVE STATISTICAL PARAMETERS FOR TUMOR MARKER LEVELS AND PATIENTS AGE REGARDING THE MALIGNANT TUMOR DIAGNOSES Dg Serous Mucinous Endometriotic Serous papillary carcinoma Mixed malignant Muller s tumor Other Statistical parameters Ca 125 Ca 19-9 CEA Ca 15.3 Age Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD Min Max Mean SD ther statistical analysis could have been done. There were 29 patients who had benign tumors and CA 125 higher than 35 IU/L. On the other hand, only one patient with malignant tumors had CA 125 less than 35 IU/L. CA 125 sensitivity was 97.87%, specificity was 65.48%, while the positive predictive value was (+PV) 61.33% and the negative predictive value was (-PV) 98.21%. There was only one patient who had false positive results of CA However, there were 25 false negative cases of CA 19-9 values. Consequently, CA 19-9 sensitivity was 46.81%, specificity was 98.81%, whereas the positive predictive value was (+PV) 95.65% and the negative predictive value was (-PV) 76.85%. There was only one patient who had benign tumors and CEA higher than the maximal referral values. In contrast, 43 patients with malignant tumors had CEA in the referral interval. Accordingly, CEA sensitivity was 8.51%, specificity was 98.81%, positive predictive value was (+PV) 80% and the negative predictive value was (-PV) 65.87%. As far as the values of CA 15-3 are concerned, there were no positive findings among investigated women of both groups. Therefore, sensitivity, specificity and predictive values were not calculated for this tumor marker. Trash hold analysis for tumor markers and the age of patients of both groups together shows that the cut-off point for CA-125 was (sensitivity 97.9%; specificity 96.4%), for 19-9 was (sensitivity 72.3%; specificity 77.4%), for CEA 1.25 (sensitivity 66%; specificity 53.6%), for CA 15-3 was 18.6 (sensitivity 80.9%; specificity 52.4%), while for

6 54 I. Liki} Ladjevi} et al. ACI Vol. LIX TABLE 4 TUMOR MARKER LEVELS REGARDING THE STAGES OF THE MALIGNANCY Dg Serous Mucinous Endometriotic Serous papillary arcinoma Mixed malignant Muller s tumor Other diagnoses Krukenberg tumor Figo stage No of patients Ca 125 (X+SD) Ca 19-9 (X+SD) CEA (X+SD) Ca 15.3 (X+SD) I II III I II III I II III I II III I II III I II III I II III IV the patients age cut-off point was 45.5 years (sensitivity 85.1%; specificity 100%) (Figure 1). Regarding the fact that CA 125 and patient s age showed significance in preoperative determination of the tumor type, further analyses were undertaken on these two parameters. For more appropriate data distribution two extreme CA 125 values were excluded from further calculations. The relationship of CA 125 levels and patients age was assessed and a highly significant correlation was determined (ρ=0.648; p=0.000) (Figure 2). Furthermore, it can be observed that 55.3% of cases could be correctly evaluated using this comparison. Moreover, there was a visible difference in data distribution between benignant and malignant tumors (Figure 3). However, simple comparison of CA 125 levels and patient s age was not proven to be very reliable specially in determining benignant from malignant cases (22.34% of malignant cases and only 8.16% of benignant could be determined in this way). Obtaining these results led us to divide the patient s

7 Br. 3 New Tumor Marker Age Score 55 TABLE 5 COORDINATES OF THE ROC (RECEIVER OPERATOR CHARACTERISTIC) CURVE TMA score Sensitivity 1-Specificity CA 125 levels with age and by doing that we have attained a Tumor Marker Age score (TMA score): TMA score = CA 125 level (IU/L) / patients age (year) This score was highly statistically correlated with the tumor type (benignant/malignant) (ρ=0,755; p=0.000). Besides there were highly statistically significant differences in score values between benignant and malignant tumors (p=0.000). In addition ROC (receiver operator characteristic) curve was made for TMA score (Figure 4). From this analysis, used for determining test accuracy and cut-off points, it could be concluded that the most appropriate cut off point for TMA score would be 2.0 with sensitivity of 94% and specificity of 90% (Table 5). This meant that if TMA score was higher than two, the ovarian tumor was most likely of malignant nature. Patients might be considered to have intermediate malignancy risk with TMA score above 1.6 and almost no risk if levels were lower. Finally, sensitivity of the TMA score according to the study results was 93.62%, specificity 88.1%, +PV 81.48%, while -PV was 96.1%. Very high sensitivity showed that TMA score could in most cases point out the positive - malignant findings, while satisfactory specificity meant that TMA score would usually disregard benignant tumors. DISCUSSION The prognosis of ovarian cancer is doubtful, mostly due to the late diagnosis with a very poor outcome in advanced stages. Prevention of the disease could improve that prognosis, but there is still no adequate screening test for ovarian cancer even though there are several screening trials going on 10. In order to diagnose the disease we still rely on pelvic examination, transvaginal ultrasound and CA 125 assay 8,11,12. Assessment for early detection of ovarian cancer can be achieved using tumor markers such as CEA, CA 19-9, CA 15-3 combined with CA 125 levels 8,9. Other tumor markers CA72-4, OVX1, M-CSF and most recently HE 4 should be respected for early detection of ovarian cancer as well 13,14. Current screening guidelines recommend the use of rectovaginal examination, pelvic or transvaginal ultrasound, and CA 125 serum levels in women at high risk. The study showed that combined multiple tumor markers can improve the overall diagnostic accuracy 8. Concerning the fact that more severe consequences can occur if the malignant tumor is not recognized, higher sensitivity of the tested tumor marker is needed 15,16. Out of all investigated tumor markers only CA 125 had a high sensitivity, which makes it the best factor that, if increased, can preoperatively predict the malignant nature of the adnexal tumor. Moreover for CA 125 negative predictive level was higher, just as expected. CA 19-9 and CEA had higher specificity than sensitivity and higher positive than the negative predictive value. This means that, if concentrations of these tumor markers are in the referral interval, they can reliably imply that patients have benign tumors. Levels of CA 125 higher than the cut-off point of and patients age above the cutoff point of 45.5 years highly significantly imply on the malignancy of adnexal mass. According to our study, as well as literature data, none of the biomarkers displayed greater diagnostic performance than CA-125. The diagnostic efficiency of CA-125 in one cohort study (all FIGO stages) was 91.1% 17, and in another one between 70 and 90% 18. Levels of CA-125 may indicate the disease extent and therefore, the likelihood of successful cytoreductive surgery 19. Also, we introduce new TMA score with high sensitivity and specificity. CONCLUSIONS With the increase of tumor marker levels and the patient s age, the malignant nature of adnexal tumors is more often. Regarding the achieved results, it can be advised that in women with adnexal masses levels of tumor markers CA 125 and CA 19-9 should always be analyzed. Those levels along with the patient s age could preoperatively predict malignant nature of the tumor. Using the TMA score can give additional data in preoperative tumor type evaluation, and especially if used together with RMI and other diagnostic methods has been confirmed of high importance in further clinical practice. The use of TMA score can prevent both under and over treatment. Further studies should be undertaken on this issue, but so far we

8 56 I. Liki} Ladjevi} et al. ACI Vol. LIX recommend preoperative calculation of TMA score in patients with adnexal masses. SUMMARY UVODJENJE NOVOG TUMOR MARKER-GODINE STAROSTI SKORA U KLINI^KU PRAKSU: EVALUACIJE VALIDNOSTI U RAZLIKOVANJU BE- NIGNIH OD MALIGNIH ADNEKSALNIH TUMORA CILJ: Cilj studije bio je ispitivanje tumorskih markera i njihove korelacije sa patohistološkim nalazima pacijenata sa adneksalnim masama. METOD: Studija je obuhvatila 139 pacijentkinja, od kojih su 84 imale benigne, 47 maligne a 8 borderline adneksalne tumore. Preoperativno su za sve pacijentkinje odredjeni nivoi tumorskih markera CA 125, CA 19-9, CEA i CA Dobijene vrednosti su postoperativno uporedjene sa patohistološkom dijagnozom tumefakta i godinama pacijentkinja. Dobijeni rezultati su nas podstakli da podelimo nivo CA 125 sa godinama pacijenta, ~ime smo dobili novi skor - Tumor marker godine starosti skor (TMG skor). REZULTATI: Pacijentkinje sa malignim u odnosu na pacijentkinje sa benignim adneksalnim tumorima su imale zna~ajno više nivoe tumor markera CEA (p<0.05), CA 125, CA 19-9 i CA 15-3 (p<0.01). TMG skor je statisti~ki zna~ajno povezan sa tipom tumora (benigni/maligni). ZAKLJU^AK: Sa porastom nivoa tumorskih markera i godinama starosti pacijentkinja ~eš}a je maligna priroda adneksalnih tumora. Rezultati naše studije naglašavaju zna~aj ispitivanja tumorskih markera (bar CA 125 i CA 19-9) kod ena sa adneksalnim masama. Njihove vrednosti zajedno sa godinama ivota pacijentkinja kao i novim TMG skorom mogli bi preoperativno da predvide malignu prirodu adneksalnih tumora. Klju~ne re~i: tumorski markeri; adneksalni tumori; borderline tumori; jajnik REFERENCES: 1. Menon U, Gentry-Maharaj A, Hallett R, Ryan A, Burnell M, et al. Sensitivity and specificity of multimodal and ultrasound screening for ovarian cancer, and stage distribution of detected cancers: results of the prevalence screen of the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Lancet Oncol 2009;10: Buamah P. Benign conditions associated with raised serum CA-125 concentration. J Surg Oncol 2000;75: Engelen MJ, Bongaerts AH, Sluiter WJ, de Haan HH, Bogchelman DH, Tenvergert EM, Willemse PH, van der Zee AG. Distinguishing benign and malignant pelvic masses: the value of different diagnostic methods in everyday clinical practice. Eur J Obstet Gynecol Reprod Biol 2008;136: He RH, Yao WM, Wu LY, Mao YY. Highly elevated serum CA-125 levels in patients with non-malignant gynecological diseases. Arch Gynecol Obstet 2011; 283: Suppl 1: Saygili U, Guclu S, Uslu T, Erten O, Demir N, Onvural A. Can serum CA-125 levels predict the optimal primary cytoreduction in patients with advanced ovarian carcinoma? Gynecol Oncol 2002;86: Partridge E, Kreimer AR, Greenlee RT, Williams C, Xu JL, Church TR, Kessel B, Johnson CC, Weissfeld JL, Isaacs C, Andriole GL, Ogden S, Ragard LR, Buys SS. Results from four rounds of ovarian cancer screening in a randomized trial. Obstet Gynecol 2009;113: Donach M, Yu Y, Artioli G, Banna G, Feng W, Bast RC, Jr., Zhang Z, Nicoletto MO. Combined use of biomarkers for detection of ovarian cancer in high-risk women. Tumour Biol 2010;31: Chan JK, Tian C, Monk BJ, Herzog T, Kapp DS, Bell J, Young RC. Prognostic factors for high-risk early-stage epithelial ovarian cancer: a Gynecologic Oncology Group study. Cancer 2008;112: Joyner AB, Runowicz CD. Ovarian cancer screening and early detection. Womens Health (Lond Engl) 2009;5: van Haaften-Day C, Shen Y, Xu F, Yu Y, Berchuck A, Havrilesky LJ, de Bruijn HW, van der Zee AG, Bast RC, Jr., Hacker NF. OVX1, macrophage-colony stimulating factor, and CA-125-II as tumor markers for epithelial ovarian carcinoma: a critical appraisal. Cancer 2001;92: Anastasi E, Marchei GG, Viggiani V, Gennarini G, Frati L, Reale MG. HE4: a new potential early biomarker for the recurrence of ovarian cancer. Tumour Biol 2010;31: Edgell T, Martin-Roussety G, Barker G, Autelitano DJ, Allen D, Grant P, Rice GE. Phase II biomarker trial of a multimarker diagnostic for ovarian cancer. J Cancer Res Clin Oncol 136: Visintin I, Feng Z, Longton G, Ward DC, Alvero AB, Lai Y, Tenthorey J, Leiser A, Flores-Saaib R, Yu H, Azori M, Rutherford T, Schwartz PE, Mor G. Diagnostic markers for early detection of ovarian cancer. Clin Cancer Res 2008;14: Vorgias G, Iavazzo C, Savvopoulos P, Myriokefalitaki E, Katsoulis M, Kalinoglou N, Akrivos T. Can the preoperative Ca-125 level predict optimal cytoreduction in patients with advanced ovarian carcinoma? A single institution cohort study.gynecol Oncol 2009;112:11-5. Acknowledgement: This work was supported by Grant from the Ministry of Science and Technological Development of the Republic of Serbia.