AIP Conference Proceedings 2193, 040010 (2019); https://doi.org/10.1063/1.5139372 2193, 040010
© 2019 Author(s).
Decrease expressions of ARID1A in non- atypical, atypical endometriosis cyst and ovarian clear cell carcinoma as a biomarker of malignancy transformations
Cite as: AIP Conference Proceedings 2193, 040010 (2019); https://doi.org/10.1063/1.5139372 Published Online: 10 December 2019
Retno Widyawati, Budiningsih Siregar and Kusmardi Kusmardi
ARTICLES YOU MAY BE INTERESTED IN
Photoelectrochemical measurement of Epstein-Barr virus DNA in peripheral blood serum of nasopharyngeal carcinoma patients
AIP Conference Proceedings 2193, 040009 (2019); https://doi.org/10.1063/1.5139371 High progesterone levels on the day of hCG administration in controlled ovarian hyperstimulation procedures reduce the expression of progesterone receptors in the endometrium Macaca nemestrina
AIP Conference Proceedings 2193, 040007 (2019); https://doi.org/10.1063/1.5139369 Intrahepatic transport of primaquine with OCT1: An in vitro study in HepG2 cells AIP Conference Proceedings 2193, 040011 (2019); https://doi.org/10.1063/1.5139373
Decrease Expressions of ARID1A in Non-atypical, Atypical Endometriosis Cyst and Ovarian Clear Cell Carcinoma as a
Biomarker of Malignancy Transformations
Retno Widyawati
1,a), Budiningsih Siregar
2), Kusmardi Kusmardi
3)1Installation of Anatomical Pathology, Prof. Dr. WZ Yohannes Hospital, Kupang, Indonesia
2Department of Anatomical Pathology, Faculty of Medicine, Universitas Indonesia/Dr. Ciptomangunkusumo National Central Hospital, Jl Salemba Raya No 6, Senen, Central Jakarta 10430, Indonesia
3Departement of Anatomical Pathology, Faculty of Medicine, Universitas Indonesia, Jl Salemba Raya No 6, Senen, Central Jakarta 10430 Indonesia
Corresponding author: a)kusmardi.ms@ui.ac.id
Abstract. Endometriosis is one of the most common gynecological abnormalities. Endometriosis cyst in the ovary also exhibited changes in epithelial cyst just like endometrium in the uterus. Nowadays, there are some research that linked endometriosis and clear cell ovarian cancer which is known with endometriosis-associated ovarian carcinoma (EAOC). It is reported that there’s a mutation that activated tumor suppressor gene (ARID1A), so protein BAF250a is not expressed in Clear Cell Carcinoma (CCC) in the ovarium. Immunohistochemistry staining of ARID1Awere done in 20 samples of non- atypical endometriosis, 20 cases of atypical endometriosis, 20 cases of CCC in the ovarium from 2012 until march 2015.
From the CCC we get 9 cases of EAOC. After that, we see if there’s any difference in the percentage of ARID1A expression in non-atypical, atypical endometriosis, CCC in the ovarium and endometriosis with CCC (EAOC). In non-atypical, atypical endometriosis and CCC groups there are significant differences on the percentage of ARID1A expression (Kruskal- Walis test p=0.0035). There are significant differences on ARID1Aexpression between non-atypical and atypical endometriosis with EAOC (p=0.001 and p=0.0015). In non-atypical, atypical endometriosis and CCC there are significant differences on the percentage of ARID1A expression (Kruskal-Walis test p=0.011). There are significant differences on ARID1Aexpression between non-atypical and atypical endometriosis with EAOC (p=0.005 and p=0.008). The expression of ARID1A in non-atypical and atypical endometriosis are significantly higher than ovarian CCC and EAOC. ARID1A may be used as a marker for malignancy transformation in endometriosis.
Keywords: ARID1A, atypical endometriosis, EAOC, ovarian clear cell carcinoma
INTRODUCTION
Ovarian cancer is the cancer with the fifth highest number of mortality rate among women in the world. According to American Cancer Society, in the year 2013 there are 22.240 new cases of ovarian cancer with total mortality of 14,030. According to Cancer Registration Bodies, Pathological Anatomy Expert Society of Indonesia, from 14 pathological center in Indonesia there are 1,110 cases of ovarian malignancy with the highest frequency among women aged 45-54 years old [1,2]. Usually patients with ovarian cancer are found in the later stages of it is disease.
This is caused by the unavailability of screening techniques and also unclear pathogenesis [3]. Besides that, the extent of the problem were magnified by the lack of effective screening to detect ovarian cancer in it is early stages [4].
Prevention methods that is available now is not effective because factor that influence gene defect that initiate cell transformation were not yet understood by the scientists [5].
Nowadays, it is known that there is a type of ovarian cancer with a special link to endometriosis known as EAOC (Endometriosis-associated ovarian carcinoma). This cancer has biological behavior similar to epithelial ovarian carcinoma. Many researchers have linked endometriosis that is located in the ovarium with ovarian carcinoma. Many
research showed that those that have endometriosis has higher risk of ovarian cancer compared to those that do not have the history of endometriosis [6,7].
Endometriosis cannot be classified as premalignant condition. But according to epidemiological data, histopathological data, and molecular biology data showed that endometriosis has the potential to evolve into malignancy [8,9].
Just like endometrium in the uterus, epithelial cells that engulf endometriosis cyst in the ovarium can have various transformation. Transformation that can happen are metaplasia, hyperplasia, atypia and even malignancies [9,10].
Endometrial cyst content many free radicals that is induced by mutagenic ferrous ion. The length of exposure on the epithelial cells of ovarian cyst may cause accumulation of somatic mutation that may help the transformation processes [11-13].
Nowadays, many research in the field of endometrosis that is linked with mutation of gene ARID1A, that is a tumor suppressor gene that coded protein BAF250a has important role in the creation of SWI/SNF chromatin remodeling complexes. This is also important in many activities that cause transcription, methylation process, synthesis and DNA repair [13,15-16]. Mutation of gene ARID1A cause a decrease in protein BAF20a, which is observed in Clear cell carcinoma (CCC) of the ovary [15].
MATERIAL AND METHODS
This study used descriptive analytical cross-sectional design and was approved by the Institutional Review Board.
This study enrolled cases 60 cases, consisted of 20 non-atypical endometriosis, 20 atypical endometriosis and 20 CCC ovarian from our hospital, in time of 2012 to March 2015. From 20 cases of CCC there are 9 cases with endometriosis and 11 cases without endometriosis. Then, from 9 CCC cases with endometriosis we determine the presence of endometriosis that is near to the tumor mass.
For ARID1A imunnostaining, 3-4μm -thick paraffin sections were cut and mounted on poly-L-lysine glass slides.
The slides were deparaffinized in xylene and rehydrated with ethanol. Endogenous peroxidase was blocked using 95 ml of methanol plus 5 ml of 3% hydrogen peroxide solution. Heat-based antigen retrieval was carried out for 10 min.
Primary antibodies used were from primary antibody of mouse monoclonal ARID1A (dilution 1:100). Slides were incubated for one night with the primary antibodies followed by staining with streptavidin-biotin peroxidase. Slides were then revealed in a diaminobenzidine solution for 2 min and stained with hematoxylin for 5 min.
A cervical tissue was used as positive control. Negative controls were obtained by omitting the primary antibodies.
Scoring of ARID1A expression were done using counting the percentage of nucleus that is stained brownish in the endometriosis cyst epithelial cells and also epithelial cells in the glands of the stroma. We measure the mean of the positive from all of the observation field. Internal control for this study include stroma cells with nucleus that is completely stained.
Descriptive data are presented in the frequency table. Numerical variables were compared using Kruskal-Walis test. Post Hoc analysis were done using Mann-Whitney test. P< 0.05 was considered statistically significant. Statistical analysis was performed using SPSS Statistics for Windows, Version17.0 (SPSS Inc, 2008, Chicago, USA).
RESULTS
In Table 1, we can see the general characteristics of the patients that is participating in this study. There are 20 cases of non-atypical endometriosis and 20 cases of atypical and 20 cases of ovarian CCC. From 20 cases of CCC there are 9 cases with endometriosis. The total samples of this study are >50 samples and the data obtained were the percentage of staining expression ARID1A which is analyzed using Kolmogorov-Smirnov test. The value of p < 0.05, so we can conclude that the distribution of the results are not normal. Because we use more than two groups of data, we continue the analysis using Kruskal-Wallis test. To see the differences between expressions of ARID1A between groups we use the Mann-Whitney test.
TABLE 1. General characteristic of patients diagnosed with endometriosis or CCC Non-Atypical
Endometriosis (N=20)
Atypical Endometriosis
(N=20)
Endometriosis with CCC
(EAOC) (N=9) CCC (N=20) Age
Mean±SD 35,15±6,9 34,7±8,5 47,22±8,8 44,74±9,1
Range 19-47 25-49 24-56 24-56
Marital status (%)
Married 16(80) 19(95) 9(100) 18(90)
Single 4(20) 1(5) 0(0) 2(10)
Side of ovarium affected (%)
Left 4(20) 1(5) 2(22,2) 3(15)
Right 7(35) 5(25) 4(44,4) 8(40)
Bilateral 9(45) 14(70) 3(33,3) 9(45)
Diagnosis (%)
Endometrial cyst 17(85) 15(75) 1(11,1) 3(15)
CON 3(15) 4(20) 2(22,2) 10(50)
Ovarian Cancer 0(0) 1(5) 4(44,4) 5(25)
SON 0(0) 0(0) 2(22,2) 2(10)
In the Figure 1, We can see the morphological histopathology of non-atypical endometriosis (A), atypical endometriosis (B), EAOC with endometriosis near the tumor mass. (↓). We can see endometriosis in transitional area is atypical (C, D).
FIGURE 1.A,B. Hematoxylin Eosin staining with 100x magnifications. A. non-atypical endometriosis B. Atypical endometriosis C and D. HE staining with 40x magnification, EAOC (↑) atypical endometriosis cells
Expressions of ARID1A
ARID1A expressions in non-atypical endometriosis have range from 40-100% with median value of 75%. In atypical endometriosis, we can get ARID1A range between 20-5% with median value of 70%. In CCC case group, expression of ARID1A showed value between 5-90% with median value of 20%. In 9 cases EAOC and 11 CCC cases
From Figure 2, we can see atypical endometriosis, 400x (HE and ARID1A staining). From ARID1A staining, we can see that half of the nucleus expressionhas brown staining. From Figure3, we can see the result of ARID1A staining in several cases, in non-atypical endometriosis with epithelial cells nucleus that has brown color, atypical endometriosis with surface epithelial cells that has brown discoloration. Tumor mass in ovarian CCC with nucleus with brownish discoloration and tumor cells without brown discoloration.
FIGURE 2. Atypical endometriosis, 100x magnifications. A. Hematoxylin Eosin staining and B. ARID1A staining. We can see that the expression of ARID1A can only be observed in several of the surface epithelial cells.
FIGURE 3. ARID1A staining, 400x magnifications. Sign ‘↑’ in the picture above showed cells that have brownish discoloration (expression of ARID1A that is scored and recorded by the resarchers). A. Case of non-atypical endometriosis B. Case of atypical
endometriosis C. Case of clear cell carcinoma of the ovary. D. Case of Clear cell carcinoma of the ovary.
ARID1A expressions of CCC Ovarium
There are significant differences of median expression of ARID1A in non-atypical endometriosis, atypical endometriosis and CCC ovarian with p value=0.0035. After this, we did post-hoc analysis (Mann-Whitney test). In atypical endometriosis and CCC ovarian we found that the median percentage of expression are statistically significant with p=0.015. This can also be found in endometriosis non-atypical and CCC ovarium, we can get a different median of expression in those values with p=0.01. In non-atypical endometriosis, we can see that the median ARID1A
expression percentage are higher compared non-atypical endometriosis, however this differences are not statistically significant with p=0.274 (Table 2).
TABLE 2. Differences of ARID1A expressions percentage in non-atypical endometriosis, atypical endometriosis and ovarian CCC
N Percentage of staining ARID1A (%)
Median (minimum-maximum) P
Histopathological diagnosis
CCC 20 20(5-90)
0,0035*
Atypical
Endometriosis 20 70(20-95)
Non-Atypical
Endometriosis 20 75(40-100)
*Kruskal-Wallis. Post-hoc Mann-Whitney test; CCC vs non-atypical with p value of 0.001; CCC vs Endometriosis atypical p= 0.0015; Endometriosis atypical and non-atypical p=0.274.
ARID1A expressions EAOC
There are significant differences of median expression of ARID1A in non-atypical endometriosis, atypical endometriosis and EAOC with p value=0.011. After this, we did post-hoc analysis (Mann-Whitney test). In non- atypical endometriosis and EAOC we found that the median percentage of expression are statistically significant with p=0.005. This can also be found in atypical and EAOC, we can get a different median of expression in those values with p=0.08 (Table 3).
TABLE 3. Differences of ARID1A expressions percentage in non-atypical endometriosis, atypical endometriosis and EAOC
N Percentage of staining ARID1A (%) Median (Minimum-Maximum) P Histopathological
diagnosis
EAOC 9 20(5-90)
0,011*
Atypical
Endometriosis 20 70(20-95)
Non-Atypical
Endometriosis 20 75(40-100)
*Kruskal-Wallis. Post-hoc (Mann-Whitney test) CCC vs non-atypical with p=0.005; CCC vs Endometriosis atypical p=0.08; Endometriosis atypical and non-atypical p=0.274
In median ARID1A expression percentage in 11 cases of CCC without endometriosis and 9 cases EAOC, we can see that there is no significant difference with p value of 0.412 (Table 4).
TABLE 4. Differences ARID1A expression between CCC without Endometriosis and EAOC N Percentage of ARID1A staining (%)
Median (Minimum-Maximum) P Morphological
type
EAOC 9 20(5-90)
0,412*
CCC without
endometriosis 11 30(5-90)
*Mann-Whitney test
DISCUSSION
In this study, 9 of 20 (45%) cases of CCC contain endometriosis. There’s some evidence that endometriosis can transform into malignancy and become ovarian cancer. This causality has been proven by epidemiological data, most patient with ovarian cancer has history of having endometriosis, it is based by the similarity of ovarian epithelial cells with endometrium. Research by Fukunaga et al (1997) stated that endometriosis cyst can develop into ovarian cancer,
clear cells subtype (54%) and endometrioid subtype (41%) [17]. Aure et al (1971) stated that 34% of 59 cases of CCC contain endometriosis [18].
Endometriosis diagnosis with atypical cells were often shrugged down upon during histopatological examinations, these atypical cells in endometriosis can cause early changes from malignant transformations. From his research, Fukunaga et al (1997) stated that epithelial atypia can be found if there’s tumor mass near the site and atypia epithelial cells is the precursor for the malignant tumor development [17]. In 9 cases EAOC, endometriosis that is found near tumor mass is atypical endometriosis. This finding is in accordance with the research that has been done by Richardo Sainz et al (1995) that stated endometriosis that is found near the tumor mass is atypical endometriosis with or without proliferation [19]. In his research, Mandai et al (2009) and Xiao et al (2012)stated that almost all of the endometriosis that is found is commonly found in transitional endometriosis area and the mass usually is atypical endometriosis [13,14].
Studying the gene ARID1A which is the precursor tumor suppressor gene that encodes BAF250a, we can see that this gene can help early diagnosis of ovarian cancer, especially clear cell subtypes. This research use CCC sample that is included in type I ovarian carcinoma, this carcinoma characteristics are indolent and with a stable genetic traits.
Type 1 ovarian carcinoma also showed relationship with benign cystic tumor in regards to carcinogenesis process [20].
The characteristic of sample
According to the data obtained from this study we can get the range and mean of age of patients in non-atypical and atypical endometriosis case. The range and mean of those two groups are quiet similar in which the non-atypical endometriosis has age range between 19 to 47 years old (mean of 34.15 years old) and range between 25-49 years old with mean of 34.7 years old. In CCC cases, the mean age is higher with 44.75 years old. This finding is similar with what Edmoson et(2001) al found in their research that stated incidence of ovarian carcinoma is lower in those younger than 40 years old, the reverse is true in patients which is older than 40 years old [21]. The incidence increased dramatically from 80-84 years old. Non-atypical and atypical endometriosis cases can be found more often in similar age, 34.15 years and 34.75 years old. This result is similar with the findings by Wei et al (2011) al that stated malignancy transformation in endometriosis took longer than 8 years to manifest [22]. Research by Yamaguchi et al (2008) reported that after 10 years, patients with endometriosis cyst has risk on having ovarian cancer [23]. Moll et al (1990) reported that transformation process in atypical endometriosis into CCC needs only 3 years to manifest itself [24].
According to marital status, most of the sample are married (with 90% of the sample have already married).
Research by Yamashita et al (2012) showed that in general, endometriosis cyst are commonly found in infertile women or women with prolonged estrogen exposure [10]. However, information of whether the patient is infertile or not are not stated in the clinical records.
In this research, we found that non-atypical and atypical endometriosis cyst are commonly found in both ovarium (bilateral) 14 (70%) and 9 (45%) of the cases. In non-atypical or atypical endometriosis unilateral, it is more often to find endometrial cyst in the right ovarium. Research result by Fukunaga et al (1997) and Farquhar (2002) in which we found unilateral ovarian endometriosis, this condition is more often found in the left ovarium [17,25]. This findings may be explained by the position of sigmoid colon in the left adnexa that is often have direct contact with the fallopian tube and the left ovary have direct contact with the pelvical area. This cause adhesion and position that is lower. When there’s a regurgitaition of endometrial cells in the tube, left ovarium and left peritoneum will have higher exposure [25]. Many confounding factors may explain these discrepancies of this research findings, some of them are race, parity or the small samples of this research.
In endometriosis case group, the most commonly made clinical diagnosis in non-atypical group and atypical group are endometriosis cyst 17(85%) and 15 (75%), the rest of the clinical diagnosis are cystic ovarian neoplasm (CON) 3(15%) and 4 (20%). In the case of CCC of ovarium, the most common clinical diagnosis is CON 10 (50%). This is followed by 2 (10%) Solid Ovarian Neoplasm (SON), 3 (15%) endometrial cyst, and 5 cases (25%) ovarian carcinoma.
This result is similar with result by Bell DA (2005) and Horiuchi et al (2003) that used ultrasonography tools to get clinical research result, 20% of endometrioid adenocarcinoma and 88% of CCC at first were diagnosed as having benign cyst or endometriosis cyst [25,27]. In EAOC group, only 44.4% were diagnosed as a carcinoma, the rest are diagnosed as having SON, CON, and endometriosis cyst. More distinct clinical definition are needed to differentiate SON and CON.
ARID1A Expression
Research by Fukunaga (1997) found close relationship between atypical endomertriosis and CCC in the ovary, this finding is solidified by research by Ogawa et al (2007) that found that endometriosis atypical can cause pre-cancer condition in the ovary [17,28]. This research is only limited to the difference of expression because of the retrospective data available. Significant differences between ARID1A expressions between 3 cases group: non-atypical endometriosis and atypical endometriosis with ovarian CCC and between non-atypical endometriosis and atypical endometriosis with EAOC. As well as comparing the expression of ARID1A level between these two groups (non- atypical and atypical endometriosis) value of p<0.05. This research has similar finding with Xiao et al (2012) in which 53.8% of CCC has ARID1A mutation that is decreased because of the expression of ARID1A staining [14]. Similar with this research, from 20 cases of CCC ovarian the general expression of ARID1A is lower compared to the median value of 20% found in this research.
In this research, we can see the difference of median value of ARID1A expression between non-atypical endometriosis group and atypical endometriosis group. This research is similar with previous research about the decrease of ARID1A expression in atypical endometrial cyst. However, in this research, the difference is not statistically significant (p=0.274). This finding showed that not all of atypical endometriosis is caused by the mutation of ARID1A gen tumor suppressor, but also caused by the mutation of the other tumor suppression genes or other gene that may be involved in the carcinogenesis process [29]. Beside mutation, other factors also involved in the carcinogenesis process including chronic inflammation and estrogen hormone.
In the case group of CCC, it was divided into two distinct case groups, that is without endometriosis and EAOC.
In group without endometriosis the median percentage ARID1A expression is higher compared to EAOC, however this differences are not statistically significant (p=0.412). This finding is similar with what Samartrzis et al (2013) found in which the frequency of ARID1A mutation is higher in EAOC compared to CCC without endometriosis [15].
Research by Scott et al (1953) and Xiao et al (2012) stated that the difficulties in getting endometriosis lesion that is found near the tumor mass of ovarian carcinoma (EAOC) is mainly caused by tissue sample that is obtained not representative for the endometriosis appearance in the tissue or because of tumor infiltration on the endometriosis lesion [14,30].
CONCLUSION
Based on the result analysis and discussion that we made above, we can say that there’s a significant difference between expression of ARID1A between non-atypical endometriosis, atypical endometriosis with CCC ovarian and EAOC group that showed ARID1A expression can be used for malignant transformation markers of endometriosis.
There is a decrease of ARID1A expression from non-atypical endometriosis to atypical endometriosis. However, statistically there are no significant differences between those two groups.
REFERENCES
1. Siegel R, Naishadham D, Jemal A. Cancer statistics 2013. CA Cancer J Clin. 2013;63:11-30.
2. Agency of Indonesian Cancer Registry. Cancer in Indonesia 2010: Histopathological Data. Jakarta.
Indonesian Pathologist Association 2013.
3. Nezhat F, Cohen C, Rahaman J, Gretz H, Cole P, Kalir T. Comparative immunohistochemical studies of bcl- 2 and p53 proteins in benign and malignant ovarian endometriotic cysts. Cancer. 2001;94:2935-40.
4. Chobanian N, Dietrich CS. Ovarian cancer. Obstet Gyn Clin N Am. 2008;88:285-99.
5. Busmar B. Ovarium Cancer. In: Aziz MF, Andrijono, Saifuddin AB, editor. National Refferences of Oncology. 1st ed. Jakarta: Yayasan Bina Pustaka Sarwono Prawirodihardjo; 2006.
6. Kennedy S, Chapron C, D’Hooghe T, Dunselman G, Greb R. ESHRE guideline for the diagnosis and treatment of endometriosis. Hum Reprod. 2005:1-7.
7. Nezhatt Farr DS, Hanon V, Pejovic T, Nezhat C. The relationship of endometriosis and ovarian malignancy:
a review. Fertil Steril. 2008:1559-70.
8. Worley MJ WW, Berkowitz RS, Ng SW. Endometriosis-Associated Ovarian Cancer: A Review of Pathogenesis. Int J Mol Sci. 2013;1:5376-9.
9. Kumar S MA, Arabi H, Bandyopadhyay S, Semaan A, Hayek K, et al. Prognostic analysis of ovarian cancer associated with endometriosis. Am J Obstet Gynecol. 2011;204:7.
10. Yamashita Y TS. Endometriosis associated ovarian cancer: The role of oxidative stress. Intech. 2012 [cited 2013 May].
11. Prefumo F FE, Venturini PL. Epithelial Abnormalities in Cystic Ovarian Endometriosis. Gynecol Oncol.
2001;84:280-4.
12. Badziad A. Endometriosis. Obstetry and Cynecology. 3rd ed. Jakarta: Yayasan Bina Pustaka Sarwono Prawirohardjo; 1999.
13. Mandai M, Yamaguchi K, Matsumura N, Baba T, Konishi I. Ovarian cancer in endometriosis: molecular biology, pathology and clinical management. Int J Clin Oncol. 2009;14:383-91.
14. Xiao W, Awadallah A, Xin W. Loss of ARID1A/BAF250a expression in ovarian endometriosis and clear cell carcinoma. Int J Clin Exp Pathol. 2012;5:642-50.
15. Samartzis EP, Samartzis N, Noske A, Fedier A, Caduff R, Dedes KJ, et al. Loss of ARID1A/BAF250a- expression in endometriosis: a biomarker for risk of carcinogenic transformation? Mod Pathol. 2012;25:885- 92.
16. Zhang Z, Xiao S, Sun GY, Liu Y, Zhang F, Yang H, et al. The clinicopathological significance of the loss of BAF250a (ARID1A) expression in endometrial carcinoma.Int J Gynecol Cancer.2014;24:534-60.
17. Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypical endometriosis:Its close association with malignant epithelial tumours. Histopathol. 1997;30:249-55.
18. Aure JC, Hoeg K, Kolstad P. Carcinoma of the ovary and endometriosis. Acta Obstet Gynecol Scand.
1971;50:63-7.
19. Cuesta RS, Eichhorn JH, Rice LW, Fuller AF, Nikrui N, Goff BA. Histologic transformation of benign endometriosis to early epithelial ovarian cancer. Gynecol Oncol. 1996;60:238-44.
20. Kurman JK, Shih IM. Pathogenesis of ovarian cancer. Lessons from Morphology and molecular biology and their clinical implications. Int J Clin Gynecol Pathol. 2008;27:151-60.
21. Edmonson RJ, Monaghan JM. The epidemiology of ovarian cancer. Int J Gynecol Cancer. 2001;11:423-9.
22. Wei JJ, William J, Bulun S. Endometriosis and ovarian cancer: a review of clinical pathologic and molecular aspects. Int J Gynecol Pathol. 2011;30:553-68.
23. Yamaguchi K, Mandai M, Tokoyuni S, Hamanishi J, Higuchi T, Takakura K Contentents of endometriosis cysts, especially the high concentration of free iron, area a possible cause of carcinogenesis in the cysts through the iron-induced persistent oidative stress. Clin Cancer. 2008;14:32-40.
24. Moll UM, Chumas JC, Challas E, Mann WJ. Ovarian carcinoma arising in atypical endometriosis. Obstet Gynecol. 1990;75:537-9.
25. Farquhar C. Endometriosis. Clin Evid. 2002;7:1654-2.
26. Bell DA. Origins and molecular pathology of ovarian cancer. Mod Pathol. 2005;18:19-32.
27. Horiuchi A, Itoh K, and Shimizu M. Toward understanding the natural history of ovarian carcinoma development: a clinicopathological approach. Gynecol Oncol. 2003;88:309-17.
28. Ogawa S,Kaku T, Amada S, Kobayashi H, Hirakawa T, Ariyoshi K, et al. Ovarian endometriosis associated with ovarian carcinoma: aclinicopathological and immunohistochemical study. Gynecol Oncol.
2000;77:298-304.
29. Seidman JD, Cho KR, Ronner BM, Kurman RJ. Surface epithelial tumors of the ovary In: Kurman RJ, Ellenson LH, Ronner BM, editor. Blaustein’s pathology of the female genital tract. 6th ed. New York:
Springer; 2011.p.679-784.
30. Scott RB. Malignant change in endometriosis. Obstet Gynecol. 1953;2:283-9.
