p130 and p27kip1 in vulvar squamous cell carcinomas

Original Contributions Expression of Cell-Cycle–Associated Proteins pRB2/p130 and p27kip1 in Vulvar Squamous Cell Carcinomas ALESSANDRA ZAMPARELLI, BS, VALERIA MASCIULLO, MD, ALESSANDRO BOVICELLI, MD, DONATELLA SANTINI, MD, GABRIELLA FERRANDINA, MD, CORRADO MINIMO, MD, PATRIZIA TERZANO, MD, SILVANO COSTA, MD, CATERINA CINTI, BS, CLAUDIO CECCARELLI, BS, SALVATORE MANCUSO, MD, GIOVANNI SCAMBIA, MD, LUCIANO BOVICELLI, MD, AND ANTONIO GIORDANO, MD, PHD Squamous cell vulvar carcinoma accounts for 4% of all gynecologic malignancies. The cause of vulvar cancer is still unclear. Identification of new biologic factors involved in vulvar carcinogenesis may be useful in clarifying the natural history of this malignancy. We investigated the immunohistochemical expression of the retinoblastoma-related proteins pRB2/p130 and CKI p27kip1 in a series of 51 invasive squamous cell carcinomas of the vulva (ISCCs) and in synchronous normal vulvar skin, non-neoplastic epithelial disorders (NNED) and vulvar intraepithelial neoplasia (VIN). Normal vulvar skin staining showed positivity for both pRB2/p130 and p27kip1. Loss of pRB2/p130 occurred in 29 (57%) of 51 specimens of ISCCs, and in 1 of 7 specimens with VIN (14%; P ⴝ .04). We also observed a significant decrease of pRB2/p130 expression from NNED to neoplastic tissues (VIN and ISCCs) (P ⴝ .004). Loss of p27kip1 expression was found in 16 of 51 specimens (31%) of invasive carcinomas, in 1 (14%) of 7 specimens of VIN, and in 2 of 18 specimens of NNED

(11%). pRB2/p130 and p27kip1 did not correlate significantly with any of the clinicopathologic parameters examined. Our data indicate that loss of pRB2/p130 and p27kip1 are frequent events in invasive vulvar carcinomas compared with synchronous premalignant lesions, nonneoplastic epithelial disorders, and normal vulvar skin. The significant progressive decrease of pRB2/p130 expression from non-neoplastic epithelial alterations through intraepithelial neoplasia to invasive vulvar carcinomas suggests a role for this tumor suppressor gene in vulvar carcinogenesis. HUM PATHOL 32:4-9. Copyright © 2001 by W.B. Saunders Company Key words: squamous vulvar carcinoma, pRB2/p130, p27kip1, prognostic factor Abbreviations: CKI, cyclin-dependent kinase inhibitor; NNED, non-neoplastic epithelial disorders; VIN, vulvar intraepithelial neoplasia; ISCCs, invasive squamous cell carcinomas; BC, basaloid cell; WC, warty cell.

The retinoblastoma gene family is composed of 3 members: the retinoblastoma gene (Rb), which is one of the most studied tumor suppressor genes, and 2

related genes, RB2/p130 and p107. The proteins encoded by these genes, pRB2/p130 and p107, have been shown to be structurally and functionally similar to pRb and to play a pivotal role as negative regulators of cell proliferation.1,2 However, the 3 proteins exhibit different growth-suppressive properties in specific cell lines, suggesting that although the different members of the retinoblastoma family may complement each other, they are not fully functionally redundant.3 The pRB2/ p130 gene product is known to associate with the complex DP-1/E2F, to bind cyclin A-cdk2 and cyclin E-cdk2 kinases, and displays a growth-suppressive activity, which is specific for the G0/G1 phases of the cell cycle. RB2/p130 has been mapped to human chromosome 16q.12.2, an area that frequently has been found altered in several human neoplasias, including breast, ovarian, hepatic, prostatic, and endometrial carcinomas, which suggests its possible role in the pathogenesis of several human cancers. In addition, we recently reported that RB2/p130 is involved in the pathogenesis and progression of lung cancer,4 mesothelioma,5 and lymphomas6 and is a strong predictor of clinical outcome in endometrial carcinomas7 and choroidal melanomas.8

From the Department of Pathology, Anatomy and Cell Biology, Thomas Jefferson University, Philadelphia, PA; the Department of Gynecology and Obstetrics, The Johns Hopkins Hospital and Medical Institutions, Baltimore, MD; the Department of Obstetrics and Gynecology, St Orsola Hospital, University of Bologna, Italy; the Department of Gynecology and Obstetrics, Catholic University, Rome, Italy; the Department of Pathology, St Orsola Hospital, Bologna, Italy; and the Institute of Normal and Pathological Cytomorphology, CNR, Bologna, Italy. Accepted for publication July 14, 2000. Supported in part by Associazione Italiana Ricerca sul Cancro (AIRC; to S.M.) and by National Institute of Health Grant RO1 CA60999/01A1, by Grant PO1-NS36466, by Grant PO1-CA56309, and by Sbarro Institute for Cancer Research and Molecular Medicine (to A.G.). V.M. is supported by a Fellowship from Consiglio Nazionale delle Ricerche (CNR). Address correspondence and reprint requests to Antonio Giordano, MD, PhD, Department of Pathology, Anatomy and Cell Biology, Jefferson Medical College of Thomas Jefferson University, Jefferson Alumni Hall Room 226, 1020 Locust St, Philadelphia, PA 19107. Copyright © 2001 by W.B. Saunders Company 0046-8177/01/3201-0002$35.00/0 doi:10.1053/hupa.2001.20371

4

pRB2/p130 AND p27kip1 EXPRESSION IN VULVAR CARCINOMAS (Zamparelli et al)

p27kip1 is a cyclin-dependent kinase inhibitor (CKI) that regulates progression from G1 into S phase by inhibiting multiple cyclin-cdk complexes, including cyclin E-cdk2, cyclin D-cdk4, and cyclin A-cdk2.9 The importance of clinical implications in decreasing p27kip1 protein levels already has been demonstrated in colorectal,10,11 breast,12 lung,13 prostate,14,15 ovarian,16 and cervical cancer.17 Squamous cell carcinomas are the most frequent type of invasive vulvar cancers and account for 4% of all female genital tract cancers. These tumors typically are keratinizing squamous carcinomas, and frequently are associated with non-neoplastic epithelial disorders (NNED) of the vulva, such as lichen sclerosus (LS), squamous cell hyperplasia, and other dermatoses.18 Several studies have identified a subset of human papillomavirus (HPV)-positive squamous cell carcinomas that exhibit a basaloid (BC) or warty (WC) histology and are associated with vulvar intraepithelial neoplasia (VIN), currently accepted as their precursor.19 The role played by HPV infection, immunodeficiency, and cigarette smoking in the causation of squamous cell carcinomas is unclear, and little research has been performed to identify the role of cell cycle regulatory proteins in the pathogenesis and progression of vulvar carcinomas. Previous studies20,21 have shown that lack of Rb or the CKI p16INK4 increased from benign lesions to invasive squamous cell carcinomas, which suggest their involvement in the pathogenesis and progression of malignant disease. In this study, we characterized by immunohistochemistry the expression pattern of pRB2/p130 and p27kip1 protein in 51 invasive squamous cell carcinomas (ISCCs) of the vulva and in other synchronous vulvar lesions, including normal tissues, LS, squamous cell hyperplasia, and VIN. We also evaluated the relationship of p27kip1 and pRB2/p130 protein expression with clinicopathologic parameters in invasive vulvar carcinomas.

Immunohistochemistry A total of 51 formalin-fixed and paraffin-embedded tumor tissue samples were processed. Sections of each specimen were cut at 3 ␮m, mounted on glass, and dried overnight at 37°C. All sections were dewaxed, rehydrated, quenched in 0.5% hydrogen peroxide and then microwave pretreated in 10 mmol/L citrate buffer, pH6.0 (10 minutes at 660 W for p27, 17 minutes at 760 W for pRB2/p130). After blocking for 1 hour at room temperature, the monoclonal antibody for p27kip1 (clone DCS-72.F6 NeoMarkers, Inc., Union City, CA) was incubated with tissue sections at 1:150 dilution at 4°C overnight. The rabbit polyclonal immune serum against pRB2/p130 (ADL1) was produced by immunizing rabbits with a peptide corresponding with the COOH terminus of the protein22 and used overnight at a 1:100 dilution. Negative controls were produced by substituting preimmune serum for primary antibodies. The strong positive immunostaining of infiltrating lymphocytes represented an internal positive control for preservation of both p27kip1 and pRB2/p130 antigenicity in the sections examined. All slides were processed by the ABC method (Vector Laboratories). Diaminobenzidine was used as the final chromogen, and Harris hematoxylin was used for counterstaining.

p27kip1 and pRB2/p130 Scoring Two pathologists (C.M., D.S.) independently evaluated each specimen and scored immunostaining for the percentage of positive nuclei. The level of concordance, expressed as the percentage of agreement between the observers, was 95% (48 of 51 specimens). In the remaining 3 samples, the score was obtained from the best judgment of the 2 observers who were in agreement. At least 20 high-power fields were chosen randomly, and 2,000 cells were counted. A number of tumors showed heterogeneous p27 and p130 expression. In these cases, the tumor was scored according to the area of highest grade because this was most likely to dictate prognosis. For p27kip1, tumor sections were considered negative if nuclear staining was absent or present in less than 5% of tumor cells.16 For pRB2/p130, a cutoff level of 10% was used, as previously described for pRB expression in vulvar carcinomas.21

Statistical Methods Fisher’s exact test for proportion and the ␹2 test were used to analyze the distribution of pRB2/p130 and p27kip1positive cases according to clinicopathologic characteristics. Statistical significance was considered at P ⬍ .05. These analyses were performed with Statistica for Windows software version 4.1 (BMDP Statistical Software, Inc, Los Angeles, CA).

MATERIALS AND METHODS Patient Populations and Clinical Data Fifty-one paraffin-embedded sections of invasive squamous carcinomas of the vulva were obtained from patients who underwent surgical resection at the Department of Gynecology and Obstetrics, S. Orsola Hospital, University of Bologna and at the Department of Gynecology of the Catholic University of Rome. Staging was carried out according to the FIGO classification: 18 patients were stage I, 20 were stage II, 12 were stage III, and 1 was stage IV. The tumors were graded as well (G1), moderately (G2), and poorly differentiated (G3). Two experienced pathologists (C.M., D.S.) confirmed the histologic diagnosis of each vulvar lesion. Among the group of vulvar carcinomas, we selected 6 sections in which adjacent histologically uninvolved vulvar epithelium was also present. LS was present at the time of tumor resection in 8 (16%) patients; 10 (20%) tumors had associated squamous cell hyperplasia. VIN (3 basaloid/warty and 4 simplex type) coexisted with the ISCC in 7 sections.

RESULTS p27kip1 and pRB2/p130 Protein Expression in Normal Skin, Non-neoplastic Epithelial Disorders (NNEDs), Intraepithelial Neoplasia (VIN), and Invasive Squamous Cell Carcinomas (ISCCs) of the Vulva The expression of pRB2/p130 and p27kip1 in vulvar tissues was determined by immunohistochemistry. Immunoreactivity for p27kip1 and pRB2/p130 was found in normal vulvar skin, NEDDs, VIN and ISCCs and is summarized in Table 1. Normal vulvar skin adjacent to the carcinoma in 5

HUMAN PATHOLOGY

Volume 32, No. 1 (January 2001)

TABLE 1. pRB2/p130 and p27kip1 Distribution in Vulvar Cancer and Synchronous Vulvar Lesions

Normal vulvar skin NNED: Lichen sclerosus Squamous hyperplasia VIN VIN I VIN II VIN III ISCC

N

p27 Positive

p27 Negative

p130 Positive

p130 Negative

NA

6 18 8 10 7 1 2 4 51

6 (100%) 16 (89%) 8 (100%) 8 (80%) 6 (86%) 1 (100%) 2 (100%) 3 (75%) 35 (69%)

— 2 (11%) — 2 (20%) 1 (14%) — — 1 (25%) 16 (31%)

6 (100%) 14 (87,5%) 8 (100%) 6 (75%) 6 (86%) 1 (100%) 2 (100%) 3 (75%) 22 (43%)

— 2 (12,5%) — 2 (25%) 1 (14%) — — 1 (25%) 29 (57%)

— 2 — 2 — — — — —

and classified as keratinizing (n ⫽ 45), basaloid (BC) (n ⫽ 2), verrucous (n ⫽ 3), or warty (WC) (n ⫽ 1). Loss of pRB2/p130 occurred in 29 (57%) of 51 specimens (Fig 1E). pRB2/p130 immunoreactivity of invasive vulvar cancer (43% of specimens) was both nuclear and cytoplasmic, and consistently lower in the expression pattern (Figs 1D and F) compared with normal vulvar skin, NNED, and VIN. In particular, a statistically significant difference was found in pRB2/p130 expression of NNED (87.5%) with respect to neoplastic tissues (VIN and ISCC; 48%) (P ⫽ .004). p130 staining also decreased significantly from VIN (86%) to ISCC (43%) (P ⫽ .04).

the same patient showed positivity for both pRB2/p130 and p27kip1 in all specimens examined. Positive nuclei were found in the basal and parabasal layers of the epithelium (Fig 1A). The pattern of p27kip1 and pRB2/ p130 immunoreactivity in LS was stronger than that seen for normal epithelium for all specimens (Fig 1B). However, squamous cell hyperplasia showed loss of pRB2/p130 in 2 of 8 specimens (25%) and loss of p27kip1 in 2 of 10 specimens (20%). All but 1 specimen of VIN (grade III) showed a weak positive nuclear staining for both pRB2/p130 and p27kip1 (Figs 1C; 2A and B). A total of 51 specimens of ISCCs were evaluated

FIGURE 1. pRB2/p130 immunohistochemical analysis in vulvar tissues. (A) Normal vulvar skin. Most of the basal and parabasal cells are positive (original magnification ⫻ 400); (B) Lichen sclerosus. Most of the basal and parabasal cells show nuclear and concomitant cytoplasmic staining (original magnification ⫻ 400); (C) Vulvar intraepithelial neoplasia. VIN III. Note stronger reactivity in larger more irregular nuclei (original magnification ⫻ 400); (D) Vulvar squamous cell carcinoma, keratinizing. Most of the tumor cells are weakly positive (original magnification ⫻ 400); (E) Vulvar squamous cell carcinoma, keratinizing. The tumor cells of this case are negative (original magnification ⫻ 400); (F) Vulvar squamous cell carcinoma, basaloid subtype. Many tumor cells are positive (original magnification ⫻ 400).

6

pRB2/p130 AND p27kip1 EXPRESSION IN VULVAR CARCINOMAS (Zamparelli et al)

FIGURE 2. p27 immunohistochemical analysis in vulvar tissues. (A) Vulvar intraepithelial neoplasia VIN III. Note high intensity of staining (original magnification ⫻ 400); (B) Vulvar intraepithelial neoplasia VIN II (moderate dysplasia). Few nuclei are positive and are weakly stained (original magnification ⫻ 400); (C) Vulvar squamous cell carcinoma, basaloid subtype. Nuclei are primarily negative. Minimal focal staining is noted in cells of a keratinizing whorl (original magnification ⫻ 400); (D) Vulvar squamous cell carcinoma, keratinizing. Note nuclear positivity in less than 50% of the tumor cells (original magnification ⫻ 400); (E, F) Vulvar squamous cell carcinoma, keratinizing. The tumor cells of this case are negative. Arrowhead points at the strong positivity of infiltrating lymphocytes (positive internal control) (original magnification ⫻ 400).

p27kip1 protein was expressed in 35 of 51 specimens (69%) (Fig 2C and D), whereas lack of p27kip1 protein expression (⬍5% positive cells) was observed in the remaining 31% of specimens (Fig 2E and F). In contrast to normal epithelium, NNED, and VIN, where p27kip1 was localized mostly in the nuclei, invasive carcinomas showed concomitant cytoplasmic staining of p27kip1 in 15% of the specimens. No statistically significant difference was found in p27kip1 expression of NNED with respect to neoplastic tissues (P ⫽ .1). The 3 verrucous carcinomas all showed high levels of p27kip1 (60% to 80% of nuclear staining) and loss of pRB2/p130.

(39%) had stage II tumors, 12 (24%) had stage III tumors, and 1 (2%) had a stage IV tumor. Histologically, 10 tumors were classified as grade 1, 30 tumors as grade 2, and 6 tumors as grade 3. The inguinal lymph nodes were examined in 36 patients, and 11 (30%) of them contained metastatic tumor. p27kip1 staining did not correlate significantly with any of the clinicopathologic parameters examined. However, we did observe a trend toward a relationship between loss of p27kip1 expression and poor grade of differentiation (G3) (P ⫽ .08). No correlations were found between pRB2/p130 immunoreactivity and clinicopathologic parameters, regardless of the pRB2/p130 cutoff level tested.7,21 As regards the relationship between p27kip1 and pRB2/p130 expression, loss of pRB2/p130 was observed more frequently in the group of p27kip1-negative patients (87%) compared with the group of p27kip1positive patients (13%), even though the statistical significance was not reached (P ⫽ .08).

Correlation of p27kip1 and pRB2/p130 Expression With Clinicopathologic Parameters The relationship between the expression of p27kip1 and pRB2/p130 and a series of clinicopathologic parameters (age, stage, grading, histotype, lymph nodes metastasis) was investigated in 51 vulvar invasive carcinomas and is summarized in Table 2. The median age of the patients was 71 years (range, 42 to 88); 14 (27%) were 65 years old or younger. According to the FIGO surgical staging system, 18 (35%) of the 51 patients had stage I tumors, 20

DISCUSSION We describe for the first time the expression pattern of the retinoblastoma-related gene pRB2/p130 7

HUMAN PATHOLOGY

TABLE 2.

Volume 32, No. 1 (January 2001)

Distribution of pRB2/p130 and p27kip1 in 51 Vulvar Squamous Cell Carcinomas According to Clinicopathologic Parameters

Variable Age ⱕ65 ⬎65 Tumor stage I II III IV Histologic grade 1 2 3 n.a. Histotype KSC BC Verrucous WC Lymph node metastases Negative Positive Not assessed

NO.

P27neg

P27pos

14 37

3 13

11 24

18 20 12 1

6 5 5 0

12 15 7 1

10 30 6 5

2 9 4

8 21 2

45 2 3 1

16 0 0 0

29 2 3 1

25 11 15

7 4

18 7

P*

P130neg

P130pos

9 28

5 9

13 15 9 1

5 5 3 0

7 20 6

3 10 0

32 1 3 1

13 1 0 0

16 9

9 2

NS

P* NS

NS

NS

NS

NS

NS

NS

NS

NS

Abbreviation: NA, not available. * Chi-square test.

and the CKI p27kip1 in normal vulvar skin, in vulvar invasive squamous cell carcinomas (ISCCs) and in synchronous vulvar epithelial alterations such as lichen sclerosus (LS), squamous cell hyperplasia, and intraepithelial neoplasia (VIN). The role played by members of the retinoblastoma gene family as tumor suppressors is well known.1 Recently, genetic alterations of the RB2 gene have been detected in several primary tumors and in human cancer cell lines.4 Furthermore, pRB2/p130 has been shown to have antiproliferative and tumor-suppressive potentials both in vivo and in vitro, using a tetracyclinedependent overexpression system. A JCV-transformed tumor cell line injected into nude mice causes lesions that regress in the presence of induced pRB2/p130.23 Similar results have been shown in a study that investigated the effects of pRB2/p130 expression in c-erbB-2– overexpressing SKOV-3 tumor cells.24 Altered pRB2/ p130 expression has been reported in several tumors, including primary lung cancer, endometrial carcinomas, and choroidal melanomas and has been associated with a poor clinical outcome.7,8,22 Lack of pRb has been reported to increase from vulvar benign lesions, to VIN, and to invasive squamous cell carcinomas of the vulva, suggesting that pRb plays an important role in tumor progression.20 A recent study, however, points out that loss of pRB does not seem to play any role in the disease initiation, because it seems to be a late event occurring only in invasive vulvar carcinomas.21 Despite the small number of samples analyzed in this study, the frequency of pRB2/p130 expression decreased from benign lesions (NNEDs) (87.5%) through VIN (86%) to invasive squamous cell carci-

noma (43%). This finding is consistent with our previous studies, which showed a progressive decrease in pRB2/p130 expression from hyperplastic endometrium through atypical hyperplasia to poorly differentiated carcinomas25 and may reflect a possible alteration of the pRB2/p130 gene function and could represent an early event in vulvar carcinogenesis. However, these data need to be confirmed in larger population studies. The high percentage of invasive vulvar cancers that exhibit loss of pRB2/p130 expression has been previously reported for other cancers, such as lung22 and endometrial7 and suggests a possible involvement of this negative cell-cycle regulator in vulvar cancer progression. Moreover, it has been recently reported26 that the HPV16 oncogene E7 binds to pRB2/ p130, leading to release of transcriptionally active E2F and progression through cell cycle. Studies are underway to verify whether pRB2/p130 binding of E7 is involved in the pathogenesis and progression of those vulvar carcinomas that are HPV-related. We found no difference in pRB2/p130 expression among the different subtypes of vulvar carcinomas in this study. It is noteworthy, however, that, as already described for pRb,21 all the verrucous carcinomas were negative for pRB2/p130 expression. Loss of pRB2/p130 in vulvar cancer could either be because of the presence of mutations, as occurs with other tumor suppressor genes such as Rb/p10527 and p53,28 or could be the effect of enhanced degradation pathways, such as the ubiquitindependent protease system, as has been shown for other cell cycle-regulatory proteins that interact with cyclins/cdks complexes, such as p27kip1.10,13 In support of the former hypothesis, genetic alterations disrupting 8

pRB2/p130 AND p27kip1 EXPRESSION IN VULVAR CARCINOMAS (Zamparelli et al) melanoma: Prognostic significance. Clin Cancer Res 5:1455-1458, 1999 9. MacLachlan TK, Sang N, Giordano A: Cyclins, cyclin-dependent kinases and cdk inhibitors: Implications in cell-cycle control and cancer. Crit Rev Eukaryot Gene Expr 5:127-156, 1995 10. Loda M, Cukor B, Tam SW, et al: Increased proteasomedependent degradation of the cyclin-dependent kinase inhibitor p27 in aggressive colorectal carcinomas. Nat Med 3:231-234, 1997 11. Thomas GV, Szigeti K, Murphy M, et al: Down-regulation of p27 is associated with development of colorectal adenocarcinoma metastases. Am J Pathol 153:681-687, 1998 12. Catzavelos C, Bhattacharaya N, Ung Y, et al: Decreased levels of the cell-cycle inhibitor p27 kip 1 protein: Prognostic implications in primary breast cancer. Nat Med 3:227-230, 1997 13. Esposito V, Baldi A, De Luca A, et al: Prognostic role of the cyclin-dependent kinase inhibitor p27 in non-small cell lung cancer. Cancer Res 57:3381-3385, 1997 14. Tsihilias J, Kapusta LR, DeBoer G, et al: Loss of cyclin-dependent kinase inhibitor p27 kip1 is a novel prognostic factor in localized human prostate adenocarcinoma. Cancer Res 58:542-548, 1998 15. Cordon-Cardo C, Koff A, Drobniak M, et al: Distinct altered patterns of p27kip1 gene expression in benign prostatic hyperplasia and prostatic carcinoma. J Natl Cancer Inst 90:1284-1291, 1998 16. Masciullo V, Sgambato A, Pacilio C, et al: Loss or altered subcellular localization of the cyclin-dependent kinase inhibitor p27 in epithelial ovarian cancer. Cancer Res 59:3790-3794, 1999 17. Skomedal H, Kristensen GB, Lie AK, et al: Aberrant expression of the cell cycle associated proteins TP53, MDM2, p21, p27, cyclin D1, RB, and EGFR in cervical carcinomas. Gynecol Oncol 73:223-228, 1999 18. Wilkinson EJ: Benign disease of vulva, in Kurman RJ (ed): Blaustein’s Pathology of the Female Genital Tract (ed 4). New York, NY, Springer Verlag, 1994, pp 31-86 19. Kurman RJ, Toki T, Shiffman MH: Basaloid and warty carcinomas of the vulva: Distinctive types of squamous cell carcinoma frequently associated with human papillomaviruses. Am J Surg Pathol 17:133-145, 1993 20. Chan May KM, Cheung TH, Chung TKH, et al: Expression of p16INK4 and retinoblastoma protein Rb in vulvar lesions of Chinese women. Gynecol Oncol 68:156-161, 1998 21. Lerma E, Matias-Guiu X, Lee JS, et al: Squamous cell carcinoma of the vulva: Study of ploidy, HPV, p53, and pRB. Int J Gynecol Pathol 18:191-197, 1999 22. Baldi A, Esposito V, De Luca A, et al: Differential expression of Rb2/p130 and p107 in normal human tissues and in primary lung cancer. Clin Cancer Res 3:1691-1697, 1997 23. Howard CM, Claudio PP, Gallia GL, et al: Retinoblastomarelated protein pRB2/p130 and suppression of tumor growth in vivo. J Natl Cancer Inst 90:1451-1460, 1998 24. Pupa SM, Howard CM, Invernizzi AM, et al: Ectopic expression of pRB2/p130 suppresses the tumorigenicity of the c-erbB-2 overexpressing SKOV3 tumor cell line. Oncogene 18:651-656, 1999 25. Susini T, Massi D, Paglierani M, et al: Retinoblastoma-related gene Rb2/p130 expression in normal, hyperplastic and malignant endometrium. HUM PATHOL 2001 (in press) 26. Smith-McCune K, Kalman D, Robbins C, et al: Intranuclear localization of human papillomavirus 16 E7 during transformation and preferential binding of E7 to the Rb family member p130. Proc Natl Acad Sci U S A 96:6999-7004, 1999 27. Cavenee WK, Hansen MF, Nordenskjold M, et al: Genetic origin of mutations predisposing to retinoblastoma. Science 228:501503, 1985 28. Finlay CA, Hinds PW, Levin AJ: The p53 proto-oncogene can act as a suppressor of transformation. Cell 57:1083-1093, 1989 29. Cinti C, Claudio PP, Howard CM, et al: Genetic alterations disrupting the nuclear localization of the retinoblastoma related gene RB2/p130 in human cell lines and primary tumors. Cancer Res 60:383-389, 2000 30. Guo Y, Sklar GN, Borkowski A, et al: Loss of the cyclin-dependent kinase inhibitor p27 (Kip1) protein in human prostate cancer correlates with tumor grade. Clin Cancer Res 3:2269-2274, 1997 31. Ciaparrone M, Yamamoto H, Sgambato A, et al: Localization and expression of p27KIP1 in multistage colorectal carcinogenesis. Cancer Res 58:114-122, 1998

the nuclear localization of the pRB2/p130 gene recently have been documented in human tumors.29 Loss of p27kip1 expression was detected in squamous cell hyperplasia (20%), VIN (14%), and squamous cell carcinomas (31%), compared with the normal vulvar skin and lichen sclerosus. The p27kip1 gene, located on chromosome 12p, rarely is affected by structural alterations. Its loss may result from increased degradation of the protein, as previously shown in other malignancies.10,13 We also observed a trend toward a decrease in the amount of p27kip1 immunoreactivity with increasing tumor grade (G3), as already reported in prostate30 and colon cancer,31 where loss of p27kip1 expression correlated significantly with poor grade of differentiation. Loss of pRB2/p130 and p27kip1 did not correlate with any of the clinicopathologic parameters used to predict clinical outcome. However, this lack of association for both proteins cannot exclude an independent prognostic role for them as already reported in breast,12 colorectal,10 and lung cancer13 for p27kip1, and in endometrial carcinomas7 and choroidal melanomas8 for p130. Studies are ongoing to assess this hypothesis on a larger series of patients with follow-up information. In summary, we show that loss of pRB2/p130 and p27kip1 are frequent events in invasive vulvar carcinomas compared with synchronous premalignant lesions, non-neoplastic epithelial disorders, and normal vulvar skin adjacent to the carcinoma, and that their expression in vulvar cancer is not related to any clinicopathologic parameter. In addition, our findings of a significant progressive decrease of pRB2/p130 expression from non-neoplastic epithelial alterations through intraepithelial neoplasia to invasive vulvar carcinomas suggest a role for this tumor suppressor gene in vulvar carcinogenesis. REFERENCES 1. Paggi MG, Baldi A, Bonetto F, et al: Retinoblastoma protein family in cell cycle and cancer: A review. J Cell Biochem 62:418-430, 1996 2. Mayol X, Grana X, Baldi A, et al: Cloning of a new member of the retinoblastoma gene family (pRb2) which binds to the E1A transforming domain. Oncogene 8:2561-2566, 1993 3. Claudio PP, Howard CM, Baldi A, et al: p130/RB2 has growth suppressive properties similar to yet distinctive from those of retinoblastoma family members pRb and p107. Cancer Res 54:5556-5560, 1994 4. Claudio PP, Howard CM, Pacilio C, et al: Mutations in the retinoblastoma-related gene RB2/p130 in lung tumors and suppression of tumor growth in vivo by retroviral-mediated gene transfer. Cancer Res 60:372-382, 2000 5. De Luca A, Esposito V, Howard CM, et al: The retinoblastoma gene family pRB/p105, p107, pRB2/p130 and Simian virus-40 large T-antigen in human mesotheliomas. Nat Med 3:913-916, 1997 6. Leoncini L, Bellan C, Cossu A, et al: Retinoblastoma related p107 and pRB2/p130 proteins in malignant lymphomas: Distinct mechanisms of cell growth control. Clin Cancer Res 5:4065-4072, 1999 7. Susini T, Baldi F, Howard CM, et al: Expression of the retinoblastoma-related gene Rb2/p130 correlates with clinical outcome in endometrial cancer. J Clin Oncol 16:1085-1093, 1998 8. Massaro-Giordano M, Baldi G, De Luca A, et al: Differential expression of the retinoblastoma gene family members in choroidal

9