Journal of Cancer Research and Therapeutics Medknow Publications on behalf of the Association of Radiation Oncologists of India (AROI) ISSN: 0973-1482 EISSN: 1998-4138 Vol. 7, Num. 1, 2011, pp. 23-28

Journal of Cancer Research and Therapeutics, Vol. 7, No. 1, January-March, 2011, pp. 23-28

Original Article

Immunohistochemical expression of progesterone receptors in pleomorphic adenoma and adenoid cystic carcinoma

Seifi Safoura¹, Eslami Mohamad², Eshghyar Nosrat-Allah², Ensani Fereshte³

¹ Department of Oral and Maxillofacial Pathology, School of Dentistry, Babol University of Medical Sciences, Babol, Iran
² Department of Oral and Maxillofacial Pathology, School of Dentistry, Babol University of Medical Sciences, Babol; Department of Oral and Maxillofacial Pathology, School of Dentistry, Tehran University of Medical Sciences, Dental Research Center, Tehran, Iran
³ Department of Pathology, School of Medicine, Tehran University of Medical Sciences, Tehran, Iran

Correspondence Address: Seifi Safoura, Department of Oral and Maxillofacial Pathology, Babol University of Medical Sciences, Babol, Iran, sf_seify@yahoo.cim

Code Number: cr11006

PMID: 21546738

DOI: 10.4103/0973-1482.80440

Abstract

Keywords: Adenoid cystic carcinoma, pleomorphic adenoma, progesterone

Introduction

The biological knowledge of salivary gland tumors (SGTs) plays a crucial role in their diagnosis, treatment and prognosis. The evaluation of the expression of different proteins can reflect facts about the biology and behavior of a tumor. ^[1]

Many studies have already reported the role of hormone treatment in melanoma, pancreas, breast cancer (carcinoid tumors), and kidney. ^{[2],[3],[4],[5]} Earlier studies on SGTs, however, have reported conflicting results. Some reported that hormone therapy was effective. ^{[1],[2],[3],[4],[5],[6],[7]} Some reported the role of progesterone receptor (PR) in the development and pathogenesis of SGT, ^[8],[9],[10] while other results did not support the application of hormone therapy. ^{[11],[12],[13]}

The present study aimed to assess the expression of progesterone marker in two common SGTs, benign pleomorphic adenoma (PA) and adenoid cystic carcinoma (ACC). We studied its expression in normal salivary glands. In addition, the association of progesterone expression with grade 2 and 3 ACC was studied.

Materials and Methods

The study was approved by local ethics committee at Tehran University of Medical Sciences.

In this retrospective, descriptive, analytical, cross-sectional study, paraffin blocks of 15 cases of ACC and 14 cases of PA were retrieved from the archive samples of School of Dentistry, Tehran University of Medical Sciences and Danesh Laboratory.

Based on the standard protocol of immunohistochemical staining, samples with bleeding, necrosis, unsuitable fixation, inflammation and samples unsuitable for quantitative analysis were excluded. The paraffin blocks of the selected ones were then retrieved from the archive and clinical information including age, sex, and lesion area was obtained from patients′ medical records.

Sections of 4 μm thickness were cut from paraffin-embedded tissue for Hematoxylin-Eosin staining and rechecked by an oral pathologist. Three micrometer sections were then prepared for immunohistochemical staining.

Immunohistochemical staining protocol

In summary, immunohistochemical staining was performed as follows.

The tissues were deparaffinized by xylene and dehydrated through graded concentrations of alcohol. Based on the manufacturer′s recommendation (Dako, Glostrup, Denmark), the slides were transferred to citrate HCl buffer (pH 6.0, 10 μL) for antigen retrieval, and heated for 10 min in a microwave oven (Panasonic 1380 W, 2 atm, 121°C). The slides were washed in phosphate buffered saline (PBS) and incubated with a progesterone antibody (1/100 dilution) for 1 hour. The slides were then washed again in PBS and incubated with biotinylated antibody for 30 min and then were washed in PBS. Finally, the slides were incubated with peroxidase labeled streptavidin for 30 min and washed in PBS. Subsequently, chromogene 3,3-diaminobenzidine (D4293; Sigma, USA) was applied for antibody staining (brown). The samples were then allowed to react with Meyer′s hematoxylin (Merck, Germany) for 30 min and dehydrated and covered with cover glass. Positive (invasive ductal carcinoma of breast) and negative (omission of the primary antibody and its replacement with saline) controls were included in all reactions.

Cell counting was performed using a conventional light microscope equipped with an ocular reticle (Olympus BX41, Olympus optical, Tokyo, Japan, Olympus Corporation Co) with 400× magnification, assessing the nuclei of 100 cells randomly selected from five distinct regions per slide. The percentage of stained cells was expressed as labeling index. Immunostaining was defined by a dark brown stain confined exclusively to the nucleus and was scored as negative when seen in fewer than 5% of the tumor cell nuclei and positive when seen in more than 5% of the tumor cell nuclei. ^[14]

In addition to tumor parenchymal cells, stromal cells (fibroblasts), inflammatory cells, vascular endothelia cells, epithelial cells around tumor, and surrounding salivary glands including acinar and ductal cells were evaluated for PR. The correlation between ACC grade and the expression of PR was also evaluated. t-Test and Mann-Whitney test were used for the statistical analysis of the results. The statistical significance level was set at 0.05.

Results

Clinical results

Of 14 PA cases, 8 were females (57%) and 6 were males (43%). The mean age of the subjects was 43.6 ± 16.2 years (range 23-72 years). There were 6 (42%) cases of major salivary glands (parotid) and 8 (57%) cases of minor salivary glands (4 cases in palate, 3 cases in maxilla and 1 case in buccal mucosa). Of the 15 cases of ACC, 5 patients were females and 10 (66/7%) were males. The mean age of the subjects was 48.5 ± 6.2 years. There were 2 (13.4%) cases of major salivary glands (parotid) and 13 (86.7%) cases of minor salivary glands (4 cases in palate, 1 case on the floor of the mouth, 1 case in mandibular symphysis, and 1 case in maxillary sinuses).

According to Neviile et al., ^[15] ACC cases were classified as 7 cases of grade 2 tumors (cribriform), 8 cases of grade 3 tumors (solid), and 5 cases of minor salivary glands were observed around tumors.

Immunohistochemical findings

Oral squamous epithelial cells adjacent to tumors

In four samples, some pieces of oral squamous cells were observed (three cases of ACC and one case of PA). Only in two cases of ACC, the nuclei in basal layer and some in parabasal layer were lightly or moderately stained with progesterone marker.

Stromal cells (fibroblasts)

Weak immunostaining of stromal cells was observed in three cases of PA. Only one case of severe immunostaining was observed. Weak immunostaining was observed in three cases of ACC.

Inflammatory cells

Of the 29 studied cases of PA and ACC, 28 did not show any immunostaining with progesterone marker. Only in one case of PA, moderate to severe immunoreactivity was observed.

Vascular endothelial cells

Of all 29 PA and ACC tumors, 28 were immunonegative, but moderate to strong immunoreactivity was noted in vascular endothelial cells of one case of PA.

Salivary glands

Salivary gland tissue adjacent to the tumors was observed in 12 cases (7 cases of PA and 5 cases of ACC). Weak to moderate immunoreactivity to PR was observed in nuclei of salivary ducts and acini cells [Figure - 1] and [Figure - 2].

Immunohistochemical study of the tumoral component in PA and ACC

Of the 14 cases of PA, strongly positive staining was observed in nuclei of tumoral cells of 1 case [Figure - 3] and [Figure - 4]. As it can be seen, stromal fibroblasts, inflammatory cells and vascular endothelial cells demonstrated moderate to strong immunostaining.

No immunoreactivity to the marker was observed in the other 13 cases of PA [Figure - 5]. No positive immunostaining with PR was observed in 15 cases of ACC [Figure - 6].

The correlation between PR expression and the grade of ACC

Of the 15 ACC cases, 7 were of grade 2 and 8 were of grade 3. Because no immunoreactivity was observed in any grade, no correlation between PR expression and the grade of ACC was found.

Discussion

Only in one of the PA cases, strong positive immunoreactivity was observed in tumoral, stromal, inflammatory and vascular endothelial cells, and no PR immunoreactivity were observed in ACC cases. The importance of hormone and endocrine therapy in ACC was reported by Ozono et al. ^[6] and Shick et al.,^[6] which is not in agreement with our observations. However, the results of Dori et al.^[12] and Pires et al.^[16] did not support a role for hormone therapy in salivary gland ACC and its tumorigenesis, which supports our findings.

Teymortash et al. studied the estrogen and progesterone receptors in PA, Warthin′s tumor (cystadenolymphomas) and surrounding normal salivary glands. They reported that progesterone positive cells could only be seen in the surrounding normal salivary glands and in Warthin′s tumor and there was no evidence for the presence of estrogen receptor in any of the studied samples. They reported that the evidence of PRs in Warthin′s tumor may implicate a potential role of endocrine factors in the development of the tumors. However, they did not propose a role for it in salivary ACC. ^[8] The latter result is in agreement with our findings.

The evaluation of some other studies also did not provide an evidence for the role of estrogen and progesterone in benign and malignant SGTs. However, possible role for androgen has been reported in tumorigenesis of some SGTs, especially the malignant types. ^[17]

Gias et al. studied the expression of estrogen and progesterone markers in patients with recurrent PA. They found low expression of estrogen receptor in all groups, but progesterone immunoreactivity in the recurrent PA group was higher (96%) compared within the control group (the PA group). They concluded that progesterone seems to be a prognostic factor in recurrent PA of the parotid gland and can be considered a potential target for hormone treatment. ^[7] Previous studies did not look at the association between progesterone and the grade of ACC. However, in the present study, no association was found between progesterone expression and the grade of ACC and progesterone expression was negative in grades 2 and 3.

Larbcharoensub et al. studied the expression of hormonal receptor in three cases with metastasizing PA of the major salivary gland. They observed positive immunohistochemical reactivity to PR but negative reactivity to estrogen receptor in all three cases. ^[10]

In order to clarify the role of the PR on cell proliferation, Yoshimura et al. used human ACC cell line established from the salivary gland. They reported that after introducing progesterone into the ACC cells, it markedly inhibited the proliferative activity of the cells. They showed that the inhibitory effect on cell proliferation was accompanied by p21 upregulation, and Id1 and c-myc downregulation. They suggested that PR reintroduction therapy might be a viable method of inhibiting human SGT progression ^[18]

The conflicting results of the previous studies can be due to the lack of a definite criterion for interpreting the immunohistochemical results.

In 2010, Tarakji et al. scored the estrogen and progesterone SGTs positive if greater than 75% of nuclei showed the staining. Some other studies scored the results between 0 and 4, where scores 0-3 were negative and score 4 was positive . ^[13] Nasser et al. also scored immunostaining as negative when seen in fewer than 5% of the tumor cell nuclei, as weak when 5-20% of tumor cell nuclei showed staining, as moderate when 21-50% of tumor cell nuclei showed staining and as strong when more than 50% of tumor cell nuclei were positive. ^[17]

When the nuclei of tumor cells were not stained by PRs, it was scored negative by Ozono et al. ^[1] Positive staining in this study was defined when seen in more than 5% of the tumor cell nuclei, which is the same criterion for breast carcinoma. ^[14] It also should be considered that in addition to what is mentioned above, other factors could also affect the results. Staining method, the applied antibody and its sensitivity, the duration of the primary fixation of paraffin blocks, age and sex of the subjects, the volume of the sample, and the study method [immunohistochemistry (IHC), polymerase chain reaction (PCR)] could have affected the results.

Other researches reported that estrogen and progesterone receptors are sensitive to proteolytic enzymes and heat; so, the negative staining of some cases could probably have resulted from the destruction of receptors during the laboratory process. ^[19]

Two hypotheses can be discussed regarding the negative results of PA and ACC. They are as follows:

1. The neoplastic process and cancer is a complicated process and different factors are involved in its formation. It is possible that tumor cells lose this marker as they become tumoral. This hypothesis is in agreement with Taymoortash et al.′s theory. ^[8]
2. It is also possible that PRs existed in the selected PA and ACC cases of this study. However, the concentration and density of PRs was below the threshold for immunohistochemical method. Further detailed immunohistochemical studies accompanied by other methods such as reverse transcriptase-polymerase chain reaction (RT-PCR) or in situ hybridization may reveal the immunoreactivity to this marker. In the present study, the existence of PRs in salivary glands surrounding PA and ACC tumors can prove their biological role and also the regulatory role of progesterone for water and electrolytes in salivary glands. Previous studies also confirmed this role. ^[19] In addition to the neurotransmitter regulation of secretion by salivary glands, hormones also may affect the regulation at different levels. Hormones like progesterone, estrogen and androgens can alter the salivary composition. However, they cannot initiate salivary secretion by themselves. ^[20] In vascular endothelial cells of all 29 PA and ACC tumors studied, 28 were immunonegative, but moderate to strong immunoreactivity was noted in 1 case of PA. The results of this study were not in agreement with those of the previous studies which showed that progesterone can induce the protein synthesis in endothelial cells and increase the permeability of vessels in gingiva. ^[19] There are some evidences that inflammatory cells can metabolize progesterone, and some other studies have reported that in comparison to healthy gingiva, the metabolism of androgen and estrogen increased in inflamed gingiva. ^[20] Of the 29 studied cases of inflammatory cells of stromal tumors, 28 cases did not show any immunostaining and only in 1 case of PA moderate to severe immunoreactivity was observed. The binding of steroid hormone to the receptor and the formation of hormone-receptor complex, its relocation to nucleolus increases the DNA and RNA synthesis and ultimately results in hypertrophy and hyperplasia. ^[21] Previous studies showed that there was higher expression of hormone receptors in patients with inflamed gingiva resulting from phenytoin. ^[22] Muscular radioactive progesterone can cause the accumulation of substance in stromal cells of the pulpal tissues. ^[23] In this study, positive progesterone staining of stromal fibroblasts was observed in only three cases of ACC and one case of PA.

Although due to the present limitations only PRs were studied and no progesterone expression was observed in PA and ACC, Ito et al. studied all the estrogen and androgen receptors in both benign and malignant tumors of the salivary glands. In all the cases, estrogen and progesterone were negative and only a low percentage of SGTs was stained with androgens. They reported that androgen may not have a role in a small subgroup of SGTs. ^[24] A summary of the evaluation of the previous studies is given in [Table - 1].

Thus, it can be concluded that no PR staining was observed in 28 cases out of 29 tumor samples. It seems that progesterone probably has no role in tumorigenesis, prognosis and hormone therapy of ACC and PA. Further studies of more cases, using the immunohistochemical method along with in situ hybridization or RT-PCR, may reveal more detailed results.

Acknowledgement

The authors of this article would like to thank all the colleagues in Danesh laboratory in Tehran for their cooperation.

References

Copyright 2011 - Journal of Cancer Research and Therapeutics

The following images related to this document are available:

Photo images