{"paper_id":"73c8c005-89ca-472b-abe0-50d522cc114e","body_text":"R E S E A R C H Open Access\nClear cell carcinomas of the ovary: a mono-\ninstitutional study of 73 cases in China with\nan analysis of the prognostic significance of\nclinicopathological parameters and IMP3\nexpression\nRui Bi1,2, Xuxia Shen 1,2, Weiwei Zhang 3, Yufan Cheng 1,2, Zheng Feng 2,4, Xu Cai 1,2 and Wentao Yang 1,2*\nAbstract\nBackground: Ovarian clear cell carcinoma (CCC) is an uncommon subtype of ovarian epithelial tumor. The prognostic\nsignificance of its clinicopathological parameters is discordant, with the exception of stage as the adverse prognostic\nfactor. The present study aimed to evaluate the prognostic significance of its clinicopathological characteristics and\nthe expression of IMP3 (Insulin-like growth factor-II mRNA-binding protein 3, IMP3 or IGF2BP3) in Chinese patients\nwith primary pure CCC.\nMethods: We collected clinicopathological data from 73 cases with a minimum of 5 years of follow-up and evaluated\nt h ee x p r e s s i o no fI M P 3b yi m m u n o h i s t o c h e m i s t r y .\nResults: In total, 49.3 % of the patients were in stage I. Advanced stages were closely related to poor prognosis of\ndisease-free survival (DFS) and overall survival (OS) (P < 0.005). Patients with CCC coexisting with endometriosis tended\nto be younger and to have unilateral involvement but did not exhibit differences in prognosis compared with patients\nwith CCC without endometriosis. Other histological features such as growth pattern, mitosis, and necrosis did not have\nprognostic significance. IMP3 was positive in 63 % of patients (46 of 73 cases); Thus, positive expression of IMP3 is an\nadverse prognostic marker in terms of OS ( P = 0.012), even in stage I patients ( P =0 . 0 3 8 ) .\nConclusions: The present study demonstrates that IMP3 expression is a prognostic marker, with the exception of\nstage. IMP3 represents a biomarker of unfavorable prognosis even in stage I patients.\nKeywords: Clear cell carcinoma, Prognosis, Stage, Endometriosis, IMP3\nBackground\nThe prevalence of ovarian clear cell carcinoma (CCC),\nwhich has significant geographic differences, accounts\nfor 5 –25 % of cases in various countries [1, 2]. Most of\nCCC in stage I [3] associated with endometriosis [4] has\na poor response to platinum-based regimens, which\nresults in an unfavorable prognosis, particularly for\nadvanced stages [3, 5, 6]. The various morphologies of\nCCC result in misclassification, which further affects the\nassessment of morbidity and prognostic factors. A\ngrading system based on CCC growth pattern corre-\nlates with survival in ovarian CCC [7]. However, this\nrelationship has not been further confirmed, and the\nimpact of the histopathological characteristics on\nprognosis has been disputed.\nInsulin-like growth factor-II mRNA-binding protein 3\n(IMP3 or IGF2BP3), a member of the IMP family, is\nhighly expressed in embryo. This protein is involved in\nthe binding, transport and stability of the embryonic\nisoforms of the IGF-II gene. IMP3 is considered as an\noncofetal protein and is related to tumor metastasis.\n* Correspondence: yangwt2000@163.com\n1Department of Pathology, Fudan University Shanghai Cancer Center,\nShanghai 200032, China\n2Department of Oncology, Shanghai Medical College, Fudan University,\nShanghai 200032, China\nFull list of author information is available at the end of the article\n© 2016 Bi et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International\nLicense (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any\nmedium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative\nCommons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://\ncreativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.\nBi et al. Diagnostic Pathology  (2016) 11:17 \nDOI 10.1186/s13000-016-0467-5\n\nThis protein is often found to be overexpressed in many\ntypes of tumors, such as gastric cancer, colorectal cancer,\nliver cancer, and pancreatic cancer, where it acts as an\nindependent prognostic factor [8 –11]. However, few\nstudies have examined the role of IMP3 in gynecologic\ntumors, such as in ovarian CCC, endometrial CCC and\ncervical squamous carcinomas [12 –14]. IMP3 expression\nin ovarian CCC was reported in a previous study; how-\never, subsequent studies with small sample size have not\nconfirmed the same results [12, 15]. Therefore, the sig-\nnificance of IMP3 expression in ovarian cancer, particu-\nlarly in CCC, requires further analysis.\nFew studies of primary pure ovarian CCC have been\nexamined in Chinese patients and the lack of effective\nprognostic indicators in addition to staging. Therefore,\nreviewing and understanding the characteristics of CCC\nand assessing IMP3 expression as a prognostic param-\neter are of great importance.\nMethods\nPatients and clinicopathological characteristics\nThe present study was approved by the Ethical Review\nCommittee of Fudan University Shanghai Cancer Center.\nData were collected from 73 patients with a primary\ndiagnosis of pure ovarian CCC who underwent their first\noperation at Fudan Cancer Center between 1999 and\n2009 and who had a minimum of 5 years of follow-up.\nHematoxylin-eosin (HE)-stained sections were reviewed\nby three gynecological pathologists. The patients\nunderwent total abdominal hysterectomy, bilateral\nsalpingo-oophorectomy, omentectomy, and appendec-\ntomy simultaneously or asynchronously. The patients\nwere then treated with platinum-based postoperative\nchemotherapy. Staging was determined according to\nthe 2014 FIGO new ovarian, tubal, peritoneal tumor\nstaging guidelines [16].\nHistological morphology was estimated approximately\naccording to the summary of DeLair et al. [17] for the\ndetermination of tumor growth pattern (tubulocystic,\nsolid, or papillary), cell type (clear cell, oxyphil cell, hob-\nnail cell or columnar cell), mitotic index (10/HPF),\nstroma (hyalinized, myxoid, fibroblastic), and necrosis.\nWe also evaluated the presence of endometriosis (ovar-\nian and/or extra-ovarian) in the specimens.\nImmunohistochemistry\nIMP3 expression was evaluated by immunohistochemis-\ntry using the EnVision method [15]. Briefly, sections\n(4 μm) were deparaffinized and treated with 0.3 %\nhydrogen peroxide, boiled in citrate buffer (pH 6.0) for\n15 min, and cooled at room temperature. Sections were\nincubated with a primary IMP3 antibody (Clone 69.1,\ndilution 1:200, Dako Glostrup, Denmark) at 4 °C over-\nnight, rinsed in PBS, and then incubated for 40 min with\nbiotinylated secondary antibody. Sections were stained\nin DAB (Dako, Glostrup, Denmark), rinsed three times\nin PBS, and then counterstained with hematoxylin. Incu-\nbation in PBS buffer in lieu of primary antibody was\nused as a negative control, and IMP3 expression in the\ngerminal center of the lymph node served as a positive\ncontrol. We defined positive IMP3 staining as convin-\ncing cytoplasmic expression in more than 10 % of tumor\ncells [18]. The positive staining intensity was recorded as\nweak, moderate, or strong.\nFollow-up\nThe patients were followed up until March 31, 2014.\nOverall survival (OS) was defined as the time from oper-\nation to either death or the last follow-up. Disease-free\nsurvival (DFS) was defined as the interval from oper-\nation to disease recurrence or the last follow-up. We\ndefined disease recurrence as a consistent elevation of\nCA125, or observation of tumor by clincial examination\nincluding physical examination and imaging studies [19].\nStatistical analyses\nThe chi-square test and Fisher ’s exact test were used to\nevaluate the association between IMP3 expression and\nmultiple clinicopathological parameters. Survival analysis\nwas determined using Kaplan –Meier univariate analysis.\nDifferences in survival curves were compared with the\nlog-rank test. P values < 0.05 were considered significant.\nFor statistical analyses, SPSS software version 19.0 (IBM\nSPSS Statistics 19) was used.\nResults\nClinicopathological characteristics\nThe ages of the 73 patients ranged from 19 to 80 years\nold (mean, 53.63 years; median, 54 years). Macroscopic-\nally, the tumor size ranged from 2 to 27 cm (mean,\n10.45 cm). Unilateral tumors were the most frequent,\noccurring in 58 cases (79.5 %), while bilateral tumors\noccurred in 15 cases (20.5 %). The serum level of CA125\nwas above normal in the 56 preoperative cases (47/56,\n83.9 %). Ascites were present in 48.5 % of patients (33/\n68) and absent in 51.5 % (35 cases) of patients in the\nrecorded data. The percentages of cases at each FIGO\nstage were as follows: stage I, 49.3 % (36 cases); stage II,\n15.1 % (11 cases); stage III, 24.7 % (18 cases) and stage\nIV, 11 % (8 cases). Follow-up information was obtained\nfrom 69 patients. The OS of patients ranged from 0.7 to\n173.8 months (mean, 61.2 m; median, 59 m), and the\nDFS ranged from 0.7 to 173.8 months (mean, 53.2 m;\nmedian, 38.2 m). The 5-year survival rate varied among\nstages as follows: stage I (26/33, 78.8 % of cases), stage II\n(5/11, 45.5 % of cases), stage III (5/18, 27.8 % of cases),\nand stage IV (0/6, 0 % of cases) ( P < 0.001).\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 2 of 8\n\nHistopathological features were summarized in Table 1.\nBasically, ovarian clear cell carcinomas showed a variable\nadmixture of papillary, tubulocystic and solid growth\npattern. A variable papillary component was present in\n82.2 % of cases (60 of 73). Pure papillary pattern, tubulo-\ncystic pattern and solid pattern were observed in 34.2 %\n(25/73), 6.8 % (5/73) and 2.7 % (2/73) cases respectively.\nMixed papillary and tubulocystic pattern were present in\n27.4 % (20/73) cases. Mixed tubulocystic and solid\npattern, mixed papillary and solid were observed in\n8.2 % (6 /73) and 8.2 % (6/73) respectively. A mixed\npattern of papillary, tubulocystic and solid were present\nin 12.3 % (9/73) cases. Ovarian clear cell carcinomas\nwere composed by variable proportions of clear cells,\noxyphilic cells, hobnail cells and colunmar cells. Hobnail\ncells were present in 56.2 % (41/73) of all cases. A mix-\nture of clear cells and oxyphilic cells was present in\n78.1 % (57/73) cases, Pure clear cell type was observed\nin 15.1 % (11/73) cases. 6.8 % (5/73) cases composed of\npure oxyphilic cell. Rare cases presented with columnar\ncells with subnuclear or supranuclear vacuoles (Fig. 1a).\nThe mean mitotic index was 4.6/10 HPFs. Mitotic\nfigures were 0 –9/10 high-power fields (HPFs) in 65 cases\n(89 %). The remaining cases had more than 10/10HPF,\nwith the highest of 17/10 HPFs. Diversity was observed\namong the stromal changes, including fibrous, hyaline\nand myxoid stroma. Pure hyaline or myxoid stroma were\nobserved in 4 cases (5.5 %) and 1 case (1.4 %) respect-\nively. Focal or patchy necrosis was observed in 39.7 %\n(29/73) of the cases. Targetoid bodies (signet ring-like\ncells with eosinophil/basophil secretions, shown in\nFig. 1b), psammoma bodies (Fig. 1c), and open tumor\ncell rings (rings of tumor cells without central stroma,\nFig. 1d) were presented in some cases. Tubulocysts\nfilled with mucinous secretions were observed in eight\ncases, one of which with severe expansile tubulocysts\nresembling thyroid follicles (Fig. 1e). Scattered nuclear\npseudoinclusions were noted in 6 cases (Fig. 1f ).\nEndometriosis was observed in the specimen in 23.3 %\n(17 of 73) of the cases.\nClinical parameters and survival (Table 2)\nFollow-up information was obtained for 69 patients,\nincluding 33 in stage I, 11 in stage II, 18 in stage III, and\n7 in stage IV. The mean OS of each stage was as follows:\nI, 85.3 months; II, 57.15 months; III, 31.5 months; and\nIV, 30.1 months. Statistically significant differences in the\nmean OS were observed ( P <0 . 0 0 1 ) . T h e m e a n D F S o f\neach stage was as follows: I, 79.9 months; II, 45.4 months;\nIII, 24 months; and IV , 14 months. The differences in the\nmean DFS were also significantly different ( P <0 . 0 0 1 ) .T h e\nsurvival of patients with stage I disease resulted in a\nremarkably favorable prognosis compared with that of\npatients in stages II –IV; we then analyzed the data from\nthe two groups in stages I and II –IV. The comparison of\nDFS and OS between the patients in stages Ia/Ib and Ic\nshowed no statistically significant differences ( P =0 . 4 4 0\nand P = 0.875, respectively).\nMoreover, patients with ovarian CCC coexisting with\nendometriosis were prone to develop unilateral tumors\n(P = 0.038), with a trend of occurrence in relatively\nyounger patients (15 patients of < 60-year-old group vs.\n2 patients of ≥ 60-year-old group, P = 0.062). Other\nclinical parameters, such as age, tumor size, preoperative\nCA125 level, ascites, and endometriosis, exhibited no\nTable 1 Morphological characteristics in 73 cases of ovarian\nclear cell carcinoma\nMorphology Cases (%)\nGrowth pattern\nTubulocystic 5 (6.8)\nSolid 2 (2.7)\nTubulocystic/Solid 6 (8.2)\nPapillary 25 (34.2)\nPapillary/Tubulocystic 20 (27.4)\nPapillary/Solid 6 (8.2)\nTubulocystic/Solid/Papillary 9 (12.3)\nat least focal papillary componen 60 (82.2)\nCell type\nClear cell 11 (15.1)\nOxyphil cell 5 (6.8)\nMixed with Clear and oxyphil cell 57 (78.1)\nHobnailing 41 (56.2)\nColumnar cell 2 (2.7)\nMitotic figures (/10 HPF)\n0–9 65 (89.0)\n10–19 8 (11.0)\nStroma\nHyalinized only 4 (5.5)\nMyxoid only 1 (1.4)\nFibroblastic only 23 (31.5)\nFibroblastic/Hyalinized 20 (27.4)\nFibroblastic/Myxoid 4 (5.5)\nHyalinized/Myxoid 1 (1.4)\nFibroblastic/Hyalinized/Myxoid 20 (27.4)\nOther miscellaneous characteristics\nFocal or patchy necrosis 29 (39.7)\nTargetoid bodies 3 (4.1)\nPsammoma bodies 8 (11.0)\nMucin secretion 8 (11.0)\nOpen tumor rings 6 (8.2)\nNuclear pseudoinclusions 6 (8.2)\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 3 of 8\n\nsignificant difference in their association with DFS and\nOS between groups (Table 2).\nIMP3 immunohistochemical staining and survival\nForty-six of the 73 cases (63 %) showed positive IMP3\nexpression (Fig. 2a and 2b), and the remaining 27 cases\n(37 %) were IMP3 negative.\nThe rate of IMP3 expression was significantly higher\nin patients with an advanced stage (II –IV), with 73.0 %\npositive cases (27 of 37), compared with the 52.8 % of\ncases observed in stage I (19 of 36) ( P = 0.046). The OS\nwas significantly shorter in IMP3-positive patients than\nin those negative cases (55.8 months vs. 71.3 months,\nP = 0.012, Fig. 3a), but no significant difference was\nobserved for DFS ( P = 0.175). Upon further stratifica-\ntion analysis of IMP3 expression in patients with stage\nI, the IMP3-positive cases (23 –133.6 months, mean\n76.0 months) had a shorter OS than did the IMP3-\nnegative cases (11.2 –173 months, mean 96.5 months)\n(P = 0.038, Fig. 3b). None of the 17 IMP3-negative cases\nof stage I died. However, 4 of 19 IMP3-positive cases\ndied from ovarian CCC metastasis, and their OS ranged\nfrom 23.0 to 76 months (mean 40.4 months). Neverthe-\nless, the DFS did not differ significantly between IMP3-\npositive and IMP3-negative cases ( P = 0.557). However,\nno statistically significant differences in OS and DFS\nwere observed in patients with stage II-IV CCC ( P =\n0.892 and P = 0.840, respectively).\nDiscussion\nCompared with common serous carcinomas, ovarian\nCCC has distinct features such as poor prognosis and\nplatinum drug resistance. A relatively high percentage\nof CCC cases are in stage I (31 –-60 %) [20 –23], which\nis significantly higher than the percentage of patients\nwho have serous carcinoma (16.6 –20 %) [3, 24]. Stage I\ncases accounted for approximately 45.8 –62.8 % of\nChinese CCC patients [25, 26]. The percentage of pure\nC C Cp a t i e n t si ns t a g eIi nt h ep r e s e n ts t u d yw a s4 9 . 3%\nFig. 1 a A solid nest is on the right, and the cystic morphologywith columnar cells resembles an endometrial gland.b Targetoid bodies are signet ring-\nlike cells contained with eosinophil/basophil secretions. Classic hobnail cells can be observed in the top left corner.c Scattered psammoma bodies.d Open\ntumor rings consist of a ring of tumor cells without a central fibrovascular stroma (A-D; original magnification, ×200).e Extremely expansive tubulocystic\nmorphology with mucinous secretions similar to a thyroid follicle (originalmagnification, ×40).f Nuclear pseudoinclusions were observed with severe cell\natypia, but mitotic figures were absent (original magnification, ×400)\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 4 of 8\n\nTable 2 The association of the clinical features and the expression of IMP3 with the prognosis of 73 patients with ovarian clear\ncell carcinoma\nDFS OS IMP3 expression Endometriosis\nPP Positive Negative P Present Absent P\nAge 0.097 0.071\n≥60y 0.084 0.073 17 5 2 19\n<60y 29 22 15 36\nTumor size 0.222 0.415\n≥15 cm 0.337 0.285 8 8 5 11\n<15 cm 38 19 12 44\nLateral 0.786 0.038\nUnilateral / / 37 21 40 17\nBilateral 9 6 15 0\nCA125 0.115 0.375\nAbove normal 0.201 0.935 29 18 10 34\nNormal 8 1 3 5\nAscites 0.347 0.521\nPresent 0.055 0.271 19 14 9 7\nAbsent 24 11 24 27\nEndometriosis 0.352 / / /\nPresent 0.534 0.424 9 8\nAbsent 36 19\nNecrosis 0.260 0.296\nPresent 0.377 0.499 16 13 5 24\nAbsent 30 14 12 31\nPapillary component 0.188 0.631\nPresent 0.123 0.180 37 18 12 42\nAbsent 9 9 5 13\nMitotic figures (/10 HPF) 0.488 0.660\n<10 0.712 0.947 40 24 12 44\n≥10 6 2 1 6\nStage 0.074 0.038\nI <0.001 <0.001 19 17 12 23\nII–IV 27 10 5 32\nDFS 44 18 0.175\nOS 44 18 0.012\nFig. 2 a Focal and moderately positive staining for IMP3 expression. b Diffuse and strong IMP3-positive staining (original magnification, ×200)\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 5 of 8\n\n(36 of 73), which is similar to the percentages that have\nbeen reported in other populations.\nAccording to a report by Higasi et al., the five-year\nDFS and OS rates of each stage of ovarian CCC were\n84 % and 88 % (stage I), 57 % and 70 % (stage II), 25 %\nand 33 % (stage III), and 0 % and 0 % (stage IV), respect-\nively [27]. The five-year OS rates of the patients in our\ngroup were similar to those reported in other studies of\nnon-Chinese populations: 78.8 % (stage I), 45.5 % (stage\nII), 27.8 % (stage III), and 0 % (stage IV). According to a\nprevious report of a Chinese population, the five-year\nOS rate of patients with stage I was 66.8 % [25]; the\nsurvival rates in the current study were higher than\nthose in previous reports. A study by Hoskins et al. indi-\ncated that the prognosis of patients with CCC in stage\nIa/Ib was more favorable than that of patients in stage Ic\n[28] and that the prognosis of patients with stage Ic was\ncloser to that of patient with stage II [29]. However, OS\nand DFS rates did not significantly differ between stages\nIa/Ib and Ic in this group; this result was comparable to\nthe report by Hoskins et al.\nOvarian CCC is closely related to endometriosis,\nwhich was found in 23.3 % of our cohort. This value is\nsimilar to the frequency of 26.8 % reported in another\nChinese population [26] but lower than that reported in\nother countries (approximately 45 %) [30, 31]. Scarfone\net al. reported that patients with CCC arising from\nendometriosis are generally younger than those with\ngeneral CCC [32]. Our results indicate that CCC cases\ncoexisting with endometriosis also had a trend of occur-\nrence among relatively younger patients and were often\nunilateral, although few cases arose directly from endo-\nmetriosis. The presence or absence of endometriosis was\nnot related to prognosis in any case in our study or in a\npreviously reported study [32]; however, some studies\nhave observed a better prognosis in CCC-associated\nendometriosis [33, 34].\nAscites were often observed during operation and\nwere a trend among the patients with an unfavorable\nprognosis compared to patients without ascites with\nrespect to DFS ( P = 0.055) but not OS ( P = 0.271).\nThe classical morphology of ovarian CCC cases is\ntubulocystic, solid and papillary or with an admixture of\ndiffernt growth pattern [17]. In the present study, 72 %\nof the cases contained papillary structures. A mixture of\nclear cells and oxyphilic cells was the most common cell\ntype. The tumor stroma was often observed to be a\nmixture of hyalinized, myxoid, or fibroblastic types,\nwhereas pure hyalinization or pure myxoid stroma was\nuncommon. In our cases, the average mitotic index was\n4.6/10 HPFs, with most cases being lower than 10/10\nHPFs (89 % of the group), which is quite consistent with\nprevious reports [31]. Furthermore, we did not observe\nany statistically significant differences in the relationship\nbetween morphology and prognosis.\nThe importance of IMP3 expression in the female re-\nproductive system is controversial. One study suggested\nthat IMP3 was a prognostic marker in metastatic ovarian\ncancer [35], Kobel et al. suggested that IMP3 expression\nwas an independent indicator of a poor ovarian CCC\nprognosis [12], and Noske et al. suggested that IMP3\nexpression was a marker of a good prognosis in ovarian\ncancer (including serous carcinoma and non-serous\ncarcinoma) [15]. No further reports regarding IMP3 as a\nprognostic biomarker have been published. The present\nstudy demonstrated that IMP3 expression is closely\nrelated to an unfavorable prognosis and that it may be\nused as a potential indicator to estimate poor prognosis.\nThe results of our study are similar to those of Kobel et\nal. In endometrial clear cell carcinoma, IMP3 was\nFig. 3 a The OS of the patients with positive IMP3 expression was worse than that of the negative cases among 73 cases. b The same trend was\nobserved in 36 cases with stage I disease\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 6 of 8\n\nreported to be closely correlated with poor prognosis\nand relapse free survival (RFS) [13].\nFurthermore, we analyzed the follow-up data of 33\npatients with stage I CCC and found that IMP3 positiv-\nity was associated with a shorter OS (76.0 months vs.\n96.5 months, P = 0.038). None of the 17 IMP3-negative\ncases of stage I died, but 4 of 19 IMP3-positive cases\ndied from ovarian CCC metastasis. The results for the\n5-year survival rate in our series show the same trend\nreported by Kobel et al. Most ovarian CCC patients\nreceived adjuvant chemotherapy after surgery. But for\nthose stage I patients, the benefit from chemotherapy\nwas not sufficient [36, 37]. However, the patients may\nhave, to endure severe side effects of chemotherapy.\nIMP3, as a potential stratification prognostic factor in\nstage I cases, may be used to stratify patients whom\nmay benefit from adjuvant therapy. However, IMP3+\ntumors in stage I patients still have a wide range of sur-\nvival (23 –133.6 months) in our study. Thus, further\nresearch are needed to verify the prognostic signifi-\ncance of IMP3 in ovarian CCC. For example, studies\nare needed to demonstrate whether IMP3 staining\ncorrelates well with molecular tests for IMP3 expres-\nsion, and furthermore, that IMP3 expression by mo-\nlecular test correlates with prognosis.\nIt had been reported that knockdown of IMP3 in cer-\nvical cancer, breast cancer and melanoma cell lines could\nreduce cell migration and invasiveness [38 –40]. In renal\ncell carcinoma, IMP3 can also promote cell migration\nand invasion by activation of NF- κB pathway [41]. As\nrecently reported, knockdown of IMP3 decreased cell\nproliferation, migration, and invasion, and increased the\nsensitivity to platinum in ovarian cancer through in-\ncreased expression of hCTR1 [42]. We presume that the\noverexpression of IMP3 can increase the resistance to\nplatinum, which may lead to poor prognosis in ovarian\nCCC, but further study is required.\nConclusion\nIn conclusion, we demonstrated that tumor stage and\nIMP3 expression are prognostic indicator in ovarian\nCCC. IMP3 overexpression is a potential marker of poor\nprognosis for ovarian CCC, even in stage I. More studies\nare required to further clarify the mechanism of IMP3\nexpression and its significance in ovarian CCC.\nCompeting interests\nThe authors declare that they have no competing interest.\nAuthors’ contributions\nRB and WY designed and revised the paper. WZ and XC carried out the\nimmunohistochemical staining. ZF collected the clinical and follow-up\ninformation and drafted the paper. RB, XS, YC and WY diagnosed the cases.\nAll authors read and approved the final manuscript.\nAcknowledgements\nThis work was supported by the key project of Science and Technology\nCommission of Shanghai Municipality (12411950300) for Xiaohua Wu.\nAuthor details\n1Department of Pathology, Fudan University Shanghai Cancer Center,\nShanghai 200032, China. 2Department of Oncology, Shanghai Medical\nCollege, Fudan University, Shanghai 200032, China. 3Psycho-Oncology\nResearch & Training (CePORT), School of Public Health, The University of\nHong Kong, Pok Fu Lam, Hong Kong. 4Department of Gynecologic\nOncology, Fudan University Shanghai Cancer Center, Shanghai 200032,\nChina.\nReceived: 12 November 2015 Accepted: 14 January 2016\nReferences\n1. Anglesio MS, Carey MS, Kobel M, Mackay H, Huntsman DG. Clear cell\ncarcinoma of the ovary: a report from the first Ovarian Clear Cell\nSymposium, June 24th, 2010. Gynecol Oncol. 2011;121:407 –15.\n2. Han G, Gilks CB, Leung S, Ewanowich CA, Irving JA, Longacre TA, et al.\nMixed ovarian epithelial carcinomas with clear cell and serous components\nare variants of high-grade serous carcinoma: an interobserver correlative and\nimmunohistochemical study of 32 cases. Am J Surg Pathol. 2008;32:955–64.\n3. Sugiyama T, Kamura T, Kigawa J, Terakawa N, Kikuchi Y, Kita T, et al.\nClinical characteristics of clear cell carcinoma of the ovary: a distinct\nhistologic type with poor prognosis and resistance to platinum-based\nchemotherapy. Cancer. 2000;88:2584 –9.\n4. del Carmen MG, Birrer M, Schorge JO. Clear cell carcinoma of the ovary:\na review of the literature. Gynecol Oncol. 2012;126:481 –90.\n5. Lee YY, Kim TJ, Kim MJ, Kim HJ, Song T, Kim MK, et al. Prognosis of ovarian\nclear cell carcinoma compared to other histological subtypes: a meta-\nanalysis. Gynecol Oncol. 2011;122:541 –7.\n6. Chan JK, Teoh D, Hu JM, Shin JY, Osann K, Kapp DS. Do clear cell ovarian\ncarcinomas have poorer prognosis compared to other epithelial cell types?\nA study of 1411 clear cell ovarian cancers. Gynecol Oncol. 2008;109:370 –6.\n7. Yamamoto S, Tsuda H, Shimazaki H, Takano M, Yoshikawa T, Kuzuya K, et al.\nHistological grading of ovarian clear cell adenocarcinoma: proposal for a\nsimple and reproducible grouping system based on tumor growth\narchitecture. Int J Gynecol Pathol. 2012;31:116 –24.\n8. Jeng YM, Wang TH, Lu SH, Yuan RH, Hsu HC. Prognostic significance of\ninsulin-like growth factor II mRNA-binding protein 3 expression in gastric\nadenocarcinoma. Br J Surg. 2009;96:66 –73.\n9. Yuan RH, Wang CC, Chou CC, Chang KJ, Lee PH, Jeng YM. Diffuse\nexpression of RNA-binding protein IMP3 predicts high-stage lymph node\nmetastasis and poor prognosis in colorectal adenocarcinoma. Ann Surg\nOncol. 2009;16:1711–9.\n10. Schaeffer DF, Owen DR, Lim HJ, Buczkowski AK, Chung SW, Scudamore CH,\net al. Insulin-like growth factor 2 mRNA binding protein 3 (IGF2BP3)\noverexpression in pancreatic ductal adenocarcinoma correlates with poor\nsurvival. BMC Cancer. 2010;10:59.\n11. Jeng YM, Chang CC, Hu FC, Chou HY, Kao HL, Wang TH, et al. RNA-binding\nprotein insulin-like growth factor II mRNA-binding protein 3 expression\npromotes tumor invasion and predicts early recurrence and poor prognosis\nin hepatocellular carcinoma. Hepatology. 2008;48:1118 –27.\n12. Kobel M, Xu H, Bourne PA, Spaulding BO, Shih Ie M, Mao TL, et al. IGF2BP3\n(IMP3) expression is a marker of unfavorable prognosis in ovarian carcinoma\nof clear cell subtype. Mod Pathol. 2009;22:469 –75.\n13. Fadare O, Liang SX, Crispens MA, Jones 3rd HW, Khabele D, Gwin K, et al.\nExpression of the oncofetal protein IGF2BP3 in endometrial clear cell\ncarcinoma: assessment of frequency and significance. Hum Pathol.\n2013;44:1508 –15.\n14. Wei Q, Yan J, Fu B, Liu J, Zhong L, Yang Q, et al. IMP3 expression is\nassociated with poor survival in cervical squamous cell carcinoma. Hum\nPathol. 2014;45:2218 –24.\n15. Noske A, Faggad A, Wirtz R, Darb-Esfahani S, Sehouli J, Sinn B, et al. IMP3\nexpression in human ovarian cancer is associated with improved survival.\nInt J Gynecol Pathol. 2009;28:203 –10.\n16. Mutch DG, Prat J. 2014 FIGO staging for ovarian, fallopian tube and\nperitoneal cancer. Gynecol Oncol. 2014;133:401 –4.\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 7 of 8\n\n17. DeLair D, Oliva E, Kobel M, Macias A, Gilks CB, Soslow RA. Morphologic\nspectrum of immunohistochemically characterized clear cell carcinoma of\nthe ovary: a study of 155 cases. Am J Surg Pathol. 2011;35:36 –44.\n18. Walter O, Prasad M, Lu S, Quinlan RM, Edmiston KL, Khan A. IMP3 is a novel\nbiomarker for triple negative invasive mammary carcinoma associated with\na more aggressive phenotype. Hum Pathol. 2009;40:1528 –33.\n19. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L,\net al. New guidelines to evaluate the response to treatment in solid tumors.\nEuropean Organization for Research and Treatment of Cancer, National\nCancer Institute of the United States, National Cancer Institute of Canada.\nJ Natl Cancer Inst. 2000;92:205 –16.\n20. Behbakht K, Randall TC, Benjamin I, Morgan MA, King S, Rubin SC. Clinical\ncharacteristics of clear cell carcinoma of the ovary. Gynecol Oncol. 1998;70:255–8.\n21. Pather S, Quinn MA. Clear-cell cancer of the ovary-is it chemosensitive?\nInt J Gynecol Cancer. 2005;15:432 –7.\n22. Takano M, Kikuchi Y, Yaegashi N, Kuzuya K, Ueki M, Tsuda H, et al. Clear cell\ncarcinoma of the ovary: a retrospective multicentre experience of 254\npatients with complete surgical staging. Br J Cancer. 2006;94:1369 –74.\n23. Jenison EL, Montag AG, Griffiths CT, Welch WR, Lavin PT, Greer J, et al.\nClear cell adenocarcinoma of the ovary: a clinical analysis and comparison\nwith serous carcinoma. Gynecol Oncol. 1989;32:65 –71.\n24. Kobel M, Kalloger SE, Huntsman DG, Santos JL, Swenerton KD, Seidman JD,\net al. Differences in tumor type in low-stage versus high-stage ovarian\ncarcinomas. Int J Gynecol Pathol. 2010;29:203 –11.\n25. Ma SK, Zhang HT, Wu LY, Liu LY. Prognostic analysis of 88 patients with\novarian clear cell carcinoma. Zhonghua Zhong Liu Za Zhi. 2007;29:784 –8.\n26. Ye S, You Y, Yang J, Cao D, Bai H, Huang H, et al. Comparison of pure and\nmixed-type clear cell carcinoma of the ovary: a clinicopathological analysis\nof 341 Chinese patients. Int J Gynecol Cancer. 2014;24:1590 –6.\n27. Higashi M, Kajiyama H, Shibata K, Mizuno M, Mizuno K, Hosono S, et al.\nSurvival impact of capsule rupture in stage I clear cell carcinoma of the ovary\nin comparison with other histological types. Gynecol Oncol. 2011;123:474–8.\n28. Mizuno M, Kikkawa F, Shibata K, Kajiyama H, Suzuki T, Ino K, et al. Long-term\nprognosis of stage I ovarian carcinoma. Prognostic importance of\nintraoperative rupture. Oncology. 2003;65:29 –36.\n29. Hoskins PJ, Le N, Gilks B, Tinker A, Santos J, Wong F, et al. Low-stage\novarian clear cell carcinoma: population-based outcomes in British\nColumbia, Canada, with evidence for a survival benefit as a result of\nirradiation. J Clin Oncol. 2012;30:1656 –62.\n30. Rauh-Hain JA, Winograd D, Growdon WB, Schorge JO, Goodman AK,\nBoruta DM, et al. Prognostic determinants in patients with uterine and\novarian clear carcinoma. Gynecol Oncol. 2012;125:376 –80.\n31. Bennett JA, Dong F, Young RH, Oliva E. Clear cell carcinomas of the ovary.\nEvaluation of prognostic parameters based on a clinicopathologic analysis\nof 100 cases. Histopathology. 2015;66:808 –15.\n32. Scarfone G, Bergamini A, Noli S, Villa A, Cipriani S, Taccagni G, et al.\nCharacteristics of clear cell ovarian cancer arising from endometriosis:\na two center cohort study. Gynecol Oncol. 2014;133:480 –4.\n33. Davis M, Rauh-Hain JA, Andrade C, Boruta 2nd DM, Schorge JO, Horowitz\nNS, et al. Comparison of clinical outcomes of patients with clear cell and\nendometrioid ovarian cancer associated with endometriosis to papillary\nserous carcinoma of the ovary. Gynecol Oncol. 2014;132:760 –6.\n34. Orezzoli JP, Russell AH, Oliva E, Del Carmen MG, Eichhorn J, Fuller AF.\nPrognostic implication of endometriosis in clear cell carcinoma of the ovary.\nGynecol Oncol. 2008;110:336 –44.\n35. Slipicevic A, Oy GF, Askildt IC, Holth A, Hellesylt E, Florenes VA, et al.\nDiagnostic and prognostic role of the insulin growth factor pathway\nmembers insulin-like growth factor-II and insulin-like growth factor binding\nprotein-3 in serous effusions. Hum Pathol. 2009;40:527 –37.\n36. Takada T, Iwase H, Iitsuka C, Nomura H, Sakamoto K, Omatsu K, et al.\nAdjuvant chemotherapy for stage I clear cell carcinoma of the ovary:\nan analysis of fully staged patients. Int J Gynecol Cancer. 2012;22:573 –8.\n37. Mizuno M, Kajiyama H, Shibata K, Mizuno K, Yamamuro O, Kawai M, et al.\nAdjuvant chemotherapy for stage i ovarian clear cell carcinoma: is it\nnecessary for stage IA? Int J Gynecol Cancer. 2012;22:1143 –9.\n38. Lu D, Yang X, Jiang NY, Woda BA, Liu Q, Dresser K, et al. IMP3, a new\nbiomarker to predict progression of cervical intraepithelial neoplasia into\ninvasive cancer. Am J Surg Pathol. 2011;35:1638 –45.\n39. Samanta S, Sharma VM, Khan A, Mercurio AM. Regulation of IMP3 by EGFR\nsignaling and repression by ERbeta: implications for triple-negative breast\ncancer. Oncogene. 2012;31:4689 –97.\n40. Kabbarah O, Nogueira C, Feng B, Nazarian RM, Bosenberg M, Wu M, et al.\nIntegrative genome comparison of primary and metastatic melanomas.\nPLoS One. 2010;5:e10770.\n41. Pei X, Li M, Zhan J, Yu Y, Wei X, Guan L, et al. Enhanced IMP3 Expression\nActivates NF-κB Pathway and Promotes Renal Cell Carcinoma Progression.\nPLoS One. 2015;10:e0124338.\n42. Hsu KF, Shen MR, Huang YF, Cheng YM, Lin SH, Chow NH, et al.\nOverexpression of the RNA-binding proteins Lin28B and IGF2BP3 (IMP3) is\nassociated with chemoresistance and poor disease outcome in ovarian\ncancer. Br J Cancer. 2015;113:414 –24.\n•  We accept pre-submission inquiries \n  Our selector tool helps you to find the most relevant journal\n  We provide round the clock customer support \n  Convenient online submission\n  Thorough peer review\n  Inclusion in PubMed and all major indexing services \n  Maximum visibility for your research\nSubmit your manuscript at\nwww.biomedcentral.com/submit\nSubmit your next manuscript to BioMed Central \nand we will help you at every step:\nBi et al. Diagnostic Pathology  (2016) 11:17 Page 8 of 8","source_license":"CC0","license_restricted":false}