Antioxidant Status in Blood of Gynaecological Patients: Influence of Diagnosis and Reproductive Factors

Cancer of the reproductive tract is an important cause of morbidity and mortality among women worldwide, with endometrial cancer (EC) as the fourth most com- mon cancer among women in developed countries. So far, it has been known that several factors may partici - pate in pathogenesis of various gynaecological diseases. Obesity is considered to be strongly associated with the risk of developing endometrial cancer and it depends on the obesity degree, expressed on the relative 2–10 scale (Olson et al., 1995; Purdie, 2003). It is also believed that the association between high body weight and EC is more pronounced in postmenopausal women (Trentham- Dietz et al., 2006); however, about 5–30 % of women are pre- or perimenopausal at the time of diagnosis (Y amazawa et al., 2000; Soliman et al., 2005). Diagnoses before the age of 50 has been linked to a number of risk factors including age of menarche, pari- ty, failure to ovulate and tamoxifen use (Straughn and Partridge 2009; Zucchetto et al., 2009). Other factors that may contribute to increased EC risk are age, late menopause, hyperandrogenaemia (Cline, 2004). Factors affecting EC and endometrial hyperplasia (EH) are known to be, at least in part, similar (Ricci et al., 2002). However, women with benign gynaecological condi- tions such as endometriosis, uterine fibroids (leiomyo- mas) or endometrial polyps may also experience in- creased risk of developing hyperplasia and malignancy (Brinton et al., 2005; Silberstein et al., 2006; Rowlands et al., 2011). Studies indicate that completed or uncom- pleted pregnancy may be protective against EC but not against EH (Parslov et al., 2000; Pike et al., 2004; Xu et al., 2004). Association of history of abortions with the EC risk is still unclear since both positive and inverse relationships were reported (Xu et al., 2004). Abnormal uterine bleeding (AUB) is one of the most common symptoms of endometrial cancer in postmeno- pausal women (Epstein and V alentin, 2004), but it also represents a risk factor for the occurrence of endome - trial hyperplasia in perimenopause (Farquhar et al., 1999). About 10 % of women who have irregular bleed- V ol. 61 27 ing in postmenopause will have a diagnosis of endome- trial cancer, the same percentage will have hyperplasia, 60 % of women will be diagnosed with uterine atrophy, and 10 % will be diagnosed with polyps (Karlsson et al., 1995). Bleeding in postmenopausal women may in- crease the risk of developing endometrial cancer up to 64 times (Gull et al., 2003). Oxidative stress may play an important role in indi- vidual risk of developing many diseases, including can- cers. Cells developed an enzymatic antioxidant (AO) pathway against reactive oxygen species (ROS), which are generated in oxidative metabolism. Superoxide dis- mutase (SOD) catalyses dismutation of superoxide ani- on (O 2 .–) to hydrogen peroxide (H2O2), which in the sec- ond step is converted to water by catalase (CA T) or glutathione peroxidase (GPx). GPx also reduces organic peroxides into alcohols, using glutathione as hydrogen donor (Halliwell, 2006). The activity of the first- and second-step AO enzymes has to be balanced to prevent potential oxidative damage in cells. V ariations in AO ca- pacity may influence individual susceptibility to patho- logical processes associated with the deleterious effects of oxidative reactions (Dalle-Donne et al., 2006; Pagliu- so et al., 2008). In our previous research, we have shown that the an- tioxidant status is altered in the blood and endometrium of women with endometrial hyperplasia and adenocarci- noma in comparison to those with polyps or leiomyo - mas. The specific changes were related to the enzyme type and diagnosis; however, the reduction in antioxi- dants and elevation of the lipid hydroperoxide level were observed in general (Pejić et al., 2006, 2009). In this study we sought to evaluate the association of age and reproductive factors such as parity, abortions and AUB with AO enzyme activities in the blood of these patients.

Subjects The material used in this study consisted of 88 blood samples of women admitted to the Department of Gynaecology and Obstetrics for gynaecological evalua- tion within routine checkups or for abnormal uterine bleeding (Metrorrhagia prolongata, Metrorrhagia recidi- vans, Metrorrhagia postmenopausi). The specimens were taken after obtaining the informed consent. The study was conducted prospectively and it was approved by the Human Studies Ethics Committee of the Clinical Centre. The protocol was consistent with the World Medical Association Declaration of Helsinki (Ethical Principles for Medical Research Involving Human Subjects). On the basis of diagnosis and histological ex- amination, subjects were divided into the following groups: patients with polypus endometrii (PE), uterus myomatosus (UM), patients with hyperplasia simplex endometrii (SH), hyperplasia complex endometrii (CH), or adenocarcinoma endometrii, stage I (ACE). With re- gard to the age, parity (nulliparous, primiparous and multiparous) and abortions, patients were also classified in appropriate categories. Samples Venous blood samples were collected into heparin - ized tubes on the same day as endometrial biopsy and centrifuged at 2500 g for 5 min. Plasma was used for LOOH concentration measurement. For SOD assay (Oxis International, Inc., Portland, OR), the pellet was resus- pended in four packed-cell volumes of ice-cold demine- ralized ultrapure water (MilliQ reagent grade water sys- tem, Millipore Corp., Bedford, MA). After addition of ethanol/chloroform extraction reagent (62.5/37.5 vol/ vol) to remove haemoglobin interference, samples were centrifuged at 3000 g for 10 min (Eppendorf centrifuge 5417, Eppendorf AG, Hamburg, Germany). The upper aqueous layer was collected and kept at -70 °C until as- say. The activities of CA T, GPx and GR were measured in blood lysates. The enzyme activities and lipid hydroperoxide (LOOH) concentration were monitored spectrophoto- metrically (Perkin Elmer Spectrophotometer, Lambda 25, Perkin Elmer Instruments, Norwalk, CT). The spe- cific enzyme activities were expressed as units (U) or mU per milligram of total cell protein (U or mU/mg pro- tein). LOOH concentration was expressed as nmol/mg protein. Determination of protein concentration was performed by the method of Lowry et al. (1951) and expressed as mg/ml. Assays Assay of SOD activity. Determination of SOD activ- ity was performed using Oxis Bioxytech® SOD-525™ Assay (Oxis International, Inc.). The method is based on a SOD-mediated increase of autoxidation of 5,6,6a11b- tetrahydro-3,9,10-tryhydroxybenzo[c]fluorene in aque- ous alkaline solution to yield a chromophore with maxi- mum absorbance at 525 nm. The SOD activity is determined from the ratio of autoxidation rates in the presence (Vs) and in the absence (Vc) of SOD. One SOD-525 activity unit is defined as the activity that dou- bles the autoxidation rate of the control blank. Assay of CAT activity. CAT activity was determined by the method of Beutler (1982). The reaction is based on the rate of H2O2 degradation by catalase contained in the examined samples. The reaction was performed in an incubation mixture containing 1 M Tris-HCl, 5 mM EDTA, pH 8.0, and monitored spectrophotometrically at 230 nm. One unit of CA T activity is defined as 1 µmol of H2O2 decomposed per minute under the assay condi- tions. Assay of GPx activity. GPx activity was assessed us- ing the Oxis Bioxytech® GPx-340™ Assay (Oxis Inter- national, Inc.), based on the principle that oxidized glu- tathione (GSSG) produced upon reduction of an organic peroxide by GPx is immediately recycled to its reduced form (GSH) with concomitant oxidation of NADPH to NADP+. The oxidation of NADPH was monitored spec- Reproductive Factors/AO Status in Gynaecological Patients 28 V ol. 61 trophotometrically as a decrease in absorbance at 340 nm. One GPx-340 unit is defined as 1 µmol of NADH oxi- dized per minute under the assay conditions. Assay of GR activity. Activity of GR was measured using the Oxis Bioxytech® GR-340™ Assay (Oxis Inter- national, Inc.). The assay is based on the oxidation of NADPH to NADP+ during the reduction of oxidized glutathione (GSSG), catalysed by a limiting concentra- tion of glutathione reductase. The oxidation of NADPH was monitored spectrophotometrically as a decrease in absorbance at 340 nm. One GR-340 unit is defined as 1 µmol of NADH oxidized per minute under the assay conditions. Lipid hydroperoxides. The concentration of LOOH was measured by Oxis Bioxytech® LPO-560™ Assay (Oxis International, Inc.), which is based on the oxida- tion of ferrous (Fe2+) ions to ferric (Fe3+) ions by hydro- peroxides under acidic conditions. Ferric ions then bind with the indicator dye, xylenol orange, and form a co- loured complex. The absorbance of the complex was measured at 560 nm. Since hydrogen peroxide content in many biological samples is much higher than that of other hydroperoxides, samples were pre-treated with catalase to decompose the existing H 2O2 and eliminate the interference. Statistical analysis Statistical analysis was conducted using the SPSS software package. The Pearson correlation method and multivariate regression analysis were used to test the as- sociation of age, parity, abortions and AUB with the ac- tivities of AO enzymes. The stepwise logistic regression model, as the most sophisticated one, was used to ensure the smallest possible set of predictor variables in the model. The principle was to enter each predictor in se- quence and to assess its value. If adding the variable contributed to the model, then it was retained, and all other variables in the model were re-tested to see if they still contributed to the success of the model. To perform this analysis, variables were assigned certain numerical scores (Table 1). Statistical significance was set at P < 0.05.

The Pearson correlation matrix obtained between 10 variables (5 predictors and 5 dependent variables) is de- picted in Table 2. It is of interest to note that, except for GR activity, all other AO enzymes and LOOH level were correlated with different predictor factors. The ta- S. Pejić et al. Table 1. Score assigned to different factors of multifactorial analysis Factor Score assigned 1 2 3 4 5 Age (years) N = 25-35 (8) 36-46 (24) 47-57 (45) 58-68 (7) 69-79 (4) Parity N = 0 (3) 1 (7) 2 (33) 3 (24) > 3 (21) Abortions N = 0 (0) 1 (35) 2 (18) 3 (17) > 3 (18) Abnormal uterine bleeding* N = None (30) MP (21) MR (16) MPM (21) Diagnosis N = PE (18) UM (12) SH (31) CH (22) ACE (5) * MP – Metrorrhagia prolongata, MR – Metrorrhagia recidivans, MPM – Metrorrhagia postmenopausi Table 2. Variable correlation matrix in the blood Variable Age Parity Abortions Bleeding Diagnosis CuZnSOD CAT GPx GR LOOH Age 1.0 0.48*** 0.25** 0.32** 0.26** -0.22* 0.06 -0.07 -0.01 0.05 Parity 1.0 0.26** 0.28** 0.29** -0.23* 0.08 -0.20* -0.10 0.13 Abortions 1.0 0.12 0.11 -0.27** 0.17* -0.07 -0.12 0.09 Bleeding 1.0 0.80*** -0.49*** 0.30** -0.55*** -0.02 0.35*** Diagnosis -0.51*** 0.34*** -0.65*** 0.18* 0.39*** CuZnSOD 1.0 -0.09 0.22* -0.08 -0.17 CAT 1.0 -0.39*** 0.23** 0.42*** GPx 1.0 -0.02 -0.34*** GR 1.0 0.21* LOOH 1.0 *P < 0.05, **P < 0.01, ***P < 0.001 V ol. 61 29 ble also shows a significant correlation between the pre- dictor variables themselves as well as between the de- pendant variables. This points to possible interactions between them in the prediction of AO enzyme activities of an individual through multiple regression. The final predictive model with multiple regression analysis for AO enzyme activities is shown in Table 3 and Fig. 1. These results show that two factors (diagno- sis and abortions) fitted the best predictive model for CuZnSOD activity (r 2 = 0.30, P < 0.001). Diagnosis alone contributed with ~6 % and abortions with ~24 % to the total variations of the CuZnSOD activity. One predictive factor alone (diagnosis) provided the best predictive model for CAT activity (r 2 = 0.11, P = 0.001) and GPx activity (r2 = 0.42, P < 0.001). It ex- plained ~11 % of total variations for the activity of CA T and 42 % of total variations for the activity of GPx. Diagnosis also scored alone for the predictive model of LOOH level (r 2 = 0.15, P < 0.001) and contributed with 15 %, while no predictors met criteria for the GR activ- ity. A large part of variations of AO enzymes and LOOH level remained unexplained, which probably points to a role of many other factors that were not considered in this study or were unknown.

The association of different clinical risk factors and various types of gynaecological pathologies is still not fully known, similarly as the influence they exert on the AO status in these patients. In this study, AO enzyme activities and the lipid peroxidation level in the blood of women with different gynaecological conditions and endometrial cancer were related to the diagnosis, AUB, age and reproductive factors (parity and abortions) to observe the strength of the relationship among them and independent association between AO enzymes and each independent variable. The relationship between antioxidants and pathologi- cal changes found in this study points to a role of the AO defence mechanisms in the aetiology of various gynae- cological disorders. A strong reverse relationship be- tween SOD and GPx activities with diagnostic catego- ries was recorded, as well as a positive one between diagnosis and CA T activity/LOOH concentration. These observations are in accordance with our previous find- ings showing a decreasing trend of SOD and GPx ac- tivities in women with endometrial hyperplasia or ade- nocarcinoma in comparison to those with endometrial polyp or leiomyoma (Pejić et al., 2006). Lowered SOD and GPx activities in the plasma of gynaecological pa- tients were also reported in other studies (Chiou and Hu, 1999; Manoharan et al., 2004). It is known that SOD, as primary scavenger of super- oxide anions, along with GPx has a protective role against lipid peroxidation. Thus, the observed reverse relationship may be due to the increased endogenous production of ROS, as also evidenced by the recorded positive relationship of pathological changes in differ - ent diagnosis and LOOH concentration. It is also known that the levels of superoxide anion and hydrogen perox- ide increase in various pathological conditions and that superoxide anion inactivates GPx (Blum and Fridovich, 1985). In support of these findings, a negative correla- tion between SOD/GPx activities and LOOH level was observed in gynaecological patients (Pejić et al., 2006). A positive relationship between diagnosis and CAT ac- tivity observed in this study indicate that CAT is less sensitive to the redox changes in the blood of the exam- ined women. Some studies point to a greater role of this enzyme in protecting erythrocytes against peroxidative stress than GPx (Mueller et al., 1997). The positive cor- relation that we previously recorded between lipid hy- droperoxides and CA T activity also supports this finding (Pejić et al., 2006). When evaluating the influence of reproductive fac- tors, we found a negative association of abortions with SOD activity only, while parity had no influence on AO enzymes or lipid peroxidation. Studies have consistent- ly shown an inverse relation between the risk of endo- metrial cancer and the number of births (Cook et al., 2006). However, data about association between one or more incomplete pregnancies, differently defined in studies as miscarriages or induced abortions, and endo- metrial cancer are mixed (Xu et al., 2004; Pocobelli et al., 2011). Since benign gynaecologic diseases and hy- perplasia may progress to cancer (Ricci et al., 2002; Brinton et al., 2005), the reproductive factors are con- Table 3. Multiple regression analysis (the strength of the model and independent contribution by significant predictor variables to the AO enzyme activities in the blood) Activities of AO enzymes Predictors B β t P value β × rxy F Model P value r2 Adjusted R2 CuZnSOD Diagnosis -0.231 -0.483 -5.298 0.000 0.059 F2,85 = 18.45 0.000 0.303 0.286Abortions -0.094 -0.218 -2.392 0.019 0.244 CAT Diagnosis 5.512 0.337 3.317 0.001 0.114 F1,86 = 11.00 0.001 0.113 0.103 GPx Diagnosis -3.210 -0.649 -7.912 0.000 0.421 F1,86 = 62.59 0.000 0.421 0.415 GR No predictors met criteria LOOH Diagnosis 0.028 0.386 3.878 0.000 0.149 F1,86 = 15.04 0.000 0.149 0.139 B = unstandardized regression coefficient, β = standardized regression coefficient, F = F statistics, which evaluates the model, r2 = variance in enzyme activity accounted for by the predictors, t = t statistics, which evaluates the predictor Reproductive Factors/AO Status in Gynaecological Patients 30 V ol. 61 sidered to be related with hyperplastic changes as well (Epplein et al., 2008). Miscarriage and pregnancy appear to be associated with increased oxidative stress. During uncomplicated pregnancies, ROS levels are elevated at a certain time- point and counterbalanced by the increased activity of antioxidants (Agarwal et al., 2012). In recurrent preg- nancy loss, studies have pointed to a role of oxidative stress in its aetiology (Poston and Raijmakers, 2004; Agarwal et al., 2008). In these patients, significantly low levels of SOD, GPx and CA T were also found, in addi- tion to an increased malondialdehyde level (El-Far et al., 2007). Spontaneous abortion is accompanied by a profound disruption of the pro-oxidant-antioxidant ho- meostasis towards oxidative stress (Lagod et al., 2001) and a first-trimester miscarriage was found to be associ- ated with significantly reduced SOD levels (Jenkins et al., 2000). Thus, a negative relationship of the SOD ac- tivity and spontaneous or induced abortions observed in this study also supports the role which oxidative stress and AO defence may have in the aetiology of gynaeco - logical disorders. Transformed tissues are known to pro- duce high levels of ROS and are constantly under oxida- tive stress (Hileman et al., 2001). The increase of ROS, such as superoxide anion, is able to stimulate cell cycle progression and promote cell proliferation by molecular mechanisms that include oncogenic signals or respira- tory chain malfunction (Pelicano et al., 2004). Cell dam- age caused by activated oxygen metabolites and altered AO capacity might be responsible for biological dif- ferences between transformed and normal tissues (Toyo- kuni, 2006). The negative relationship that indicates a lower SOD activity and increased superoxide concen- trations, observed in our study, implies that patients with benign, premalignant and malignant gynaecological dis- eases are likely to be under oxidative stress. Fig. 1. Graphs showing the standard predictive value of the linear composite of predictors vs. CuZnSOD (A), CA T (B), GPx (C) activities and LOOH (D) level S. Pejić et al. V ol. 61 31 A large part of the examined correlations remained unexplained, which probably points to a role of other factors that were not considered in this study or were unknown. However, this study shows that in gynaeco- logical patients with various diagnoses, the reproductive and other factors may be associated with antioxidant ca- pacity and ability to defend against oxidative damage. The correlations that were established between the pre- dictor variables also indicate possible interactions in the prediction of antioxidant enzyme activities.

Agarwal, A., Gupta, S., Sekhon, L., Shah R. (2008) Redox considerations in female reproductive function and assisted reproduction: from molecular mechanisms to health impli- cations. Antioxid. Redox Signal. 10, 1375-1403. Agarwal, A., Aponte-Mellado, A., Premkumar, B. J., Shaman, A., Gupta S. (2012) The effects of oxidative stress on female reproduction: a review. Reprod. Biol. Endocrinol. 10, 49. Beutler, E. (1982) Catalase. In: Red Cell Metabolism, a Ma- nual of Biochemical Methods, ed. Beutler, E., pp. 105-106. Grune and Stratton, New York. Blum, J., Fridovich, I. (1985) Inactivation of glutathione per- oxidase by superoxide dismutase radical. Arch. Biochem. Biophys. 240, 500-508. Brinton, L. A., Sakoda, L. C., Sherman, M. E., Frederiksen, K., Kjaer, S. K., Graubard, B. I., Olsen, J. H., Mellemkjaer, L. (2005) Relationship of benign gynecologic diseases to subsequent risk of ovarian and uterine tumors. Cancer Epidemiol. Biomarkers Prev. 14, 2929-2935. Chiou, J. F., Hu, M. L. (1999) Elevated lipid peroxidation and disturbed antioxidant enzyme activities in plasma and erythrocytes of patients with uterine cervicitis and myoma. Clin. Biochem. 32, 189-192. Cline, J. M. (2004) Neoplasms of the reproductive tract: the role of hormone exposure. ILAR J. 45, 179-188. Cook, L. S., Weiss, N. S., Doherty, J. A. Chen, C. (2006) Endometrial cancer. In: Cancer Epidemiology and Preven- tion, eds. Schottenfeld, D., Fraumeni J. F. Jr., pp. 1027- 1044. Oxford University Press, New Y ork. Dalle-Donne, I., Rossi, R., Colombo, R., Giustarini, D., Milzani, A. (2006) Biomarkers of oxidative damage in hu- man disease. Clin. Chem. 52, 601-623. El-Far, M., El-Sayed, I. H., El-Motwally, Ael G., Hashem, I. A., Bakry, N. (2007) Tumor necrosis factor-α and oxidant status are essential participating factors in unexplained re- current spontaneous abortions. Clin. Chem. Lab. Med. 45, 879-883. Epplein, M., Reed, S. D., V oigt, L. F., Newton, K. M., Holt, V . L. Weiss, N. S. (2008) Risk of complex and atypical endo- metrial hyperplasia in relation to anthropometric measures and reproductive history. Am. J. Epidemiol. 168, 563-570. Epstein, E., Valentin, L. (2004). Managing woman with the post-menopausal bleeding. Best Pract. Res. Clin. Obstet. Gynaecol. 18, 125-143. Farquhar, C. M., Lethaby, A., Sowter, M., Verry, J. Baranyai, J. (1999) An evaluation of risk factors for endometrial hy- perplasia in premenopausal women with abnormal men- strual bleeding. Am. J. Obstet. Gynecol. 181, 525-529. Gull, B., Karlsson, B., Milsom, I., Granberg, S. (2003) Can ultrasound replace dilation and curettage? A longitudinal evaluation of postmenopausal bleeding and transvaginal sonographic measurement of the endometrium as predic- tors of endometrial cancer. Am. J. Obstet. Gynecol. 188, 401-408. Halliwell, B. (2006) Reactive species and antioxidants. Redox biology is a fundamental theme of aerobic life. Plant Physiol. 141, 312-322. Hileman, E. A., Achanta, G., Huang, P . (2001) Superoxide dis- mutase: an emerging target for cancer therapeutics. Expert Opin. Ther. Targets 5, 697-710. Jenkins, C., Wilson, R., Roberts, J., Miller, H., McKillop, J. H., Walker, J. J. (2000) Antioxidants: their role in preg- nancy and miscarriage. Antiox. Redox Signal. 2, 623-628. Karlsson, B., Granberg, S., Wikland, M., Ylostalo, P., Torvid, K., Marsal, K., V alentin, L. (1995) Transvaginal ultra- sonography of the endometrium in women with postmeno- pausal bleeding – a Nordic multicenter study. Am. J. Obstet. Gynecol. 172, 1488-1494. Lagod. L., Paszkowski, T., Sikorski, R., Rola, R. (2001). The antioxidant-prooxidant balance in pregnancy complicated by spontaneous abortion. Ginekol. Pol. 72, 1073-1078. (in Polish) Lowry, O. H., Rosebrough, N. J., Farr, A. L., Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. Manoharan, S., Kolanjiappan, K., Kayalvizni, M. (2004) Enhanced lipid peroxidation and impaired enzymic anti- oxidant activities in the erythrocytes of patients with cervi- cal carcinoma. Cell. Mol. Biol. Lett. 9, 699-707. Mueller, S., Riedel, H. D., Stremmel, W. (1997) Direct evi- dence for catalase as the predominant H 2O2-removing en- zyme in human erythrocytes. Blood 90, 4973-4978. Olson, S. H., Trevisan, M., Marshall, J. R., Graham, S., Zielezny, M., Vena, J. E., Hellmann, R., Freudenheim, J. L. (1995) Body mass index, weight gain, and risk of endome- trial cancer. Nutr. Cancer 23, 141-149. Pagliuso, R. G., Abbud-Filho, M., Alvarenga M. P. S., Fer - reira-Baptista M. A. S., Biselli J. M., Biselli P. M. Goloni- Bertollo E. M., Pavarino-Bertelli, E. C. (2008) Role of glu- tathione S-transferase polymorphisms and chronic allograft dysfunction. Transplant. Proc. 40, 743-745. Parslov, M., Lidegaard, Ø., Klintorp, S., Pedersen, B., Jønsson, L., Eriksen, P . S., Ottesen, B. (2000) Risk factors among young women with endometrial cancer: a Danish case- control study. Am. J. Obstet. Gynecol. 182, 23-29. Pejić, S., Kasapović, J., Todorović, A., Stojiljković, V ., Pajović, S. B. (2006) Lipid peroxidation and antioxidant status in blood of patients with uterine myoma, endome - trial polypus, hyperplastic and malignant endometrium. Biol. Res. 39, 619-629. Pejić, S., Todorović, A., Stojiljković, V ., Kasapović, J., Pajović, S. B. (2009) Antioxidant enzymes and lipid per- oxidation in endometrium of patients with polyps, myoma, hyperplasia and adenocarcinoma. Reprod. Biol. Endocrin. 7, 149. Pelicano, H., Carney, D., Huang, P . (2004) ROS stress in can- cer cells and therapeutic implications. Drug Resist. Updat. 7, 97-110. Reproductive Factors/AO Status in Gynaecological Patients 32 V ol. 61 Pike, M. C., Pearce, C. L., Wu, A. H. (2004) Prevention of cancers of the breast, endometrium and ovary. Oncogene 23, 6379-6391. Pocobelli, G., Doherty, J. A., V oigt, L. F., Beresford, S. A., Hill, D. A., Chen, C., Rossing, M. A., Holmes, R. S., Noor, Z. S., Weiss, N. S. (2011) Pregnancy history and risk of endometrial cancer. Epidemiology 22, 638-645. Poston, L., Raijmakers, M. T. (2004) Trophoblast oxidative stress, antioxidants and pregnancy outcome – a review. Placenta 25(Suppl A), S72–S78. Purdie, D. M. (2003) Epidemiology of endometrial cancer. Rev. Gynaecol. Pract. 3, 217-220. Ricci, E., Moroni, S., Parazzini, F., Surace, M., Benzi, G., Salerio, B., Polverino, G., La V ecchia, C. (2002) Risk fac- tors for endometrial hyperplasia: results from a case-con- trol study. Int. J. Gynecol. Cancer 12, 257-260. Rowlands, I. J., Nagle, C. M., Spurdle, A. B., Webb, P. M., Australian National Endometrial Cancer Study Group, Austra lian Ovarian Cancer Study Group (2011) Gyne colo- gical conditions and the risk of endometrial cancer. Gynecol. Oncol. 123, 537-541. Silberstein, T., Saphier, O., van V oorhis, B. I., Plosker, S. M. (2006) Endometrial polyps in reproductive-age fertile and infertile women. IMAJ J. 8, 192-195. Soliman, P. T., Oh, J. C., Schmeler, K. M., Sun, C. C., Slomovitz, B. M., Gershenson, D. M., Burke, T. W., Lu, K. H. (2005) Risk factors for young premenopausal women with endometrial cancer. Obstet. Gynecol. 105, 575-580. Straughn, J. M. Jr., Partridge, E. E. (2009) Endometrial can- cer. In: General Surgery, Principles and International Practice, eds. Bland, K. I., Sarr, M. G., Büchler, M. W., Csendes, A., Garden, O. J., Wong, J., pp. 1761-1771. Springer-Verlag, London Ltd. Toyokuni, S. (2006) Novel aspects of oxidative stress-associ- ated carcinogenesis. Antioxid. Redox Signal. 8, 1373-1377. Trentham-Dietz, A., Nichols, H. B., Hampton, J. M., New- comb, P . A. (2006) Weight change and risk of endometrial cancer. Int. J. Epidemiol. 35, 151-158. Xu, W. H., Xiang, Y . B., Ruan, Z. X., Zoeng, W., Cheng, J. R., Dai, Q., Gao, Y . T., Shu, X.O. (2004) Menstrual and repro- ductive factors and endometrial cancer risk: results from a population-based case-control study in urban Shanghai. Int. J. Cancer 108, 613-619. Yamazawa, K., Seki, K., Matsui, H., Kihara, M., Sekiya, S. (2000) Prognostic factors in young women with endome- trial carcinoma: a report of 20 cases and review of litera- ture. Int. J. Gynecol. Cancer 10, 212-222. Zucchetto, A., Serraino, D., Polesel, J., Negri, E., De Paoli, A., Dal Maso, L., Montella, M., La Vecchia, C., Franceshi, S., Talamini, R. (2009) Hormone-related factors and gyneco- logical conditions in relation to endometrial cancer risk. Eur. J. Cancer Prev. 18, 316-321. S. Pejić et al.