Tumor Angiogenesis: Determined By VEGF Expression, MAGS Scoring, Doppler Study, As Prognostic Indicator In Carcinoma Breast

ISPUB.COM The Internet Journal of Surgery Volume 8 Number 1 Tumor Angiogenesis: Determined By VEGF Expression, MAGS Scoring, Doppler Study, As Prognostic Indicator In Carcinoma Breast R Shankar, S Tiwary, R Khanna, M Kumar, A Khanna Citation R Shankar, S Tiwary, R Khanna, M Kumar, A Khanna. Tumor Angiogenesis: Determined By VEGF Expression, MAGS Scoring, Doppler Study, As Prognostic. The Internet Journal of Surgery. 2005 Volume 8 Number 1. Abstract Introduction: Tumor angiogenesis have been newly emerging prognostic indicator in many tumors including carcinoma breast. Various methods of assessment of angiogenesis are VEGF expression, MAGS scoring,chalky counting, MVD, and Colour Doppler. We assessed VEGF, MAGS scoring and Doppler Ultrasound in patients of carcinoma breast and fibroadenoma. Material and Methods: 30 samples of carcinoma breast and 10 fibroadenomas had MAGS scoring and immunohistochemical staining for VEGF. All patients of carcinoma breast had Color Doppler examination by 7.5 MHz probe and Peak systolic velocity, Peak diastolic velocity, Resistivity index and Pulsatility index were studied. The values were related with various well known clinicopathologic parameters of carcinoma breast and also among themselves. Result : VEGF expression is significantly more diffuse and intense in lymph node positive and higher grade carcinoma breast (p<0.001). MAGS score was also higher in lymph node positive and higher grade carcinoma breast (p<0.001). Color Doppler was not related to any of the clinicopathologic parameters of breast cancer. Conclusion : Higher VEGF expression and higher MAGS score can be used as a poor prognostic indicator in carcinoma breast while colour Doppler failed to yield results as a prognostic marker. INTRODUCTION The tumor metastasis depends on capacity of tumor cells to induce angiogenesis. Tumor growth beyond 1 to 2 mm3 depends on angiogenesis (Folkman et al, 1966). Major angiogenic factors are VEGF (Vascular Endothelial Growth Factor), basic fibroblast growth factor, platelet derived endothelial cell growth factor (Fidler et al, 2000). Estimation of vascularity of tumor and tumor angiogenesis have been newly emerging prognostic indicatior in many tumors. Measuring VEGF in solid tumors provides new and sensitive markers for tumor progression, metastasis and prognosis (Gasparini et al, 1995). Consistent with its proposed role in the pathogenesis of tumors and metastasis, VEGF has prognostic value in many cancers including carcinoma breast, esophagus, colorectal, ovarian, melanoma, lung carcinomas. VEGF is a unique mitogen for vascular endothelial cells which has many splice variants, VEGF 121, 145, 165, 183, 189, 206. All these variants are expressed in carcinoma breast however expression is highest for VEGF 121 (Margit et al, 2002). In some studies VEGF expression has been demonstrated to be an independent prognostic marker for survival in carcinoma breast. In this study, we examined the expression of VEGF in breast carcinoma and investigated its relation with size of tumor, lymph node status, metastasis, stage of tumor, grade of tumor and vascular invasion. Also a correlation was made between various Doppler parameters and MAGS. MAGS (Microscopic angiogenesis grading system) scoring of tumor tissues was described by Brem et al in 1972 is a quantitative technique of measuring degree of angiogenesis in a tumor. It is based on 3 parameters, vasoproliferation, endothelial cell hyperplasia and endothelial cytology. Brem et al 1972 in their pioneer study on wide range of human neoplasms found that chondrosarcomas were most 1 of 8

endothelial poor tumors with MAGS score of 10-20, fibrosarcomas had higher scores of 20-30. Adenocarcinomas had still more intense neovascularization, of MAGS 40-70. Kumar M et al (2004) reported that MAGS scoring was significantly correlated with grade of tumor and gleasons score in prostatic cancers and poorly differentiated tumors had higher MAGS score. on intensity of staining-+, ++ and +++ (c) focal staining was given 1 mark and diffuse staining was given 2 marks, + intensity was given 1 mark, ++ as 2 marks, and +++ as 3 marks. The type of staining was multiplied by intensity of staining and VEGF Multiplication score was obtained. We studied MAGS scoring in breast cancers and fibroadenomas and correlations have been looked with various other clinicopathologic parameters as well as color Doppler and VEGF expression. Vasoproliferation (N), Endothelial Cell Hyperplasia (E), Endothelial Cytology(X) were studied under microscope on 4 m paraffin sections stained with Hematoxylin and Eosin under 500 times magnification and score was calculated as MATERIAL AND METHODS MAGS SCORE = KnN + KeE + KxX The study was conducted from March 2003 to March 2005 in an University hospital, Varanasi, India. FNAC proven 30 patients of carcinoma breast and 10 patients of fibroadenoma were taken in study. All patients underwent detailed history and examination and investigations to look for metastasis. All patients of carcinoma breast underwent color Doppler study of the breast by 7.5 MHz probe and their Peak Systolic Velocity (PSV), Peak Diastolic Velocity (PDV), Resistivity Index (RI) and Pulsatility Index (PI) were determined. 16 patients underwent modified radical mastectomy while 14 underwent Tru-cut needle biopsy followed by neoadjuvant chemotherapy (CAF). The tissue samples were collected in 10% formalin and were finally processed to paraffin wax. Detailed histology regarding type of tumor, grade, vascular invasion, lymph node status was performed. All the samples of carcinoma breast and fibroadenomas underwent MAGS scoring and immunohistochemical staining with VEGF. (The Kit for VEGF assessment was obtained from Lab Vision ) VEGF STAINING Poly-L-Lysine coated slides were used to fix the tissues over it. Immunohistochemical staining was done for VEGF on both carcinoma breast and fibroadenoma samples. Ultra V Block was used to block non specific background staining. Primary antibody Ab-3 (JH-121) mouse monoclonal antibody was applied for 60 minutes at room temperature. Biotinylated goat anti polyvalent was applied for 30 minutes at room temperature. DAB chromogen was used for staining, and counterstaining was done with hematoxylin. Positive control for VEGF staining was Angiosarcoma stained with VEGF which took diffuse and deep intensity staining. Three parameters were taken into consideration (a) depending on type of staining- focal or diffuse (b) depending 2 of 8 MICROSCOPIC ANGIOGENESIS GRADING SYSTEM (MAGS): Vasoproliferation has a maximum value set at 40 with Kn a constant for microscope and N is the number of capillaries seen under 500 times magnification in each field. Endothelial cell hyperplasia has a value set at 30 with Ke = 3. E is the number of cells lining the cross section of the capillary. Endothelial cytology was set a maximum value of 30. X ranged from 0-5 and stood for histological appearance of endothelial cell in hyperplastic zone. Kx was a constant set to allow maximal cytological change of endothelium to correspond numerically to previous two parameters. The MAGS score using the above mentioned formula ranges from 0 to 100. STATISTICS All the parameters were correlated with well known prognostic parameters of carcinoma breast as well as with themselves. Statistical analysis was done by SPSS version 10.0. Chi square test, paired T test, ANOVA test and Pearson Correlation were determined to assess the level of significance. P<0.05 were taken as significant. RESULTS Mean age of all the cancer patients was 43.1 ± 11.5 year (range 28-81 yrs). None of the patient were in stage I, 5 (16.7%)were in stage II, 21 (70%) in stage III and 4 (13.3%) in stage IV. Distribution of various Doppler parameters in carcinoma breast patients is shown in table 1.

Figure 1 Figure 2 Table 1: Distribution of various Doppler parameters in carcinoma breast patients Table 2: Distribution according to VEGF expression and MAGS score in benign and carcinoma breast patients 21 cases of carcinoma breast had diffuse staining for VEGF while 9 had focal. In case of fibroadenoma all the cases had only focal expression. Maximum cases of carcinoma breast had ++ or +++ scores while only 2 cases of fibroadenoma had ++ intensity. The multiplication score in carcinoma breast was 6 in 13 patients. While in fibroadenomas highest multiplication score was 2 which was seen only in 2 patients (Table 2). All carcinoma breast tissues had MAGS score more than 20 but only one fibroadenoma tissue had MAGS score more than 20. 19 cancer patients (63.3%) had MAGS score more than 40 (Table 2). Highest MAGS score of carcinoma was 55 & of fibroadenoma was 22. On comparasion of MAGS score and VEGF expression between fibroadenoma and carcinoma patients,it was found that all the scores were significantly higher in cancer group than the fibroadenoma group (Table 3). Figure 3 Table 3: Correlation of MAGS, VEGF intensity and VEGF multiplication score between carcinoma breast and fibroadenomas All the parameter were higher in tumors with vascular invasion but these differences were not statistically significant except for PSV and PDV (Table 4). Tumor size was not related with MAGS score, PSV, PDV, RI, PI, focal / diffuse VEGF expression and multiplication scores. However size showed a statistically significant relation with intensity of VEGF staining. Larger tumors had more intense VEGF expression (Table 4). Patients with lymph node metastasis had more diffuse expression of VEGF than without lymph nodes and this was statistically significant. There was significant difference in MAGS score, VEGF intensity, multiplication score between positive and negative lymph node groups. But there was no significant difference in the two categories in any of the Doppler parameters 3 of 8

(Table 4). Higher mean MAGS score (50.5 ± 4.04) was associated with stage IV tumor than stage III i.e. 41.21 ± 7.81 and stage II 33.0 ± 7.07 which was statistically significant. Other variables did not show any significant difference between stages (Table 4). With increasing grade, tumor showed more diffuse staining for VEGF. Grade was significantly related to intensity of expression of VEGF, multiplication score and MAGS score. There was no correlation between grade and Doppler parameters (Table 4). Tumor with distant metastasis had higher MAGS score and VEGF multiplication scores than tumor without distant metastasis. But there was no significant relation between VEFG focal/diffuse expression, intensity of VEGF and any of the Doppler parameters with metastasis (Table 4). Figure 4 Table 4: Correlation of VEGF, MAGS score and Color doppler with various clinicopathologic parameters of breast cancer Highly significant correlation was found, between MAGS score and VEGF intensity; MAGS score and VEGF multiplication score (p<0.001); VEGF intensity and VEGF multiplication score; VEGF intensity and VEGF multiplication score; PSV & PDV and RI and PI (Table 5). 4 of 8

Figure 5 Figure 8 Table 5: Correlation among various angiogenesis parameters Figure 3: Microphotograph of Invasive ductal carcinoma with high MAGS score (x 400) Figure 6 Figure 1: Microphotograph of Carcinoma Breast with diffuse VEGF expression and 3+ intensity (x 400) DISCUSSION Figure 7 Several studies have indicated tumor angiogenesis as a prognostic parameter in various cancers including carcinoma breast. Using immunohistochemical and biochemical methods, several studies have shown a worse prognosis for those patients with tumors with high angiogenic activity. Tumor angiogenesis can be studied by various parameters like estimation of expression of VEGF, MVD, Chalky counting, MAGS scoring, Color Doppler study. We looked into a combination of three modes of estimating angiogenesis; VEGF by Immunohistochemial staining; MAGS on Histology; and Color Doppler study. All these three parameters have been correlated with other well known prognostic parameters of carcinoma breast and also among themselves. Figure 2: Microphotograph of Carcinoma Breast with diffuse VEGF expression and 3+ intensity (x 200) 5 of 8 Peters Engl et al (1995) reported that RI more than 0.7 was characteristic of malignant tumors with a sensitivity of 82% and specificity of 81%. Again in 1999 same group of authors stated that patients with an intratumoral blood flow velocity more than 25 cm/s have 4.33 fold increased risk of death from breast cancer. In our study mean PSV was 23.56 ± 15.49. Yang et al (2002) did not find any correlation between PSV and tumor size, lymph node status, stage. In our study PSV was higher in larger, higher grade, advanced stage and metastatic tumors. But these relations were not found statistically significant. Mehta et al (2000) and Lee WJ et al (1995) stated that presence of high flow tumor signals is associated with axillary lymph node metastasis but we did not find any correlation of PSV with lymph node metastasis. We found statistically significant correlation between vascular invasion and PSV and PDV. PDV also did

not show any correlation with other parameters. Similarily RI and PI also did not show any correlation with any of the clinicopathologic parameters. cheap method for assessment of neoangiogenesis and also has very good reproducibility. But very limited number of studies are available in literature. Stoeltzing et al (2003) showed that the expression of VEGF was higher in patients with metastatic colonic cancers as compared to those with localized disease and it was directly correlated with the extent of proliferation and neovascularization. Similar findings are also reported on other cancers like hepatocellular carcinoma, melanoma, gastric carcinoma, lung carcinoma and all suggest that VEGF overexpresion is associated with poorer prognosis. We did MAGS scoring of all the cases of carcinoma and fibroadenoma. and found that malignant tumors had much higher MAGS score than benign tumors. Large number of studies have been done on carcinoma breast also. Margit et al (2002) found that of all the VEGF variants (121,145,165,183,189,206) VEGF 121 show highest level of expression. In our study also we have seen for VEGF 121 expression. We did VEGF staining of both carcinoma breast and fibroadenoma cases and found that carcinomas expressed more diffuse and intense staining, also the multiplication score we calculated was significantly higher in carcinoma than fibroadenoma. Linderholm and Colleagues (1998, 2000) reported that VEGF expression was significantly associated with tumor size and histological grade. Similar study on carcinoma breast was also conducted by Gottfried et al (2004), they also confirmed that VEGF is a significant marker for overall survival. They found that VEGF is associated significantly with tumor size, and high nuclear grade and positive axillary lymph node status. We also found that larger tumor showed more intense VEGF expression. Also, the metastatic lymph node tumor expressed more diffuse and intense staining, the multiplication score was also higher in them as compared to tumor which did not have metastatic lymph node. When grade I, II & III were compared regarding VEGF, significant difference was observed between them (p<0.001) in all the VEGF parameters. The mean scores were increasing as grade was increasing similar to the observation made by linderholm (2000) and Gottfried (2004). We did not observe any relation of VEGF expression with stage of disease and vascular invasion. But multiplication score was found significantly higher in patients with distant metastasis than those without metastasis. MAGS score was first demonstrated by stress Brem et al (1972) they found that higher scores were consistently displayed by more malignant brain neoplasms. It provides a composite score ranging from 0 to 100. It is a very easy and 6 of 8 Kumar M et al (2004) reported that higher MAGS score was associated with poorly differentiated prostatic carcinoma. In our study also higher MAGS score was associated with higher grade tumor and the difference between them was statistically significant. We also found positive significant correlation of VEGF with, lymph node status, distant metastasis and stage of tumor. Higher scores were found in tumors with lymph node metastasis, distant metastasis and advanced stage tumors. On comparison between Doppler, VEGF and MAGS, we found that MAGS was positively correlated to VEGF (focal / diffuse, intensity and multiplication scores). But no correlation was observed between any of the Doppler parameter and MAGS or VEGF expression. CONCLUSION Our study was designed to evaluate the prognostic importance of VEGF expression, MAGS score and Doppler parameters in breast cancers. We compared all these parameters with tumor size, stage, histological grade, lymph node status, vascular invasion and metastasis. It was found that higher grade, and lymph node positive tumors expressed more VEGF and they had more MAGS score as well. Also, MAGS score is higher in metastatic and advanced stage tumors. Intensity of VEGF expression also increased significantly as the size of tumor increased. A significant positive correlation was observed between VEGF expression and MAGS score but not with Doppler parameters. Therefore, our study suggests that expression of VEGF and MAGS score are good markers for tumor angiogenesis for prognostication in carcinoma breast and they are markers of poor prognosis. CORRESPONDENCE TO Prof. A K Khanna Department of Surgery, Institute of Medical Sciences, Banaras Hindu University, Varanasi221005 (India) Phone: 91-542 - 2318418 Fax: 91-542-2367568 Email: akk_dr@sify.com

References 1. Fiddler IJ. Angiogenesis and cancer metastasis. Cancer Journal 2000;6(Suppl 2): S134-41. 2. Folkman J, Cole P, Zimmerman S. Tumor behavior in isolated perfused organs: in vitro growth and metastases of biopsy material in rabbit thyroid and canine intestinal segment. Ann Surg 1966;164:491-502. 3. Gasparini G, Harris AL. Clinical importance of the determination of tumor angiogenesis in breast carcinoma: much more than a new prognostic tool. J Clin Oncol 1995;13:765-782. 4. Margit Stimpfl, Dan Tong, Barbara Fasching, Eva Schuster, Andreas Obermair, Sepp Leodolter and Robert Zeillinger1 Vascular Endothelial Growth Factor Splice Variants and Their Prognostic Value in Breast and Ovarian Cancer. Clinical cancer research Vol. 8,2253-2259,July 2002 5. Brem S, Contran R, Folkman J. Tumor angiogenesis : A quantitative method for histologic grading. J Nat Cancer Inst 1972;48:347-356. 6. Kumar M, Tandon V, Singh PB, Mitra S, Vyas N. Angiogenesis and Histologic scoring in prostatic carcinoma carcinoma - A valuable cost effective prognostic indicator. Uro Oncol 2004;4: 139-143. 7. Peters-Engl C, Frank W, Leodolter S and Medl M. Tumor flow in malignant breast tumors measured by Doppler ultrasound: an independent predictor of survival. Breast Cancer Res Treat 1995;54:65-71 7 of 8 8. Yang WT, Tse GM, Lam PK, Metreweli C, Chang J. Correlation between color power Doppler sonographic measurement of breast tumor vasculature and immunohistochemical analysis of microvessel density for the quantitation of angiogenesis. J Ultrasound Med 2002;21:1227-35. 9. Mehta TS, Raza S, Baum JK. Use of Doppler ultrasound in the evaluation of breast carcinoma. Semin Ultrasound CT MR 2000;21:297-307. 10. Lee WJ, Chu JS, Houng SJ, Chung MF, Wang SM. Breast cancer angiogenesis : a quantitative morphologic and Doppler imaging study. Ann Surg Oncol 1995;2:246-251. 11. Stoeltzing O, Liu W, Reinmuth N, Parikh A, Ahmad SA et al. Angiogenesis and antiangiogenic therapy of colon cancer liver metastasis. Ann Surg Oncol 2003;10(7): 722-33. 12. Linderholm B, Tavelin B, Grankvist K et al. Vascular endothelial growth factor is of high prognostic value in node-negative breast carcinoma. J Clin Oncol 1998;16:3121-3128. 13. Linderholm B, Lindh B, Tavelin B. p53 and vascular endothelial growth factor (VEGF) expression predicts outcome in 833 patients with primary breast carcinoma. Int J Cancer 2000;89:51-62. 14. Gottfried E. Konecny, Glosia Y. Meng, Michael Untch, Melinda Epstein Marh D. Pegram. Association between HER-2/neu and vascular endothelial growth factor expression predicts clinical outcome in primary breast cancer patients. Clinical Cancer Res 2004;10:1706-1716.

Author Information Ravi Shankar, MBBS Junior Resident, Department of General Surgery, Institute of Medical Sciences, Banaras Hindu University Satyendra K. Tiwary, MS Senior Registrar, Department of General Surgery, Institute of Medical Sciences, Banaras Hindu University Rahul Khanna, MS, PhD Associate Professor, Department of General Surgery, Institute of Medical Sciences, Banaras Hindu University Mohan Kumar, MD Professor, Department of Pathology, Institute of Medical Sciences, Banaras Hindu University A. K. Khanna, MS, FACS Professor, Department of General Surgery, Institute of Medical Sciences, Banaras Hindu University 8 of 8