| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Gynecology and Obstetrics, Division of Gynecology, Department of Pathology, Gynecopathological Unit, and Department of Medical Computer Sciences/Clinical Biometrics, University Hospital of Vienna, Vienna, Austria.
Address reprint requests to: Andreas Obermair, MD, Department of Gynecology and Obstetrics, University Hospital of Vienna, Währinger Gürtel 18-20, A-1090 Vienna, Austria, E-mail: andreas.obermair{at}akh-wien.ac.at
 |
Abstract |
|---|
Top
 Abstract MethodsResultsDiscussionReferences |
|---|
Methods: We reviewed 93 consecutive patients treated surgically for endometrial cancer at the University Hospital of Vienna between 1983 and 1989. Histologic sections were obtained from original paraffin-embedded blocks and stained immunohistochemically for CD34 antigen. Microvessel density was determined by enumeration of intra-tumor CD34-positive cells under a light microscope at 200x magnification using an examination area of 0.74 mm2. Log-rank test and Cox proportional-hazards models (univariate and multivariate) were applied for overall survival analysis.
Results: Overall, the 25% quantile of survival was reached at 37.9 months. The 5-year survival rate was 82.2% in 69 patients whose tumors had microvessel counts no more than 100/0.74 mm2 field, and 52.0% in 24 patients whose tumors had microvessel counts of more than 100/0.74 mm2 field (log-rank P = .004). In the multiple Cox model, high microvessel counts (relative risk [RR] 1.2; 95% confidence interval [CI] 1.1, 1.4) as well as undifferentiated tumors (RR 6.1; CI 2.2, 16.8), and advanced stage of disease (RR 2.6; CI 1.3, 5.1) independently exerted an adverse influence on the survival of patients with endometrial cancer.
Conclusion: High intratumor microvessel count is associated with poor survival of patients with endometrial cancer.
Angiogenesis, essential for tumor growth and development of metastases, is mediated by angiogenic molecules, inducing the growth of a close capillary network that surrounds and invades tumors.1,2 Even when there are no metastases at the time of primary surgery, the microvessel count within the primary tumor affects the clinical outcome of patients with a variety of malignancies.3
First reports found that high microvessel counts had an adverse effect on prognoses of patients with endometrial cancer.4,5 In a series of consecutively treated endometrial cancer patients, the present study compared microvessel counts with established prognostic factors, and tested microvessel counts for usefulness as an independent prognostic factor in that disease.
|
Methods |
|---|
|
Top
AbstractMethodsResultsDiscussionReferences |
|---|
Ninety-three slides of patients with endometrial cancer, clinical stages I–III, were stained for CD34 antigen, which is known to stain positively for vascular endothelial cells in routinely fixed paraffin sections.8,9 The staining procedure was as follows: after dehydration in xylol and graded alcohol, sections were exposed to 3% H2O2 for 15 minutes. After pretreatment in a microwave oven for 10 minutes, they were incubated in a humidity chamber at room temperature with the prediluted primary antibody (Monoclonal Antibody QBEnd/10 to Endothelium, BioGenex, San Ramon, CA) for 1 hour, followed by the second antibody (biotinylated anti-mouse antibody; BioGenex). As chromogen 3-amino-9-ethylcarbazole (AEC substrate pack; BioGenex) was used. Counterstaining was done with hematoxylin.
After scanning the immunostained section at low magnification (40x), the area of clear-cut cancer tissue with the greatest number of distinctly highlighted microvessels was selected. Microvessel count was determined according to guidelines published by Weidner, by counting all vessels at a total magnification of 200x, in an examination area of 0.74 mm2,10 Determination of the staining reaction for CD34 antigen was strictly confined to the area of highest microvessel density within or immediately adjacent to each tumor. In the following, microvessel counts are given as the value for one field of 0.74 mm2. Any endothelial cell was regarded as a single countable microvessel, regardless of whether a lumen was visible or not. Unstained lumina were considered artefacts even if they contained blood or tumor cells. Microvessel density was determined by one investigator (RW) who was blinded to the clinical course of the patient. The positive control slide for CD34 antigen was prepared from paraffin-fixed breast cancer tissue that was known to contain high microvessel density. The negative control slide was prepared from the same tissue block. Instead of the primary antibody, a normal nonimmune serum supernatant was used.
Overall survival was defined as the period from primary surgery until date of cancer-related death of the patient. Data on patients who had survived were censored at the last follow-up visit. By using the non-parametric Kruskal-Wallis test,11 microvessel count was compared with various clinical and histopathologic parameters. Microvessel counts were given per field (0.74 mm2) as median (25%–75% quantiles). Microvessel count was not only analyzed as a continuous variable, but as a dichotomous variable, defining equal or less than 100 microvessels per field as less vascularized and higher than 100 microvessels per field as highly vascularized. Because no cutoff level is given for the determination of CD34-positive cells in endometrial cancer, the cutoff level of 100 counts per field was used, representing the 75% quantile of the microvessel count of the total patient population. Due to the favorable outcome of patients with endometrial cancer, dichotomization at a relatively high cutoff level should help identify patients at high risk for relapse. Univariate analysis of overall survival was performed as outlined by Kaplan and Meier.12 Estimated overall survival probabilities are given as percentage ± standard error. Survival curves were compared using the log-rank test.13 The Cox proportional-hazards model was used for univariate and multivariate analyses.14
|
Results |
|---|
|
Top
AbstractMethodsResultsDiscussionReferences |
|---|
Median microvessel count was 72 microvessels per 0.74 mm2 field (range 45–99 microvessels per 0.74 mm2 field) for all patients. Microvessel count did not correlate with patient age at diagnosis, histologic subtype, histologic grade, stage of disease, or depth of myometrial invasion. Microvessel density was 63 microvessels per 0.74 mm2 field (42–99 microvessels per 0.74 mm2 field) in patients who received postoperative radiation therapy and 97 microvessels per 0.74 mm2 field (78–116 microvessels per 0.74 mm2 field) in those who did not receive radiation therapy (Kruskal-Wallis, P = .195). Patient characteristics are given in Table 1
.
|
. For 5 years, the overall survival probability was 82.2% (± 7.6) in 69 patients whose tumors had microvessel counts at or below 100/field, and 52.0% (± 13.2) in 24 patients whose primary tumors had microvessel counts above 100/field (log-rank; P = .004). Univariate analysis among the covariates stages of disease (stage II versus I, III versus II), microvessel count (continuous), and histologic grading (G2 versus G1, G3 versus G2) showed that the stage of disease, microvessel count, and histologic grading were prognostic factors for overall survival (Table 2). After adjustment for stage of disease and histologic grade, microvessel count proved to be an independent prognostic factor in a multiple Cox model (Table 2).
|
|
|
Discussion |
|---|
|
Top
AbstractMethodsResultsDiscussionReferences |
|---|
Previous reports compared microvessel count with neoplasia in human endometrium, or with established prognostic factors in endometrial carcinoma. Morgan et al15 and Abulafia et al16 reported microvessel counts significantly higher in endometrial hyperplasia and endometrial carcinoma than normal endometrium. In patients with stage I endometrial cancer, Abulafia et al16 found that greater depth of invasion and higher tumor grade directly correlated to angiogenic activity, whereas Morgan et al15 reported no significant correlation between stromal vascular density and patient age at diagnosis, histologic grade, tumor type, vascular space involvement, or depth of myometrial invasion. By measuring vascular density with an image analysis system, Willemse et al17 found that higher vascular lumen diameter of immunostained microvessels was correlated strongly with increased serum levels of tissue-polypeptide-specific antigen. Increased serum CA 125 levels were correlated significantly with an increased number of vessels per mm2.
Our data were consistent with an early report that found significant differences in tumor microvessel density with disease-free survival. By using a matched-pairs design, Kirschner et al4 compared 25 patients with recurrent endometrial cancer to 25 who lived tumor-free during follow-up periods. After staining for factor VIII-related antigen, patients with low microvessel counts in their primary tumors had significantly increased disease-free and overall survival. Kaku et al18 reported that microvessel count was correlated significantly with tumor grade, myometrial invasion, and lymphatic vascular space involvement. However, in this series of stage I and II endometrial cancers, patients with high microvessel counts (at least 60/field) had significantly worse outcomes than those with low microvessel counts. Recently, tumor angiogenesis, determined by high microvessel count, was shown to adversely influence the clinical course of unselected patients with endometrial cancer.19,20 Although the measurements of microvessels differed according to the antigens used (factor VIII-related antigen versus CD34), examination area, and number of fields examined, our study and that of Salvesen et al19 found similar rates of probability of survival. Salvesen et al19 reported 5-year survival probability rates of 57% and 90% for patients with high and low microvessel counts, respectively. In contrast, Wagatsuma et al20 reported a 0% survival rate in patients with high microvessel counts.
For pure chance, patients who did not have postoperative radiation therapy had higher microvessel counts than those who did. Because low microvessel count is related to a 30% increase in overall survival, and because radiation therapy has never been associated with such a huge increase in survival probability, we assume that better survival in patients with low microvessel counts is a result of a biologic difference, rather than a difference in treatment. However, we could not add radiation therapy as an additional covariate to multivariate analysis to prevent an overfit of the Cox model.
Comparison of previous publications is made difficult by differences in the following: 1) constitution of patient population, 2) magnification and examined area, 3) follow-up period, and 4) methods of microvessel enumeration. In endometrial cancer the mean microvessel count per mm2 varies from 77/mm2 to 842/mm2 when sections were stained for factor VIII-related antigen and enumerated at 400x magnification.5,16 By using immunostaining for CD34 antigen, a highly sensitive and reliable antigen for highlighting vascular endothelial cells,8,9 we found a median microvessel count of 72/mm2 in patients with stage I–III endometrial carcinoma. In keeping with the guidelines of Weidner, one histologic section was immunostained and microvessel count was determined by one investigator who was blinded to the clinical outcome in the series of 93 consecutive patients.
Prognostic factors identify patients at low or high risk of relapse, influencing therapeutic decisions. Adjuvant pelvic radiation therapy is considered unnecessary in patients with grade 1 or 2 lesions confined to the inner half of the myometrium, because they have excellent prognoses. In all other cases, the effectiveness of adjuvant radiation therapy is scientifically debated. Usually radiation therapy is administered after surgery in patients with moderately or poorly differentiated International Federation of Gynecology and Obstetrics (FIGO) stage Ib tumors and in those with tumors spread outside the inner half of the myometrium. In the literature, this treatment is believed to reduce the risk of local recurrence, although its effect on overall survival is still unclear.21,22
Our results emphasized the need for further clinical trials to test angiogenic intensity as a stratification criterion. It is possible that endometrial cancer patients with organ-confined disease and less vascularized tumors do not benefit from postoperative radiation therapy.
While cytotoxic agents are directed against tumor cells, anti-angiogenic molecules address the stromal component of the tumor. This might be the reason clinical trials report high response rates to anti-angiogenic therapy in a variety of solid tumors, including those that commonly do not respond, or respond poorly, to conventional chemotherapy.23,24 Our data gave indirect evidence that the growth and spread of endometrial cancer is angiogenesis-dependent, emphasizing the need for clinical trials to study the influence of anti-angiogenic substances on carcinoma of the uterine corpus.
|
Footnotes |
|---|
Received April 22, 1998. Received in revised form August 25, 1998. Accepted September 10, 1998.
|
References |
|---|
|
Top
AbstractMethodsResultsDiscussionReferences |
|---|
2. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nature Med 1995;1:27–31.[Medline]
3. Weidner N. Intratumoral microvessel density as a prognostic factor in cancer. Am J Pathol 1995;147:9–19.[Medline]
4. Kirschner CV, Alanis-Amezcua JM, Martin VG, Luna N, Morgan E, Yang JJ, et al. Angiogenesis factor in endometrial carcinoma: A new prognostic indicator? Am J Obstet Gynecol 1996;174:1879–84.[Medline]
5. Morgan KG, Wilkinson N, Buckley CH. Angiogenesis in endometrial carcinoma. Int J Gynecol Cancer 1996;6:385–8.
6. Malviya VK, Deppe G, Malone JM. Reliability of frozen section examination in identifying poor prognostic indicators in stage I endometrial adenocarcinoma. Gynecol Oncol 1989;34:299–304.[Medline]
7. Morrow CP, Bundy BN, Kurmann RJ. Relationship between surgical-pathological risk factors and outcome in clinical stage I and II carcinoma of the endometrium: A Gynecologic Oncology Group study. Gynecol Oncol 1991;40:55–65.[Medline]
8. Fina L, Molgaard HV, Robertson D, Bradley NJ, Monaghan P, Delai D, et al. Expression of the CD34 gene in vascular endothelial cells. Blood 1990;75:2417–26.
9. Kuzu I, Bicknell R, Harris AL, Jones M, Gatter KC, Mason DY. Heterogeneity of vascular endothelial cells with relevance to diagnosis of vascular tumours. J Clin Pathol 1992;45:143–8.
10. Weidner N. Current pathologic methods for measuring intratumoral microvessel density within breast carcinoma and other solid tumors. Breast Cancer Res Treat 1995;36:169–80.[Medline]
11. Kruskal WH, Wallis WA. Use of ranks in one-criterion variance analysis. J Am Statist Assoc 1952;47:583–621.
12. Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Statist Assoc 1985;53:457–81.
13. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1966; 50:163–70.[Medline]
14. Cox DR. Regression models and life tables. J Roy Stat Soc [B] 1972;34:187–220.
15. Morgan KG, Wilkinson N, Buckley CH. Angiogenesis in normal, hyperplastic and neoplastic endometrium. J Pathol 1996;179:317–20.[Medline]
16. Abulafia O, Triest WE, Sherer DM, Hansen CC, Ghezzi F. Angiogenesis in endometrial hyperplasia and stage I endometrial carcinoma. Obstet Gynecol 1995;86:479–85.[Abstract]
17. Willemse F, Nap M, Bruijn HWA, Hollema H. Quantification of vascular density and of lumen and vessel morphology in endometrial carcinoma. Analyt Quant Cytol Histol 1997;19:1–7.
18. Kaku T, Kamura T, Kinukawa N, Kobayashi H, Sakai K, Tsuruchi N, et al. Angiogenesis in endometrial cancer. Cancer 1997;80:741–7.[Medline]
19. Salvesen HB, Iversen OE, Akslen LA. Independent prognostic importance of microvessel density in endometrial carcinoma. Br J Cancer 1998;77:1140–4.[Medline]
20. Wagatsuma S, Konno R, Sato S, Yajima A. Tumor angiogenesis, hepatocyte growth factor, and c-met expression in endometrial carcinoma. Cancer 1998;82:520–30.[Medline]
21. Hacker NF. Uterine cancer. In: Hacker NF, Berek JS, eds. Practical gynecologic oncology. 2nd ed. Baltimore, MD: Williams and Wilkins, 1994:285–326.
22. Barakat RR, Park RC, Grigsby PW, Muss HD, Norris HJ. Corpus: Epithelial tumors. In: Hoskins WJ, Perez CA, Young RC, eds. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia, PA: Lippincott-Raven, 1997:859–96.
23. Lindner DJ, Borden EC. Effects of tamoxifen and interferon-beta or the combination on tumor-induced angiogenesis. Int J Cancer 1997;71:456–61.[Medline]
24. Yamamoto T, Terada N, Nishizawa Y, Petrow V. Angiostatic activities of medroxyprogesteron acetate and its analogues. Int J Cancer 1994;56:393–9.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
