p27 and Cyclin D1 Abnormalities in Uterine Papillary Serous Carcinoma

Gynecologic Oncology 77, 439 – 445 (2000) doi:10.1006/gyno.2000.5814, available online at http://www.idealibrary.com on

p27 and Cyclin D1 Abnormalities in Uterine Papillary Serous Carcinoma 1 Mary Jo Schmitz, M.D., Denver T. Hendricks, Ph.D.,* John Farley, M.D., Robert R. Taylor, M.D., Joseph Geradts, M.D.,† G. Scott Rose, M.D., and Michael J. Birrer, M.D., Ph.D.* ,2 Division of Gynecologic Oncology, Walter Reed Army Medical Center, Washington, DC 20307; *Molecular Mechanisms Section, Cell and Cancer Biology Department, Medicine Branch, Division of Clinical Sciences, National Cancer Institute, National Institutes of Health, Rockville, Maryland 20850; and †Nuffield Department of Pathology and Bacteriology, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom Received December 13, 1999

INTRODUCTION

Objective. The expression status of p27 and cyclin D1 was examined in 21 uterine papillary serous carcinoma (UPSC) specimens to determine the role of these genes in the development of this disease. The status of p53, p16, Rb, and K-ras was also determined in these tissues so that a marker profile for UPSC could be compared with the published marker profile for other forms of endometrial and ovarian cancer. Methods. Immunohistochemistry was performed on 21 UPSC tissue sections to determine the expression status of p27, cyclin D1, p53, p16, and Rb. K-ras mutations were identified by restriction fragment length polymorphism analysis of DNA isolated from the UPSC sections. Results. All specimens displayed at least one molecular abnormality. A high incidence of p27 alterations were observed, with reduced p27 expression measured in 16 of 21 (76%) tumors, followed by p53 alterations observed in 13 of 21 (62%) tumors. The p27 abnormalities occur at an early stage of the disease, with 63% (5/8) of Stage I cases displaying reduced p27 expression. Cyclin D1 overexpression was observed in 4 of 21 (19%) specimens, whereas p16, Rb, and K-ras abnormalities were each observed in 2 of 21 specimens (10%). Both K-ras mutations were at codon 12. The p16 and Rb abnormalities coexisted in the same specimens. Conclusion. UPSC tumors display a high incidence of p27 abnormalities, suggesting that p27 abnormalities play an important role in the development of this disease. Our results also indicate that cyclin D1 overexpression is involved in the development of a small number of UPSC cases. A comparison of our results with reports by other authors suggests that UPSC shares molecular marker alterations with both ovarian cancer and endometrioid adenocarcinoma. © 2000 Academic Press Key Words: uterine papillary serous carcinoma; endometrial cancer; p27; cyclin D1; p53; p16; Rb; K-ras.

1

Endometrial adenocarcinoma is the most common malignancy of the female genital tract with an annual incidence of approximately 36,000 [1]. It accounts for more than 6300 deaths annually in the United States. The most common histologic type of this disease is endometrioid adenocarcinoma, which makes up 75– 80% of cases. Typically, endometrioid adenocarcinoma presents at early stage which results in a good prognosis with an overall 5-year survival rate of 75%. Several other pathologic subtypes of endometrial cancer exist, with uterine papillary serous cancer (UPSC) being the next most common, accounting for approximately 10% of endometrial cancers. This histologic subtype generally presents at more advanced stages, resulting in a much poorer prognosis with an overall 5-year survival of 33% [2]. The histologic appearance and prognosis of UPSC more closely resemble those of ovarian cancer than endometrioid adenocarcinoma of the endometrium. Although the basis for the aggressiveness and high mortality rate associated with UPSC is unknown, it is anticipated that an understanding of the molecular alterations associated with this disease may provide some insight into its clinical behavior. Deregulated cellular proliferation is frequently due to alterations in genes and gene products involved in cell cycle control or spurious growth signals in the cell cycle regulatory machinery. Specifically, genes associated with the regulation of the G 1 checkpoint in the cell cycle are frequent targets for alterations in cancer. p27 belongs to the kinase inhibitor protein (KIP) family and this protein inhibits the cyclin-dependent kinases associated with cyclins E and D [3]. These particular cyclin kinase complexes regulate cell proliferation by stimulating the phosphorylation of the retinoblastoma protein (Rb), causing the release and activation of E2F (elongation factor 2), a transcriptional factor. E2F facilitates the traversal of the G 1 checkpoint in the cell cycle. p27 expression is frequently reduced in cancers including gastric, breast, prostate, and nonsmall cell lung carcinomas (see Ref. 3 for review). Cyclin D1, which was isolated as the PRAD1 oncogene [4], is frequently

This study was supported in part by funding from the U.S. Navy Bureau of Medicine and Surgery, Clinical Investigation Program, Study B91-097. The opinions expressed herein are those of the authors and do not reflect the official positions of the Department of Defense or any of its Uniformed Service Branches. 2 To whom correspondence should be addressed at Cell and Cancer Biology Department, Medicine Branch, National Cancer Institute, 9610 Medical Center Drive, Room 300, Rockville, MD 20850. Fax: (301) 402 4422. E-mail: [email protected]. 439

0090-8258/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

440

SCHMITZ ET AL.

overexpressed in several cancers including ovarian [5], esophageal [6], pancreatic [7], and breast [8]. Studies in a transgenic murine model have confirmed that cyclin D1 can act as an oncogene [9]. The objective of this study was to measure p27 and cyclin D1 levels in 21 UPSC tumors to evaluate the status of these cell cycle regulatory proteins in this disease. We also examined the status of p16, K-ras, Rb, and p53 to profile the alterations of these marker genes in the context of p27 and cyclin D1 alterations. MATERIALS AND METHODS Case Selection Twenty-two tumor specimens from 1985 to 1997 were obtained from the pathology archives of the National Naval Medical Center, Bethesda, Maryland, and Walter Reed Army Medical Center, Washington, DC, under active institutionally approved protocols. All patients entered into the study had surgery performed by a gynecologic oncologist. Surgical staging or restaging (pre-1988) was assigned based on the 1988 criteria established by the International Federation of Gynecology and Obstetrics (FIGO). All specimens were reviewed in the Tumor Board by members of the Department of Pathology and Division of Gynecologic Oncology. Patient 8 was excluded because insufficient tumor was available for the study; only a focus of tumor was present in a polyp. Inpatient and outpatient charts were reviewed to determine stage, treatment modality, and survival. DNA Extraction and Analysis All paraffin-embedded tissue blocks were serially sectioned for H&E staining, DNA analysis, and immunohistochemical staining. Tissue used for molecular analysis was sandwiched between two H&E sections to ensure the presence of the pathologic abnormality. Five-micrometer sections were placed on gelatin-coated slides for immunohistochemical analysis. Eight micron unstained sections were deparaffinized and digested for DNA analysis as previously described [10]. Immunohistochemistry p27. p27 was detected using mouse monoclonal antibody K2050 (Transduction Laboratories, Lexington, KY) as previously described [11]. Immunohistochemical (IHC) staining was evaluated by visual counting of cells from at least 10 random high-power fields (100⫻) with a minimum of 1000 cells counted by three authors (M.J.S., M.J.B., J.F.). Samples with reduced staining showed ⬍50% of the assessed cells stained positive for p27. Cyclin D1. DCS-6, a mouse monoclonal anti-human cyclin D1 antibody was used to detect cyclin D1, followed by the VectaStain Elite ABC kit for IHC visualization. A cutoff value

of 5% nuclear staining was used to separate normal staining (⬍5%) from cyclin D1-overexpressing cells (⬎5%). p53. A mouse monoclonal antibody, DO-7 (Dako Corp., Carpenteria, CA), directed against human p53 was used for IHC staining as described by Teneriello et al. [10]. Results of the IHC staining were reviewed without knowledge of the molecular data by three authors (M.J.S., R.R.T., M.J.B.). Each specimen was also assessed for the presence of tumor on H&E slides. Rb. A murine monoclonal antibody, 3C8 (QED BioScience Inc., San Diego, CA), directed against human Rb was used for immunohistochemical detection of Rb [12]. Normal Rb expression was scored as nuclear staining in all areas of the tissue, with cell-to-cell heterogeneity and variable staining intensity. Abnormal nuclear staining was identified by the absence of staining in all or a portion of the tumor with preserved reactivity in admixed nonneoplastic tissue. All specimens were reviewed without prior knowledge of patient history or molecular analysis (J.G., M.J.B.). p16. Sections were incubated with p16 antiserum (Pharmingen, San Diego, CA) as described previously [13]. The nuclear staining observed in normal ovarian stromal tissue served as a positive control for p16 staining, and abnormal p16 staining was identified by the absence of staining in tumor tissue in the presence of normal immunoreactivity in adjacent nonneoplastic tissue. Polymerase Chain Reaction (PCR) Amplification and Detection of K-ras Gene Mutations PCR amplification for K-ras was performed using a Perkin– Elmer Cetus 9600 thermocycler (Perkin–Elmer Cetus, Norwalk, CT). A 5-␮l aliquot of the DNA solution was PCR amplified in a final volume of 50 ␮l as previously described [10], using oligonucleotides obtained from Midland Labs (Midland, TX). Mutations were detected by restriction fragment length polymorphism (RFLP) analysis using mismatched primers [10]. Restriction enzyme digestion of the PCR mixture was performed according to the manufacturer’s recommendations (BanI, BclI, and EarI; New England Biolabs, Beverly, MA). DNA fragments were visualized by electrophoresis. The undigested and digested PCR products were paired side by side during electrophoresis to facilitate RFLP identification. RESULTS The tumor specimens of 21 patients were analyzed in our study. Eight patients were stage I, 5 Stage III, and 8 Stage IV. Patients with Stage I disease were termed early stage and those with Stage III and IV disease were called advanced stage for the purpose of comparison in this study. Two patients received adjuvant radiation and chemotherapy, seven had chemotherapy only, ten had adjuvant radiation therapy, one patient died prior to the initiation of therapy, and one patient received no further

441

p27 AND CYCLIN D1 EXPRESSION IN UPSC

TABLE 1 Staging, Clinical History, and Molecular Marker Profile for UPSC Samples Used in Study

Specimen

Stage

p27

Cyclin D1

p53

Rb

K-ras 12

p16

Survival (months)

PS22 PS5 PS11 PS17 PS21 PS3 PS6 PS13 PS2 PS4 PS7 PS12 PS20 PS18 PS1 PS9 PS10 PS14 PS15 PS16 PS19

IA IB IB IB IB IC IC IC IIIC IIIC IIIC IIIC IIIC IVA IVB IVB IVB IVB IVB IVB IVB

—a A A A A A — — A A A A A — A A A A A A —

— A — — A — A — — — — — — — — — — — A — —

A — — — A A — — A A A A — A A A A — — A A

— — — — A — — — — — — — — — — — A — — — —

— — — — — — — A — — — — A — — — — — — — —

— — — — A — — — — — — — I — — — A — — — —

31* 38* 127* 40* 60* 31 70* 28 73 70* 26 14 62* 11 35 11 13 9 15 DD 6

a —, Normal; A, abnormal; I, insufficient tissue; DD, died of other disease; *patient alive. PS10 and PS21 displayed loss of Rb staining in 90 and 50% of tumor tissue, respectively.

therapy after surgery. Six of the eight patients with early-stage disease are surviving, while only 2 of 13 patients with advanced-stage disease remain alive.

tumor cells (T, see arrow), with normal cyclin D1 levels in the adjacent stromal cells (S, see arrow). Other Molecular Markers

p27 and Cyclin D1 The UPSC samples displayed a high incidence of altered p27 expression levels, with 76% (16/21) of samples displaying reduced p27 staining, relative to adjacent normal tissue (Table 1, Fig. 1). Based on our criteria, samples with reduced staining showed ⬍50% of the assessed cells stained positive for p27, with a minimum of 1000 cells counted each time by three independent assessors. Figure 1 illustrates UPSC tumor tissue with normal (A) and reduced (B, C) p27 staining, respectively. In Figs. 1B and 1C, the tumor tissue displayed reduced staining for p27, in contrast to normal levels of staining for p27 in the adjacent stromal tissue (see arrows). As shown in Table 1, reduced p27 staining was not confined to late-stage tumors, with 63% (5/8) of Stage I tumors displaying reduced staining for p27. Elevated cyclin D1 was detected in 19% (4/21) of the tumors (PS5, PS6, PS15, PS21) (Fig. 1). Three of the tumors with elevated cyclin D1 levels also displayed reduced p27 levels whereas sample PS6 showed no alterations in any of the other markers analyzed in this study (Table 1). Figure 1D shows a tissue section with elevated staining for cyclin D1. The figure clearly shows the elevated cyclin D1 staining confined to the

Sixty-two percent (13/21) of the UPSC samples analyzed displayed elevated p53 levels, and this was the second most common marker altered in this study. Thirty-eight percent (3/8) of Stage I tumors overexpressed p53 compared with 77% (10/13) of advanced-stage tumors (Table 1). A representative tumor sample staining positive for p53 is shown in Fig. 1E. All tumor samples showing altered p53 levels in Table 1 displayed strong nuclear staining ⬎10% of tumor tissue in the section analyzed. From Table 1, 48% (10/21) of tumor sections showed alterations for both p27 and p53, 3 with alterations for p53 alone and 3 with alterations for p27 only. All other markers displayed a low incidence of alterations, with Rb (2/21), K-ras (2/21), and p16 (2/21) each altered in 10% of the tumors analyzed (Table 1). Figure 1F shows one of the tumor samples with reduced p16 staining in the tumor cells (T, see arrow), and normal p16 staining in the epithelial cells lining a cyst (N, see arrow). All K-Ras mutations were observed at codon position 12 and none at position 61. Rb (2/21) and p16 (2/21) abnormalities were present in the same tissue samples (PS10 and PS21), and these samples also displayed abnormal p53 and p27 levels. Interestingly, these were Stage IVB and IB tumors, respectively.

442

SCHMITZ ET AL.

FIG. 1. (A) Normal p27 immunostaining in tumor cells (T) in specimen PS13. (B) Reduced p27 staining in tumor cells (T) in PS 1. Stromal cells (S) show normal stain for p27. (C) Reduced p27 staining in tumor cells (T), with stromal cells (S) showing normal staining for p27 in PS1. (D) Overexpression of cyclin D1 in tumor cells (T) compared with negative cyclin D1 staining in stromal cells (S) in PS21. (E) Positive immunostaining for p53 in PS10. (F) Absence of p16 staining in tumor cells (T) compared with intense p16 staining in normal epithelium (N) in specimen PS10.

DISCUSSION This study identified reduced p27 levels as one of the most frequently observed molecular abnormalities in UPSC. More than 75% of the tumors analyzed here had reduced p27 expression. In addition, the reduced p27 expression was also observed in early-stage tumors, suggesting that this alteration

is an early event in the etiology of UPSC. These observations suggest that p27 plays an important role in the development and/or progression of UPSC, presumably by facilitating a proliferative signal in those tumors with reduced p27 levels. Previous reports show that reduced p27 expression levels correlate with poor prognosis in several other cancers (see Ref. 3 for review). The data set in this study was unfortunately too

443

p27 AND CYCLIN D1 EXPRESSION IN UPSC

small for any meaningful statistical analysis. Current evidence indicates that alterations in p27 function rarely result from mutational events in this gene [14] but rather from decreased protein expression due to accelerated proteosomal degradation of this cell cycle protein [3]. However, there is a paucity of data concerning the upstream events leading to increased p27 turnover. This is also the first report on the status of cyclin D1 in UPSC samples and we show that this protein was overexpressed in 19% (4/21) of the tumors analyzed here. Three of the cases with elevated cyclin D1 levels also displayed reduced p27 levels. Although altered cyclin D1 levels correlated with prognosis in cancers of the esophagus [6] and the pancreas [7], the small numbers in this study preclude an accurate analysis of prognostic significance for cyclin D1 on UPSC survival. From our data, it would appear that cyclin D1 is involved in the development of a small number of UPSC cases, but it remains to be seen whether cyclin D1 alterations have any prognostic significance in this infrequent, but devastating disease. From a clinical perspective, UPSC presents a unique picture as an endometrial neoplasm. It is typically diagnosed at late stage and has a poor prognosis, in contrast to other subtypes of endometrial cancer which are diagnosed much earlier and have more favorable prognoses [2]. Histologically, UPSC resembles papillary serous carcinoma of the ovary, and the pattern of metastatic spread and the clinical progression of UPSC also resemble those of ovarian cancer rather than endometrial cancer [15]. In addition, the treatment regimen for UPSC is very similar to the treatment used for ovarian cancer [15]. In this context, investigators have closely examined the profile of molecular alterations in UPSC and other subtypes of endometrial cancer and papillary serous ovarian cancer to better understand the relationship between these disease entities. In a recent report, Carduff et al. [16] demonstrated the close similarities between the molecular alterations in UPSC and ovarian cancer and the differences between UPSC and endometrioid uterine cancer. Table 2 summarizes some of the molecular alterations that have been published for UPSC, ovarian papillary serous cancer (OPSC), and uterine endometrioid adenocarcinoma (UEA). The profile of altered markers in UPSC more closely resembles the profile seen in OPSC rather than that in UEA, particularly for p53, Rb, K-Ras, and p16 (Table 2). The high incidence of p53 alterations (62%) in this study is consistent with previous reports for p53 alterations in UPSC tumor samples [25–27]. Our data suggest that p53 alterations (like p27 alterations) are an early event in the development of UPSC. Ovarian papillary serous carcinomas also display a high frequency of p53 alterations, in contrast to the lower frequency of p53 alterations reported for uterine endometrioid adenocarcinomas (Table 2). Other workers have shown that endometrioid endometrial carcinomas display a higher incidence of p16 and K-Ras alterations and a lower incidence of p53 alterations than observed for UPSC (this study). This underscores the histologic and

TABLE 2 Comparison of Altered Molecular Markers in Papillary Serous Carcinomas and Endometrioid Adenocarcinomas of the Uterus and Papillary Serous Carcinomas of the Ovary % Tumor samples with abnormal marker Marker

UPSC

UEA

OPSC

p27

76 (this study)

Cyclin D1

19 (this study)

81 [17] 68 a 41 [20] 56 [21]

p53

48 [24] 62 (this study) 71 [25] 75 [16] 78 [26] 85 [27] 100 [28] 0 [16] 10 (this study)

5 [16] 29 [25] 34 [29] 41 [20] 57 [28]

50 [18] 65 [19] 13 [5] 32 [19] 33 [23] 39 [22] 58 [30] 60 [32] 62 [16] 81 [31]

p16

10 (this study)

Rb

10 (this study)

20 [35] 20 [37] 34 [36] 4 [40] 3 [37]

K-ras

a

13 [16] 18 [33]

0 [16] 0 [10] 4 [12] 5 [34] 10 [38] 24 [39] 14 [12] 4 [41]

Unpublished data from this laboratory.

clinical differences between these two types of endometrial cancer (Table 2). Although the molecular marker profile in UPSC resembles the marker profile for OPSC rather than that for UEA, this similarity depends on the markers analyzed. The clustering of UPSC and OPSC is less dramatic if we include the results for p27 and cyclin D1 generated in this study. Recent reports show a high incidence of p27 alterations in OPSC (50% [18], 65% [19], and 60% [Farley, unpublished data]) as well as UEA [17]. In addition, although UEA was reported to show a higher incidence of cyclin D1 alterations (41% [20] and 56% [21]) than observed in the present study, ovarian cancer displays a wide range of cyclin D1 alterations (13% [5], 32% [19], 33% [23], and 39% [22]). It would be difficult to group UPSC tumors with OPSC tumors (rather than UEA tumors) on the basis of reduced p27 levels and cyclin D1 overexpression levels. This suggests that UPSC and UEA display similarities and differences in their profile of altered molecular markers that may reflect similarities and differences in the cell type of origin, exposure to carcinogens, and/or hormonal effects. This observation does not reduce the significance of the observed clinical, histological, and molecular similarities between UPSC and ovarian cancer. The evidence suggests that UPSC and serous ovarian cancer share some features that may have important implications for

444

SCHMITZ ET AL.

the diagnostic and therapeutic strategies that we consider for these two types of cancers. Furthermore, the high incidence of p27 alterations observed in UPSC tumors in this study suggests that this gene plays an important role in the tumorigenesis of UPSC. ACKNOWLEDGMENTS We thank Dr. L. Smith, Dr. M. Reid, and Mr. S. Wise for their technical assistance, and Dr. J. Carlson for his critical reading of the manuscript.

Practice of Gynecologic Oncology. 2nd ed. Philadelphia, Lippincott– Raven, 1997 16. Caduff RF, Svoboda-Newman SM, Bartos RE, Ferguson AW, Frank TS: Comparative analysis of histologic homologues of endometrial and ovarian carcinoma. Am J Surg Pathol 22:319 –326, 1998 17. Bamberger A, Riethdorf L, Milde-Langosch K, Bamberger CM, Thuneke I, Erdmann I, Schulte HM, Lo¨ning T: Strongly reduced expression of the cell cycle inhibitor p27 in endometrial neoplasia. Virchow’s Arch 434: 423– 428, 1999

REFERENCES

18. Masciullo V, Sgambato A, Pacilio C, Pucci B, Ferrandina G, Palazzo J, Carbone A, Cittadina A, Mancuso S, Scambia G, Giordano A: Frequent loss of expression of the cyclin-dependent kinase inhibitor p27 in epithelial ovarian cancer. Cancer Res 59:3790 –3794, 1999

1. Landis SH, Murray T, Bolden S, Wingo PA: Cancer statistics, 1998. CA Cancer J Clin 48:6 –30, 1998

19. Sui L, Tokuda M, Ohno M, Hatase O, Hando T: The concurrent expression of p27 (kip1) and cyclin D1 in epithelial ovarian tumors. Gynecol Oncol 73:202–209, 1999

2. Wilson TO, Podratz KC, Gaffey TA, Malkasian GD Jr, O’Brien PC, Naessens JM: Evaluation of unfavorable histologic subtypes in endometrial adenocarcinoma. Am J Obstet Gynecol 162:418 – 423, 1990

20. Nikaido T, Li SF, Shiozawa T, Fujii S: Coabnormal expression of cyclin D1 and p53 protein in human uterine endometrial carcinomas. Cancer 78:1248 –1253, 1996

3. Lloyd RV, Erickson LA, Jin L, Kulig E, Qian X, Cheville JC, Scheithauer BW: p27kip1: A multifunctional cyclin-dependent kinase inhibitor with prognostic significance in human cancers. Am J Pathol 154:313–323, 1999

21. Ito K, Sasano H, Yoshida Y, Sato S, Yajima A: Immunohistochemical study of cyclins D and E and cyclin dependent kinase (cdk) 2 and 4 in human endometrial carcinoma. Anticancer Res 18:1661–1664, 1998

4. Arnold A, Kim HG, Gaz RD, Eddy RL, Fukusima Y, Kronenberg HM: Molecular cloning and chromosomal mapping of DNA rearranged with the parathyroid hormone gene in a parathyroid adenoma. J Clin Invest 83: 2034 –2040, 1989

22. Shigemasa K, Tanimoto H, Parham GP, Parmley TH, Ohama K, O’Brien TJ: Cyclin D1 overexpression and p53 mutation status in epithelial ovarian cancer. J Soc Gynecol Invest 6:102–108, 1999

5. Worsley SD, Ponder BAJ, Davis RR: Overexpression of cyclin D1 in epithelial ovarian tumors. Gynecol Oncol 64:189 –195, 1997 6. Anayama T, Furihata M, Ishikawa T, Ohtsuki Y, Ogoshi S: Positive correlation between p27Kip1 expression and progression of human esophageal squamous cell carcinoma. Int J Cancer 79:439 – 443, 1998 7. Kornmann M, Ishiwata T, Itakura J, Tangvoranuntakul P, Beger HG, Korc M: Increased cyclin D1 in human pancreatic cancer is associated with decreased postoperative survival. Oncology 55:363–369, 1998 8. Michalides RF, Hageman P, van Tinteren H, Houben L, Weintjens E, Klompmaker P, Pterse J: A clinicopathological study on overexpression of cyclin D1 and of p53 in a series of 248 patients with operable breast cancer. Br J Cancer 73:728 –734, 1996

23. Hung W, Chai C, Huang J, Chuang L: Expression of cyclin D1 and c-Ki-ras gene product in human epithelial ovarian tumors. Hum Pathol 27:1324 –1328, 1996 24. Bancher-Todesca D, Gitsch G, Williams KE, Kohlberger P, Neunteufel W, Obermair A, Heinze G, Breitenecker G, Hacker NF: p53 protein overexpression: A strong prognostic factor in uterine papillary serous carcinoma. Gynecol Oncol 71:59 – 63, 1998 25. Zheng W, Cao P, Zheng M, Kramer EE, Godwin TA: p53 overexpression and bcl-2 persistence in endometrial carcinoma: Comparison of papillary serous and endometrioid subtypes. Gynecol Oncol 61:167–174, 1996 26. Kovalev S, Marchenko ND, Gugliotta BG, Chalas E, Chumas J, Moll UM: Loss of p53 function in uterine papillary serous carcinoma. Hum Pathol 29:613– 619, 1998

9. Wang TC, Cardiff RD, Zukerberg L, Lees E, Arnold A, Schmidt EV: Mammary hyperplasia and carcinoma in MMTV-cyclin D1 transgenic mice. Nature 369:669 – 671, 1994

27. Moll UM, Chalas E, Auguste M, Meaney D, Chumas J: Uterine papillary serous carcinoma evolves via a p53-driven pathway. Hum Pathol 27: 1295–1300, 1996

10. Teneriello MG, Ebina M, Linnoila RI, Henry M, Nash JD, Park RC, Birrer MJ: p53 and ki-ras gene mutations in epithelial ovarian neoplasm. Cancer Res 53:3103–3108, 1993

28. Geisler JP, Geisler HE, Wiemann MC, Zhou Z, Miller GA, Crabtree W: p53 expression as a prognostic indicator of 5-year survival in endometrial cancer. Gynecol Oncol 74:468 – 471, 1999

11. Castavelos C, Bhattacharya N, Ung Y, Wilson J, Roncari L, Sandhu C: Decreased levels of the cell-cycle inhibitor p27kip1 protein: Prognostic implications in primary breast cancer. Nat Med 3:227–230, 1997

29. Shiozawa T, Xin L, Nikaido T, Fujii S: Immunohistochemical detection of cyclin A with reference to p53 expression in endometrial endometrioid carcinomas. Int J Gynecol Pathol 16:348 –353, 1997

12. Taylor RR, Linnoila RI, Gerardts J, Teneriello MG, Nash JD, Park RC, Birrer MJ: Abnormal expression of the retinoblastoma gene in ovarian neoplasma and correlation to p53 and K-ras mutations. Gynecol Oncol 58:307–311, 1995

30. Eltabbakh GH, Belinson JL, Kennedy AW, Biscotti CV, Casey G, Tubbs RR, Blumenson LE: p53 overexpression is not an independent prognostic factor for patients with primary ovarian epithelial cancer. Cancer 80:892– 898, 1997

13. Geradts J, Kratzke RA, Niehans GA, Lincoln CE: Immunohistochemical detection of cyclin-dependent kinase inhibitor 2/multiple tumor suppressor gene 1(CDKN2/MTS1) product p16ink4a in archival human solid tumors: Correlation with retinoblastoma protein expression. Cancer Res 55:6006 – 6011, 1995

31. Wen WH, Reles A, Runnebaum IB, Sullivan-Halley J, Bernstein L, Jones LA, Felix JC, Kreienberg R, El-Naggar A, Press MF: p53 and expression in ovarian cancers: Correlation with overall survival. Int J Gynecol Pathol 18:29 – 41, 1999

14. Kawamata N, Morosetti R, Miller CW: Molecular analysis of the cyclin dependent kinase inhibitor gene p27/Kip1 in human malignancies. Cancer Res 55:2266 –2269, 1995

32. Dong Y, Walsh MD, McGuckin MA, Cummings MC, Gabrielli BG, Wright GR, Hurst T, Khoo SK, Parsons PG: Reduced expression of retinoblastoma gene product (pRB) and high expression of p53 are associated with poor prognosis in ovarian cancer. Int J Cancer 74:407– 415, 1997

15. Barakat RR, Park RC, Grigsby PW, Muss HD, Norris HJ: Corpus: Epithelial tumors, in Hoskins WJ, Perez CA, Young RC (Eds), Principles and

33. Sasaki H, Nishii H, Takahashi H, Tada A, Furusato M, Terashima Y, Siegal GP, Parker SL, Kohler MF, Berchuk A, Boyd J: Mutation of the

p27 AND CYCLIN D1 EXPRESSION IN UPSC

34.

35.

36.

37.

Ki-Ras protooncogene in human endometrial hyperplasia and carcinoma. Cancer Res 53:1906 –1910, 1993 Mandai M, Konishi I, Kuroda H, Komatsu T, Yamamoto S, Nanbu K, Matshushita K, Fukumoto M, Yamabe H, Mori T: Heterogeneous distribution of K-ras-mutated epithelia in mucinous ovarian tumors with special reference to histopathology. Hum Pathol 28:34 – 40, 1998 Nakashima R, Fujita M, Enomoto T, Haba T, Yoshino K, Wada H, Kurachi H, Sasaki M, Wakasa K, Inoue M, Buzard G, Murata Y: Br J Cancer 80:458 – 467, 1999 Shiozawa T, Nikaido T, Shimizu M, Zhai Y, Fujii S: Immunohistochemical analysis of the expression of cdk4 and p16 INK4 in human endometrioidtype endometrial carcinoma. Cancer 80:2250 –2256, 1997 Milde Langosch K, Riethdorf L, Bamberger A, Lo¨ning T: p16/MTS1 and pRB expression in endometrial carcinomas. Virchow’s Arch 434:23–28, 1999

445

38. Milde-Langosch K, Ocon E, Becker G, Loning T: p16/MTS1 inactivation in ovarian carcinomas: High frequency of reduced protein expression associated with hyper-methylation or mutation in endometrioid and mucinous tumors. Int J Cancer 79:61– 65, 1998 39. Fujita M, Enomoto T, Haba T, Nakashima R, Sasaki M, Yoshino K, Wada H, Buzard GS, Matszaki N, Wakasa K, Murata Y: Alteration of p16 and p15 genes in common epithelial ovarian tumors. Int J Cancer 74:148 –155, 1997 40. Niemann TH, Yilmaz AG, McGaughy VR, Vaccarello L: Retinoblastoma protein expression in endometrial hyperplasia and carcinoma. Gynecol Oncol 65:232–236, 1997 41. Dodson MK, Cliby WA, Xu H, DeLacey KA, Hu S, Keeney GL, Li J, Podratz KC, Jenkins RB, Benedict WF: Evidence of functional Rb protein in epithelial ovarian carcinomas despite loss of heterozygosity at the Rb locus. Cancer Res 54:610 – 613, 1994