Patients and Methods
Between October 1995 and November 1997, 104 consecutive women with clinical stages IB–IIA cervical carcinoma, who had radical abdominal hysterectomies and pelvic lymph node dissections, were enrolled. Another 30 consecutive women with leiomyomata uteri who had simple hysterectomies without evidence of cervical disease served as controls. Women treated with radiotherapy or chemotherapy before surgery were excluded. The operative method was the standard type III Wertheim hysterectomy plus dissection of lymph nodes en bloc, including common iliac nodes and pelvic lymph nodes. We recorded clinical and pathologic data including demographics, clinical stage, tumor size, histologic type, depth of cervical stromal invasion, parametrial involvement, lymphovascular emboli, and status of pelvic lymph nodes.
Tissue specimens were taken during surgery, immediately frozen in liquid nitrogen, and stored at −70C until analyzed. One part of the tissue specimens was used for the study, and the other part was sent for pathologic examination. Preparation of cancerous tissue was done as described by Toi et al,9 with modifications. After stripping away blood and necrotic tissue, the samples were thawed on ice, placed in 10 vol of ice-cold cell lysis buffer (pH 7.8, containing 100 mM K2HPO4-KH2PO4, 1 mM DTT, 2 mM EDTA, 1% Triton X-100, and 0.75 μg/mL leupeptin), and homogenized. Benign cervical specimens were excised from the squamocolumnar junction of the cervices and prepared according to the same procedure as for cancer specimens. 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). Then vascular endothelial growth factor levels of the extract were measured by enzyme immunoassay, using a commercially available kit (Quantikine Human Vascular Endothelial Growth Factor Immunoassay; R&D Systems, Minneapolis, MN), as described by Obermair et al.10 The measurements of total protein and vascular endothelial growth factor in the extract were duplicated, and the average value of each sample was recorded. Sensitivity of the assay was less than 5 pg/mL, and intra- and interassay variation was less than 10%. Vascular endothelial growth factor protein level is presented as picograms per milligram of protein.
The Student t test, Kruskal-Wallis test, Mann-Whitney U test, Fisher exact test, and Spearman correlation coefficient analysis were used for statistical analysis. Vascular endothelial growth factor concentration was analyzed not only as a continuous variable, but as a categorical variable, defining values below the 75% quantile as normal expression and above the 75% quantile as overexpression.10 P < .05 was statistically significant.
The average age of women in the cancer and control groups was 50.6 years (range 29–70 years) and 43.5 years (range 33–51 years), respectively, and 45 (42.9%) and three (10.0%), respectively, were menopausal. The cancer group had higher gravidity (5.4 ± 6.4 versus 3.6 ± 2.3, P = .02) and higher parity (3.5 ± 1.7 versus 2.5 ± 1.5, P = .002) than controls. The diameter of tumors in the cancer group ranged from 0.8 to 6.0 cm (median 2.5 cm). Eighty-seven (83.6%) patients presented with stage I cervical cancer and 17 (16.4%) with stage II disease. The histologic types were squamous cell carcinoma (n = 79), adenocarcinoma (n = 20), and adenosquamous cell carcinoma (n = 5). Thirty-three (31.7%) tumors with lymphovascular emboli were noted. Tumors with parametrial invasion were found in 32 patients (30.8%), and pelvic lymph node metastases were found in 20 patients (19.2%).
Vascular endothelial growth factor concentrations in cervical cancer tissues ranged from 0 to 4050 pg/mg of protein (median 180.0) and those in normal cervix ranged from 0 to 120 pg/mg of protein (median 0.0), a significant difference (P < .001).
As shown in Table 1, there were no significant differences in clinical stage, menopausal status, or histologic type with respect to vascular endothelial growth factor concentration. The correlation between the patient age and vascular endothelial growth factor concentration was not significant (P = .13). Tumors larger than 4 cm (1030.0 versus 118.0 pg/mg of protein, P < .001) and with deep stromal invasion (364.0 versus 111.0 pg/mg of protein, P = .016) had higher median vascular endothelial growth factor levels than those smaller than 4 cm or with superficial stromal invasion. Tumors with lymphovascular emboli (568.0 versus 118.0 pg/mg of protein, P = .006), parametrial invasion (582.0 versus 117.0 pg/mg of protein, P = .04), or pelvic lymph node metastasis (759.5 versus 121.0 pg/mg of protein, P = .002) also had higher vascular endothelial growth factor levels than tumors without them.
We defined values of vascular endothelial growth factor concentration between 0 and 75% (0–750 pg/mg of protein) as normal expression and values above the 75% quantile (more than 750 pg/mg of protein) as overexpression. Tumors with overexpressed vascular endothelial growth factor were larger (3.35 ± 1.17 versus 2.13 ± 1.28 cm, P < .001) and had higher incidence of deep stromal invasion (20 of 57 versus 6 of 47, P = .009), lymphovascular emboli (15 of 33 versus 11 of 71, P = .011), parametrial invasion (15 of 32 versus 11 of 72, P = .002), and lymph node metastasis (10 of 20 versus 16 of 84, P = .004) than those with normal expression (Table 2).
The VEGF protein levels showed a positive correlation with tumor sizes by Spearman correlation coefficiency (r = 0.340, P < .001).
Toi et al9 reported that intratumoral vascular endothelial growth factor concentrations were higher in more vascularized than less vascularized malignant breast tumors. Obermair et al10 observed significantly higher intratumoral vascular endothelial growth factor concentrations in breast cancer tissue than normal epithelial breast tissue. Our investigation showed that vascular endothelial growth factor concentrations were significantly higher in cervical cancer tissues than normal cervical tissues, as Obermair et al10 reported for breast cancer. Dobbs et al11 found that secretion of vascular endothelial growth factor could be detected as early as intraepithelial lesions. Guidi et al7 also found that neoplastic cells strongly expressed mRNA, encoding the angiogenic cytokine vascular endothelial growth factor in most invasive cervical carcinomas. Epithelial cells in most examples of benign epithelia expressed only focal low-level vascular endothelial growth factor mRNA.8 Those findings showed that onset of angiogenesis was an early event in the premalignant changes of the cervix, which enhanced the expression of vascular endothelial growth factor by the abnormal epithelium and increased the amount of vascular endothelial growth factor as the disease progressed. This present investigation agrees with previous studies that cervical cancer tissues have markedly increased protein levels of vascular endothelial growth factor over those of nonneoplastic cervical epithelial tissues. It is noteworthy that vascular endothelial growth factor expression was also observed in benign cervical tissues. We postulate that vascular endothelial growth factor might affect normal physiologic or inflammatory processes of the cervix, or that there are abnormal epithelia of cervices that can secrete vascular endothelial growth factor but not be detected as abnormal on histopathologic examination.
Tokumo et al12 reported that the intensity of vascular endothelial growth factor expression was significantly stronger in adenocarcinoma than in squamous cell carcinomas. We found no difference in the vascular endothelial growth factor concentrations between different histologic types. Different measuring methods might influence sensitivity of detection of vascular endothelial growth factor expression. Further study to evaluate the correlation between enzyme immunoassay and immunohistochemical staining could determine whether different histologic types of cancer express different amounts of vascular endothelial growth factor.
Tumor sizes correlated well with vascular endothelial growth factor concentrations. Angiogenesis is essential for tumor growth beyond a critical size, and numerous angiogenic factors have been identified.13,14 Vascular endothelial growth factor regulates tumor angiogenesis. We found that tumor burden might influence production of vascular endothelial growth factor. The mechanisms responsible for tumor angiogenesis are complex, and whether secretion of vascular endothelial growth factor induces tumor growth or tumor cells secrete vascular endothelial growth factor during tumor growth is still unknown. Understanding the effect of vascular endothelial growth factor in cervical neoplasia requires further investigation.
In our series, tumors with overexpressed vascular endothelial growth factor showed higher percentages of deep stromal invasion, parametrial invasion, lymphovascular emboli, and lymph node metastases than those with superficial stromal invasion and without parametrial invasion, lymphovascular emboli, or lymph node metastases. Those data suggest that concentration of vascular endothelial growth factor could indicate the local invasive activity and metastatic potential of cervical carcinoma. However, we also found some tumors expressing low vascular endothelial growth factor concentrations that had invasive and metastatic activity. Our explanations are that some inherent cancer behavior, not merely the size of tumors, regulates production of vascular endothelial growth factor from tumor tissues and that angiogenic cytokines or growth factors other than vascular endothelial growth factor influence the invasive and metastatic potential of cervical carcinoma.
Radical hysterectomy with pelvic lymphadenectomy is frequently used to treat stage IB and IIA cervical cancer.15,16 The presence of pelvic lymph node metastasis is one of the most important prognostic factors in cervical carcinoma and might be crucial for treatment planning.17 We found 58% (15 of 26) and 38% (11 of 26) of tumors overexpressing vascular endothelial growth factor (>750 pg/mg of protein) associated with parametrial invasion and lymph node metastasis, respectively. Tumors with overexpressed vascular endothelial growth factor might have higher incidences of advanced local invasion and lymph node metastasis. Preoperative vascular endothelial growth factor concentration might be used to differentiate high-risk patients with parametrial invasion or lymph node metastasis and might be helpful in the treatment planning of early-staged cervical cancer patients. We suggest that early-stage tumors with overexpressed vascular endothelial growth factor be treated with radiation therapy or chemoradiation rather than with an operation because of high-risk of parametrial invasion or lymph node metastasis. A large investigation would be necessary to evaluate this clinical application.
1. Folkman J. Tumor angiogenesis: Therapeutic implications. N Engl J Med 1971;285:1182–6.
2. Weidner N, Semple JP, Welch WR, Folkman J. Tumor angiogenesis and metastatic correlation in invasive breast carcinoma. N Engl J Med 1991;324:1–8.
3. Ferrara N, Heinsohn H, Waldner CE, Bunting S, Thomas GR. The regulation of blood vessel growth by vascular endothelial growth factor. Ann N Y Acad Sci 1995;752:246–56.
4. Klasbrun M, Soker S. VEGF/VPF: The angiogenic factor found? Curr Biol 1993;3:699–702.
5. Dvorak HF, Brown LF, Detmar M, Dvorak AM. Vascular permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 1995;146:1029–39.
6. Boocock CA, Charnock-Jones DS, Sharkey AM, McLaren J, Barker PJ, Wright KA, et al. Expression of vascular endothelial growth factor and its receptors Flt and KDR in ovarian carcinoma. J Natl Cancer Inst 1995;87:506–16.
7. 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.
8. Brown LF, Berse B, Jackman RW, Tognazzi K, Guidi AJ, Dvorak HF, et al. Expression of vascular permeability factor (vascular endothelial growth factor) and its receptors in breast cancer. Hum Pathol 1995;26:86–91.
9. Toi M, Kondo S, Suzuki H, Yamamoto Y, Inada K, Imazama T, et al. Quantitative analysis of vascular endothelial growth factor in primary breast cancer. Cancer 1996;77:1101–6.
10. Obermair A, Kucera E, Mayerhofer K, Speiser P, Seifert M, Czerwenka K, et al. Vascular endothelial growth factor (VEGF) in human breast cancer: Correlation with disease-free survival. Int J Cancer 1997;74:455–8.
11. Dobbs SP, Hewett PW, Johnson IR, Carmichael J, Murray JC. Angiogenesis is associated with vascular endothelial growth factor expression in cervical intraepithelial neoplasia. Br J Cancer 1998; 76:1410–5.
12. Tokumo K, Kodama J, Seki N, Nakanishi Y, Miyagi Y, Kamimura S, et al. Different angiogenic pathways in human cervical cancers. Gynecol Oncol 1998;68:38–44.
13. Folkman J. What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 1990;82:4–6.
14. Folkman J. The role of angiogenesis in tumor growth. Semin Cancer Biol 1992;3:65–71.
15. Delgado G, Bundy B, Zaino R, Sevin BU, Creasman WT, Major F. Prospective surgical-pathological study of disease-free interval in patients with stage IB squamous cell carcinoma of the cervix: A Gynecologic Oncology Group study. Gynecol Oncol 1990;38:352–7.
16. Inoue T, Morita K. The prognostic significance number of positive nodes in cervical carcinoma stage IB, IIA and IIB. Cancer 1990;65:1923–7.
17. Lin HH, Cheng WF, Chan KW, Chang DY, Chen CK, Huang SC. Risk factors for recurrence in patients with stage IB, IIA, and IIB cervical carcinoma after radical hysterectomy and postoperative pelvic irradiation. Obstet Gynecol 1996;88:274–9.