ORIGINAL ARTICLE: ENDOMETRIOSIS

Comparative study of ovarianclear cell carcinoma with andwithout endometriosis in People'sRepublic of China

Shuang Ye, M.D.,a Jiaxin Yang, M.D.,a Yan You, M.D.,b Dongyan Cao, M.D.,a Huimin Bai, M.D.,a

Jinghe Lang, M.D.,a Jie Chen, M.D.,b and Keng Shen, M.D.a

a Department of Obstetrics and Gynecology and b Department of Pathology, Peking Union Medical College Hospital,Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, People's Republic of China

Objective: To analyze and compare the clinicopathological features and prognosis of ovarian clear cell carcinoma (CCC) with orwithout endometriosis in Chinese patients.Design: Comparative study based on a retrospective review of medical charts.Setting: A general university hospital.Patient(s): Two hundred ten patients diagnosed and treated with ovarian CCC between 2000 and 2012.Intervention(s): Patients were divided into two groups depending on coexisting endometriosis. A comparison of clinicopathologicalcharacteristics was performed. The Kaplan-Meier model and Cox regression were employed in survival analysis.Main Outcome Measure(s): Clinicopathological parameters and survival outcomes.Result(s): Of 210 patients, 79 (37.6%) were confirmed to have concurrent endometriosis. Patients with endometriosis were 8 yearsyounger than those without. They were more likely to present at early stage (78.5%) with resectable tumors in primary surgery (withoptimal cytoreduction rate at 89.9%) and platinum-sensitive disease (51.7%). Median overall survival for patients withendometriosis was 115 months, an increase of 52 months when compared with 63 months for patients without endometriosis. The5-year survival rate in patients with endometriosis was 70.2%, while it was 52.6% for those without. Univariate and multivariateanalysis showed that coexisting endometriosis was not an independent predictor of survival outcome. Tumor stage and optimaldebulking were the independent prognostic factors for both overall survival and progression-free survival.Conclusion(s): Patients with ovarian CCC and coexisting endometriosis had distinct clinicopathological features and better survival

Use your smartphone

outcome. However, endometriosis per se did not confer improved survival. (Fertil Steril�2014;102:1656–62. �2014 by American Society for Reproductive Medicine.)Key Words: Epithelial ovarian cancer, clear cell adenocarcinoma, endometriosis,clinicopathological features, prognosis

Discuss: You can discuss this article with its authors and with other ASRM members at http://fertstertforum.com/yes-ovarian-ccc-endometriosis-china/

to scan this QR codeand connect to thediscussion forum forthis article now.*

* Download a free QR code scanner by searching for “QRscanner” in your smartphone’s app store or app marketplace.

ndometriosis is a common gyne- immune response abnormality and study involving 4,331 women with

E cologic disorder, which is char-acterized by ectopic growth of

endometrial glands and stroma. Previ-ous publications pointed out that

Received June 3, 2014; revised July 30, 2014; accepted2014.

S.Y. has nothing to disclose. J.Y. has nothing to discloto disclose. H.B. has nothing to disclose. J.L. hasK.S. has nothing to disclose.

This study was supported by the National High Techn(863 program, grant no. 2012AA02A507) and th(grant nos. 81172482 and 81372780).

Reprint requests: Keng Shen, M.D., No. 1 Shuaifuyuanple's Republic of China (E-mail: shenkeng@vip.s

Fertility and Sterility® Vol. 102, No. 6, December 201Copyright ©2014 American Society for Reproductivehttp://dx.doi.org/10.1016/j.fertnstert.2014.08.008

1656

alteration as well as inflammation inwomen with endometriosis might pre-dispose them to have cancer and infec-tions (1–3). In a large cross-sectional

August 4, 2014; published online September 23,

se. Y.Y. has nothing to disclose. D.C. has nothingnothing to disclose. J.C. has nothing to disclose.

ology Research Development Program of Chinae National Natural Science Foundation of China

Road, Dongcheng District, Beijing 100730, Peo-ina.com).

4 0015-0282/$36.00Medicine, Published by Elsevier Inc.

surgically diagnosed endometriosis, ahigher prevalence of ovarian cancerwas noted (3). A pooled analysis ofcase-control studies included 13,226controls and 7,911 women withovarian malignancy and concludedthat endometriosis is associated withcertain histology subtypes includingclear cell carcinoma (CCC) (20.2%), en-dometrioid carcinoma (13.9%), andlow-grade serous carcinoma (9.2%) (4).

Emerging studies have shown thatendometriosis-associated ovarian car-cinoma (EAOC) might deviate fromthe non-EAOC in many key biological

VOL. 102 NO. 6 / DECEMBER 2014

Fertility and Sterility®

characteristics (5–7). However, most studies aggregatedovarian cancer cases with different histological subtypes,making comparisons between various studies suboptimal(4–6, 8–10).

Our previous retrospective studies aimed to analyze theclinical features of ovarian carcinoma with coexisting pelvicendometriosis (6) and endometriosis-associated endometrioidcarcinoma (7). Based onour data, CCC is themost commonhis-tology (47.1%), and patients with EOAC tend to be youngerand diagnosed at an earlier stagewhen comparedwith patientswith non-EOAC. Up to now, two different studies, one fromJapan (11) and the other from the United States (12), havebeen undertaken to investigate the possible prognostic impli-cation of endometriosis in ovarian CCC. However, thepublished studies that mainly concentrate on endometriosis-associated ovarian CCC in Chinese patients are comparativelyfew. Therefore, in the present study, we aimed to evaluate andcompare the clinical and pathologic characteristics as well assurvival outcomes of ovarian CCC patients with or withoutendometriosis in China.

MATERIALS AND METHODSStudy Subjects

The Ovarian Clear Cell Carcinoma Database was set up andmaintained by research faculty members in our department.All the CCC patients diagnosed and treated in our hospitalsince 1982 were included, and basic information was recordedin the database. After obtaining Institutional Review Boardapproval, we identified all the patients between the years2000 and 2012. Case selection was based on original diag-nosis by the pathologists via pathology reports in the medicaldocumentation. To be included in this study, patients had tofulfill the following criteria: [1] patients underwent primarycomprehensive staging surgery or cytoreductive surgery(CRS) in our hospital; [2] diagnosis of ovarian CCC wasconfirmed by histopathology; [3] a complete medical recordand pathology report were available.

Data Collection

Data collection included age at diagnosis, menopausal state,personal history, family cancer history, serum level of cancerantigen 125 (CA125), ascites, date and type of primary surgery,primary tumor size, International Federation of Gynecologyand Obstetrics (FIGO) stage, lymph node metastasis, endome-triosis, residual disease, adjuvant chemotherapy, date of dis-ease progression or recurrence, and tumor status at the dateof last contact. All patients with stage I and II ovarian CCC un-derwent complete staging surgery, and patients with stage IIIand IV ovarian CCC received CRS. Adjuvant chemotherapywas routinely administered after primary surgery with fewexceptions. In the study period, all the patients received aplatinum-based chemotherapy regimen, including paclitaxeland carboplatin and paclitaxel and cisplatin. Specific chemo-therapy dosage was listed as follows: TC regimen (paclitaxel175 mg/m2 þ carboplatin area under curve 5) or TP regimen(paclitaxel 175 mg/m2 þ cisplatin 70 mg/m2). Both wererepeated every three weeks. The number of cycles rangedfrom six to nine after tailoring to different individuals.

VOL. 102 NO. 6 / DECEMBER 2014

Microscopic slides were reviewed and confirmed by a sin-gle experienced gynecologic pathologist (Dr. You). Patholog-ical features of ovarian CCC with and without endometriosisare shown in Figure 1. Specifically, EAOC was defined as fol-lows: [1] presence of CCC and endometriosis in the sameovary, [2] presence of endometriosis in one ovary and ofCCC in the contralateral ovary, [3] presence of CCC and extra-ovarian endometriosis.

All the patients were staged by the FIGO 1999 stagingsystem according to CRS and pathological findings. Theywere further divided into two groups for statistical analysis:early stage (FIGO I-II) and advanced stage (FIGO III-IV).Optimal CRS was defined as residual disease less than (orincluding) 1 cm after primary debulking. The tumor sizewas measured by gynecologic oncologists during operationand recorded in a surgery note. Patients were consideredto have platinum-sensitive disease if the interval timewas >6 months from the completion of the last platinum-based chemotherapy to disease recurrence. The normal upperlimit of serum CA 125 was 35 U/mL.

Progression-free survival (PFS) was defined as the timeinterval from the date of primary surgery to the date of diseaseprogression or recurrence. Overall survival (OS) was definedas the time interval from the date of the primary surgery tothe date of death or last contact. Patients with incompletefollow-up information were excluded from the survivalanalysis.

Statistical Analyses

Statistical analyses were performed using SPSS, version 17.0(SPSS, Inc.), and GraphPad Prism, version 5.0 (GraphPadSoftware, Inc.). Comparisons between the two groups (CCCwith or without endometriosis) were performed by using thec2-test and parametric Student's t tests. PFS and OS timeswere estimated using the Kaplan-Meier model, while Coxregression was used for multivariate analysis. Variableswith statistical significance in univariate analyses wereincluded in the multivariate one. All P values reported weretwo tailed, and P< .05 was considered statistically significant.

RESULTSDuring the study period 2000–2012, a total of 210 patientsfulfilled the inclusion criteria. Of them, 79 (37.6%) patientswere associated with endometriosis and allocated to group1, while the other 131 (62.4%) without endometriosis were as-signed to group 2. In group 1, 20 (25.3%) had been diagnosedwith endometriosis before admission. Among them, 17 pa-tients had a surgical history of ovarian endometriosis (cystec-tomy) and three were clinically diagnosed with endometriosison the basis of dysmenorrhea, pelvic mass detected bybimanual exam or ultrasound, painful nodularity in thepouch of Douglas, and slightly elevated serum CA 125.

Comparison of Clinical and Pathological Features

The clinical and pathological variables between the twogroups are shown in Table 1. For all patients, the mean ageat the time of diagnosis was 51 � 10.9 years (range, 30–

1657

FIGURE 1

Pathological features of ovarian CCCwith (panels B–D) and without endometriosis (panelA, magnification�200). Panel (B) shows the presence ofCCC and endometriosis in the same ovary (white and black arrows indicate CCC and endometriosis, respectively; magnification�40). Panels (C andD) magnify the arrow-pointing areas in panel (B) (magnification �200). Hematoxylin and eosin stain.Ye. Ovarian clear cell carcinoma and endometriosis. Fertil Steril 2014.

ORIGINAL ARTICLE: ENDOMETRIOSIS

81 years). Patients with endometriosis were approximately8 years younger than those without endometriosis, whichwas statistically significant (P< .001). Consistent with age,69.6% of the patients in group1 and only 36.6% in the group

TABLE 1

Clinicopathological features of patients with ovarian CCC.

Variable Group 1 (n [

Mean age, y (range) 46 (30–60Premenopausal (%) 55 (69.6)Gravid <2 (%) 41 (51.9)Personal breast cancer history (%) 2 (2.5)CA 125 in normal range (%) 30 (42.2) (Serum CA 125 (U/mL), mean � SD 213.0 � 794FIGO stage (%)

I 56 (70.9)II 6 (7.6)III 17 (21.5)IV 0

Primary tumor size (cm), mean� SD (range) 11.3 � 5.9 (2–30)With ascites (%) 30 (38.0)Residual disease (%)

%1 cm (optimal) 71 (89.9)>1 cm (suboptimal) 8 (10.1)

Lymph node metastasis 11 (14.5) (Status (%)

Alive 54 (68.4)Dead 23 (29.1)Unknown 2 (2.5)

Note: Group 1 patients had coexisting endometriosis, while group 2 patients did not.

Ye. Ovarian clear cell carcinoma and endometriosis. Fertil Steril 2014.

1658

2 were premenopausal (P< .001). The two groups also had asignificant difference in gravidity number: patients withendometriosis tended to have lower gravidity than thosewithout endometriosis. There was no difference between these

79) Group 2 (n [ 131) P value

) 54 (30–74) < .00148 (36.6) < .00138 (29.0) < .0013 (2.2) 1.000

n ¼ 71) 25 (19.8) (n ¼ 126) .001.1 1,279.9 � 2,359.9 < .001

42 (32.1)7 (5.3)

71 (54.2)11 (8.4)

(n ¼ 77) 10.5 � 5.5 (1–30) (n ¼ 140) .48785 (64.9) < .001

75 (57.3) < .00156 (42.7)

n ¼ 76) 37 (34.3) (n ¼ 108) .003

79 (60.3)44 (33.6)8 (6.1)

VOL. 102 NO. 6 / DECEMBER 2014

Fertility and Sterility®

two groups in terms of personal breast cancer history. Thepreoperative serum CA 125 level was available in 197 pa-tients: 27.9% (55/197) of patients had normal preoperativeserum CA 125 level, with 42.2% (30/71) in group 1 and19.8% (25/126) in group 2. Patients in group 2 were notonly more likely to have elevated CA 125 but also to have amuch higher value with statistical significance (P< .001).Thirty patients (38.0%) in group 1 and 85 (64.9%) in group2 had ascites when the patients were admitted to the hospital(P< .001). There was no difference between these two groupsin terms of primary tumor size, with the average size being11 cm. In general, as shown in Table 1, 52.9% of patientswith ovarian CCC were diagnosed at an early stage, while47.1% were diagnosed at an advanced stage. Of note, 78.5%of the patients with endometriosis were at an early stagecompared with 37.4% of patients without endometriosis(P< .001).

Comparison of Treatment Response

Optimal debulking was achieved in 89.9% (71/79) of thepatients in group 1 and in only 57.3% (75/131) of those ingroup 2 (P< .001). In the entire cohort, 184 (87.6%) patientsunderwent pelvic and or para-aortic lymphadenectomy. Pa-tients in group 2 had more cases of lymph node metastasiscompared with the patients in group 1 (34.3% vs. 14.5%;P¼ .003). After primary surgery, 201 (95.7%) patients receivedplatinum-based chemotherapy, while nine (4.3%) patients did

FIGURE 2

Kaplan-Meier survivals for ovarian CCC patients. Panels (A and B) plot overcohort. Panels (C and D) show survival outcomes of all patients based on cYe. Ovarian clear cell carcinoma and endometriosis. Fertil Steril 2014.

VOL. 102 NO. 6 / DECEMBER 2014

not owing to different reasons such as poor general condition,personal refusal, and economic issues. Ten patients were lostto follow-up after operation or chemotherapy.

Among 193 patients with both adjuvant chemotherapyand follow-up information, tumor recurrence occurred in38.7% (29/75) of the patients with endometriosis and in56.8% (67/118) of the patients without endometriosis(P¼ .014, Pearson c2-test). For these patients with recurrence,51.7% (15/29) of the patients in group 1 had platinum-sensitive disease, compared with 38.8% (26/67) in group 2,which was not statistically significant (P¼ .240).

Comparison of Prognosis and Survival Outcome

After a mean follow-up time of 51 months (range, 1–165 months), 67 (23 in group 1 and 44 in group 2) disease-specific deaths were observed and 133 (66.5%) were censoredat last follow-up in the entire study cohort. The specific distri-bution of censored patients based on stage was 90 for stage I,8 for stage II, 31 for stage III, and 4 for stage IV. Figure 2 dem-onstrates survival curves for all the patients and subgroupanalysis based on concurrent endometriosis. Estimated fromKaplan-Meier survival curves, 5-year overall survival was90.5% in stage I, 56.8% in stage II, 32.8% in stage III, and0% in stage IV. As clearly seen from the curves, there was asignificant difference between the two groups (with orwithout endometriosis) in both PFS and OS. Median survivalwas 115 months and 63 months for group 1 and group 2,

all survival and progression-free survival for all the patients in the studyoexisting endometriosis.

1659

ORIGINAL ARTICLE: ENDOMETRIOSIS

respectively, with statistical significance (P¼ .036). Patientswith endometriosis also had better PFS outcome comparedwith those without (median PFS not achieved vs. 35 months;P¼ .006). In addition, a significant difference was alsoobserved concerning 5-year survival (70.2% vs. 52.6%;P¼ .036).

Table 2 is the result of univariate and multivariate sur-vival analysis. In univariate analysis, CA 125, stage, residualdisease, and endometriosis were significant prognostic factorsfor both OS and PFS. However, after these factors wereentered into multivariate analysis, only FIGO stage and resid-ual disease were confirmed to be independent predictors of OSand PFS, while coexisting endometriosis was not.

DISCUSSIONThe association between endometriosis and ovarian carci-noma has been extensively investigated in the literature sincethe first series of cases of ovarian cancer arising in the contextof endometriosis reported by Sampson in 1925 (13). A largepopulation-based study conducted in Japan illustrated thatendometriosis led to a nine-fold increase in the risk of devel-oping ovarian carcinoma (14). Ovarian CCC is one of the mostcommon histological subtypes that is reported to be in asso-ciation with endometriosis, with a frequency ranging from9% to 70% (5, 9–12, 15–18). Endometriosis is considered bysome researchers to be a specific risk factor for ovarian

TABLE 2

Significant predictors of survival in univariate and multivariate survival an

A. Univariate analysis

Variable Category n

Age (y) <50 97R50 103

Menopausal status Pre 98Post 102

Gravidity <2 76R2 124

CA 125 (U/mL) <200 112R200 78

Stage Early 108Advanced 92

Residual disease <1 cm 141R1 cm 59

Endometriosis Yes 77No 123

B. Multivariate analysis

Variable

OS

HR 95% CI P

CA 125 0.968 0.543–1.724Stage 4.588 2.146–9.808 <Residual disease 2.819 1.546–5.139Endometriosis 0.592 0.326–1.077Note: HR ¼ hazard ratio; CI ¼ confidence interval.a Ten patients (two with endometriosis and eight without) were lost to follow-up and thus not incl

Ye. Ovarian clear cell carcinoma and endometriosis. Fertil Steril 2014.

1660

carcinoma, particularly for the endometrioid and clear cellsubtypes (4, 9, 19, 20).

Accurate diagnosis of endometriosis requires surgicalintervention, type of procedure, and thoroughness and famil-iarity of the surgeon with different appearances of endometri-osis (20). The surgical method, laparoscopy or laparotomy,might influence the intraoperative diagnosis of the disease.Laparoscopy was considered as a standard operative proce-dure in the diagnosis and treatment of endometriosis, but indebulking surgery for ovarian cancer laparotomy has beenthe traditional method owing to the advantage of better oper-ative field and good palpation in which pelvic endometriosisor abnormal lesion is more likely to be found. In our institu-tion, open surgery is the routine approach for patients suspi-cious for ovarian malignancy. Almost all the patientsunderwent laparotomy for debulking surgery. Laparoscopicsurgery to early-stage ovarian carcinoma is still consideredan investigational method among many gynecologic oncolo-gists (21).

The criteria for the diagnosis of EAOC varied betweendifferent studies. Some included only cases with evidence ofmalignant transformation in the endometriosis glands lead-ing to carcinoma (22, 23), while in other studies tumorswere considered as EAOC as long as endometriosis wasfound within a surgical specimen coexisting with ovarianCCC (5, 6, 11, 12). We decided to adopt the latter criteriamainly for two purposes. First, excluding extraovarianendometriosis might inevitably underscore the incidence,

alysis.a

PercentageOS

P valuePFS

P value

48.5 .722 .39951.549.0 .935 .48951.038.0 .068 .10262.058.9 < .001 < .00141.154.9 < .001 < .00146.070.5 < .001 < .00129.538.5 .036 .00661.5

PFS

value HR 95% CI P value

.911 0.864 0.510–1.464 .587

.001 5.435 2.726–10.838 < .001

.001 2.601 1.509–4.482 .001

.086 0.587 0.332–1.037 .587

uded in the survival analysis.

VOL. 102 NO. 6 / DECEMBER 2014

Fertility and Sterility®

given that in advanced cases, some endometriotic lesionsmight not be visible as the cancer becomes more severe.Second, the latter criteria increase the comparability withtwo previous studies on ovarian CCC and endometriosis(11, 12).

A retrospective study from Japan (11) pointed out that38% of patients of the total 53 ovarian CCC patients had car-cinoma accompanied by pelvic endometriosis and that thosepatients were found to have a better prognosis. However,there was no significant difference in the proportion ofearly-stage tumors in ovarian CCC with or without endome-triosis in their study. Orezzoli et al. (12) found endometriosisin association with 49% of ovarian CCC (n ¼ 84). Accordingto their study, patients with endometriosis were younger, hadan earlier stage disease at diagnosis, and had a betteroutcome. Of note, their study concluded that advanced tumorstage and absence of endometriosis were the only significantfactors associated with poor survival. Erzen et al. (5) conduct-ed a large study (n ¼ 232) including various histological sub-types to demonstrate that EAOC had a lower stage atpresentation and a better OS. However, this was true of endo-metrioid carcinoma but not of the CCC in their study.A similar study by Cuff and Longacre (9) illustrated that endo-metriosis was associated with 53% of ovarian CCC and thatendometriosis did not have a significant relationship withPFS. In other words, endometriosis per se did not appear topredict prognosis in CCC.

We compared the clinical and pathological characteristicsand prognosis in ovarian CCC with or without endometriosisin the current study, which was the largest cohort study inChina up to now. Based on our data, 37.6% of the patientswith ovarian CCC (n ¼ 210) had endometriosis by the criteriasimilar to those of the two previous studies (11, 12) and ourown study (8). In our study, ovarian CCC patients withcoexisting endometriosis presented distinct clinical andpathological features when compared with those withoutendometriosis. Endometriosis-associated ovarian CCC tendedto occur in younger women who were more often nulliparous,diagnosed at early stage, and more sensitive to platinum-based chemotherapy regimen and who had a longer survival(both PFS and OS). What was worth mentioning here was thatthe difference in survival was not that large overall, whereasthe prognosis of stage I was much better and the percentage ofstage I disease was much higher in the endometriosis group(70.9% vs. 32.1%). In addition, after univariate and multivar-iate survival analysis, endometriosis was not an independentsignificant prognostic factor for PFS or OS in patients withovarian CCC, which was a different result from the study byOrezzoli and colleagues (12). In our study, FIGO stage of thetumor and residual disease were proved to be the only signif-icant independent predictors of survival outcomes. Therefore,it might lead to a hypothesis that endometriosis-associatedovarian CCC is an entity with better survival outcome, mainlyowing to the higher prevalence of patients diagnosed at theearly stage. Because of endometriosis, patients usually pre-sented some symptoms such as dysmenorrhea, dyspareunia,and/or pelvic mass and often went to hospital for treatmentso that EAOC was diagnosed at an early stage. Some studieshave revealed that the difference in FIGO stage at diagnosis

VOL. 102 NO. 6 / DECEMBER 2014

of the patients with EAOCmight be the result of the symptom-atic character of the coexisting condition (12, 23). In theclinical setting, patients having a higher chance ofendometriosis malignant transformation should be closelyfollowed (24). High-risk factors might include a long historyof endometriosis, early onset, endometriosis-associated infer-tility history, and/or history of infertility treatment andovarian endometrioma (24).

Ovarian CCC is thought to be a separate disease fromother histological subtypes of epithelial ovarian carcinoma(EOC) owing to the unique clinical features, different prog-nosis, and underlying molecular pathways (25–27).Deligdisch et al. conducted a retrospective study including76 cases of stage I ovarian cancer, with 71% (54/76) of thecases being nonserous, to find that histologically confirmedovarian endometriosis was present in 52.6% (40/76) of stageI ovarian carcinoma cases, of which 39 were associated withnonserous cancers (28). Two published studies with a largesample size were undertaken to compare the prognosis ofpatients with ovarian CCC with those with serous EOC(29, 30). In general, ovarian CCC has a survival outcomecomparable to that of serous EOC in terms of early-stage dis-ease (29, 30). However, ovarian CCC patients have muchworse survival than their serous counterparts with respectto advanced-stage disease. Low survival rates in ovarianCCC might in part reflect the tumor's lack of sensitivity toplatinum-based chemotherapy (25, 26, 29). Based on ourdata in Chinese patients, the overall 5-year survival ratewas 90.5% in stage I, 56.8% in stage II, 32.8% in stage III,and 0% in stage IV, which was quite consistent with datafrom the United States and Japan (29, 31). However, ourstudy demonstrated that ovarian CCC patients withendometriosis had a better prognosis, with 70.2% 5-year sur-vival compared with 52.6% in patients without endometri-osis. Possible explanations might be more early-stagediseases and sensitivity to platinum-based chemotherapy inthese group patients. Therefore, we suggested that ovarianCCC should be stratified by coexisting endometriosis for thepurpose of individualized treatment strategies. The standardapproach might be applied in ovarian CCC patients withendometriosis, while a more effective treatment regimen ishighly needed for patients without endometriosis throughrandomized clinical trials.

Our results also revealed that ovarian CCC-associatedendometriosis was only a morphologically defined concept,and whether it proves to be a distinct disease entity or notshould go beyond morphological and clinical analysis. Tobe more specific, the hypothesis of endometriosis-associatedCCC should be tested at the molecular level (32). We haveone ongoing study aimed to construct a tissue microarrayfor ovarian CCC in which molecular markers related toovarian CCC would be much more easily and efficientlyinvestigated.

When interpreting the data of this study, several thingsmust be pointed out. First, as mentioned, advanced diseasemight make the diagnosis of endometriosis more difficult,which would inevitably change the percentages of locallyand advanced disease and might thus affect the study resultsand overall interpretation. Therefore, a detailed patient

1661

ORIGINAL ARTICLE: ENDOMETRIOSIS

history and multiple-site biopsy during operation might helpto better illustrate endometriosis-associated CCC in advancedcases. Second, the cohort was limited by the selection and sur-veillance biases often associated with studies from a singleacademic institution. Furthermore, owing to disease rarity,we included only 79 cases of endometriosis-associated CCC.A multicenter study in China would make a large samplesize possible.

Conclusion

Endometriosis-associated ovarian CCC had distinct clinicaland pathological features. These patients had better survivalcompared with those without endometriosis. Endometriosisitself did not confer improved survival in ovarian CCC.

REFERENCES1. Brinton LA, Gridley G, Persson I, Baron J, Bergqvist A. Cancer risk after a hos-

pital discharge diagnosis of endometriosis. Am J Obstet Gynecol 1997;176:572–9.

2. Bulun SE. Endometriosis. N Engl J Med 2009;360:268–79.3. Gemmill JA, Stratton P, Cleary SD, Ballweg ML, Sinaii N. Cancers, infections,

and endocrine diseases in women with endometriosis. Fertil Steril 2010;94:1627–31.

4. Pearce CL, Templeman C, RossingMA, Lee A, Near AM,Webb PM, et al. As-sociation between endometriosis and risk of histological subtypes of ovariancancer: a pooled analysis of case-control studies. Lancet Oncol 2012;13:385–94.

5. Erzen M, Rakar S, Klancnik B, Syrjanen K. Endometriosis-associated ovariancarcinoma (EAOC): an entity distinct from other ovarian carcinomas as sug-gested by a nested case-control study. Gynecol Oncol 2001;83:100–8.

6. Wang S, Qiu L, Lang JH, Shen K, Yang JX, Huang HF, et al. Clinical analysis ofovarian epithelial carcinoma with coexisting pelvic endometriosis. Am J Ob-stet Gynecol 2013;208:413.e1–5.

7. Wang S, Qiu L, Lang JH, Shen K, Huang HF, Pan LY, et al. Prognostic analysisof endometrioid epithelial ovarian cancer with or without endometriosis: a12-year cohort study of Chinese patients. Am J Obstet Gynecol 2013;209:241.e1–9.

8. Aris A. Endometriosis-associated ovarian cancer: a ten-year cohort studyof women living in the Estrie Region of Quebec, Canada. J Ovarian Res2010;3:2.

9. Cuff J, Longacre TA. Endometriosis does not confer improved prognosis inovarian carcinoma of uniform cell type. Am J Surg Pathol 2012;36:688–95.

10. Kumar S, Munkarah A, Arabi H, Bandyopadhyay S, Semaan A, Hayek K,et al. Prognostic analysis of ovarian cancer associated with endometriosis.Am J Obstet Gynecol 2011;204:63.e1–7.

11. Komiyama S, Aoki D, Tominaga E, Susumu N, Udagawa Y, Nozawa S. Prog-nosis of Japanese patients with ovarian clear cell carcinoma associated withpelvic endometriosis: clinicopathologic evaluation. Gynecol Oncol 1999;72:342–6.

12. Orezzoli JP, Russell AH, Oliva E, Del Carmen MG, Eichhorn J, Fuller AF. Prog-nostic implication of endometriosis in clear cell carcinoma of the ovary. Gy-necol Oncol 2008;110:336–44.

1662

13. Sampson JA. Endometrial carcinoma of the ovary, arising in endometrial tis-sue in that organ. Arch Surg 1925;10:1–2.

14. Kobayashi H, Sumimoto K, Moniwa N, Imai M, Takakura K, Kuromaki T,et al. Risk of developing ovarian cancer among women with ovarian endo-metrioma: a cohort study in Shizuoka, Japan. Int J Gynecol Cancer 2007;17:37–43.

15. Aure JC, Hoeg K, Kolstad P. Mesonephroid tumors of the ovary. Clinical andhistopathologic studies. Obstet Gynecol 1971;37:860–7.

16. Oral E, Ilvan S, Tustas E, Korbeyli B, Bese T, Demirkiran F, et al. Prevalence ofendometriosis in malignant epithelial ovary tumours. Eur J Obstet GynecolReprod Biol 2003;109:97–101.

17. Ogawa S, Kaku T, Amada S, Kobayashi H, Hirakawa T, Ariyoshi K, et al.Ovarian endometriosis associated with ovarian carcinoma: a clinicopatho-logical and immunohistochemical study. Gynecol Oncol 2000;77:298–304.

18. Rauh-Hain JA, Winograd D, Growdon WB, Schorge JO, Goodman AK,Boruta DM, et al. Prognostic determinants in patients with uterine andovarian clear carcinoma. Gynecol Oncol 2012;125:376–80.

19. Nagle CM, Olsen CM, Webb PM, Jordan SJ, Whiteman DC, Green AC. En-dometrioid and clear cell ovarian cancers: a comparative analysis of risk fac-tors. Eur J Cancer 2008;44:2477–84.

20. Nezhat F, Datta MS, Hanson V, Pejovic T, Nezhat C, Nezhat C. The relation-ship of endometriosis and ovarian malignancy: a review. Fertil Steril 2008;90:1559–70.

21. Ghezzi F, Malzoni M, Vizza E, Cromi A, Perone C, Corrado G, et al. Laparo-scopic staging of early ovarian cancer: results of a multi-institutional cohortstudy. Ann Surg Oncol 2012;19:1589–94.

22. Valenzuela P, Ramos P, Redondo S, Cabrera Y, Alvarez I, Ruiz A. Endome-trioid adenocarcinoma of the ovary and endometriosis. Eur J Obstet GynecolReprod Biol 2007;134:83–6.

23. Sainz de la Cuesta R, Eichhorn JH, Rice LW, Fuller AF Jr, Nikrui N, Goff BA.Histologic transformation of benign endometriosis to early epithelial ovariancancer. Gynecol Oncol 1996;60:238–44.

24. Nezhat F, Apostol R, MahmoudM, El DaoukM.Malignant transformation ofendometriosis and its clinical significance. Fertil Steril 2014;102:342–4.

25. Anglesio MS, Carey MS, Kobel M, Mackay H, Huntsman DG. Clear cell car-cinoma of the ovary: a report from the first Ovarian Clear Cell Symposium,June 24th, 2010. Gynecol Oncol 2011;121:407–15.

26. del Carmen MG, Birrer M, Schorge JO. Clear cell carcinoma of the ovary: areview of the literature. Gynecol Oncol 2012;126:481–90.

27. Kennedy AW, Biscotti CV, Hart WR, Webster KD. Ovarian clear cell adeno-carcinoma. Gynecol Oncol 1989;32:342–9.

28. Deligdisch L, Penault-Llorca F, Schlosshauer P, Altchek A, Peiretti M,Nezhat F. Stage I ovarian carcinoma: different clinical pathologic patterns.Fertil Steril 2007;88:906–10.

29. Chan JK, Teoh D, Hu JM, Shin JY, Osann K, Kapp DS. Do clear cell ovariancarcinomas have poorer prognosis compared to other epithelial cell types?A study of 1411 clear cell ovarian cancers. Gynecol Oncol 2008;109:370–6.

30. Lee YY, Kim TJ, Kim MJ, Kim HJ, Song T, Kim MK, et al. Prognosis of ovarianclear cell carcinoma compared to other histological subtypes: a meta-anal-ysis. Gynecol Oncol 2011;122:541–7.

31. Takano M, Kikuchi Y, Yaegashi N, Kuzuya K, Ueki M, Tsuda H, et al. Clearcell carcinoma of the ovary: a retrospective multicentre experience of 254patients with complete surgical staging. Br J Cancer 2006;94:1369–74.

32. Munksgaard PS, Blaakaer J. The association between endometriosis andovarian cancer: a review of histological, genetic and molecular alterations.Gynecol Oncol 2012;124:164–9.

VOL. 102 NO. 6 / DECEMBER 2014