Secondary Logo

Journal Logo

Role of combination of insulin-like growth factor II messenger RNA-binding protein 3 and claudin-1 immunostaining in differentiation between endometrial endometrioid carcinoma and uterine serous carcinoma

Shamloula, Maha M.a; Mohamed, Dareen A.a; El-Dorf, Aymanb

doi: 10.1097/01.XEJ.0000436660.83774.67
ORIGINAL ARTICLES
Free

Background and objectives Endometrial carcinoma is a common cancer among women. The endometrioid type of carcinoma usually has a better prognosis than uterine serous carcinoma (USC). Diagnosis may be challenging as not all USCs have a papillary architecture, and some endometrioid adenocarcinomas (EACs) may manifest as solid sheets of tumor cells, especially of high grade. The aim of this study was to evaluate the combination of insulin-like growth factor II messenger RNA-binding protein 3 (IMP3) and claudin-1 immunostaining in differentiation between both tumors.

Materials and methods Histopathological evaluation of 86 archival EAC, 20 USC, 10 endometrial hyperplasia, eight atypical endometrial hyperplasia, and two endometrial intraepithelial carcinoma patients was carried out. Immunohistochemical evaluation of IMP3 and claudin-1 expression was carried out by progesterone (PR) and P53 immunostaining in challenging cases.

Results There was a significant difference between IMP3 and claudin-1 in EAC patients, whereas all patients with USC were IMP3+/claudin-1+. In one of the patients with a challenging case of EAC, the diagnosis was changed to USC as the patient was IMP3+/claudin-1+, as well as P53+/PR−. All challenging cases of USC were IMP3+/claudin-1+ and P53+/PR−; hence, their diagnosis was confirmed. Both endometrial intraepithelial carcinoma patients were IMP3+/claudin-1+, whereas all nonatypical endometrial hyperplasia patients were IMP3−/claudin-1−.

Conclusion Combined use of IMP3 and claudin-1 is useful in differentiating between EAC and USC, especially in diagnostically challenging cases.

Departments of aPathology

bGynecology and Obstetrics, Faculty of Medicine, Tanta University, Tanta, Egypt

Correspondence to Maha M. Shamloula, Department of Pathology, Faculty of Medicine, Tanta University, 31111 Tanta, Egypt Tel: +20 100 057 1583; fax: +0403337544; e-mail: mshamloula@gmail.com

Received September 1, 2013

Accepted September 15, 2013

Back to Top | Article Outline

Introduction

Worldwide, endometrial cancer is the seventh most common type of malignant tumor; however, its incidence varies among regions. In less developed countries, risk factors are less common and endometrial cancer is rare, although specific mortality is higher. The incidence in North America and Europe is 10 times higher than in less developed countries; in these regions, endometrial cancer is the commonest cancer of the female genital tract and the fourth commonest cancer after breast, lung, and colorectal cancers (Amant et al., 2005).

Carcinoma of the endometrium is uncommon among women younger than 40 years of age. The peak incidence is in the 55–65-year age group. A higher frequency of this form of neoplasia is seen among obese, diabetic, and infertile women. Women who develop cancer of the endometrium are generally single, nulliparous, and with a history of functional menstrual irregularities consistent with anovulatory cycles (Linkov et al., 2008).

It is well recognized that endometrial carcinomas fall into two major subdivisions: well or moderately differentiated estrogen-dependent subtypes that tend to have a good prognosis (type I), such as endometrioid and mucinous carcinomas, and aggressive non-estrogen-dependent subtypes, such as clear-cell and serous categories (type II; Amant et al., 2005).

Back to Top | Article Outline

Type I

This type constitutes the majority of sporadic endometrial carcinomas (∼70–80%). These carcinomas follow the estrogen-related pathway. They arise in the background of unopposed estrogenic stimulation, as they are associated with endometrial hyperplasia, and express estrogen receptors (ERs) and progesterone receptors (PRs). Histologically, most tumors show an endometrial differentiation and are of low grade. The rare mucinous adenocarcinomas are also considered to be type I carcinomas, as they usually express ER and/or PR and are of low histopathological grade. However a subset of low-grade endometrial carcinomas with ER and PR expression, clearly type I carcinomas, seem unrelated to estrogen in the background of the atrophic endometrium (Nofech-Mozes et al., 2008).

Back to Top | Article Outline

Type II

This type constitutes about 10–20% of endometrial carcinomas. These follow the estrogen unrelated pathway and arise in the background of the atrophic endometrium. These tumors usually occur at a higher age, ∼5–10 years later than type I tumors. They are typically high-grade carcinomas of nonendometrioid differentiation, most frequently serous and less frequently clear-cell, small-cell, undifferentiated, and squamous cell carcinomas. ER and PR expression is usually negative or weakly positive (Veral et al., 2002).

Although differentiating between endometrioid adenocarcinomas (EACs) and uterine serous carcinomas (USCs) is usually easily accomplished, there are multiple variations that should be considered. For example, not all USCs have papillary architecture because they may manifest in a glandular or tubuloglandular pattern. Similarly, EACs could manifest with papillary architecture. Finally, high-grade EACs may manifest as solid sheets of tumor cells, and differentiating them from USCs can be very difficult (Silverberg, 2007).

Claudins, the main transmembrane proteins of tight junctions, were discovered in 1998, with 24 currently known members in the human body. Claudins are crucial for the maintenance of cellular polarity and paracellular transportation of molecules. Claudin proteins can be upregulated and mislocalized in cancer cells. Claudins have been suggested to be involved in the growth and differentiation of cells, in carcinogenesis, and in cancer progression in several tumors. Altered expression of claudins (increased or decreased expression) may characterize the invasion and metastasis of different tumors (Hahn et al., 2006).

The expression of claudin-1 increases during tumorigenesis of colon cancer (Dhawan et al., 2005), melanoma (Leotlela et al., 2007), oral squamous cell carcinoma (Dos Reis et al., 2008), and hepatocellular carcinoma (Yoon et al., 2010).

An insulin-like growth factor II messenger RNA-binding protein 3 (IMP3) is expressed in many cells of the developing fetus. In contrast, it is not expressed in most adult tissues except in the gonads. However, IMP3 is expressed in a number of cancers. It has been implicated in tumor growth, migration, and invasion (Yantiss et al., 2005). An antibody specific to IMP3 has been developed and used successfully in routine immunohistochemical methods to study the overexpression of the protein in malignant pancreatic, renal, and uterine endocervical lesions (Jiang et al., 2006).

The aim of this work was to evaluate the combined use of IMP3 and claudin-1 immunostaining in archival cases of USCs, EAC, and endometrial hyperplasia to re-evaluate the histological diagnosis at the background of their expression and to throw light on their relation with tumor grade.

Back to Top | Article Outline

Materials and methods

Case selection and histological review

A retrospective study wascarried out on hysterectomy and endometrial curetting specimens from EAC patients and selected patients with endometrial hyperplasia, operated at the Gynecology and Obstetrics Department, during the period from 2010 to 2012; the specimens were retrieved from the archives of the Department of Pathology, Tanta University.

All cases were revised to confirm the histological subtypes and were divided into the following groups:

  • Group I: 86 EAC patients.
  • Group II: 20 USC patients.
  • Group III: 20 patients with endometrial hyperplasia and endometrial intraepithelial carcinoma (EIC).

The histological grades of endometrial carcinoma patients were determined according to architectural pattern, nuclear grade, and mitotic index (Silverberg et al., 2003; Prat, 2004). All USCs are, by definition, of grade 3.

Back to Top | Article Outline

Immunohistochemical staining

Immunohistochemical staining was performed on a 4-μm thick section retrieved from the archives of the Pathology Department. Briefly, the tissue section was deparaffinized and rehydrated. Slides were incubated in 3% H2O2 for 10 min to reduce nonspecific background staining due to endogenous peroxidase. For epitope retrieval, specimens were heated for 20 min in 10 mmol/l citrate buffer (pH 6.0) in a microwave oven (700 W). After incubating with Ultra V Block (Lab Vision Corporation, Fremont, California, USA) for 7 min at room temperature to block background staining, slides were incubated with a rabbit polyclonal antibody specific to IMP3 (prediluted, Cat No. ab 82044; Abcam, Cambridge Science Park, Cambridge, England) for 1 h at room temperature, and antibody binding was detected on the Ultra Vision LP Detection System (Lab Vision Corporation) according to the manufacturer’s recommendations. Color was developed using 3,3′-diaminobenzidine and by counterstaining with hematoxylin (Song et al., 2011). For negative controls, the primary antibody was replaced with non-immune IgG.

The expression of IMP3 is identified through cytoplasmic staining; the number of positive tumor cells and staining intensity were evaluated by image analysis (Leica Q win software); the cutoff value for positivity for IMP3 was 5%. Scores of 1 and 2represented focal positivity and a score of 3 represented diffuse positivity [1 (6–21% positivity), 2 (21–50% positivity), and 3 (>50% positivity)]. The intensity of the staining was evaluated as weak, intermediate, or strong (Li et al., 2007). Positive controls of pancreatic carcinoma were used to confirm correct immunohistochemical staining.

Similar steps were carried out for claudin-1 staining with the application of ready-to-use rabbit polyclonal anti-claudin-1 antibody (Cat No. RB-9209-R7; Lab Vision Corporation). The expression of claudin-1 is identified through membranous staining. Immunoreactivity was evaluated by image analysis as follows: 0, fewer than 5% tumor cells reactive; 1+, 5--25% tumor cells reactive; 2+, 25--50% tumor cells reactive; and 3+, more than 50% tumor cells reactive. For claudin-1, the intensity of staining also was graded as weak, moderate, or strong (Hahn et al., 2006). Normal skin was used as a positive control for claudin-1 immunostaining.

P53 ready-to-use mouse monoclonal anti-p53 Ab-1 (clone: PAB 240, Cat No. MS104R7; Lab Vision Corporation) and ready-to-use rabbit monoclonal anti-progesterone receptor (PR) antibodies (clone: SP2, Cat No. RM9102RQ; Lab Vision Corporation) were used in similar steps, only in diagnostically challenging cases. The expression was considered positive with nuclear expression for both markers.

Back to Top | Article Outline

Statistical study

Independent sample t-tests and χ2-tests were used to compare mean values of continuous variables in different groups. P less than 0.05 was considered to indicate significance.

Back to Top | Article Outline

Results

Clinicopathologic features

Group I

Group I included 86 patients with EAC. The age of the patients ranged from 46 to 70 years. The median age was 58 years. Histological revaluation revealed 10 patients with difficult diagnosis because the tumor had a papillary architecture, an eosinophilic syncytial pattern with some bizarre cells that can be mistaken for USC (Fig. 1). The histological features that were used to classify them as EACs were cells with low nuclear grade and inconspicuous nucleoli; two of the tumors were associated with atypical endometrial hyperplasia (AEH).

Fig. 1

Fig. 1

The tumors were graded as follows: 33 (38%) of grade 1, 29 (34%) of grade 2, and 24 (28%) of grade 3.

Back to Top | Article Outline

Group II

Group II included 20 patients with USC. The age of the patients ranged from 60 to 85 years, with a median age of 72 years. Histological re-evaluation revealed eight diagnostically challenging cases, because of the glandular architecture, which could be a pitfall in the differential diagnosis with EAC. The histological features used to classify the tumors as USCs were discordance between a predominantly glandular architecture associated with high cytological nuclear grade as defined by bizarre nuclear forms and prominent nucleoli (Fig. 2). All tumors were considered to be of grade 3.

Fig. 2

Fig. 2

Back to Top | Article Outline

Group III

Group III included 20 patients: 10 with non-AEH (cystic and simple endometrial hyperplasia), eight with AEH (Fig. 3), and two with EIC, with atypical nuclei lining the surfaces and glands of the atrophic endometrium. The surface shows a slightly papillary contour and some cells show hobnail morphology and smudged, hyperchromatic nuclei.

Fig. 3

Fig. 3

Back to Top | Article Outline

Immunohistochemical expression

IMP3 expression is shown in Table 1.

Table 1

Table 1

Back to Top | Article Outline

Group I

Among the EAC patients, 72/86 (84%) were negative for IMP3, whereas 14 (16%) showed immunohistochemical expression of the protein (P<0.05): 13/14 patients (93%) showed focal positive staining in 6–21% of tumor cells and 1/14 patients (7%) showed diffuse positivity in more than 50% of the tumor cells. IMP3 staining intensity was strong in 1/14 patients (7%), intermediate in 1/14 patients (7%), and weak in 12/14 patients (86%). Only 1/10 challenging cases of EAC showed strong and diffuse positivity of IMP3 in both papillary and glandular patterns (Fig. 4).

Fig. 4

Fig. 4

Back to Top | Article Outline

Group II

All serous carcinoma patients showed positive cytoplasmic staining for IMP3 (P<0.05): 18/20 patients (90%) showed diffuse positive staining for IMP3 in more than 50% of tumor cells, and 2/20 patients (10%) showed focal staining in 21–50% of the tumor cells. Immunohistochemical staining intensity for IMP3 was found to be strong in 14/20 patients (70%; Fig. 5) and intermediate in 6/20 (30%) patients with serous carcinoma. All challenging cases (8/8) showed strong and diffuse IMP3 positivity.

Fig. 5

Fig. 5

Back to Top | Article Outline

Group III

Among AEH and EIC cases, IMP3 expression was found to be significantly positive in EIC (100%), with strong and diffuse positivity seen in more than 50% of the cells (Fig. 6; P<0.05), whereas 2/8 of AEH cases (25%) showed IMP3 positivity with focal expression in 6-21% of cells: one case showed moderate and the other showed weak expression. All cases (100%) of non AEH showed negative expression of IMP3 (P<0.05).

Fig. 6

Fig. 6

Claudin-1 expression is shown in Table 2.

Table 2

Table 2

Back to Top | Article Outline

Group I

In Group I, 61/86 (71%) EAC patients were negative for claudin-1 expression (Fig. 7), whereas 25/86 (29%) patients showed immunohistochemical expression of the protein (P<0.05). Of the 25 patients, 17 (68%) showed positive staining in 5–25% of tumor cells, five (20%) in 25–50% of the tumor cells, and three (12%) in greater than 50% of the tumor cells. Claudin-1 staining intensity was strong in 1/25 patients (4%), intermediate in 2/25 patients (8%), and weak in 22/25 patients (88%). The same challenging case of EAC that showed strong and diffuse positivity of IMP3 also showed strong and diffuse positivity of claudin-1 in both glandular and papillary patterns.

Fig. 7

Fig. 7

Back to Top | Article Outline

Group II

All of the serous carcinoma patients showed membranous staining for claudin-1 (P<0.05): 16/20 patients (80%) showed positive staining for claudin-1 in greater than 50% of tumor cells (Fig. 8), 3/20 patients (15%) showed staining in 25–50% of tumor cells, and 1/20 patients (5%) showed positive staining in 5–25% of tumor cells. Claudin-1 staining intensity was strong in 18/20 patients (90%) and intermediate in 2/20 patients (10%). All challenging cases were claudin-1 positive (Fig. 9).

Fig. 8

Fig. 8

Fig. 9

Fig. 9

Back to Top | Article Outline

Group III

Claudin-1 expression was positive in all patients with EIC (100%; P<0.05), and positivity of strong staining intensity was seen in greater than 50% of cells (Fig. 10), whereas 3/8 patients (38%) with AEH showed positivity for claudin-1, all whom showed weak staining intensity: one patient showed expression in 25–50% of the cells and two patients showed expression in 5–25% of the cells with no significant differences. All non-AEH patients (100%) showed significant negative expression on claudin-1 immunostaining (P<0.05).

Fig. 10

Fig. 10

Back to Top | Article Outline

P53 and PR expression in diagnostically challenging cases

Among the 10 challenging cases of group I, P53 expression was positive in one case, which showed strong and diffuse positivity for IMP3 and claudin-1, whereas PR immunostaining was negative in this case. In group II all eight challenging cases were positive for P53 immunostaining and negative for PR immunostaining.

On examination of the expression of markers according tumor grade, it was found that IMP3 and claudin-1 expression was significantly correlated with higher tumor grade (Table 3).

Table 3

Table 3

All patients with USC who showed positivity for IMP3 were also positive for claudin-1, whereas among EAC patient, of the 14 patients with IMP3 positivity, only two patients (8%) showed claudin-1 positivity and 12 patients were negative for claudin-1, representing 20% of claudin-1 negative patients, with no significant differences between the groups (Table 4).

Table 4

Table 4

Back to Top | Article Outline

Discussion

Endometrial serous carcinoma is the most common aggressive form of endometrial malignancy. It has a strong predilection for extrauterine spread and poor clinical prognosis. The mechanism of its aggressive clinical behavior is unknown, although recent research has focused on the potential role of cellular adhesion molecules in this process (Leblanc et al., 2001).

In this study, we have demonstrated that IMP3 is highly expressed in USC and EIC as compared with EAC (P<0.0001), and expression of IMP3 in the latter form, although low in frequency, is correlated with higher tumor grades, suggesting that IMP3 may be of diagnostic value in differential diagnosis, a marker of tumor aggressive clinical behavior, and important in the pathogenesis of endometrial malignancy.

The expression pattern of IMP3 in USC and EIC patients was diffuse and of strong intensity, whereas staining was typically patchy and of moderate or weak intensity in EAC patients. Non-AEH patients showed 100% negativity for IMP3, and only 25% of AEH patients showed weak and focal positivity. We concluded that expression of IMP3 is closely associated with type II endometrial cancer. Strong and diffuse IMP3 expression is highly sensitive to endometrial USCs, including their putative precursor lesions. Therefore, IMP3 may be a useful diagnostic marker in the assessment of endometrial cancers and their precursor lesions. This finding is the same as that of Li et al. (2007).

These results are of particular interest because IMP3 has been found to be a biomarker associated with the progression of other unrelated cancers. Thus its expression progressively increases with advancing stage in pancreatic cancers (Yantiss et al., 2005). Moreover, in renal cell carcinoma, IMP3 expression helps identify the majority of cases that go on to metastasize and ultimately kill patients (Jiang et al., 2006). Together these findings raise the possibility that IMP3 promotes aggressive behavior in tumors and might explain why most of the more highly aggressive serous carcinomas express this molecule, as well as why it is present more frequently in high-grade compared with low-grade endometrial carcinomas.

The same finding was observed for claudin-1 expression, which had a strong and diffuse expression in USC, and its precursor, with mild and focal expression in high-grade EAC, with significant differences between the two; hence, claudin-1 could differentiate between EAC and USC. This was agreement with the findings of Sobel et al. (2006), who concluded in their study that the two types of EACs were well distinguished by claudins 1 and 2 on immunohistochemical analysis.

In addition, we found that all USC patients who express IMP3 and its precursors also express claudin-1, whereas among EAC patients, 92% of claudin-1-positive patients were significantly negative for IMP3. Hence, it is suspected that claudin-1+/IMP3+ is a good combination for correct diagnosis of USCs, especially if their expression were diffuse and strong; however, this suspicion needs more studies using univariate and multivariate exact logistic regression analyses and follow-up of patients as in the study by Mhawech-Fauceglia et al. (2010).

On reevaluation of the histological diagnosis, there was a misdiagnosed USC case. It was incorrectly diagnosed as EAC but with careful examination there was focal papillary architecture. On immunohistochemical analysis it was strongly and diffusely positive for IMP3 and claudin-1 in both glandular and papillary areas. This case was P53 positive and PR negative.

The archival previously diagnosed challenging USC cases were IMP3 +, claudin-1 +, P53 +, and PR negative. This implies that the presence of papillae and high nuclear grade is still the gold standard in the diagnosis of USC and that the combined usage of IMP3 and claudin-1 is a good confirmatory test.

Back to Top | Article Outline

Acknowledgements

Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline

References

Amant F, Moerman P, Neven P, Timmerman D, Van Limbergen E, Vergote I.Endometrial cancer.Lancet2005;366:491–505.
Dhawan P, Singh AB, Deane NG, No Y, Shiou S-R, Schmidt C, et al..Claudin-1 regulates cellular transformation and metastatic behavior in colon cancer.J Clin Invest2005;115:1765–1776.
Dos Reis PP, Bharadwaj RR, Machado J, MacMillan C, Pintilie M, Sukhai MA, et al..Claudin 1 overexpression increases invasion and is associated with aggressive histological features in oral squamous cell carcinoma.Cancer2008;113:3169–3180.
Hahn HP, Bundock EA, Hornick JL.Immunohistochemical staining for claudin-1 can help distinguish meningiomas from histologic mimics.Am J Clin Pathol2006;125:203–208.
Jiang Z, Chu PG, Woda BA, Rock KL, Liu Q, Hsieh C-C, et al..Analysis of RNA-binding protein IMP3 to predict metastasis and prognosis of renal-cell carcinoma: a retrospective study.Lancet Oncol2006;7:556–564.
Leblanc M, Poncelet C, Soriano D, Walker-Combrouze F, Madelenat P, Scoazec JY, Darai E.Alteration of CD44 and cadherins expression: possible association with augmented aggressiveness and invasiveness of endometrial carcinoma.Virchows Arch2001;438:78–85.
Leotlela PD, Wade MS, Duray PH, Rhode MJ, Brown HF, Rosenthal DT, et al..Claudin-1 overexpression in melanoma is regulated by PKC and contributes to melanoma cell motility.Oncogene2007;26:3846–3856.
Li C, Zota V, Woda BA, Rock KL, Fraire AE, Jiang Z, et al..Expression of a novel oncofetal mRNA-binding protein IMP3 in endometrial carcinomas: diagnostic significance and clinicopathologic correlations.Mod Pathol2007;20:1263–1268.
Linkov F, Edwards R, Balk J, Yurkovetsky Z, Stadterman B, Lokshin A, Taioli E.Endometrial hyperplasia, endometrial cancer and prevention: gaps in existing research of modifiable risk factors.Eur J Cancer2008;44:1632–1644.
Mhawech-Fauceglia P, Herrmann FR, Rai H, Tchabo N, Lele S, Izevbaye I, et al..IMP3 distinguishes uterine serous carcinoma from endometrial endometrioid adenocarcinoma.Am J Clin Pathol2010;133:899–908.
Nofech-Mozes S, Ghorab Z, Ismiil N, Ackerman I, Thomas G, Barbera L, et al..Endometrial endometrioid adenocarcinoma: a pathologic analysis of 827 consecutive cases.Am J Clin Pathol2008;129:110–114.
Prat J.Prognostic parameters of endometrial carcinoma.Hum Pathol2004;35:649–662.
Silverberg SG, Kurman RJ, Nogales F, Mutter GL, Kubik-Huch RATavassoli FA, Devilee P.WHO histological classification of tumours of the uterine corpus.Pathology and genetics of tumours of the breast and female genital organs2003.Lyon, France:IARC Press;221–232.
Silverberg SG.The endometrium: pathologic principles and pitfalls.Arch Pathol Lab Med2007;131:372–382.
Sobel G, Németh J, Kiss A, Lotz G, Szabó I, Udvarhelyi N, et al..Claudin 1 differentiates endometrioid and serous papillary endometrial adenocarcinoma.Gynecol Oncol2006;103:591–598.
Song DH, Yang JW, Kim DC, Lee JS, Lee JH, Jeong SH, Ko GH.Insulin-like growth factor II mRNA-binding protein 3 expression in benign and premalignant lesions and carcinomas of the stomach.Korean J Pathol2011;45:379–385.
Veral A, Zekioǧlu O, Nart D, Çiris M, Özdemir N, Erhan Y.P53, C-ERB, KI-67, estrogen and progesterone receptor expression and histological parameters in types I and II endometrial carcinomas.Ann Saudi Med2002;223–4242–244.
Yantiss RK, Woda BA, Fanger GR, Kalos M, Whalen GF, Tada H, et al..KOC (K homology domain containing protein overexpressed in cancer): a novel molecular marker that distinguishes between benign and malignant lesions of the pancreas.Am J Surg Pathol2005;29:188–195.
Yoon C-H, Kim M-J, Park M-J, Park I-C, Hwang S-G, An S, et al..Claudin-1 acts through c-Abl-protein kinase Cδ (PKCδ) signaling and has a causal role in the acquisition of invasive capacity in human liver cells.J Biol Chem2010;285:226–233.
©2013Egyptian Journal of Pathology