- Email: [email protected]
Angiogenesis of endometrial carcinomas assessed by measurement of intratumoral blood flow, microvessel density, and vascular endothelial growth factor levels
Angiogenesis of Endometrial Carcinomas Assessed by Measurement of Intratumoral Blood Flow, Microvessel Density, and Vascular Endothelial Growth Factor Levels CHIEN-NAN LEE, MD, WEN-FANG CHENG, MD, PhD, CHI-AN CHEN, MD, JAN-SHOW CHU, MD, CHANG-YAO HSIEH, MD, MSPH, AND FON-JOU HSIEH, MD Objective: To evaluate the relationship between blood flow in the tumor assessed by color Doppler ultrasound, microvessel density, and vascular endothelial growth factor levels in endometrial carcinoma. Methods: Forty-nine patients undergoing surgery for endometrial carcinoma were enrolled. Transvaginal color Doppler ultrasound was performed preoperatively and the lowest resistance index (RI) in the tumor was recorded for analysis. Vascular endothelial growth factor in the tumor was quantified by enzyme immunoassay. The microvessel density of the excised tumor was assessed immunohistochemically. The relationships between the corresponding RI, microvessel density, and vascular endothelial growth factor level of the tumor tissues and clinical and pathologic parameters were analyzed. Results: Significantly lower RIs were noted in tumors of stage II or greater (0.37 compared with 0.50, P < .001), of high histologic grade (grade 3) (0.34 compared with 0.49, P ⴝ .004), with deep myometrial invasion (one-half depth or greater) (0.39 compared with 0.49, P ⴝ .002), with lymphovascular emboli (0.38 compared with 0.49, P < .001), or with lymph node metatasis (0.30 compared with 0.49, P < .001) compared with stage I tumors and tumors of histologic grade 1 or 2, with superficial myometrial invasion, without lymphovascular emboli, or with no lymph node metastasis. Increased vascular endothelial growth factor levels and microvessel density (ⴛ ⴛ 200 field) also were detected in tumors of stage II or greater (975 compared with 129 pg/mg, P ⴝ .014; and 88 compared with 61, P ⴝ .018, respectively), with lymphovascular emboli (1138 compared with 120 pg/mg, P ⴝ .002; and 86 compared with 63, P ⴝ .023), or with lymph node metastasis (1011 compared with 95 pg/mg, P < .001; and 98 compared with 61, P ⴝ .019). Resistance index, microvessel density, and vascular endothelial growth factor From the Departments of Obstetrics and Gynecology and Pathology, College of Medicine, National Taiwan University, Taipei, Taiwan. Supported by grant NSC-88-2314-B-002-381 from the National Science Committee of Taiwan.
VOL. 96, NO. 4, OCTOBER 2000
levels in the tumor showed linear correlations (RI compared with microvessel density: r ⴝ ⴚ .32, P ⴝ .03; RI compared with vascular endothelial growth factor levels: r ⴝ ⴚ .40, P ⴝ .004; microvessel density compared with vascular endothelial growth factor levels: r ⴝ .36, P ⴝ .011). Conclusion: Blood flow assessed by color Doppler ultrasound has histologic and biologic correlations with angiogenesis and vascular endothelial growth factor levels and might play an important role in predicting tumor progression and metastasis in endometrial carcinoma. (Obstet Gynecol 2000;96:615–21. © 2000 by The American College of Obstetricians and Gynecologists.)
Angiogenesis plays an important role in the pathogenesis of cancer.1 Recent studies have related angiogenesis to cancer growth and metastasis.2,3 The growth of solid tumors and their metastatic spread is angiogenesisdependent and this has been confirmed in several experimental and clinical studies.4,5 Vascular endothelial growth factor, also known as vascular permeability factor, which induces the growth of a capillary network surrounding the tumor and acts as a highly specific mitogen on endothelial cells, is regarded as an important angiogenic factor in tumor angiogenesis.6 Overexpression of vascular endothelial growth factor was found to correlate with survival in several patients with malignant tumors.7,8 However, relatively little is known about the clinical correlation between vascular endothelial growth factor and tumor progression and metastasis in endometrial carcinoma. Ramos et al9 demonstrated the possibility of detection of tumor vasculature by Doppler ultrasound even in a small amount of malignant tissue in an animal model. Color Doppler ultrasound shows that neovascularized vessels have blood flow characteristics that differ from
0029-7844/00/$20.00 PII S0029-7844(00)00976-5
615
those of normal vessels. We also have demonstrated, using color Doppler ultrasound, that increased vasculature tends to have increased malignant potential in ovarian and endometrial neoplasms.10,11 However, the histologic or biologic correlations with blood flow assessed by color Doppler ultrasound still are lacking. Determining microvessel density is regarded as a standard procedure to quantitate tumor angiogenesis. The progression from endometrial hyperplasia, complex hyperplasia, atypical hyperplasia, and stage IA endometrial carcinoma to invasive endometrial carcinoma has been noted.12 However, microvessel density always has been assessed retrospectively and in vitro. Preoperative prediction of the microvessel density would be of value clinically for evaluating severity and progression of disease. The aims of this study were to determine whether there are correlations among the corresponding resistance index (RI), microvessel density, and vascular endothelial growth factor concentration in the tumor and to evaluate whether vascular endothelial growth factor levels reflect the severity of endometrial carcinoma. The information obtained might be useful in the search for histologic and biologic associations with blood flow assessed by color Doppler ultrasound.
Materials and Methods Between January 1993 and December 1997, 55 patients with endometrial carcinoma were enrolled in this study. In all cases, the diagnosis was made by fractional dilation and curettage. Six patients with undetectable blood flow by color Doppler ultrasound were excluded. The other 49 patients (89%) with detectable blood flow were recruited. All 49 women underwent surgery, including total abdominal hysterectomy, bilateral salpingo-oophorectomy, and pelvic and/or para-aortic lymph node dissection or sampling. Patients with lymph node metastasis or deep myometrial invasion (more than one-half depth) underwent postoperative adjuvant irradiation. Surgical specimens were evaluated for tumor size, histologic grading, depth of myometrial invasion, and presence of lymphovascular emboli and lymph node metastasis. Stage was determined using the International Federation of Gynecology and Obstetrics classification.13 Tumors of histologic types other than adenocarcinoma and adenoacanthoma were excluded. The carcinomas were classified using a threegrade system: grade 1 carcinomas showed glandular formation in more than 95% of the tumor, grade 2 carcinomas showed a solid growth pattern in 5–50%, and grade 3 carcinomas showed a solid pattern in more than 50%. We used transvaginal sonography to detect arterial
616 Lee et al
Angiogenesis in Endometrial Cancer
blood flow in the tumor on the day before hysterectomy and at least 10 days after diagnostic uterine curettage. The equipment used was a color Doppler ultrasound unit (Ultramark 9 or HDI 3000; Advanced Technology Laboratories, Inc., Bothell, WA) with a high-pass filter set on 100 Hz to eliminate low-frequency signals resulting from vessel-wall motion. The pulse-repetitive frequency was 1–25 kHz, depending on the blood flow velocity under investigation. The waveforms of blood flow velocity in the artery of the tumor were obtained, and the RI was calculated as the difference between the systolic and end-diastolic frequency shifts divided by the systolic frequency shift. A series of similar arterial waveforms with maximal clarity were obtained reproducibly over three consecutive cardiac cycles and regarded as satisfactory at the site of sampling. The lowest RI was used for analysis. Tissue specimens were obtained during surgery, immediately frozen in liquid nitrogen, and stored at ⫺70C until analyzed. The remaining tissue specimens were sent for pathologic examination. Cancer tissues were prepared as described.14 Briefly, the tissue samples, after being stripped of blood and necrotic tissue, were thawed on ice placed in 10 volumes of ice-cold cell lysis buffer (pH 7.8, containing 100 mM K2HPO4-KH2PO4, 1 mM dithiothreitol, 2 mM ethylenediaminetetra-acetic acid, 1% Triton X-100, and 0.75 g/mL leupeptin) and were homogenized. The lysate was centrifuged and the supernatant recovered and stored at ⫺70C until analysis. The total protein in the prepared supernatant was measured by protein assay (Bio-Rad, Hercules, CA), and then vascular endothelial growth factor levels of the extract were determined by enzyme immunoassay. For determination of vascular endothelial growth factor levels, a commercially available enzyme immunoassay kit was used (Quantikine human vascular endothelial growth factor immunoassay; Research and Diagnostic Systems Inc., Minneapolis, MN) as described.8 Measurement of total protein and vascular endothelial growth factor level in the extract was repeated, and the average values for each sample were recorded. The concentration of vascular endothelial growth factor in the tumor was represented as picogram per milligram of protein. Formalin-fixed, paraffin-embedded tissue blocks were recut, the paraffin was removed, and the tissue sections were rehydrated. Endogenous peroxidase activity was blocked by hydrogen peroxide. Specimens then were incubated with mouse anti-CD34 monoclonal antibody diluted to a 1:20 ratio (BioGenex, San Ramon, CA). Next, they were incubated with biotinylated horse antimouse immunoglobulin (Ig) G and treated with the preformed avidin-biotin complex. The substrate solution for peroxidase contained 0.2% diaminobenzidine
Obstetrics & Gynecology
three areas were counted. The average in the three ⫻200 fields in each tumor was calculated and thereafter was referred to as density count. Large vessels with thick muscular walls and lumina greater than approximately eight blood cells were excluded from the count. All measurements were performed by a single pathologist who was unaware of any clinical or pathologic data before counting. We used SPSS (Statistical Package for Social Sciences) 6.0 (SPSS Inc., Chicago, IL) for Windows to perform the Mann-Whitney U test and Spearman correlation analysis.
Figure 1. Arterial blood flow in the tumor of one patient with endometrial carcinoma (resistance index 0.35).
tetrahydrochloride. Normal mouse IgG was substituted for primary antibody as a negative control. Sections were counterstained lightly with hematoxylin. Microvessels in the tumors were highlighted by antiCD34 immunostaining in formalin-fixed, paraffinembedded sections. Microvessel density was quantified as previously described.15 Briefly, the stained slides were examined at low-power magnification (⫻40 and ⫻100 total magnification) to identify the areas of highest neovascularization (so-called hot spots) of the tumor. The three most vascular areas were chosen in each section. Microvessels on a ⫻200 field (⫻20 objective and ⫻10 ocular [Olympus BH-2 microscope; Olympus, Tokyo, Japan], 0.74 mm2 per field, with the field size measured with an ocular micrometer) in each of the
Results Patients were between 25 and 74 years of age (mean 47.4), with a parity of 0 – 8 and a maximal tumor diameter of 0.5–7.0 cm (mean 3.10). Twenty-seven women were postmenopausal. The RI in the tumor ranged from 0.19 to 0.73 (median 0.47) and a representative case is shown in Figure 1. The microvessel density ranged from 8 to 164 (median 63). The concentration of vascular endothelial growth factor in the tumor ranged from 0 to 4313 pg/mg (median 220). Significantly lower RIs were noted in tumors of stage II or greater (0.37 compared with 0.50, P ⬍ .001), of high histologic grade (grade 3) (0.34 compared with 0.49, P ⫽ .004), with deep myometrial invasion (one-half depth or greater) (0.39 compared with 0.49, P ⫽ .002), with lymphovascular emboli (0.38 compared with 0.49, P ⬍ .001), or with lymph node metastasis (0.30 compared with 0.49, P ⬍ .001) compared with stage I tumors or tumors of histologic grade 1 or 2, with superficial
Table 1. Correlation Between Resistance Index, Microvessel Density, and Vascular Endothelial Growth Factor Levels in the Tumor and Pathologic Parameters in 49 Endometrial Carcinoma Patients
Surgical stage I ⬎I Histologic grade 1 or 2 3 Myometrial invasion ⬍1⁄2 depth ⱖ1⁄2 depth Lymphovascular emboli No Yes Pelvic lymph node metastasis No Yes
VEGF (pg/mg) (25–75%)
P*
MVD (⫻200 field) (25–75%)
n
RI (25–75%)
P*
P*
35 14
.50 (.42–.65) .37 (.29 –.45)
⬍.001
129 (0 –542) 975 (69 –2280)
.014
61 (37–98) 88 (58 –113)
.018
39 10
.49 (.40 –.62) .34 (.26 –.44)
.004
161 (11– 802) 438 (22–1649)
.38
64 (39 –100) 69 (53–104)
.66
35 14
.49 (.42–.62) .39 (.30 –.45)
.002
129 (29 –542) 867 (6 –1589)
.06
61 (38 –98) 77 (56 –113)
.16
37 12
.49 (.40 –.64) .38 (.27–.46)
⬍.001
120 (0 –529) 1138 (358 –2280)
.002
63 (37–99) 86 (58 –114)
.023
37 12
.49 (.45–.61) .30 (.25–.39)
⬍.001
95 (0 – 450) 1011 (587–2225)
⬍.001
61 (36 – 89) 98 (64 –108)
⬍.019
RI ⫽ resistance index; VEGF ⫽ vascular endothelial growth factor; MVD ⫽ microvessel density. * Mann-Whitney U test.
VOL. 96, NO. 4, OCTOBER 2000
Lee et al
Angiogenesis in Endometrial Cancer
617
Figure 2. The correlation between resistance index (RI) and microvessel density (MVD) in the tumor was a negative linear one (r ⫽ ⫺.32, P ⫽ .03).
myometrial invasion, without lymphovascular emboli, or with no lymph node metastasis (Table 1). Increased vascular endothelial growth factor levels and microvessel density also were detected in tumors of stage II or greater (975 compared with 129 pg/mg, P ⫽ .014; 88 compared with 61, P ⫽ .018, respectively), with lymphovascular emboli (1138 compared with 120 pg/mg, P ⫽ .002; and 86 compared with 63, P ⫽ .023), or with lymph node metastasis (1011 compared with 95 pg/mg, P ⬍ .001; and 98 compared with 61, P ⫽ .019). No statistically significant difference could be found in vascular endothelial growth factor levels or microvessel density in tumors of high histologic grade (438 compared with 161 pg/mg, P ⫽ .38; and 69 compared with 64, P ⫽ .66, respectively) or deep myometrial invasion (867 compared with 129 pg/mg, P ⫽ .06; 77 compared with 61, P ⫽ .16) (Table 1). Spearman correlation analysis revealed negative linear correlations between RI in the tumor and microvessel density and vascular endothelial growth factor concentrations (RI compared with microvessel density: r ⫽ ⫺.32, P ⫽ .03, formula: y ⫽ 105.421288 ⫺ 77.796124x; RI compared with vascular endothelial growth factor levels: r ⫽ ⫺.40, P ⫽ .004, formula: y ⫽ 2104.7413 ⫺ 3176.1170x) (Figures 2 and 3). There was a positive linear correlation between vascular endothelial growth factor levels and microvessel density (vascular endothelial growth factor levels compared with microvessel density: r ⫽ .36, P ⫽ .011, formula: y ⫽ 57.831659 ⫹ 0.018078x) (Figure 4).
Figure 3. The correlation between resistance index (RI) and vascular endothelial growth factor (VEGF) concentration in the tumor was a negative linear one (r ⫽ ⫺.40, P ⫽ .004).
investigators to evaluate tumor-associated microvessel density.2–5 The findings of these studies suggest that the greater the number of vessels, the more likely it is that tumor cells will enter the circulation or the lymphatics. Counting microvessels in the tumor is the most commonly used method to quantify tumor angiogenesis. Immunohistochemical staining for endothelial-specific markers such as factor VIII–related antigen, CD31 antigen, or CD34 antigen clearly highlights the vascular endothelium. However, two main problems are encountered when microvessel density measurement is used to quantitate angiogenesis. First, methodologic problems such as inter- and intraobserver variability, tumor heterogeneity, and selection of the area of most intense neovascularization (hot spot) remain unsolved. Absolute vessel counts also are influenced by total magnification, the examination area selected, and the skill and experience of the investigator. Extensive efforts have been made to assess angiogenesis objectively. Second, microvessel
Discussion Potential tumor growth and metastasis depend on the ability of a tumor to induce angiogenesis.1–3 The use of immunohistochemical methods to determine whether a tumor has entered an angiogenic phase has allowed
618 Lee et al
Angiogenesis in Endometrial Cancer
Figure 4. The correlation between microvessel density (MVD) and vascular endothelial growth factor (VEGF) concentration was a positive linear one (r ⫽ .36, P ⫽ .011).
Obstetrics & Gynecology
density is determined retrospectively and in vitro and therefore might not play a role in the preoperative evaluation of and treatment planning for patients with endometrial carcinoma. To improve the accuracy of evaluation of angiogenesis, we attempted to use resistance value of blood flow in the tumor to represent in vivo angiogenesis and vascular endothelial growth factor levels to quantitate angiogenesis more objectively. Resistance index values in the tumor correlated well with microvessel density in this survey. Therefore, blood flow assessed by color Doppler ultrasound could represent the angiogenic activity of endometrial carcinoma. Magnetic resonance imaging (MRI) and transvaginal ultrasound are fairly accurate methods for preoperatively assessing depth of myometrial invasion, tumor staging, and extracorporeal spread.10,16 Color Doppler ultrasound has some advantages over MRI, being less time-consuming and less expensive. This study demonstrates that lower arterial RIs are found in endometrial carcinomas of advanced stages, of high histologic grade, with deep myometrial invasion, with lymphovascular emboli, or with lymph node metastasis. Color Doppler ultrasound has been used to detect neovascularization in solid tumors arising from the ovary.10,16 We have observed incremental angiogenesis in endometrial neoplasms with increased malignant potentials by color Doppler ultrasound.11 Resistance index values in the tumors reflect gross angiogenic activity from feeding arteries of the endometrial tumors. The tumor becomes more angiogenic as the disease progresses. We suggest that endometrial cancer patients with low RIs might have deep myometrial invasion, lymph node metastasis, and advanced-stage tumors. In vivo tumor angiogenesis may be demonstrated by color Doppler ultrasound. However, its histologic and biologic associations with neovascularization observed by color Doppler ultrasound are not yet clear. Microvessel density and vascular endothelial growth factor levels could explain this phenomenon, because both microvessel density and vascular endothelial growth factor levels were linearly correlated with RI in the tumors. This relation showed that blood flow assessed by color Doppler ultrasound had histologic and biological associations with angiogenesis. Conventionally, microvessel density was determined by immunohistochemical staining using antibodies against various endothelial cell–related antigens such as factor VIII, CD31, and CD34.4 Both microvessel density and vascular endothelial growth factor levels were determined in the in vitro surgical specimen. The present study has revealed close correlations between RI, microvessel density, and vascular endothelial growth factor levels in endometrial tumors. Thus, tumor vascularity now can
VOL. 96, NO. 4, OCTOBER 2000
be assessed in vivo and in situ at the time of ultrasound. Importantly, the in vivo RI in the tumor could be determined preoperatively, thus indicating the microvessel density of tumor tissues, and such an assessment could provide useful information to clinicians for evaluating disease progression in and conducting preoperative planning for endometrial carcinoma patients. Patients with lower tumor RIs might have deep myometrial invasion, lymph node metastasis, and distant metastasis. Vascular endothelial growth factor, an endotheliumspecific mitogen and potent mediator of vascular permeability, although expressed in normal tissues such as the endometrium,17 is frequently found in high levels in solid tumors.14 Vascular endothelial growth factor proteins consist of four splicing variants: vascular endothelial growth factor121, vascular endothelial growth factor165, vascular endothelial growth factor189, and vascular endothelial growth factor206.18 The two shorter forms are estimated to be free to diffuse because of weak heparin-binding activity. The vascular endothelial growth factor enzyme immunoassay used in this study detected vascular endothelial growth factor165 isoform. Therefore, actual vascular endothelial growth factor concentrations might be higher than the vascular endothelial growth factor concentrations measured in this study. The time of secretion of vascular endothelial growth factor could be regarded as the critical point at which tumors switch from nonangiogenic to angiogenic status. The high microvessel density reflects only late angiogenesis of the tumor with neovascularization. Therefore, the presence of vascular endothelial growth factor protein could be representative of the early stage of tumor angiogenesis, and high microvessel density could represent the late stage of tumor angiogenesis. Higher vascular endothelial growth factor concentrations and higher microvessel densities were seen only in advanced-stage tumors, tumors with lymphovascular emboli, and tumors with lymph node metastasis. No significant difference in vascular endothelial growth factor levels or microvessel densities was noted in tumors of high histologic grade or with deep myometrial invasion. Tumors of high histologic grade showed a higher incidence of deep myometrial invasion (P ⫽ .003), which indicates that some inherent cancer behavior other than merely tumor angiogenesis may influence the local invasive or metastatic ability. In our study, microvessel density correlated significantly with vascular endothelial growth factor concentration. This finding is compatible with the finding that strong vascular endothelial growth factor messenger RNA expression is associated with increased microvessel density in squamous intraepithelial lesions and invasive squamous cell carcinomas of the cervix.19
Lee et al
Angiogenesis in Endometrial Cancer
619
Tumor cells secrete vascular endothelial growth factor to induce proliferation of endothelial cells, and the proliferative endothelium results in neovascularization, which is represented by microvessel density. Some tumors in our study had high histologic microvessel densities but expressed low levels of vascular endothelial growth factor. The role of the in vivo angiogenic effect is complex, and other angiogenic factors such as fibroblast growth factor and platelet-derived growth factor should be investigated further to elucidate angiogenesis in endometrial cancers. Emoto et al20 evaluated the relationship between in vivo angiogenesis RI determined by transvaginal color Doppler ultrasound and in vitro angiogenesis microvessel density determined by immunohistochemistry. They concluded that blood flow in the tumor may correlate with endothelial cell activity of tumor vessels but not with microvessel density. They demonstrated an indirect relationship between in vivo and in vitro parameters of angiogenesis. Our results showing that RIs in the tumor correlated with histologic microvessel density were different from theirs, perhaps because endometrial tumors are smaller than ovarian tumors. The area where we determined microvessel density is closer to the artery (or arteries) assessed by color Doppler ultrasound for endometrial carcinoma. Therefore, the correlation between microvessel density and RI would be expected to be stronger in endometrial carcinoma than in ovarian carcinoma. Also, vascular endothelial growth factor levels correlated with the proliferation of endothelium; therefore, vascular endothelial growth factor concentrations might indicate proliferative activity of endothelial cells in tumor vessels. Resistance index values in the tumors also correlated well with the vascular endothelial growth factor levels in our study; therefore, blood flow assessed by color Doppler ultrasound might reflect histologic and biologic correlations with angiogenesis and vascular endothelial growth factor levels in tumor angiogenesis. Recently, Kana et al21 described the antitumor and antimetastatic effects of vascular endothelial growth factor–neutralizing antibody in an animal model. If inhibition of vascular endothelial growth factor offers an opportunity for the treatment of endometrial carcinoma, then determination of RI in the tumor by color Doppler ultrasound might be a useful way to assess the magnitude of response. Further study is necessary to clarify the mechanism of promoting angiogenesis by vascular endothelial growth factor in endometrial carcinoma.
References 1. Folkman J, Shing Y. Angiogenesis. J Biol Chem 1992;267:10931– 4.
620 Lee et al
Angiogenesis in Endometrial Cancer
2. Folkman J. The role of angiogenesis in tumor growth. Semin Cancer Biol 1992;3:65–71. 3. Ellis LM, Fidler IJ. Angiogenesis and metastasis. Eur J Cancer 1996;32A:2451– 60. 4. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis—Correlation in invasive breast carcinoma. N Engl J Med 1991;324:1– 8. 5. Folkman J. Tumor angiogenesis. In: Holland F, Frei E III, Bast RC Jr, Kufe DW, Morton DL, Weichselbaum RR, eds. Cancer medicine. Philadelphia: Lea & Febiger, 1989:153–70. 6. Senger DR, Connolly DT, Van de Water L, Feder J, Dvorak HF. Purification and NH2-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor. Cancer Res 1990;50: 1774 – 8. 7. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hypersensitivity, and angiogenesis. Am J Pathol 1995;146:1029 –39. 8. Obermair A, Kucera E, Mayerhofer K, Speiser P, Seifert M, Czerwenka K, et al. Vascular endothelial growth factor in human breast cancer: Correlation with disease-free survival. Int J Cancer 1997;74:455– 8. 9. Ramos I, Fernandez LA, Morse SS, Fortune KL, Taylor KJ. Detection of neovascular signals in a 3-day Walker 256 rat carcinoma by Doppler ultrasound. Ultrasound Med Biol 1988;14:123– 6. 10. Wu CC, Lee CN, Chen TM, Shyu MK, Hsieh CY, Chen HY, et al. Incremental angiogenesis assessed by color Doppler ultrasound in the tumorigenesis of ovarian neoplasm. Cancer 1994;73:1251– 6. 11. Cheng WF, Chen TM, Chen CA, Wu CC, Huang KT, Hsieh CY, et al. Clinical application of intratumoral blood flow study in patients with endometrial carcinoma. Cancer 1998;82:1881– 6. 12. Abulafia O, Triest WE, Sherer DM, Hansen CC, Chezzi F. Angiogenesis in endometrial hyperplasia and stage I endometrial carcinoma. Obstet Gynecol 1995;86:479 – 85. 13. International Federation of Gynecology and Obstetrics (FIGO) stages—1988 revision. Gynecol Oncol 1989;35:125–7. 14. Cheng WF, Chen CA, Lee CN, Chen TM, Hsieh FJ, Hsieh CY. Vascular endothelial growth factor in cervical carcinoma. Obstet Gynecol 1999;93:761–5. 15. Cheng WF, Lee CN, Chu JS, Chen CA, Chen TM, Shau WY, et al. Vascularity index as a novel parameter for the “in vivo” assessment of angiogenesis in cervical carcinoma. Cancer 1999;85:651–7. 16. Fleischer AC, Rodgers WH. Ovarian masses. In: Fleischer AC, Emerson DS, eds. Color Doppler sonography in obstetrics and gynecology. New York: Churchill Livingstone, 1993:59 –121. 17. Charnock-Jones DS, Sharkey AM, Rajput-Williams J, Burch D, Schofield JP, Fountain SA, et al. Identification and localization of alternately spliced mRNAs for vascular endothelial growth factor in human uterus and estrogen regulation in endometrial carcinoma cell lines. Biol Reprod 1993;48:1120 – 8. 18. Ferrara N, Houck KA, Jakeman LB, Winer J, Leung DW. The vascular endothelial growth factor family of polypeptides. J Cell Biochem 1991;47:211– 8. 19. Guidi AJ, Abu-Jawdeh G, Berse B, Jackman RW, Tognazzi K, Dvorak HF, et al. Vascular permeability factor (vascular endothelial growth factor) expression and angiogenesis in cervical neoplasia. J Natl Cancer Inst 1995;87:1237– 45. 20. Emoto M, Iwasaki H, Mimura K, Kawarabayashi T, Kikuchi M. Differences in the angiogenesis of benign and malignant ovarian tumors, demonstrated by analyses of color Doppler ultrasound, immunohistochemistry, and microvascular density. Cancer 1997; 80:899 –907. 21. Kana T, Konno H, Tanaka T, Baba M, Matsumoto K, Nakamura S, et al. Anti-tumor and anti-metastatic effects of human-vascularendothelial-growth-factor-neutralizing antibody on human colon
Obstetrics & Gynecology
and gastric carcinoma xenotransplanted orthotopically into nude mice. Int J Cancer 1998;77:933– 6.
Received December 13, 1999. Received in revised form March 20, 2000. Accepted March 30, 2000.
Address reprint requests to:
Fon-Jou Hsieh, MD Department of Obstetrics and Gynecology National Taiwan University Hospital No. 7, Chung-Shan South Road Taipei Taiwan E-mail: [email protected]
NEW,
Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
THIRD EDITION OF WRITING GUIDE AVAILABLE
To help prospective authors, especially those who are beginning in medical journal writing, to produce better papers, this journal has developed “A Guide to Writing for Obstetrics & Gynecology.” Copies of the third edition of this booklet are available, free of charge, from: Editorial Office, Obstetrics & Gynecology, 10921 Wilshire Boulevard, Suite 403, Los Angeles, CA 90024-3908. Requests may also be submitted by FAX: (310) 208-2838, or E-mail, [email protected].
VOL. 96, NO. 4, OCTOBER 2000
Lee et al
Angiogenesis in Endometrial Cancer
621
VOL. 96, NO. 4, OCTOBER 2000
levels in the tumor showed linear correlations (RI compared with microvessel density: r ⴝ ⴚ .32, P ⴝ .03; RI compared with vascular endothelial growth factor levels: r ⴝ ⴚ .40, P ⴝ .004; microvessel density compared with vascular endothelial growth factor levels: r ⴝ .36, P ⴝ .011). Conclusion: Blood flow assessed by color Doppler ultrasound has histologic and biologic correlations with angiogenesis and vascular endothelial growth factor levels and might play an important role in predicting tumor progression and metastasis in endometrial carcinoma. (Obstet Gynecol 2000;96:615–21. © 2000 by The American College of Obstetricians and Gynecologists.)
Angiogenesis plays an important role in the pathogenesis of cancer.1 Recent studies have related angiogenesis to cancer growth and metastasis.2,3 The growth of solid tumors and their metastatic spread is angiogenesisdependent and this has been confirmed in several experimental and clinical studies.4,5 Vascular endothelial growth factor, also known as vascular permeability factor, which induces the growth of a capillary network surrounding the tumor and acts as a highly specific mitogen on endothelial cells, is regarded as an important angiogenic factor in tumor angiogenesis.6 Overexpression of vascular endothelial growth factor was found to correlate with survival in several patients with malignant tumors.7,8 However, relatively little is known about the clinical correlation between vascular endothelial growth factor and tumor progression and metastasis in endometrial carcinoma. Ramos et al9 demonstrated the possibility of detection of tumor vasculature by Doppler ultrasound even in a small amount of malignant tissue in an animal model. Color Doppler ultrasound shows that neovascularized vessels have blood flow characteristics that differ from
0029-7844/00/$20.00 PII S0029-7844(00)00976-5
615
those of normal vessels. We also have demonstrated, using color Doppler ultrasound, that increased vasculature tends to have increased malignant potential in ovarian and endometrial neoplasms.10,11 However, the histologic or biologic correlations with blood flow assessed by color Doppler ultrasound still are lacking. Determining microvessel density is regarded as a standard procedure to quantitate tumor angiogenesis. The progression from endometrial hyperplasia, complex hyperplasia, atypical hyperplasia, and stage IA endometrial carcinoma to invasive endometrial carcinoma has been noted.12 However, microvessel density always has been assessed retrospectively and in vitro. Preoperative prediction of the microvessel density would be of value clinically for evaluating severity and progression of disease. The aims of this study were to determine whether there are correlations among the corresponding resistance index (RI), microvessel density, and vascular endothelial growth factor concentration in the tumor and to evaluate whether vascular endothelial growth factor levels reflect the severity of endometrial carcinoma. The information obtained might be useful in the search for histologic and biologic associations with blood flow assessed by color Doppler ultrasound.
Materials and Methods Between January 1993 and December 1997, 55 patients with endometrial carcinoma were enrolled in this study. In all cases, the diagnosis was made by fractional dilation and curettage. Six patients with undetectable blood flow by color Doppler ultrasound were excluded. The other 49 patients (89%) with detectable blood flow were recruited. All 49 women underwent surgery, including total abdominal hysterectomy, bilateral salpingo-oophorectomy, and pelvic and/or para-aortic lymph node dissection or sampling. Patients with lymph node metastasis or deep myometrial invasion (more than one-half depth) underwent postoperative adjuvant irradiation. Surgical specimens were evaluated for tumor size, histologic grading, depth of myometrial invasion, and presence of lymphovascular emboli and lymph node metastasis. Stage was determined using the International Federation of Gynecology and Obstetrics classification.13 Tumors of histologic types other than adenocarcinoma and adenoacanthoma were excluded. The carcinomas were classified using a threegrade system: grade 1 carcinomas showed glandular formation in more than 95% of the tumor, grade 2 carcinomas showed a solid growth pattern in 5–50%, and grade 3 carcinomas showed a solid pattern in more than 50%. We used transvaginal sonography to detect arterial
616 Lee et al
Angiogenesis in Endometrial Cancer
blood flow in the tumor on the day before hysterectomy and at least 10 days after diagnostic uterine curettage. The equipment used was a color Doppler ultrasound unit (Ultramark 9 or HDI 3000; Advanced Technology Laboratories, Inc., Bothell, WA) with a high-pass filter set on 100 Hz to eliminate low-frequency signals resulting from vessel-wall motion. The pulse-repetitive frequency was 1–25 kHz, depending on the blood flow velocity under investigation. The waveforms of blood flow velocity in the artery of the tumor were obtained, and the RI was calculated as the difference between the systolic and end-diastolic frequency shifts divided by the systolic frequency shift. A series of similar arterial waveforms with maximal clarity were obtained reproducibly over three consecutive cardiac cycles and regarded as satisfactory at the site of sampling. The lowest RI was used for analysis. Tissue specimens were obtained during surgery, immediately frozen in liquid nitrogen, and stored at ⫺70C until analyzed. The remaining tissue specimens were sent for pathologic examination. Cancer tissues were prepared as described.14 Briefly, the tissue samples, after being stripped of blood and necrotic tissue, were thawed on ice placed in 10 volumes of ice-cold cell lysis buffer (pH 7.8, containing 100 mM K2HPO4-KH2PO4, 1 mM dithiothreitol, 2 mM ethylenediaminetetra-acetic acid, 1% Triton X-100, and 0.75 g/mL leupeptin) and were homogenized. The lysate was centrifuged and the supernatant recovered and stored at ⫺70C until analysis. The total protein in the prepared supernatant was measured by protein assay (Bio-Rad, Hercules, CA), and then vascular endothelial growth factor levels of the extract were determined by enzyme immunoassay. For determination of vascular endothelial growth factor levels, a commercially available enzyme immunoassay kit was used (Quantikine human vascular endothelial growth factor immunoassay; Research and Diagnostic Systems Inc., Minneapolis, MN) as described.8 Measurement of total protein and vascular endothelial growth factor level in the extract was repeated, and the average values for each sample were recorded. The concentration of vascular endothelial growth factor in the tumor was represented as picogram per milligram of protein. Formalin-fixed, paraffin-embedded tissue blocks were recut, the paraffin was removed, and the tissue sections were rehydrated. Endogenous peroxidase activity was blocked by hydrogen peroxide. Specimens then were incubated with mouse anti-CD34 monoclonal antibody diluted to a 1:20 ratio (BioGenex, San Ramon, CA). Next, they were incubated with biotinylated horse antimouse immunoglobulin (Ig) G and treated with the preformed avidin-biotin complex. The substrate solution for peroxidase contained 0.2% diaminobenzidine
Obstetrics & Gynecology
three areas were counted. The average in the three ⫻200 fields in each tumor was calculated and thereafter was referred to as density count. Large vessels with thick muscular walls and lumina greater than approximately eight blood cells were excluded from the count. All measurements were performed by a single pathologist who was unaware of any clinical or pathologic data before counting. We used SPSS (Statistical Package for Social Sciences) 6.0 (SPSS Inc., Chicago, IL) for Windows to perform the Mann-Whitney U test and Spearman correlation analysis.
Figure 1. Arterial blood flow in the tumor of one patient with endometrial carcinoma (resistance index 0.35).
tetrahydrochloride. Normal mouse IgG was substituted for primary antibody as a negative control. Sections were counterstained lightly with hematoxylin. Microvessels in the tumors were highlighted by antiCD34 immunostaining in formalin-fixed, paraffinembedded sections. Microvessel density was quantified as previously described.15 Briefly, the stained slides were examined at low-power magnification (⫻40 and ⫻100 total magnification) to identify the areas of highest neovascularization (so-called hot spots) of the tumor. The three most vascular areas were chosen in each section. Microvessels on a ⫻200 field (⫻20 objective and ⫻10 ocular [Olympus BH-2 microscope; Olympus, Tokyo, Japan], 0.74 mm2 per field, with the field size measured with an ocular micrometer) in each of the
Results Patients were between 25 and 74 years of age (mean 47.4), with a parity of 0 – 8 and a maximal tumor diameter of 0.5–7.0 cm (mean 3.10). Twenty-seven women were postmenopausal. The RI in the tumor ranged from 0.19 to 0.73 (median 0.47) and a representative case is shown in Figure 1. The microvessel density ranged from 8 to 164 (median 63). The concentration of vascular endothelial growth factor in the tumor ranged from 0 to 4313 pg/mg (median 220). Significantly lower RIs were noted in tumors of stage II or greater (0.37 compared with 0.50, P ⬍ .001), of high histologic grade (grade 3) (0.34 compared with 0.49, P ⫽ .004), with deep myometrial invasion (one-half depth or greater) (0.39 compared with 0.49, P ⫽ .002), with lymphovascular emboli (0.38 compared with 0.49, P ⬍ .001), or with lymph node metastasis (0.30 compared with 0.49, P ⬍ .001) compared with stage I tumors or tumors of histologic grade 1 or 2, with superficial
Table 1. Correlation Between Resistance Index, Microvessel Density, and Vascular Endothelial Growth Factor Levels in the Tumor and Pathologic Parameters in 49 Endometrial Carcinoma Patients
Surgical stage I ⬎I Histologic grade 1 or 2 3 Myometrial invasion ⬍1⁄2 depth ⱖ1⁄2 depth Lymphovascular emboli No Yes Pelvic lymph node metastasis No Yes
VEGF (pg/mg) (25–75%)
P*
MVD (⫻200 field) (25–75%)
n
RI (25–75%)
P*
P*
35 14
.50 (.42–.65) .37 (.29 –.45)
⬍.001
129 (0 –542) 975 (69 –2280)
.014
61 (37–98) 88 (58 –113)
.018
39 10
.49 (.40 –.62) .34 (.26 –.44)
.004
161 (11– 802) 438 (22–1649)
.38
64 (39 –100) 69 (53–104)
.66
35 14
.49 (.42–.62) .39 (.30 –.45)
.002
129 (29 –542) 867 (6 –1589)
.06
61 (38 –98) 77 (56 –113)
.16
37 12
.49 (.40 –.64) .38 (.27–.46)
⬍.001
120 (0 –529) 1138 (358 –2280)
.002
63 (37–99) 86 (58 –114)
.023
37 12
.49 (.45–.61) .30 (.25–.39)
⬍.001
95 (0 – 450) 1011 (587–2225)
⬍.001
61 (36 – 89) 98 (64 –108)
⬍.019
RI ⫽ resistance index; VEGF ⫽ vascular endothelial growth factor; MVD ⫽ microvessel density. * Mann-Whitney U test.
VOL. 96, NO. 4, OCTOBER 2000
Lee et al
Angiogenesis in Endometrial Cancer
617
Figure 2. The correlation between resistance index (RI) and microvessel density (MVD) in the tumor was a negative linear one (r ⫽ ⫺.32, P ⫽ .03).
myometrial invasion, without lymphovascular emboli, or with no lymph node metastasis (Table 1). Increased vascular endothelial growth factor levels and microvessel density also were detected in tumors of stage II or greater (975 compared with 129 pg/mg, P ⫽ .014; 88 compared with 61, P ⫽ .018, respectively), with lymphovascular emboli (1138 compared with 120 pg/mg, P ⫽ .002; and 86 compared with 63, P ⫽ .023), or with lymph node metastasis (1011 compared with 95 pg/mg, P ⬍ .001; and 98 compared with 61, P ⫽ .019). No statistically significant difference could be found in vascular endothelial growth factor levels or microvessel density in tumors of high histologic grade (438 compared with 161 pg/mg, P ⫽ .38; and 69 compared with 64, P ⫽ .66, respectively) or deep myometrial invasion (867 compared with 129 pg/mg, P ⫽ .06; 77 compared with 61, P ⫽ .16) (Table 1). Spearman correlation analysis revealed negative linear correlations between RI in the tumor and microvessel density and vascular endothelial growth factor concentrations (RI compared with microvessel density: r ⫽ ⫺.32, P ⫽ .03, formula: y ⫽ 105.421288 ⫺ 77.796124x; RI compared with vascular endothelial growth factor levels: r ⫽ ⫺.40, P ⫽ .004, formula: y ⫽ 2104.7413 ⫺ 3176.1170x) (Figures 2 and 3). There was a positive linear correlation between vascular endothelial growth factor levels and microvessel density (vascular endothelial growth factor levels compared with microvessel density: r ⫽ .36, P ⫽ .011, formula: y ⫽ 57.831659 ⫹ 0.018078x) (Figure 4).
Figure 3. The correlation between resistance index (RI) and vascular endothelial growth factor (VEGF) concentration in the tumor was a negative linear one (r ⫽ ⫺.40, P ⫽ .004).
investigators to evaluate tumor-associated microvessel density.2–5 The findings of these studies suggest that the greater the number of vessels, the more likely it is that tumor cells will enter the circulation or the lymphatics. Counting microvessels in the tumor is the most commonly used method to quantify tumor angiogenesis. Immunohistochemical staining for endothelial-specific markers such as factor VIII–related antigen, CD31 antigen, or CD34 antigen clearly highlights the vascular endothelium. However, two main problems are encountered when microvessel density measurement is used to quantitate angiogenesis. First, methodologic problems such as inter- and intraobserver variability, tumor heterogeneity, and selection of the area of most intense neovascularization (hot spot) remain unsolved. Absolute vessel counts also are influenced by total magnification, the examination area selected, and the skill and experience of the investigator. Extensive efforts have been made to assess angiogenesis objectively. Second, microvessel
Discussion Potential tumor growth and metastasis depend on the ability of a tumor to induce angiogenesis.1–3 The use of immunohistochemical methods to determine whether a tumor has entered an angiogenic phase has allowed
618 Lee et al
Angiogenesis in Endometrial Cancer
Figure 4. The correlation between microvessel density (MVD) and vascular endothelial growth factor (VEGF) concentration was a positive linear one (r ⫽ .36, P ⫽ .011).
Obstetrics & Gynecology
density is determined retrospectively and in vitro and therefore might not play a role in the preoperative evaluation of and treatment planning for patients with endometrial carcinoma. To improve the accuracy of evaluation of angiogenesis, we attempted to use resistance value of blood flow in the tumor to represent in vivo angiogenesis and vascular endothelial growth factor levels to quantitate angiogenesis more objectively. Resistance index values in the tumor correlated well with microvessel density in this survey. Therefore, blood flow assessed by color Doppler ultrasound could represent the angiogenic activity of endometrial carcinoma. Magnetic resonance imaging (MRI) and transvaginal ultrasound are fairly accurate methods for preoperatively assessing depth of myometrial invasion, tumor staging, and extracorporeal spread.10,16 Color Doppler ultrasound has some advantages over MRI, being less time-consuming and less expensive. This study demonstrates that lower arterial RIs are found in endometrial carcinomas of advanced stages, of high histologic grade, with deep myometrial invasion, with lymphovascular emboli, or with lymph node metastasis. Color Doppler ultrasound has been used to detect neovascularization in solid tumors arising from the ovary.10,16 We have observed incremental angiogenesis in endometrial neoplasms with increased malignant potentials by color Doppler ultrasound.11 Resistance index values in the tumors reflect gross angiogenic activity from feeding arteries of the endometrial tumors. The tumor becomes more angiogenic as the disease progresses. We suggest that endometrial cancer patients with low RIs might have deep myometrial invasion, lymph node metastasis, and advanced-stage tumors. In vivo tumor angiogenesis may be demonstrated by color Doppler ultrasound. However, its histologic and biologic associations with neovascularization observed by color Doppler ultrasound are not yet clear. Microvessel density and vascular endothelial growth factor levels could explain this phenomenon, because both microvessel density and vascular endothelial growth factor levels were linearly correlated with RI in the tumors. This relation showed that blood flow assessed by color Doppler ultrasound had histologic and biological associations with angiogenesis. Conventionally, microvessel density was determined by immunohistochemical staining using antibodies against various endothelial cell–related antigens such as factor VIII, CD31, and CD34.4 Both microvessel density and vascular endothelial growth factor levels were determined in the in vitro surgical specimen. The present study has revealed close correlations between RI, microvessel density, and vascular endothelial growth factor levels in endometrial tumors. Thus, tumor vascularity now can
VOL. 96, NO. 4, OCTOBER 2000
be assessed in vivo and in situ at the time of ultrasound. Importantly, the in vivo RI in the tumor could be determined preoperatively, thus indicating the microvessel density of tumor tissues, and such an assessment could provide useful information to clinicians for evaluating disease progression in and conducting preoperative planning for endometrial carcinoma patients. Patients with lower tumor RIs might have deep myometrial invasion, lymph node metastasis, and distant metastasis. Vascular endothelial growth factor, an endotheliumspecific mitogen and potent mediator of vascular permeability, although expressed in normal tissues such as the endometrium,17 is frequently found in high levels in solid tumors.14 Vascular endothelial growth factor proteins consist of four splicing variants: vascular endothelial growth factor121, vascular endothelial growth factor165, vascular endothelial growth factor189, and vascular endothelial growth factor206.18 The two shorter forms are estimated to be free to diffuse because of weak heparin-binding activity. The vascular endothelial growth factor enzyme immunoassay used in this study detected vascular endothelial growth factor165 isoform. Therefore, actual vascular endothelial growth factor concentrations might be higher than the vascular endothelial growth factor concentrations measured in this study. The time of secretion of vascular endothelial growth factor could be regarded as the critical point at which tumors switch from nonangiogenic to angiogenic status. The high microvessel density reflects only late angiogenesis of the tumor with neovascularization. Therefore, the presence of vascular endothelial growth factor protein could be representative of the early stage of tumor angiogenesis, and high microvessel density could represent the late stage of tumor angiogenesis. Higher vascular endothelial growth factor concentrations and higher microvessel densities were seen only in advanced-stage tumors, tumors with lymphovascular emboli, and tumors with lymph node metastasis. No significant difference in vascular endothelial growth factor levels or microvessel densities was noted in tumors of high histologic grade or with deep myometrial invasion. Tumors of high histologic grade showed a higher incidence of deep myometrial invasion (P ⫽ .003), which indicates that some inherent cancer behavior other than merely tumor angiogenesis may influence the local invasive or metastatic ability. In our study, microvessel density correlated significantly with vascular endothelial growth factor concentration. This finding is compatible with the finding that strong vascular endothelial growth factor messenger RNA expression is associated with increased microvessel density in squamous intraepithelial lesions and invasive squamous cell carcinomas of the cervix.19
Lee et al
Angiogenesis in Endometrial Cancer
619
Tumor cells secrete vascular endothelial growth factor to induce proliferation of endothelial cells, and the proliferative endothelium results in neovascularization, which is represented by microvessel density. Some tumors in our study had high histologic microvessel densities but expressed low levels of vascular endothelial growth factor. The role of the in vivo angiogenic effect is complex, and other angiogenic factors such as fibroblast growth factor and platelet-derived growth factor should be investigated further to elucidate angiogenesis in endometrial cancers. Emoto et al20 evaluated the relationship between in vivo angiogenesis RI determined by transvaginal color Doppler ultrasound and in vitro angiogenesis microvessel density determined by immunohistochemistry. They concluded that blood flow in the tumor may correlate with endothelial cell activity of tumor vessels but not with microvessel density. They demonstrated an indirect relationship between in vivo and in vitro parameters of angiogenesis. Our results showing that RIs in the tumor correlated with histologic microvessel density were different from theirs, perhaps because endometrial tumors are smaller than ovarian tumors. The area where we determined microvessel density is closer to the artery (or arteries) assessed by color Doppler ultrasound for endometrial carcinoma. Therefore, the correlation between microvessel density and RI would be expected to be stronger in endometrial carcinoma than in ovarian carcinoma. Also, vascular endothelial growth factor levels correlated with the proliferation of endothelium; therefore, vascular endothelial growth factor concentrations might indicate proliferative activity of endothelial cells in tumor vessels. Resistance index values in the tumors also correlated well with the vascular endothelial growth factor levels in our study; therefore, blood flow assessed by color Doppler ultrasound might reflect histologic and biologic correlations with angiogenesis and vascular endothelial growth factor levels in tumor angiogenesis. Recently, Kana et al21 described the antitumor and antimetastatic effects of vascular endothelial growth factor–neutralizing antibody in an animal model. If inhibition of vascular endothelial growth factor offers an opportunity for the treatment of endometrial carcinoma, then determination of RI in the tumor by color Doppler ultrasound might be a useful way to assess the magnitude of response. Further study is necessary to clarify the mechanism of promoting angiogenesis by vascular endothelial growth factor in endometrial carcinoma.
References 1. Folkman J, Shing Y. Angiogenesis. J Biol Chem 1992;267:10931– 4.
620 Lee et al
Angiogenesis in Endometrial Cancer
2. Folkman J. The role of angiogenesis in tumor growth. Semin Cancer Biol 1992;3:65–71. 3. Ellis LM, Fidler IJ. Angiogenesis and metastasis. Eur J Cancer 1996;32A:2451– 60. 4. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastasis—Correlation in invasive breast carcinoma. N Engl J Med 1991;324:1– 8. 5. Folkman J. Tumor angiogenesis. In: Holland F, Frei E III, Bast RC Jr, Kufe DW, Morton DL, Weichselbaum RR, eds. Cancer medicine. Philadelphia: Lea & Febiger, 1989:153–70. 6. Senger DR, Connolly DT, Van de Water L, Feder J, Dvorak HF. Purification and NH2-terminal amino acid sequence of guinea pig tumor-secreted vascular permeability factor. Cancer Res 1990;50: 1774 – 8. 7. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hypersensitivity, and angiogenesis. Am J Pathol 1995;146:1029 –39. 8. Obermair A, Kucera E, Mayerhofer K, Speiser P, Seifert M, Czerwenka K, et al. Vascular endothelial growth factor in human breast cancer: Correlation with disease-free survival. Int J Cancer 1997;74:455– 8. 9. Ramos I, Fernandez LA, Morse SS, Fortune KL, Taylor KJ. Detection of neovascular signals in a 3-day Walker 256 rat carcinoma by Doppler ultrasound. Ultrasound Med Biol 1988;14:123– 6. 10. Wu CC, Lee CN, Chen TM, Shyu MK, Hsieh CY, Chen HY, et al. Incremental angiogenesis assessed by color Doppler ultrasound in the tumorigenesis of ovarian neoplasm. Cancer 1994;73:1251– 6. 11. Cheng WF, Chen TM, Chen CA, Wu CC, Huang KT, Hsieh CY, et al. Clinical application of intratumoral blood flow study in patients with endometrial carcinoma. Cancer 1998;82:1881– 6. 12. Abulafia O, Triest WE, Sherer DM, Hansen CC, Chezzi F. Angiogenesis in endometrial hyperplasia and stage I endometrial carcinoma. Obstet Gynecol 1995;86:479 – 85. 13. International Federation of Gynecology and Obstetrics (FIGO) stages—1988 revision. Gynecol Oncol 1989;35:125–7. 14. Cheng WF, Chen CA, Lee CN, Chen TM, Hsieh FJ, Hsieh CY. Vascular endothelial growth factor in cervical carcinoma. Obstet Gynecol 1999;93:761–5. 15. Cheng WF, Lee CN, Chu JS, Chen CA, Chen TM, Shau WY, et al. Vascularity index as a novel parameter for the “in vivo” assessment of angiogenesis in cervical carcinoma. Cancer 1999;85:651–7. 16. Fleischer AC, Rodgers WH. Ovarian masses. In: Fleischer AC, Emerson DS, eds. Color Doppler sonography in obstetrics and gynecology. New York: Churchill Livingstone, 1993:59 –121. 17. Charnock-Jones DS, Sharkey AM, Rajput-Williams J, Burch D, Schofield JP, Fountain SA, et al. Identification and localization of alternately spliced mRNAs for vascular endothelial growth factor in human uterus and estrogen regulation in endometrial carcinoma cell lines. Biol Reprod 1993;48:1120 – 8. 18. Ferrara N, Houck KA, Jakeman LB, Winer J, Leung DW. The vascular endothelial growth factor family of polypeptides. J Cell Biochem 1991;47:211– 8. 19. Guidi AJ, Abu-Jawdeh G, Berse B, Jackman RW, Tognazzi K, Dvorak HF, et al. Vascular permeability factor (vascular endothelial growth factor) expression and angiogenesis in cervical neoplasia. J Natl Cancer Inst 1995;87:1237– 45. 20. Emoto M, Iwasaki H, Mimura K, Kawarabayashi T, Kikuchi M. Differences in the angiogenesis of benign and malignant ovarian tumors, demonstrated by analyses of color Doppler ultrasound, immunohistochemistry, and microvascular density. Cancer 1997; 80:899 –907. 21. Kana T, Konno H, Tanaka T, Baba M, Matsumoto K, Nakamura S, et al. Anti-tumor and anti-metastatic effects of human-vascularendothelial-growth-factor-neutralizing antibody on human colon
Obstetrics & Gynecology
and gastric carcinoma xenotransplanted orthotopically into nude mice. Int J Cancer 1998;77:933– 6.
Received December 13, 1999. Received in revised form March 20, 2000. Accepted March 30, 2000.
Address reprint requests to:
Fon-Jou Hsieh, MD Department of Obstetrics and Gynecology National Taiwan University Hospital No. 7, Chung-Shan South Road Taipei Taiwan E-mail: [email protected]
NEW,
Copyright © 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
THIRD EDITION OF WRITING GUIDE AVAILABLE
To help prospective authors, especially those who are beginning in medical journal writing, to produce better papers, this journal has developed “A Guide to Writing for Obstetrics & Gynecology.” Copies of the third edition of this booklet are available, free of charge, from: Editorial Office, Obstetrics & Gynecology, 10921 Wilshire Boulevard, Suite 403, Los Angeles, CA 90024-3908. Requests may also be submitted by FAX: (310) 208-2838, or E-mail, [email protected].
VOL. 96, NO. 4, OCTOBER 2000
Lee et al
Angiogenesis in Endometrial Cancer
621