The role of angiogenesis and COX-2 expression in the evolution of vulvar lichen sclerosus to squamous cell carcinoma of the vulva

Transcription

1 Gynecologic Oncology 106 (2007) The role of angiogenesis and COX-2 expression in the evolution of vulvar lichen sclerosus to squamous cell carcinoma of the vulva Maria Rosaria Raspollini, Grazia Asirelli, Gian Luigi Taddei Department of Human Pathology and Oncology, School of Medicine University of Florence, Viale G.B. Morgagni, Florence, Italy Received 2 March 2007 Available online 8 June 2007 Abstract Objectives. We aimed to determine whether premalignant changes in vulvar lichen sclerosus (LS) could be identified by analysing markers of angiogenesis and the expression of the enzyme cyclooxygenase-2 (COX-2). Methods. Eight cases of histologically diagnosed vulvar LS, which showed an evolution to carcinoma of the vulva histologically documented, were compared to 10 cases of vulvar LS, for which follow-up information was available for at least 9 years, and to 10 cases of LS adjacent to squamous cell carcinoma (SCC) of the vulva. The microvessel density (MVD), and the expression of vascular endothelial growth factor (VEGF) and of COX-2 were analysed. Results. Difference of MVD between unchanged LS cases and LS cases evolving to SCC and LS adjacent to SCC cases was statistically significant (P = 0.008, Wilcoxon Mann Whitney test). Difference of VEGF and COX-2 expression between unchanged LS cases and LS cases evolving to SCC and LS adjacent to SCC cases were statistically significant (P = and P = 0.01, respectively; Fisher's exact test). Conclusions. Our study addresses the possibility that immunohistochemical studies may add information to permit the identification of LS as a precursor lesion that has a greater potential to evolve into SCC. These data may identify characteristics of vulvar LS disclosing alterations that indicate the further development to cancer; therefore, it may allow the identification of a group of LS patients who need a careful follow-up and adjunctive biopsies Elsevier Inc. All rights reserved. Keywords: Lichen sclerosus; Vulva; Angiogenesis; COX-2; Squamous cell carcinoma Introduction Vulvar carcinogenesis is a multistep process in which clinically recognizable pre-stages occasionally occur. Histogical criteria, such as the presence of epithelial dysplasia are commonly used to predict the behaviour of these lesions, but the assessment of dysplasia, in some cases, is highly subjective [1]. Moreover, the presence of dysplasia does not always indicate malignant transformation and its absence does not preclude it. The immunohistochemical study of precursor lesions may disclose some of the alterations that dictate the development of cancer, independently of recognizable morphological alterations. Genital lichen sclerosus (LS), a chronic inflammatory disease, has also been associated with SCC with a reported Corresponding author. Fax: address: mariarosaria.raspollini@unifi.it (M.R. Raspollini). incidence of 5% of prospective risk of malignant transformation [2,3]. The importance of angiogenesis in tumor progression has been highlighted by studies showing that the angiogenic potential of tumors assessed by microvessel density (MVD) directly correlates with poor prognosis [4]. Among the numerous promoters of angiogenesis, VEGF, a homodimeric, kda, heparin-binding glycoprotein, which stimulates endothelial cell proliferation in vitro and has angiogenic activity in vivo, is one of the most potent [5]. A significant relation between angiogenesis, evaluated with MVD, and the expression of vascular endothelial growth factor (VEGF) has been observed in tumor specimens, with clinical outcome in vulvar squamous cell carcinoma (SCC) patients [6]. At the moment, there are no known lesions linking the normal vulvar epithelium to SCC. In addition, in vulvar pathology the role of angiogenesis is still unclear [7 10], and /$ - see front matter 2007 Elsevier Inc. All rights reserved. doi: /j.ygyno

2 568 M.R. Raspollini et al. / Gynecologic Oncology 106 (2007) there is a lack of studies focusing on variability of angiogenesis, if any, in different vulvar lesions evolving into vulvar SCC. In the last years, studies have been focused on the involvement of cyclooxygenase-2 (COX-2) in critical steps of tumor progression [11]; and recently, the expression of COX-2 has been studied also in vulvar cancer [12,13]. Moreover, several researches have recently highlighted the association the COX-2 over-expression with angiogenesis [14,15]. The aim of this study was to analyse the presence of markers of angiogenesis (MVD and VEGF) and the COX-2 expression in three series of cases: 1) vulvar LS in women who, after a long follow-up period, did not develop invasive tumor, 2) vulvar LS in women who, in the follow-up period, developed invasive SCC of the vulva, and 3) LS adjacent to SCC. We aimed to determine whether the pre-malignant changes in LS could be identified by analysing the role of angiogenesis and COX-2 expression, in order to identify, among women with vulvar LS, those patients who may benefit from a careful clinical follow-up and from adjunctive vulvar punch biopsies. Materials and methods Patients We selected all cases diagnosed as vulvar LS with available microscopic slides and paraffin blocks from 1995 to We found 148 cases documented by vulvar biopsy in the files of the Department of Human Pathology and Oncology of the University of Florence. We combined these with cases that had a histological diagnosis of squamous cell carcinoma (SCC) of the vulva. We identified eight cases characterized by a previous biopsy with an histological diagnosis of vulvar LS, which showed, after a period of 10 months to 9 years, an evolution to carcinoma of the vulva, histologically documented. The slides and tissue blocks of all eight biopsies and all eight surgical specimens were reviewed to confirm the histologic diagnosis and were utilized for immunohistochemical analysis. We also searched 10 vulvar LS cases, for whom follow-up information was available for a minimum of 9 years. In addition, we searched from the files all cases diagnosed as vulvar SCC in the same period. From all the cases we selected 10 cases that showed LS adjacent to SCC. We evaluated the histologic features and the immunohistochemical staining in all 10 cases. All squamous cell carcinomas were of the keratinizing histotype and LS was observed in the skin near cancer. Therefore, we compared the histological features and the immunohistochemical staining of three group of LS cases: 1) LS unchanged, 2) LS evolved to SCC, and 3) LS adjacent to SCC. The mean age of vulvar LS patients was 58.8 years (range, 44 to 77 years), while the mean age of the patients with LS which evolved to SCC was 61.3 years (range, 56 to 79 years), and the mean age of the patients with LS adjacent to SCC was 73.4 years (range, 56 to 86 years). Diagnosis of LS was done according to established histological characteristics: hyperkeratosis, thin epidermis and loss of rete ridges, vacuolar interface changes, edema and/or sclerosis of dermal collagen, chronic inflammatory cell infiltrate of dermis. We also looked for the presence of atypical LS, i.e. lichen showing cells with nuclear enlargement, enlarged nucleoli, and nuclear crowding in lower portions of the epidermis, as previously described [16]. The histological type of each SCC was determined according to the criteria of Kurman et al. [17] Tissue specimens and immunohistochemistry Hematoxylin eosin stained sections from each histological specimen were reviewed by two pathologists (GLT, MRR) to confirm the histological diagnosis. We selected one representative paraffin block from each case for further studies. For immunohistochemical analysis, 3 μm sections were prepared, each section was deparaffinized using xylene and subsequently hydrated. Intratumoral microvessel studies were performed with a specific endothelial marker, the anti- CD34 monoclonal antibody (clone QBEnd/10, Ventana, Tucson, Arizona). VEGF expression studies were performed with a specific anti-vegf monoclonal antibody (clone C1, Santa Cruz, Santa Cruz, CA). The antibodies were placed on the slides and incubated according the Immunostainer Ventana Benchmark. When the staining run was complete, the tissue sections were removed from the stainer, counterstained with hematoxylin, dehydrated and mounted with Permount. The cyclooxygenase-2 expression was evaluated with a monoclonal antibody anti-cox-2 (clone COX229, Histo-line, Hannover, Germany; 1:50 dilution, at 4 C), and with antigen rescue in microwave with TEC buffer (tris EDTA citrate), ph 8 for 35. The immunohistochemical studies were performed by the streptavidin biotin peroxidase method (UltraVision kit, LAB VISION; Fremont, CA) with diaminobenzidine as chromogen and hematoxylin as nuclear counterstain. Sections of positive CD34 hemangioendothelioma, and VEGF positive breast carcinoma were used as positive control tissues, respectively for antibody CD34 and antibody VEGF. Sections of strongly positive COX-2 colon carcinoma were used as positive control tissues. Negative controls were performed by substituting the primary antibody with non-immune sera. The immunohistochemically stained sections were evaluated without previous knowledge of the clinical outcome of each patient. Evaluation of MVD, and VEGF and COX-2 expression MVD was quantified using the method of Weidner et al. [18]. For microvessel quantitation, all slides were examined at low-power magnification ( 40) to identify areas of dermal tissue in the area within 200 μm just beneath the basement membrane of the epidermis, with the greatest density of stained microvessels ( hot spot ). Three fields with the highest number of hot spot were selected. The number of stained microvessels was quantified in each section at high magnification ( 400). In this area, the average value of MVD was considered in the three most vascularized areas at 400 field. The epidermis, in accordance with previous studies [9], was excluded by the MVD count because this structure is avascular. Single endothelial cells or clusters of endothelial cells, with or without a lumen, were considered to be individual vessels. The counts were expressed as total number of microvessels in 400 fields. The MVD was considered a continuous variable for the Kruskal Wallis non-parametric test. Vulvar epithelium showing brown staining of the antibody specific VEGF and COX-2 of cytoplasm were scored as positive. The proportion of immunostained cells was scored at low magnification (5 objective lens) by evaluating the entire specimen. When vulvar epithelium area with positive immunostaining was N10% of the total epithelium of the specimen, the case was scored as positive. Intensity of staining was also evaluated subjectively using a range from 0 (none) to 1 (faint) to 2 (strong). Cases in which the intensity of staining was scored b2 were considered negative. To assess the interobserver variability in the evaluation of MVD, and VEGF and COX-2 expression, all sections were analysed by two independent observers (GLT, MRR). Initially the slides were evaluated independently, and those graded diversely were subsequently re-evaluated by the two pathologists together for a final decision at a conference microscope. Statistical analysis The average vessel density per 400 fields for each tissue specimen of the groups of patients were analysed according to the Wilcoxon Mann Whitney test [19]. Non parametric analysis was used because of the skewed data distribution. The relationship between the presence of VEGF and COX-2 positivity of vulvar LS epithelium in the three groups of patients was calculated according to Fisher's exact test. Data analysis was performed using the SPSS Version 14.1 (Chicago, IL) statistical package. A P value 0.05 was considered to be statistically significant.

3 M.R. Raspollini et al. / Gynecologic Oncology 106 (2007) Results Pathologic features The histopathologic analysis of all cases showed hyperkeratosis, thin epidermis, a variable edema and homogenization of dermal collagen, and a variable chronic inflammatory cell infiltrate of derma. In 2 cases of unchanged vulvar LS, we observed nuclear atypia in epidermal keratinocytes. In 5 cases of patients with LS evolving to SCC, we observed nuclear atypia in epidermal keratinocytes. Immunohistochemistry In vulvar unchanged LS cases the mean MVD was 11.3% (range, 7 to 15) (Fig. 1), while in LS cases evolving to SCC and in the LS adjacent to SCC the mean resulted higher; 29.6% (range, 12 to 68) (Fig. 2), and 23.4 (range, 14 to 36). The difference of MVD in the three groups of patients was statistically significant (P = 0.008, Wilcoxon Mann Whitney test) (Fig. 3). Both differences of MVD between unchanged LS cases and LS cases evolving to SCC and between unchanged LS cases and LS adjacent to SCC cases were statistically significant as well (P = and P = 0.003, respectively); while the difference of MVD between LS cases evolving to SCC and LS adjacent to SCC cases did not result significant (P=0.34). We observed VEGF expression in 4 out of 10 vulvar unchanged LS cases (Fig. 4), while in LS cases evolving to SCC and in those adjacent to SCC, VEGF was expressed in 75% and 100% of cases, respectively, the difference being statistically significant (P = 0.007, Fisher's exact test). We did not observe COX-2 expression in any case of vulvar unchanged LS cases, while in LS cases evolving to SCC and in those adjacent to SCC cases, COX-2 was expressed in 5 cases (62.5%) (Fig. 5) and in 3 cases (30%), respectively. The difference was again statistically significant (P = 0.01, Fisher's exact test). We also observed few unchanged LS cases with a MVD value near the lowest MVD value of LS cases which evolved to SCC. However, these unchanged LS cases did not show any positive staining to VEGF and COX-2, while the LS cases Fig. 2. A high number of microvessels immunostained with CD34 in vulvar LS evolving to SCC. Microvessels are more dense within the subepithelial dermis compared to unchanged vulvar LS cases. evolving to SCC showed at least one, or both, positive expressions of VEGF and COX-2 (Table 1). Discussion In the present study we have found that angiogenesis is poorly expressed in the LS cases remaining unchanged for years. The median MVD was significantly different in statistics in respect to both LS evolving to SCC and LS adjacent to SCC. We also observed that the unchanged cases did not express VEGF and COX-2 compared to both evolving LS or LS adjacent to SCC. Vulvar keratinizing SCC, the predominant type of vulvar carcinoma, is associated either with LS and simplex vulvar intraepithelial neoplasia (VIN): both lesions have also been found adjacent to vulvar SCC [20]. LS is a common disease in the genital area in either middleaged or elderly women. Different considerations support a relationship between LS and SCC: a personal study describes the presence of LS, histologically reported, associated to SCC in one-third of the specimens excised for vulvar SCC [21]. In Fig. 1. A low number of microvessels immunostained with CD34 in unchanged vulvar LS. Fig. 3. In unchanged vulvar LS cases the mean MVD was 11.3% (range, 7 to 15), while in LS cases evolving to SCC it was 29.6% (range, 12 to 68) and in the LS adjacent to SCC cases it was 23.4% (range, 14 to 36). The difference of MVD in the three groups of patients with LS was statistically significant (P=0.008, Wilcoxon Mann Whitney test).

4 570 M.R. Raspollini et al. / Gynecologic Oncology 106 (2007) Table 1 Microdevessel density (MVD), VEGF, and COX-2 expression in LS unchanged, LS evolving, and LS adjacent to SCC LS unchanged LS evolving LS adjacent to SCC MVD Mean 11.3% 29.6% 23.4% Range (7 15) (12 68) (14 36) Positive VEGF expression 4/10 6/8 10/10 Positive COX-2 expression 0/10 5/8 3/10 The few LS unchanged cases with a MVD value near to the lowest MDV value of LS cases evolving to SCC did not show any positive staining to VEGF and COX-2; while LS cases evolving to SCC with lowest MDV value showed almost one, or both, positive expression of VEGF and COX-2. Fig. 4. Immunohistochemical staining for VEGF. Diffuse immunoreactivity is observed in evolving vulvar LS. addition, the incidence of SCC is higher in women presenting LS in comparison to women without LS [22,23]. A mean time interval of 10 years from LS to vulvar SCC is reported. Simplex VIN does not represent a common lesion, since it accounts for only about 2 to 10% of VIN biopsies, generally occurring in older women compared to classic VIN [24]. Simplex VIN is less common than vulvar LS, yet it is the most likely precursor of keratinizing SCC of the vulva [1,25]. However, the presence of VIN adjacent to invasive SCC of the vulva was not reported in a high percentage of cases [26 28]. The cytological similarities of simplex VIN to welldifferentiated keratinizing invasive SCC is unmistakable. Both simplex VIN and well-differentiated keratinizing invasive SCC show keratinocytes with enlarged and atypic nuclei with prominent nucleoli and eosinophilic cytoplasm. Invasive tumors can replace the precursor lesions as VIN, but perhaps there is also the possibility that SCC without VIN may arise directly from some LS, possibly atypical or dysplastic, that may evolve to invasive SCC. To our knowledge this is the first study reporting an analysis of markers of angiogenesis and COX-2 expression in a series of unchanged LS cases compared to LS evolving to vulvar SCC cases and LS observed adjacent to SCC cases. A previous study [12], analysing the role of COX-2 expression in non-neoplastic and neoplastic vulvar epithelial lesions, suggested that its overexpression may contribute to vulvar tumorigenesis. The authors, however, reported a strong COX-2 expression in high-grade VIN, with respect to low-grade VIN, suggesting that up-regulation of COX-2 can play a role in tumor onset and progression, but not in LS cases. According to a recent study [9], MVD was confirmed as a useful parameter in determining potential malignant progression, but the role of this marker has not been demonstrated for determining the risk of malignant progression in LS cases. However, it should be underlined that in this study, the authors did not describe any evolution of LS cases to vulvar SCC. The particular value of our study arises from the comparison of the eight cases with an histologic diagnosis of vulvar LS evolved to squamous carcinoma, to the 10 LS cases adjacent to SCC, and to the 10 vulvar LS unchanged cases with a long follow-up. The staining pattern of this series may indicate a pathway to carcinogenesis in some LS cases independently of recognizable morphological alterations. Our study addresses the possibility that careful immunohistochemical studies may add some information to allow identification of those LS which are precursor lesions with a great potential to evolve into SCC. This fact has been addressed also by a recent review in which some immunohistochemical markers have been found to be relevant in solving differential problems in diagnostic gynaecological pathology [29]. The results of this study, associated with our previous data [30], may identify characteristics of vulvar LS disclosing alterations that indicate the further development to cancer; therefore, it may allow the identification of a group of LS patients who need a careful follow-up and adjunctive biopsies. Acknowledgments Fig. 5. Immunohistochemical staining for COX-2. Diffuse immunoreactivity is observed in evolving vulvar LS. We are indebted to Prof. Vieri Boddi for his invaluable help in the statistical analysis and with Chiara Lucarelli for the precious technical assistance in the preparation of specimens.

5 M.R. Raspollini et al. / Gynecologic Oncology 106 (2007) References [1] Yang B, Hart WR. Vulvar intraepithelial neoplasia of the simplex (differentiated) type: a clinicopathological study including analysis of HPV and p53 expression. Am J Surg Pathol 2000;24: [2] Hart WR, Norris NJ, Helving EB. Relation of lichen sclerosus et atrophicus of the vulva to development of carcinoma. Obstet Gynecol 1975;45: [3] Carlson JA, Ambros R, Malfetano J, et al. Vulvar lichen sclerosus and squamous cell carcinoma: a cohort, case control and investigation study with historically perspective; implications for chronic inflammation and sclerosis in the development of neoplasia. Hum Pathol 1998;29: [4] Folkman J. What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 1990;82:4 6. [5] Lutgendorf SK, Johnsen EL, Cooper B, et al. Vascular endothelial growth factor and social support in patients with ovarian carcinoma. Cancer 2002;95: [6] Obermair A, Kohlberger P, Bancher-Todesca D, et al. Influence of microvessel density and vascular permeability factor/vascular endothelial growth factor expression on the prognosis in vulvar cancer. Gynecol Oncol 1996;63: [7] MacLean AB, Reid WM, Rolfe KJ, Gammell SJ, Pugh HE, Gatter KC, et al. Role of angiogenesis in benign, premalignant and malignant vulvar lesions. J Reprod Med 2000;45: [8] Bancher-Todesca D, Obermair A, Bilgi S, Kohlberger P, Kainz C, Breitenecker G, et al. Angiogenesis in vulvar intraepithelial neoplasia. Gynecol Oncol 1997;64: [9] Saravanamuthu J, Reid W, Gorge DS, et al. The role of angiogenesis in vulvar cancer, vulvar intraepithelial neoplasia, and vulvar lichen sclerosus as determined by microvessel density analysis. Gunecol Oncol 2003;89: [10] Doldi N, Origoni M, Bassan M, et al. Vascular endothelial growth factor. Expression in human vulvar neoplastic and nonneoplastic tissues. J Reprod Med 1996;41: [11] Tsujii M, DuBois RN. Alterations in cellular adhesion and apoptosis in epithelial cells overexpressing prostaglandin endoperoxide synthase 2. Cell 1995;83: [12] Ferrandina G, Ranelletti FO, Salutati V, et al. Expression of ciclooxygenase-2 (COX-2) in non neoplastic and neoplastic vulvar epithelial lesions. Gynecol Oncol 2004;92: [13] Nofech-Mozes S, Kupets R, Rasty G, et al. Cyclooxygenase-2 (COX-2) immunostaining does not correlate with the degree of vulvar neoplasia. J Obstet Gynaecol Can 2006;28: [14] Lehay KM, Ornberg RL, Wang Y, et al. Cyclooxygenase-2 inhibition by celecoxib reduces proliferation and induces apoptosis in angiogenic endothelial cells in vivo. Cancer Res 2002;62: [15] Raspollini MR, Amunni G, Villanucci A, et al. COX-2 status in relation to tumor microvessel density (MVD) and VEGF expression. Analysis in ovarian carcinoma patients with low and high survival. Oncol Rep 2004;11: [16] Chiesa-Vottero A, Dvoretsky P. Histopathologic study of thin vulvar squamous cell carcinomas and associated cutaneous lesions: a correlative study of 48 tumors in 44 patients with analysis of adjacent vulvar intraepithelial neoplasia types and lichen sclerosus. Am J Surg Pathol 2006;30: [17] Kurman RJ, Toki T, Schiffman MH. Basaloid and warty carcinomas of the vulva: distinctive types of squamous cell carcinoma frequently associated with human papillomaviruses. Am J Surg Pathol 1993;17: [18] Weider N, Semple JP, Welch WR, Folkman J. N Engl J Med 1991;324:1 8. [19] Fisher LD, Van Belle G. Biostatistics: a methodology for the health sciences. 2nd edition, 3rd series. A Wiley Interscience Publication; [20] Rouzier R, Morice P, Haie-Meder C, Lhomme C, et al. Prognostic significance of epithelial disorders adjacent to invasive vulvar carcinomas. Gynecol Oncol 2001;81: [21] Carli P, De Magnis A, Mannone F, et al. Vulvar carcinoma associated with lichen sclerosus. Experience at the Florence, Italy, Vulvar Clinic. J Reprod Med 2003;48: [22] Carli P, Cattaneo A, De Magnis A, et al. Squamous cell carcinoma arising in vulval lichen sclerosus: a longitudinal cohort study. Eur J Cancer Prev 1995;4: [23] Jones RW, Sadler L, Grant S, et al. Clinically identifying women with vulvar lichen sclerosus at increased risk of squamous cell carcinoma. J Reprod Med 2004;49: [24] Poulsen H, Junge J, Vyberg M, et al. Small vulvar squamous cell carcinomas and adjacent tissues. A morphologic study. APMIS 2003;111: [25] Hart WR. Vulva intraepithelial neoplasia: historical aspects and current status. Int J Gynecol Pathol 2001;20: [26] Gomez Roueda N, Garcia A, Vighi S, et al. Epithelial alteration adjacent to invasive squamous carcinoma of the vulva. J Reprod Med 1994;39: [27] Jones RW, Baranyai J, Stables S. Trends in squamous cell carcinoma of the vulva: the influence of vulvar intraepithelial neoplasia. Obstet Gynecol 1997;90: [28] Rouzier R, Morice P, Haie-Meder C, et al. Prognostic significance of epithelial disorders adjacent to invasive vulvar carcinomas. Gynecol Oncol 2001;81: [29] Deavers MT, Malpica A, Silva G. Immunohistochemistry in gynaecological pathology. Int J Gynecol Cancer 2003;13: [30] Raspollini MR, Asirelli G, Moncini D, Taddei GL. A comparative analysis of lichen sclerosus of the vulva and lichen sclerosus that evolves in vulvar squamous cell carcinoma. Am J Obstet Gynecol 2007:197 (in press).