AbstractPurpose To evaluate the expression of c-kit (CD117) inendometrial hyperplasia and endometrial cancer.Methods Expression of c-kit in 10 normal endometrium,18 simple endometrial hyperplasia, 16 complex endome-trial hyperplasia (10 cases with atypia and 6 cases withoutatypia), and 6 endometrial cancer were investigated byimmunohistochemistry.Results c-Kit expression decreased as the lesion pro-gressed to endometrial cancer. Immunostaining was mostlyfocal and weak in the normal endometrium and was mostlydiVuse and strong in the simple and complex endometrialhyperplasia.Conclusions Simple and complex hyperplastic endome-trial tissues express diVuse cytoplasmic staining for c-kitand the expression decreases with the progression of thelesion.
Endometrial hyperplasia is related to prolonged estrogenstimulation in association with a diminished to absent pro-gestational activity. The World Health Organization, basedon previous studies by Kurman et al., has accepted a classi-Wcation that includes four diagnostic categories: simplehyperplasia, complex hyperplasia, simple atypical hyper-plasia, and complex atypical hyperplasia [1]. Hyperplasticglands may retain some of their architectural abnormalities,but the glandular cells show attenuation [2]. The proto-oncogene c-kit is the cellular homologue of the oncogenev-Kit of HZ4 feline sarcoma virus. It is located on chromo-some 4 (4q11–12) in the human genome [3]. c-Kit encodesa 145-kDa transmembrane tyrosine kinase receptor(CD117), the ligand of which is stem cell factor (SCF)[4, 5].
c-Kit receptor-ligand system has been shown to have animportant role in the embryonic development of migratorycell lineages, including primordial germ cells [4]. It hasbeen suggested that c-kit, together with its ligand SCF, mayalso contribute to tumor development via stimulation ofgrowth in an autocrine and/or paracrine fashion [5]. Suchinteractions between c-kit and SCF have been shown tohave a mitogenic eVect in various tumors including ovary,breast carcinomas, and small cell carcinomas of the lung [6,7]. Expression of the KIT receptor has also been reported innormal tissues of the human female genital tract, includingthe uterus [5] and the ovaries [8]. Several gynecologicalmalignancies, such as endometrial [5] and ovarian cancerappear to produce the KIT receptor. Because of its role inregulating cell diVerentiation and tissue morphogenesis, theloss of c-kit expression may be a required step in malignanttransformation and tumor progression in endometrialhyperplasia. With this study, we sought to evaluate c-kit
E. Yilmaz (&) · O. Celik · Y. Simsek · I. Turkcuoglu · E. CelikDepartment of Obstetrics and Gynecology, Inonu University School of Medicine, 44100 Malatya, Turkeye-mail: [email protected]
M. GülDepartment of Histology and Embryology, Inonu University School of Medicine, Malatya, Turkey
S. HascalikDivision of Obstetrics and Gynecology, Park Hospital, Malatya, Turkey
E. AydinDepartment of Pathology, Inonu University School of Medicine, Malatya, Turkey
123
198 Arch Gynecol Obstet (2012) 286:197–200
expression in endometrium of women with endometrialhyperplasia and endometrial cancer.
Materials and methods
Four-micrometer thick paraYn-embedded sections of for-malin-Wxed tissues from 10 normal endometrium, 18 sim-ple endometrial hyperplasia and 16 complex endometrialhyperplasia (10 cases with atypia and 6 cases without aty-pia), and 6 endometrial cancer cases were immunohisto-chemically analyzed for CD117 expression. Sections weredeparaYnated in xylene and rehydrated through a gradedseries of ethanol solutions. After hydration with phosphate-buVered saline, primary antibody CD117 (ready-to-usemouse monoclonal antibody, Novocastra, Newcastle, UK)was applied for 20 min. Using labeled streptavidin–biotinkit (Ultravision, Lab Vision Corp, Fremont CA, USA)immunostaining was performed. Diaminobenzidine wasused as chromogen, and the sections were counterstainedwith hematoxylin-eosin. Immunostained cytoplasmic mastcells were accepted as positive control group. The immuno-staining was graded as 1+ through 3+ in respect of intensityof staining and as focal (10% of cells), intermediate (>10and <50% of cells), and diVuse (>50% of cells) staining inrespect of distribution of staining. CD117 expression wascompared with univariate analysis of variance (ANOVA)using the Statistical Package for Social Sciences software15.0 for Windows package software (SPSS, Inc., Chicago,IL, USA). Tukey test was used for multiple comparisons ofthe groups. A p value <0.05 was accepted as signiWcant.
Results
The immunohistochemical results for CD117 expression informalin-Wxed, paraYn-embedded hyperplastic endome-trial and endometrial cancer tissue are summarized inTable 1. CD 117 immunostaining was detected in 60% (6/10) of normal endometrial tissue (2 patients +2 intensity, 4patients +1 intensity). In all the cases with simple atypical
hyperplasia (18/18), CD117 staining was observed (6patients +3, 5 patients +2 and 7 patients +1 intensity). How-ever, CD117 staining was observed in 83% (5/6) of thecases with complex non-typical hyperplasia (1 patient +3, 2patients +2 and 2 patients +1 intensity), in 70% (7/10) ofcomplex atypical hyperplasia (2 patients +3, 3 patients +2and 2 patients +1 intensity), and in 16% (1/6) of endome-trial cancer patients (+1 intensity). CD117 expression inpatients diagnosed with simple non-atypical hyperplasia(p = 0.0001) and complex non-atypical hyperplasia(p = 0.025) was signiWcantly higher compared with casesdiagnosed with endometrial cancer. CD117 expression washigher in cases with complex atypical endometrial hyper-plasia, compared with cases with endometrial cancer; how-ever, the diVerence was not statistically signiWcant. Of sixnormal endometrial tissue stained with CD117, staining wasfocal in four cases (66.7%) and intermediate in two (33.3%)cases. Of 15 cases with simple non-atypical endometrialhyperplasia, 12 cases stained diVuse (80%) and 3 casesstained focally (20%). Among 5 complex non-atypical hyper-plasia cases, stained CD117, 4 diVuse (80%) and 1 (20%)staining were observed, whereas 5 diVuse (71.4%), 1 interme-diate (14%), and 1 focal (14%) staining were seen among 7cases with complex atypical hyperplasia (Fig. 1). Excludingone case, CD117 staining could not be demonstrated inpatients with established endometrial cancer (Fig. 2). Inpatients with detected endometrial cancer, staining in cyto-plasmic mast cells was accepted as positive control (Fig. 3).
Discussion
We found the c-kit expression both in normal and hyper-plastic endometrium. Previous studies which used cryopre-served endometrial tissue were not able to detect CD117immunostaining in normal endometrial tissue [9, 10]. How-ever, we and others using formalin Wxed, paraYn embed-ded endometrial tissue detected CD117 immunostaining innormal endometrium [11]. Similar to Elmore et al. [11] wefound the CD117 immunostaining mostly focal and weak innormal endometrium. The non-detection of c-kit expression
Table 1 CD117 expression in simple, complex hyperplastic endometrium and endometrial cancer
Tissue type No. of positive cases (%)
Immunostaining distribution Immunostaining intensity
in the cryopreserved endometrial tissue may be due to diY-culty in the interpretation of the focal and weak immuno-staining in compromised tissue architecture.
In the current study, c-kit expression was higher in nor-mal endometrium and simple endometrial hyperplasia;however, it decreased in the complex endometrial hyperpla-sia. Also the immunostaining was mostly diVuse and strongin the simple and complex endometrial hyperplasia cases.There are no data about the c-kit expression in simple andcomplex endometrial hyperplasia separately; however,Elmore et al. [11] found the c-kit expression to be 57% andimmunostaining mostly diVuse and strong in a pooled sim-ple and complex hyperplasia cases.
c-Kit expression has been found in various normalhuman tissues, including spermatogonia, testicular andovarian interstitial and surface epithelial cells, melanocytes,
tissue mast cells, astrocytes, renal tubules, parotid cells,thyrocytes, dermal sweat glands, esophageal glands, andbreast epithelium [8–10]. Further, the c-kit expression wasdetected in various non-lymphoid solid tumors, includingseminoma, lung tumors, and melanoma of low invasive-ness, cervix and ovarian carcinomas, suggesting a role forc-kit signaling as functional autocrine and/or paracrinegrowth regulation in these tumors [9, 12]. However, severalstudies comparing the c-kit expression between corre-sponding normal and malignant nonlymphoid tissues, spe-ciWcally breast epithelium, melanocytes, and thyroidepithelium have found loss of expression. Natali et al. [13]demonstrated a loss of c-kit expression in breast cancerswhen compared with normal breast epithelium. Similarly,the same investigators demonstrated a loss of expression in
Fig. 1 Immunoreactivity of c-kit in endometrial hyperplasias. a Weakcytoplasmic c-kit immunostaining in the cells of a simple hyperplasia,£400. b Moderate cytoplasmic c-kit staining in a simple hyperplasia,£200. c Strong cytoplasmic c-kit staining in another simple hyperpla-
sia, £200. d Strong cytoplasmic staining in complex hyperplasia withatypia, £200. e Strong cytoplasmic staining in complex hyperplasiawithout atypia, £400
Fig. 2 c-Kit immunoreactivity of endometrial cancer £40 Fig. 3 c-Kit immunoreactivity in cytoplasmic mast cells used as pos-itive control group £20
123
200 Arch Gynecol Obstet (2012) 286:197–200
more deeply invasive melanomas and in papillary and fol-licular thyroid carcinomas as compared with normal thy-roid epithelium [14].
The aforementioned previous studies indicate a dual roleof c-kit in the carcinogenesis, exerting a potent mitogeniceVect in some cell lines and regulation of cell diVerentiationand tissue morphogenesis in others. In our study, lower c-kitexpression was observed in patients diagnosed with endome-trial cancer. However, c-kit expression was high in the sim-ple endometrial hyperplasia and it was decreased in thecomplex endometrial hyperplasia. This Wnding may suggest amitogenic eVect of the c-kit in the initiation of tumorogenesisin the endometrium, but later loss of cell diVerentiation dueto decreased c-kit expression, in the progression of the lesion.A similar mechanism was proposed for the low-invasive anddeeply invasive melanomas, previously [9].
In conclusion, IHC studies with subjective basis areopen to errors. We showed the c-kit expression in simpleand complex endometrial hyperplasia. c-Kit seems to initi-ate endometrial tumorigenesis through mitogenic eVect andlead to lesion progression by dysregulation of cell diVeren-tiation. Further studies evaluating coexpression of c-kit andSCF, dysregulation of c-kit expression, or c-kit gene muta-tions in endometrial hyperplasias are needed to further elu-cidate the role of c-kit in endometrial tumorigenesis.
ConXict of interest The authors declare that they have no conXict ofinterest.
References
1. Tresserra F, Lopez-Yarto M, Grases PJ, Ubeda A, Pascual MA,Labastida R (2003) Endometrial hyperplasia with secretory chang-es. Gynecol Oncol 88:386–393
2. Kurman RJ, Kaminski PF, Norris HJ (1985) The behavior of endo-metrial hyperplasia. A long-term study of “untreated” hyperplasiain 170 patients. Cancer 56:403–412
3. Wang L, Felix JC, Lee JL, Tan PY, Tourgeman DE, O’Meara AT,Amezcua CA (2003) The proto-oncogene c-kit is expressed inleiomyosarcomas of the uterus. Gynecol Oncol 90:402–406
4. McCulloch EA, Minden MD (1993) The cell surface receptor en-coded by the proto-oncogene KIT and its ligand. Cancer Treat Res64:45–77
5. Vliagoftis H, Worobec AS, Metcalfe DD (1997) The protoonco-gene c-kit and c-kit ligand in human disease. J Allergy Clin Immu-nol 100:435–440
6. Williams DE, de Vries P, Namen AE, Widmer MB, Lyman SD(1992) The steel factor. Dev Biol 151:368–376
7. Hibi K, Takahashi T, Sekido Y, Ueda R, Hida T, Ariyoshi Y, Tak-agi H (1991) Coexpression of the stem cell factor and the c-kitgenes in small-cell lung cancer. Oncogene 6:2291–2296
8. Parrott JA, Kim G, Skinner MK (2000) Expression and action ofkit ligand/stem cell factor in normal human and bovine ovariansurface epithelium and ovarian cancer. Biol Reprod 62:1600–1609
9. Natali PG, Nicotra MR, Sures I, Santoro E, Bigotti A, Ullrich A(1992) Expression of c-kit receptor in normal and transformed hu-man nonlymphoid tissues. Cancer Res 52:6139–6143
10. Lammie A, Drobnjak M, Gerald W, Saad A, Cote R, Cordon-Car-do C (1994) Expression of c-kit and kit ligand proteins in normalhuman tissues. J Histochem Cytochem 42:1417–1425
11. Elmore LW, Domson K, Moore JR, Kornstein M, Burks RT(2001) Expression of c-kit (CD117) in benign and malignant hu-man endometrial epithelium. Arch Pathol Lab Med 125:146–151
12. Inoue M, Kyo S, Fujita M, Enomoto T, Kondoh G (1994) Coex-pression of the c-kit receptor and the stem cell factor in gynecolog-ical tumors. Cancer Res 54:3049–3053
13. Natali PG, Nicotra MR, Sures I, Mottolese M, Botti C, Ullrich A(1992) Breast cancer is associated with loss of the c-kit oncogeneproduct. Int J Cancer 52:713–717
14. Natali PG, Berlingieri MT, Nicotra MR, Fusco A, Santoro E, Big-otti A, Vecchio G (1995) Transformation of thyroid epithelium isassociated with loss of c-kit receptor. Cancer Res 55:1787–1791
