Endometriosis-associated ovarian carcinoma

Share Embed


Descrição do Produto

1658

Endometriosis-Associated Ovarian Carcinoma Differential Expression of Vascular Endothelial Growth Factor and Estrogen/Progesterone Receptors

Marcela G. del Carmen, M.D.1 Anne E. Smith Sehdev, M.D.2 Amanda Nickles Fader, M.D.3 Marianna L. Zahurak, M.D.4 Michael Richardson, M.D.5 John P. Fruehauf, M.D.5 F. J. Montz, M.D., K.M.1* Robert E. Bristow, M.D.1 1

The Kelly Gynecologic Oncology Service, Department of Gynecology and Obstetrics/Department of Oncology, The Johns Hopkins Hospital and Medical Institutions, Baltimore, Maryland.

2

Division of Gynecologic Pathology, Department of Gynecology and Obstetrics/Department of Pathology, The Johns Hopkins Hospital and Medical Institutions, Baltimore, Maryland.

3

Department of Obstetrics, Gynecology, and Reproductive Sciences, Magee-Women’s Hospital, Pittsburgh, Pennsylvania.

4

Department of Biostatistics, The Johns Hopkins Oncology Center, The Johns Hopkins Hospital and Medical Institutions, Baltimore, Maryland.

5

Oncotech Incorporated, Irvine, California.

BACKGROUND. Multiple epidemiologic and histologic studies have suggested that ovarian endometriosis can give rise to malignant ovarian tumors, primarily those of epithelial origin. The progression of endometriosis to endometriosis-associated ovarian carcinoma (EAOC) has not been investigated thoroughly and is poorly understood at best. Using immunohistochemical methods, we compared the differential expression patterns of various cytokines and growth factors in atypical endometriosis (AE) and EAOC. METHODS. Using the Johns Hopkins Pathology Data Bank, tissue blocks from patients diagnosed with EAOC or AE were identified. Tissue blocks were stained for 4 markers: vascular endothelial growth factor (VEGF), Ki-67, estrogen receptor (ER), and progesterone receptor (PR). RESULTS. Seventeen cases of EAOC and 8 cases of AE were identified. Staining for VEGF was documented in 16 of 17 (94%) EAOC tissue blocks and in only 1 of 8 (12.5%) AE tissue blocks (P ⬍ 0.0001). Only 4 of the 17 (23%) EAOC tissue blocks exhibited positive staining for ER, compared with 8 of 8 (100%) AE tissue blocks (P ⫽ 0.0005). Positive staining for PR was noted in only 6 of 17 (35%) EAOC samples but was present in 8 of 8 (100%) AE samples (P ⫽ 0.003). Seventy percent of EAOC samples exhibited positive staining for Ki-67, compared with 37.5% of AE samples (P ⫽ 0.19). CONCLUSIONS. EAOC appears to be associated with overexpression of VEGF and reduced expression of both ER and PR. Variations in VEGF expression may be associated with the malignant transformation of endometriosis and may present both diagnostic and therapeutic options for the treatment of ovarian malignancies. Cancer 2003;98:1658 – 63. © 2003 American Cancer Society.

Marcela G. del Carmen’s current address: Gynecologic Oncology Service, Gillette Center for Women’s Cancers, The Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts.

KEYWORDS: ovarian carcinoma, endometriosis, vascular endothelial growth factor, estrogen receptor, progesterone receptor.

* Deceased.

O

Address for reprints: Marcela G. del Carmen, M.D., Gynecologic Oncology Service, Gillette Center for Women’s Cancers, Cox 5, 100 Blossom Street, Boston, MA 02114; Fax: (617) 726-1949; E-mail: [email protected] John P. Fruehauf is chief scientific officer of Oncotech Incorporated (Irvine, California), the commercial reference lab that performed the immunohistochemical tests described in this article. Therefore, he has a related financial interest in this work. Received April 23, 2003; accepted July 21, 2003. © 2003 American Cancer Society DOI 10.1002/cncr.11714

varian carcinoma is the leading cause of death due to gynecologic malignancy and the fourth most common cause of death due to cancer in the United States.1 The high mortality associated with this malignancy is due in part to the presence of widely metastatic disease in most patients at the time of initial diagnosis. To date, there are no adequate methods available for early detection. Ovarian carcinogenesis is poorly understood. Elucidation of the mechanism of this malignancy may be achieved by better understanding endometriosis. Endometriosis, the presence of viable endometrial tissue outside the uterus, affects 5–15% of premenopausal women and 3–5% of postmenopausal women.2 Although endometriosis is not a frank malignancy, its behavior is characterized by certain features that also are seen in malignant lesions.3 Like malignant disease, endometriotic implants can metastasize to local and distant sites. Endometriotic

Endometriosis-Associated Ovarian Carcinoma/del Carmen et al.

lesions also can attach to, invade, and damage other tissue, but unlike malignant lesions, endometriosis does not result in a catabolic state and rarely is fatal.3 In 1925, Sampson4 theorized that endometriotic lesions could undergo malignant transformation. In recent years, both histologic and epidemiologic evidence has accumulated suggesting that ovarian endometriosis can give rise to malignant ovarian tumors, primarily those that are epithelial in origin, which are known as endometriosis-associated ovarian carcinoma (EAOC).3,5– 8 Recent data suggest that cytologic and/or structural atypia in endometriotic implants may be associated with ovarian malignancies and that a transformative route from atypical endometriosis (AE) to carcinoma may exist.9,10 Ki-67 indices have been shown to be higher in AE than in typical endometriotic implants. The Ki-67 index in AE, however, is lower than in malignant ovarian lesions.10 Vascular endothelial growth factor (VEGF) also appears to play a role in angiogenesis and tumor progression in ovarian carcinoma. VEGF is a dimeric, endothelial cell–specific glycoprotein that induces microvascular permeability and leads to the extravasation of plasma proteins and to proangiogenic stromal changes.11 VEGF expression has been documented in a series of solid tumors, including ovarian malignancies. Its role in ovarian carcinogenesis may be related to angiogenesis and autocrine/paracrine autoregulation.11,12 Both estrogen receptors (ER) and progesterone receptors (PR) are expressed in endometriotic tissue.2,13,14 Loss of steroid receptor expression in EAOC may also be part of carcinogenesis and may be indicative of cell dedifferentiation. Our hypothesis is that a transformative pathway exists between AE and EAOC. Malignant changes may be heralded by the expression of certain proteins that are not present or are expressed at a lower level in endometriosis or atypical endometriosis than in EAOC. We proposed VEGF and Ki-67 as candidate proteins that may be involved in carcinogenesis and may be markers of characteristic immunohistochemical differences between premalignant and malignant lesions. The current study also investigated the expression of ER and PR in AE and EAOC.

MATERIALS AND METHODS Approval for the current study was obtained from the Johns Hopkins Medical Institutions (JHMI; Baltimore, MD) Clinical Research Committee and the JHMI Joint Committee on Clinical Investigation. All patients underwent initial surgical treatment at the Johns Hopkins Hospital. Gynecologic pathologists at our institution made the initial histologic diagnoses. Our institution’s pathology database was used to identify

1659

tissue blocks from patients diagnosed with either AE or EAOC. A total of 25 patient samples were studied (17 EAOC and 8 AE). All slides were re-reviewed independently by a single pathologist (A. S.) to verify diagnoses. Diagnoses of AE were based on the presence of large, hyperchromatic nuclei with moderate or marked pleomorphism, cellular crowding, increased nuclear-to-cytoplasmic ratio, and tufting.15,16 Immunohistochemical (IHC) staining for VEGF, Ki-67, ER, and PR was performed on fixed, paraffinembedded tissue blocks. All IHC blocks were assessed by a single pathologist (M. R.). Sections (thickness, 3– 4 ␮m) were deparaffinized in Histoclear威 (Raymond A. Lamb Laboratories, Apex, NC) and then rehydrated in descending grades (100% to 70%) of ethanol. Antigen retrieval for VEGF was performed with protease digestion. For the Ki-67 procedure, antigen retrieval was performed with microwaving to 20 minutes, at 5-minute intervals. For ER and PR antigen retrieval, slides were kept in a pressure cooker for 20 minutes. Automated IHC procedures were performed using I6000 (BioGenex, San Ramon, CA) and NexES (Ventana Medical Systems, Phoenix, AZ) immunostainers. Endogenous peroxidase activity was blocked with 3% hydrogen peroxidase in distilled water during a 10minute treatment. For monoclonal antibodies, sections were preincubated with normal goat serum (BioGenex) to minimize background staining. Slides were incubated for 30 – 60 minutes with primary antibodies. Immunoperoxidase staining was performed using a supersensitive streptavidin-biotin detection kit (BioGenex) or a universal avidin-biotin detection system (DAB Detection Kit; Ventana Medical Systems). Next, counterstaining was performed with hematoxylin or methyl green, a cover slip was applied to the slide, and the slide was viewed via light microscopy. All slides were reviewed by a single histopathologist (M. R.) who was blinded to the original diagnoses. Entire tissue sections were evaluated for staining intensity and the percentage of tumor cells staining positively. In each case, only the malignant component was assigned a score. Positive and negative controls were run for each type of staining. The control tissue varied depending on the IHC stain. Staining intensity was scored as negative (⬍ 5% of tumor cells exhibiting staining), 1⫹ (mild intensity), 2⫹ (moderate intensity), 3⫹ (intensity equal to that of the positive control), or 4⫹ (intensity greater than that of the positive control).17,18 HScore values, which are objective measurements of tumor heterogeneity that reflect the variability in the percentage of cells stained and in the intensity of staining within a tumor and provide a single quantity describing the amount of marker present in the field examined, were calculated by mul-

1660

CANCER October 15, 2003 / Volume 98 / Number 8

FIGURE 1. (A,B) Diffuse and strong immunostaining, respectively, for vascular endothelial growth factor (VEGF) in endometriosis-associated ovarian carcinoma. (C,D) In atypical endometriosis, low VEGF staining intensity was noted.

tiplying the percentage of cells with positive staining by the staining intensity score plus 1 (HScore ⫽ [% positive] ⫻ [intensity ⫹ 1]). In the AE tissue samples, epithelial and stromal cells were stained together. The percentage of stained cells was assessed by direct observation of the entire tissue sample at 100⫻ and 200⫻ magnification. All staining results were confirmed by a second observer. Markers associated with histology were selected based on cross-tabulation and Fisher exact test results. Logistic regression models were used to estimate odds ratios (ORs) and 95% confidence intervals (CIs). The small sample sizes in the current study necessitated the use of exact methods for these logistic regressions. All statistical computations were performed using LogXact (Cytel Software Corporation, Cambridge, MA) and the SAS system (SAS Institute, Cary, NC).19 –21 All reported P values are two-sided. P values less than 0.05 were considered statistically significant.

RESULTS Sixteen of the 17 EAOC specimens (94.1%) stained positively for VEGF, compared with only 12.5% of the AE specimens (P ⬍ 0.0001). The OR was calculated to be 70.64 (95% CI, 4.54 –5332) for VEGF staining in these samples (i.e., compared with AE samples, EAOC samples were 70.64 times more likely to overexpress VEGF); this finding was statistically significant (P ⫽ 0.0003). The width of the CI for the OR is a reflection of our small sample size. The mean VEGF HScore for EAOC samples was 280 (range, 0 –360); the HScore of

the AE sample that stained positively for VEGF was 90 (Fig. 1). Ki-67 expression was observed in 12 of 17 EAOC samples and in only 3 of 8 AE samples (P ⫽ 0.19). The OR for Ki-67 staining was 3.67 (95% CI, 0.50 –34.68; P ⫽ 0.26). The mean Ki-67 HScore for EAOC samples was 180 (range, 0 –210), compared with 80 (range, 0 –150) for AE samples. Positive staining for ER was noted in 23.5% of EAOC tissue blocks and in all AE tissue blocks (P ⫽ 0.0005). The OR for ER expression was 0.04 (i.e., compared with AE samples, EAOC samples were only 0.04 times as likely to express ER); this finding was statistically significant (P ⫽ 0.0009). The mean ER HScore for EAOC samples was 90 (range, 0 –180), compared with 180 (range, 140 –270) for AE samples. Thirty-five percent of EAOC tissue blocks and all AE tissue blocks stained positively for PR (P ⫽ 0.003). The OR for PR expression was 0.06 (i.e., compared with AE samples, EAOC samples were only 0.06 times as likely to express PR); this finding also was statistically significant (P ⫽ 0.006). The mean PR HScore for EAOC samples was 90 (range, 0 –150), compared with 270 (range, 180 –360) for AE samples. Tables 1 and 2 summarize the IHC staining results and HScore data, respectively.

DISCUSSION Little doubt exists that ovarian endometriosis can lead to malignant transformation. Although this malignant predisposition was first described by Sampson4 in 1925, the mechanism of the transformation remains

Endometriosis-Associated Ovarian Carcinoma/del Carmen et al. TABLE 1 Immunohistochemical Staining Results No. of patients with positive staining (%)

EAOC (n ⫽ 17) AE (n ⫽ 8) P valuea

VEGF

Ki-67

ER

PR

16/17 (94.1) 1/8 (12.5) ⬍0.0001

12/17 (70.5) 3/8 (37.5) 0.19

4/17 (23.5) 8/8 (100) 0.0005

6/17 (35.3) 8/8 (100) 0.003

VEGF: vascular endothelial growth factor; ER: estrogen receptor; PR: progesterone receptor; EAOC: endometriosis-associated ovarian carcinoma; AE: atypical endometriosis. a P values less than 0.05 were considered statistically significant.

TABLE 2 HScoresa for Vascular Endothelial Growth Factor, Ki-67, Estrogen Receptor, and Progesterone Receptor Mean HScore (range)

EAOC (n ⫽ 17) AE (n ⫽ 8)

VEGF

Ki-67

ER

PR

280 (0–360) 90 (—)

180 (0–210) 80 (0–150)

90 (0–180) 180 (140–270)

90 (0–150) 270 (180–360)

VEGF: vascular endothelial growth factor; ER: estrogen receptor; PR: progesterone receptor; EAOC: endometriosis-associated ovarian carcinoma; AE: atypical endometriosis. a HScores were calculated by multiplying the percentage of cells exhibiting positive staining by the staining intensity plus 1 (HScore ⫽ [% positive] ⫻ [intensity ⫹ 1]).

unclear.22 Among patients with endometriosis, 0.3– 1.6% will develop a malignant ovarian tumor, most often of endometrioid or clear-cell histology.3,5,6 In a series of 556 patients undergoing surgery for ovarian malignancies, the frequency of concurrent endometriosis was investigated; it was found that the frequency of endometriosis in patients with serous and mucinous tumors ranged from 3.6% to 5.6%, and endometriosis was noted in 26.3% of patients with endometrioid malignancies and 21.2% of patients with clear cell malignancies.7 Brinton et al.8 reported on a series of 20,686 Swedish women initially hospitalized for endometriosis; these women were found to have a 2-fold excess risk of developing an ovarian malignancy, which increased to a nearly 4-fold excess risk after a follow-up period of greater than 10 years. Patients with endometriosis who develop EAOC have a distinct clinical profile. In particular, patients with EAOC are younger, present with earlier-stage disease, and have longer disease-free survival than do patients with epithelial ovarian carcinoma.23 In fact, the biologic behavior of ovarian malignancies that arise in association with endometriosis may be more favorable than the behavior of other ovarian malignancies. Future studies aimed at better understanding which patients with endometriosis will go on to de-

1661

velop EAOC are warranted. The current investigation attempted to elucidate the nature of the progression from AE to carcinoma by exploring the presence and expression of certain molecular markers. The biologic behavior of AE has been the subject of much investigation. Chalas et al.24 noted that the nuclear organizer region counts in AE fell between those found in endometriosis and those found in carcinoma. In the study performed by Chalas et al., three of four patients with AE and high nuclear organizer region counts developed malignant disease. DNA aneuploidy also has been noted in AE, whereas normal epithelium is characterized by diploidy.25 These findings support the theory that AE has the potential to be transformed into EAOC. Identification of prognostic factors that, in conjunction with established clinicopathologic parameters, can predict which patients are at risk of experiencing tumor progression may lead to improvements in the treatment and, ideally, in the overall survival of patients with ovarian carcinoma. Angiogenesis, an essential component of tumor growth and metastasis, is dependent on the production of certain factors by host and/or tumor cells. In the absence of angiogenesis, tumor growth in animal models occurs in situ, with tumor size limited to 2–3 mm in diameter.26 Overexpression of VEGF has been implicated in the pathogenesis of various human malignancies, including ovarian malignancies.11,12,26 VEGF-neutralizing antibodies have been shown to block the production of ascites and tumor growth in immunocompromised mice with xenotransplanted lesions.15 In syngeneic mice with ovarian tumors, in vitro inhibition of VEGF-induced stimulation of endothelial cells prevented VEGF-dependent tumor growth, vascularization, and metastasis.16 VEGF is an angiogenic factor that is expressed by tumor cells and plays a major role in the proliferation and migration of endothelial cells, providing adequate nourishment to tumors and allowing them to establish continuity with the vasculature of the host.26 In this way, VEGF promotes the establishment of tumor vasculature and facilitates solid tumor growth and metastatic spread.27–29 Deguchi et al.30 documented the presence of VEGF in 16 of 25 patients with a history of endometriosis and clear cell carcinoma of the ovary. To our knowledge, our curent study is the first to characterize the difference in VEGF staining between AE and EAOC. We found VEGF expression in 16 of 17 EAOC specimens and in only 1 of 8 AE specimens. These results were highly statistically significant (P ⬍ 0.0001). VEGF expression in EAOC may be an indicator of neoplastic activity, and its presence may signify a step in the transition from atypia to carcinoma in patients with endometriosis. The

1662

CANCER October 15, 2003 / Volume 98 / Number 8

relatively selective expression of VEGF in malignant tissue may have diagnostic value, and the inhibition of tumor growth via receptor blockade may possess therapeutic significance. Atypia has been reported in 1.7% of cases of benign endometriosis, compared with 61% of endometriotic lesions associated with ovarian malignancies.9 The transition from endometriosis to AE to carcinoma may be explained by differential Ki-67 expression.10 Studies have reported mean Ki-67 indices of 23.1 in ovarian carcinoma, 9.9 in AE, and 2.7 in typical endometriosis.10 Ki-67 interacts with nuclear proteins present in proliferating cells throughout the entire cell cycle, except in the G0 and early G1 phases.31 Ki-67 indices have been shown to be lower in premalignant or in situ lesions than in carcinomas of the head, neck, larynx, breast, or uterine cervix.32–36 The Ki-67 index in cervical intraepithelial neoplasia was higher than in normal tissue and was correlated with the degree of epithelial dysplasia.35,36 These observations indicate that the level of proliferative activity in premalignant and in situ lesions is lower than the level of proliferative activity in carcinomas but higher than the level in benign tissue. In the current study, positive Ki-67 staining was more common in EAOC than in AE. Although this finding did not achieve statistical significance, a definite trend was noted. The values reported in the current series are higher than those reported by Ogawa et al.10; this discrepancy may be attributable to staining differences and to the smaller sample size in the current study. Our findings, however, confirmed the results of previous investigations that showed the Ki-67 index in AE to be higher than in typical endometriosis but lower than in EAOC.10,32,33,35,36 The reported difference in Ki-67 staining suggests that AE lesions have proliferative activity that is greater than that of benign, typical endometriotic lesions but less than that of EAOC lesions. These findings may have both diagnostic and prognostic significance. The loss of ER expression observed in EAOC specimens as compared with AE specimens may be the result of dedifferentiation. In other words, as AE progresses to carcinoma, receptor expression may be lost, resulting in uncontrolled cellular proliferation and metastatic disease. Recent studies have documented the differential expression of ER in normal ovarian epithelium compared with primary tumors and metastatic lesions;30 varying amounts of the ␣ and ␤ subunits were noted in normal ovarian epithelium, lower levels of ␤ subunit expression were documented in primary tumors, and no ␤ subunit expression was reported in metastatic lesions. ER status in EAOC may have significant prognostic value. Subsequent studies

should attempt to uncover differences in receptor subunit expression among endometriosis, AE, and EAOC. One of the limitations of the current study is the small sample sizes. For example, the small sample size may account for a type II error in the Ki-67 staining results. Nonetheless, based on the information collected in the current pilot study, we hope to concentrate our future research efforts on investigating the differential expression of vascular markers in endometriosis samples and in mucinous, serous, clear cell, and endometrioid primary ovarian malignant lesions that are not associated with endometriosis. This broader comparison will facilitate a better understanding of the possible transition from endometriosis to AE and EAOC, as characterized by differential marker expression. The other types of ovarian malignancies not associated with endometriosis will serve as controls for the EAOC samples. Another limitation brought about by the absence of an endometriosisonly sample involves the documented VEGF expression in this type of tissue. Although VEGF overexpression in EAOC may be due in part to an increased inflammatory response or to the presence of endometriosis in EAOC tissue, VEGF overexpression is well documented in angiogenesis and progression of ovarian malignancies. Our future efforts in making these distinctions will include the examination of a group of endometriosis-only samples to better characterize the differential VEGF expression in endometriosis and EAOC samples. Better understanding and characterization of the markers that may be involved in the transition from AE to EAOC may serve as a diagnostic tool and facilitate the early identification of patients with endometriosis who are at risk of developing malignant disease. In addition, the knowledge that differences in molecular marker expression exist between AE and EAOC may be used to triage patients with endometriosis, AE, or EAOC to different treatment algorithms. For example, surgical intervention may be offered to the patient who has completed childbearing and who has AE that exhibits the expression of markers, such as VEGF and Ki-67, that are associated with the malignant behavior of EAOC. The selective expression of certain markers, such as VEGF, in malignant tissue also may have therapeutic value. It is conceivable that antiangiogenic agents could be used to treat patients with EAOC whose tumors demonstrate VEGF overexpression. Further research efforts must include investigations of the presence of these markers in endometriosis as compared with AE and EAOC and of differences in the expression of these markers in endometriotic lesions adjacent to carcinomatous regions in patients with EAOC. The staining pattern of the endometriotic le-

Endometriosis-Associated Ovarian Carcinoma/del Carmen et al.

sion adjacent to the carcinoma may be indicative of the progression from endometriosis to EAOC, as well as the biology of EAOC. Once the transition from benign to atypical and malignant tissue is better elucidated through the characterization of molecular markers, marker expression can be analyzed for correlations with clinical outcome.

REFERENCES 1. 2.

3. 4. 5. 6. 7. 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics, 2001. CA Cancer J Clin. 2001;51:15–36. Stenchever MA, Droegemueller W, Herbst AL, Mishell DR. Comprehensive gynecology (4th edition). St Louis: Mosby, 2001. Swiersz LM. Role of endometriosis in cancer and tumor development. Ann N Y Acad Sci. 2002;955:281–292. Sampson J. Endometrial carcinoma of the ovary, arising in endometrial tissue in that organ. Arch Surg. 1925;10:1–72. Seidman JD. Prognostic importance of hyperplasia and atypia in endometriosis. Int J Gynecol Pathol. 1999;15:1–9. Mostoufizadeh M, Scully RE. Malignant tumors arising in endometriosis. Clin Obstet Gynecol. 1980;23:951–963. Vercellini P, Parazzini F, Bolis G, et al. Endometriosis and ovarian cancer. Am J Obstet Gynecol. 1993;169:181–182. Brinton L, Gridley M, Persson I, Baron J, Berqvist A. Cancer risk after a hospital discharge diagnosis if endometriosis. Am J Obstet Gynecol. 1997;176:572–579. Fukunaga M, Nomura K, Ishikawa E, Ushigome S. Ovarian atypical endometriosis: its close association with malignant epithelial tumours. Histopathology. 1997;30:249 –255. Ogawa S, Kau T, Amada S, et al. Ovarian endometriosis associated with ovarian carcinoma: a clinicopathological and immunohistochemical study. Gynecol Oncol. 2000;77:298–304. Brustmann H, Naude S. Vascular endothelial growth factor expression in serous ovarian carcinoma: relationship with high mitotic activity and high FIGO stage. Gynecol Oncol. 2002;84:47–52. Boocock CA, Charnock-Jones DS, Sharkey AM, et al. Expression of vascular endothelial growth factor and its receptors flt and KDR in ovarian carcinoma. J Natl Cancer Inst. 1995; 87:506 –516. Fujishita A, Nakane PK, Koji T, Masuzaki H. Expression of estrogen and progesterone receptors in endometrium and peritoneal endometriosis: an immunohistochemical and in situ hybridization study. Fertil Steril. 1997;67:856 – 864. Nisolle M, Casanas-Roux F, Donnez J. Immunohistochemical analysis of proliferative activity and steroid receptor expression in peritoneal and ovarian endometriosis. Fertil Steril. 1997;68:912–919. Olson TA, Mohanraj D, Ramakrsihnan S. In vivo neutralization of vascular endothelial growth factor (VEGF)/vascular permeability factor (VPF) inhibits ovarian carcinoma-associated ascites formation and tumor growth. Int J Oncol. 1996;8:505–511. Mu J, Abe Y, Tsutsui T, et al. Inhibition of growth and metastasis of ovarian carcinoma by administering a drug capable of interfering with vascular endothelial growth factor activity. Jpn J Cancer Res. 1996;87:963–971. Grant SW, Kyshtoobayeva A, Kurosaki T, Jacowatz J, Fruehauf JP. Mutant p53 correlates with reduced expression of thrombospondin-1, increased angiogenesis and metastatic progression in melanoma. Cancer Detect Prev. 1998;22:185–194. Poller DN, Roberts EC, Bell JA, Elston CW, Blamey RW, Ellis IO.

19. 20. 21. 22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

1663

p53 protein expression in mammary ductal carcinoma in situ: relationship to immunohistochemical expression of estrogen receptor and c-erb-B-2 protein. Hum Pathol. 1993;24:463– 468. Cox DR. The analysis of binary data. London: Methuen, 1970. Cytel Software Corporation. LogXact user’s manual. Cambridge, MA: Cytel, 1992. SAS Institute. SAS user’s guide: statistics (Version 5 edition). Cary, NC: SAS Institute, 1985. Erze M, Rakar S, Klancar B, Syrjanen K. Endometriosisassociated ovarian carcinoma (EAOC): an entity distinct from other ovarian carcinomas as suggested by a nested case-control study. Gynecol Oncol. 2001;83:100 –108. McMeekin DS, Burger RA, Manetta A. Endometrioid adenocarcinoma of the ovary and its relationship to endometriosis. Gynecol Oncol. 1995;59:81– 86. Chalas E, Chumas J, Barbieri R, Mann WJ. Nuclear organizer regions in endometriosis, atypical endometriosis, and clear cell and endometrioid carcinomas. Gynecol Oncol. 1991;40: 260 –263. Ballouck F, Ross JS, Wolf BC. Ovarian endometriotic cysts: an analysis of cytologic atypia and DNA ploidy patterns. Am J Clin Pathol. 1994;102:415– 419. Shen GH, Ghazizadeh M, Kawamani O, et al. Prognostic significance of vascular endothelial growth factor expression in human ovarian carcinoma. Br J Cancer. 2000;83:196 –203. Hu YL, Tee MK, Goetzl EJ, et al. Lysophosphatidic acid induction of vascular endothelial growth factor expression in human ovarian cancer cells. J Natl Cancer Inst. 2001;93:762–768. Mesiano S, Ferrara N, Jaffe RB. Role of vascular endothelial growth factor in ovarian cancer: inhibition of ascites formation by immunoneutralization. Am J Pathol. 1998;153:1249–1256. Folkman J. A new link in ovarian cancer angiogenesis: lysophosphatidic acid and vascular endothelial growth factor expression. J Natl Cancer Inst. 2001;93:734 –735. Deguchi M, Matsumoto Y, Ishiko O, Ogita S. Angiogenesis in ovarian clear cell carcinoma and its relation to endometriosis. Oncol Rep. 2000;7:651– 654. Gerdes J, Schwab U, Lemke H, Stein H. Production of a mouse monoclonal antibody reactive with a human nuclear antigen associated with cell proliferation. Int J Cancer. 1983;31:13–20. Silvestri F, Bussani R, Pavletic N, Mannone T, Bosatra A. From epithelial dysplasia to squamous carcinoma of the head and neck region: evolutive and prognostic histopathological markers. Acta Otolaryngol Suppl. 1997;527:49 –51. Ioachim E, Assimakopoulos D, Agnantis NJ, Peschos D, Zissi A, Skevas A. Altered patterns of retinoblastoma gene product expression in benign, premalignant and malignant epithelium of the larynx: an immunohistochemical study including correlation with p53, bcl-2, and proliferating indices. Anticancer Res. 1999;19:541–545. Shen KL, Harn HJ, Ho LI, Yu CP, Chiu SC, Lee WH. The extent of proliferative and apoptotic activity in intraductal and invasive ductal breast carcinomas detected by Ki-67 labeling and terminal deoxynucleotidyl transferase-mediated digoxigenin-11-dUTP nick end labeling. Cancer. 1998; 82:2373–2381. Payne S, Kernohan NM, Walker F. Proliferation in the normal cervix and in preinvasive cervical lesions. J Clin Pathol. 1996;49:667– 671. Bulten J, van der Laak JA, Gemmink JH, Pahlplatz NM, de Wilde PC, Hanselaar AG. MIB1, a promising marker for the classification of cervical intraepithelial neoplasia. J Pathol. 1996;178:268 –273.

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.