Angiogenesis 7: 53–58, 2004. 2004 Kluwer Academic Publishers. Printed in the Netherlands.
53
Glutathione S-transferase genotype GSTM1 as a predictor of elevated angiogenic phenotype in patients with early onset breast cancer Rui Medeiros1, Raquel Soares1, 2, Andre´ Vasconcelos1, Fernando Schmitt2 & Carlos Lopes1 1
Molecular Oncology Unit – Department of Pathology, Instituto Portugueˆs de Oncologia, Porto, Portugal; 2Institute of Molecular Pathology and Immunology of Porto, Porto, Portugal
Received 25 January 2002; accepted in revised form 7 January 2004
Key words: angiogenesis, breast cancer, genotypes, glutathione, GST, pharmacogenomic
Abstract The genes coding for separate isoforms of both the human glutathione-S-transferase class (GST) mu and class theta enzymes (GSTM1 and GSTT1) are polymorphic with a percentage of normal individuals exhibiting a homozygous deletion of the genes. An association between glutathione, proliferation and tumour angiogenesis has been observed. The aim of the present study was to analyse GST polymorphisms and to determine its correlation with the angiogenesis status of the tumoral tissue of patients with breast cancer. For each case, immunohistochemistry of tumour tissue and DNA genotyping by PCR on genomic DNA isolated from blood cells were performed. The mean intratumoral microvessel density (MVD index) was higher for the cases with GSTM1 wild-type genotype in comparison with the cases with the GSTM1-null genotype (89.6 ± 10.0 vs. 60.9 ± 6.7; P ¼ 0.022). This was even more evident for women with a breast cancer onset before the age of 35 (106.9 ± 11.9 vs. 61.8 ± 9.8; P¼0.011). Multivariate logistic regression analysis of GSTM1 and GSTT1 genotypes, histologic grade, axillary node status and age at diagnosis demonstrate the independent association between GSTM1 genotypes and angiogenesis and the association of GSTM1-wild type genotype with high MVD index (adjusted OR ¼ 5.98, 95% CI 1.28–28.10, P ¼ 0.023). A role of this enzyme in the hypoxia-induced metabolic pathway can be a connection for its association with angiogenesis. Further studies on the enzymatic components of the glutathione biosynthetic pathway in other cancers could define a pharmacogenomic profile of human neoplasia and help identify targets for the development of therapeutic strategies.
Introduction Breast cancer is the most common malignancy in women and is the major cause of death in the industrialised countries. Despite adequate primary treatment at the time of diagnosis, breast cancer remains a major cause of morbidity and early death in women: In 25% to 30% of axillary node-negative and 75% to 80% of axillary node-positive patients with breast cancer, the disease will recur within 10 years and these patients, subsequently die. The importance of angiogenesis in tumour progression and metastasis is now well recognised. The growth of a tumour beyond 1–2 mm3 requires the induction of new capillary blood vessels [1]. During tumour development, new vessel formation occurs as an early event, which precedes tumour invasion and is Correspondence to: Dr Rui Medeiros, Unit of Molecular Oncology, Instituto Portugueˆs de Oncologia, R. Dr. Ant. Bernardino Almeida, 4200-072 Porto, Portugal. Tel: +351-22-5502011; Fax: +351-225084001; E-mail:
[email protected]
a discrete component of the tumour phenotype, rather than resulting from tumour hypoxia as the growing mass outgrows its blood supply [2]. Tumour cells appear to stimulate angiogenesis in the neighbouring stroma before tumour invasion occurs [3]. Solid tumour growth beyond a volume of 2–3 mm3 cannot be sustained by diffusion and depends on the establishment of a vascular network for supplying nutrients and removing metabolic waste products [4]. An association between glutathione (GSH), proliferation and modulation of angiogenesis has been suggested. During angiogenesis, formerly differentiated human microvascular endothelial cells (HMECs) return to a proliferative growth state. Many fundamental questions regarding HMEC function, such as how HMECs adapt to changes in bioenergetic requirements upon return to proliferative growth, remained unanswered. Cultured HMECs maintain high rates of oxidative metabolism [5, 6]. Proliferative HMECs were shown to possess significantly higher (relative to both large vessel endothelial cells, and differentiated HMECs) levels of glutathione (GSH) [6].
54 The glutathione S-transferases (GSTs) comprise a supergene family of phase-2 detoxifying enzymes that catalyse a variety of reduced glutathione-dependent reactions with compounds containing an electrophilic centre. The GST substrates include products of oxidative metabolism. Polymorphisms in many xenobiotic metabolising enzymes occur leading to variation in the level of enzyme expression in vivo. The gene coding for separate isoforms of both the human glutathione-Stransferase class mu and class theta enzymes (GSTM1 and GSTT1) are polymorphic with a variable ethnic distribution [7, 8]. A significant percentage of normal individuals exhibit genetic polymorphism with a homozygous deletion of the genes. The aim of the present study is to analyse GSTM1 polymorphisms in the genomic DNA isolated from peripheral blood of patients with breast cancer and to determine its correlation with the angiogenesis status of the tumoral tissue.
Materials and methods Samples and patients We studied a series of 41 cases of women with breast cancer. The mean age at diagnosis for patients was 41.2 ± 14.6. The histological grade was evaluated according to modified Bloom and Richardson criteria, and two cases were grade 1, 23 were grade 2 and 16 cases were classified as grade 3. The presence of metastasis in the axillary node was detected in 20 cases, 20 cases were negative, and in one case, this information was not available. Immunohistochemistry of tumour tissue and DNA genotyping on DNA isolated from peripheral blood were performed according to the following protocols: To characterise the frequency of GSTM1 and GSTT1 polymorphisms in cancer-free women within the Portuguese healthy population, we analysed DNA isolated from 123 samples. Tissue analysis: immunohistochemistry Angiogenesis was evaluated by immunohistochemical staining of tumour vessels for factor VIII-related antigen according to a previous report [9]. The avidin– biotin–peroxidase complex method was used, and factor VIII-related antigen immunostaining was preceded by pepsin digestion, at room temperature, for 30 min. The primary antibody, F VIII diluted 1 : 60 (Dakopatts, Copenhagen, Denmark) was applied to the sections and incubated overnight at 4 C. This was followed by incubation with a 1 : 200 dilution of biotin-labelled antimouse secondary antibody (Dakopatts, Copenhagen, Denmark) for 30 min and ABC (Dakopatts) for 60 min. Careful rinses were done with PBS between each step of the procedure. The colour was developed with diaminobenzidine, and the sections were lightly counterstained with haematoxilin, dehydrated and mounted.
R. Medeiros et al. Negative controls for the immunostaining were carried out by omission of the primary antibody. As a positive control, sections, from previously detected highly angiogenic breast carcinomas were used. Evaluation of immunohistochemical data Any positive staining single cell or cluster of cells clearly separated from adjacent clusters and background, with or without lumen, was considered an individual vessel, as recommended in previous studies [9–13]. Areas of fibrosis, necrosis and inflammation, as well as vessels with a muscle wall, were excluded from the counting. Microvessels were counted in the three most vascularised areas in a 200 · field (0.74 mm2) by 4 observers simultaneously. Results were analysed for the average microvessel count (MVD index). Genotype analysis Blood samples were collected in EDTA-containing tubes. Genomic DNA was extracted from the white blood cell fraction from each study subject by a salting out procedure [14]. GSTM1 genotypes were determined by the use of a multiplex polymerase chain reaction (PCR) method adapted from a previously established protocol [7]. This technique does not distinguish between heterozygote and homozygote GSTM1 and GSTT1 genotype, but it conclusively identifies null genotypes (homozygous deletion). GSTM1 primers (sense G5-5¢ GAA CTC CCT GAA AAG CTA AAG C; antisense G6-5¢GTT GGG CTC AAA TAT ACG GTG G) anneal with the GSTM1 gene at 2401–2422 and 2598–2619, respectively, and B-globin primers (sense GH20-5¢ GAA GAG CCA AGG ACA GGT AC; antisense PC04-5¢CAA CTT CAT CCA CGT TCA CC) anneal with the B-globin gene at 61992–62011 and 62240–62259, respectively. PCR products were submitted to eletrophoresis on a 2% ethidium bromide-stained agarose gel. The absence of amplifiable 219-bp or 480bp (in the presence of 268-bp B-globin PCR product) indicates the GSTM1-null genotype or the GSTT1-null genotype, respectively (Figure 1). Statistical analysis Statistical differences between qualitative data were compared by the v2 test when appropriate. Quantitative data were expressed as mean ± SEM and compared by the analysis of variance. Further, we included a multivariate logistic regression model with the high MVD index as dependent variable. A high MVD index was defined as a mean microvessel count higher than 90 after the analysis of a histogram with all results. Early onset breast cancer was defined as cases of women with an onset of breast cancer before the age of 35 years. A level of two-tailed value P < 0.05 was considered significant.
GSTM1 genotypes and angiogenesis
55
MVD Index (microvessel density)
110
Figure 1. Electrophoresis of PCR products. Detection of PCR amplification of GSTT1 (480bp fragment), B-Globin (268bp fragment) and GSTM1 (219bp fragment) genes. Absence of the PCR product indicates the null genotype. Ethidium bromide-stained electrophoresed representative PCR products samples: 100bp ladder (lane M); GSTM1 wild type (lane 2); GSTM1 null (lanes 1, 3, 5); GSTT1 null genotype (lane 4).
100
p=0.022
90
80
70
60
50
wild-type
null
(n=20)
(n=21)
GSTM1 genotype Figure 2. Graphical representation of the comparison of the MVD index microvessel density (mean ± s.e.m) in breast cancer cases regarding the presence of a genotype GSTM1 null or GSTM1 wild type.
Results In this study, we analysed GSTM1 genotypes of 41 cases of women with breast cancer. We found GSTM1-null genotype in 51.2% (21 out of 41) and GSTM1-wild type in 48.8% of cases. The clinicopathological parameters of the analysed cases according to GSTM1 genotypes are shown in Table 1. There were no statistically significant differences between the group of patients with the GSTM1 wild-type genotype (GSTM1-wt) and the GSTM1-null genotype, concerning to age at diagnosis, tumour size, histological grade, axillary node status. A statistically significant association was found between angiogenesis (FVIII microvessel count: MVD index) and GSTM1 genotypes. We observed a lower mean microvessel count in the GSTM1 null genotype cases (Figure 2). The mean intratumoral microvessel count was 89.6 ± 10.0 for the cases with GSTM-wt genotype
and 60.9 ± 6.7 for the cases with the GSTM1-null genotype (Table 1). These differences were statistically significant (P¼0.022). When we evaluate our results regarding the early onset of breast cancer (under the age of 35), a stronger association was found between angiogenic index (MVD) and GSTM1 genotypes. The MVD index was 106.9 ± 11.9 for GSTM1-wt compared to 61.8 ± 9.8 for GSTM1-null cases (P¼0.011). We found no differences in the comparison of GSTT1 genotypes and the age at diagnosis, tumour size, histological grade, axillary node status and MVD index. Multivariate logistic regression analysis confirmed that GSTM1-wt is independently correlated with high MVD index among age, histologic grade, axillary node
Table 1. GSTM1 genotype as a predictor of increased angiogenic phenotype (microvessel density (MVD) index). Clinicopathological features
GSTM1 wt (n = 20)
GSTM1 null (n = 21)
P-values*
Age (years) Size (mm) Histologic gradea I/II III Axillary node status Negative Positive MVD index All breast cancer Early onset breast cancer
41.1 ± 3.0 3.5 ± 0.8
40.1 ± 2.9 3.5 ± 0.4
0.815 0.995
12 (60.0%) 8 (40.0%)
13 (61.9%) 8 (38.1)
0.900
11 (55.0%) 9 (45.0%)
9 (45.0%) 11 (55.0%)
0.527
89.6 ± 10.0 106.9 ± 11.9
60.9 ± 6.7 61.8 ± 9.8
0.022 0.011
Summary of the clinicopathological features of breast cancer cases according to GSTM1 genotypes. According to the modified Bloom and Richardson criteria. *Statistical data were obtained by chi-square test for all categorical variables (histologic grade and axillary node status) and analysis of variance for continuous variables (age, size, and MVD index). a
56
R. Medeiros et al.
Table 2. Multivariate analysis of GSTM1 and GSTT1 genotypes, histologic grade, axillary node status and age at diagnosis and its association with high MVD index in breast tumour tissue.
GSTM1 GSTT1 Age Histological grade Axillary node status
Adjusted odds ratio (95% CI)
P-value
5.98 1.77 0.93 0.83 0.74
0.023 0.583 0.054 0.825 0.715
(1.28–28.10) (0.22–13.96) (0.87–1.00) (0.17–4.00) (0.14–3.75)
MVD: microvessel density; CI: confidence interval.
status and GSTT1 (adjusted OR ¼ 5.98, 95% CI 1.28– 28.10, P ¼ 0.023) as shown in Table 2. Regarding GST polymorphisms in cancer-free women, we detected GSTM1-wt in 41.5% (51 out of 123), GSTM1-null in 58.5%, GSTT1-null in 32.7% and GSTT1-wt in 67.3% of samples.
Discussion Glutathione (GSH) and its associated enzymes GSTs, play a protective role within the cell. GSH is an important intracellular antioxidant and is the most abundant non-protein thiol present in the cell. The conjugation of GSH with a compound, catalysed by GST, renders the compound less toxic against cellular targets, and more hydrophilic and thus more readily excretable [15]. GSH and its related conjugating enzymes play a crucial role in cellular protection against chemical and oxidative stress. GSH is activated to form the oxidised GSH dimer (GSSG) by several compounds such as hydrogen peroxide in response to oxidative stress [16]. Two of the four cytosolic subfamilies of GSTs (a, l or M, p, h or T) exhibit genetic polymorphisms in their population distribution, with a large percentage of individuals displaying a homozygous deletion of the structural genes (GSTM1-null and GSTT1-null) [7, 8]. The GSTs comprise a supergene family of phase-2 detoxifying enzymes that catalyse a variety of reduced glutathione-dependent reactions with compounds containing an electrophilic centre. The GSTs substrates include products of oxidative stress. In this study, we analysed GSTM1 and GSTT1 genetic polymorphism in women with breast cancer, and examined association with tumour angiogenesis and other clinicopathological parameters. The frequency of the GSTM1-null genotype was 51.2% and the frequency of the GSTM1 wild type was 48.8%. Regarding GST, we found GSTT1-null in 14.6% (6 out of 41) of the cases. These results are consistent with the frequencies observed in the literature for GSTM1-null and GSTT1-null genotypes [7, 16]. Breast cancer cases presenting GSTM1-wt genotype had higher microvessel counts (89.6 ± 10.0) than cases presenting GSTM1-null geno-
type (60.9 ± 6.7), and this difference was significant (P ¼ 0.022). These findings were even more impressive in women with an early onset of breast cancer (Table 1). Multivariate analysis confirmed that GSTM1-wt genotype is independently correlated to high MVD index. An association between glutathione, proliferation and angiogenesis has been previously reported [5, 6], and it has been suggested that invasive breast carcinoma-induced angiogenesis is age-dependent and that intratumoral microvessels count is higher in younger women with breast cancer [9]. Results obtained by Weidner et al. [12, 13] using anti-endothelial-cell antibodies to identify tumour vasculature demonstrate that MVD was a useful prognostic marker for human breast cancer. Many other studies, using a similar methodology, have demonstrated the value of using microvessel density (MVD) as a prognostic indicator for a wide range of cancers [17– 21]. Recently, the role of angiogenesis in the biological behaviour of human breast cancer micrometastasis has been suggested [22]. Therefore, measurement of MVD facilitates assessments of disease stage and the likelihood of recurrence and helps in guiding treatment decisions [23]. There are few reports in which microvessel density did not correlate with the risk of metastasis or mortality. However, this can be explained by other factors produced by the tumours [25]. During angiogenesis, formerly differentiated human microvascular endothelial cells (HMECs) return to a proliferative growth state with high rates of oxidative metabolism. In comparison to both large vessel endothelial cells and differentiated HMECs, proliferative HMECs were shown to possess significantly higher levels of GSH [5]. Studying perinatal lung epithelium, Haddad et al. [25] have recently shown that GSH regulates the activity of transcription factors in response to changes in the availability of oxygen. Virtually all tumours are in a state of both hypoxia and oxidative stress, and these two primary stresses initiate gene expression programmes leading to angiogenesis [26]. One of the factors influenced by oxidative conditions is hypoxia-inducible factor-1 a (HIF-1 a). HIF-1 a is selectively stabilised during hypoxia, whereupon it is translocated to the nucleus inducing the transcription of several genes involved in glycolytic metabolism, cell cycle events and vascular proliferation. HIF-1 a is probably the best-characterised regulator of vascular endothelial growth factor (VEGF) expression, apparently inducing VEGF mRNA, by binding to hypoxiaresponse elements located in the promotor region of this gene. VEGF, on the contrary, is a well known proangiogenic factor, since it acts directly on endothelial cells stimulating their proliferation. By promoting the anti-oxidant activity of GSH, GSTs are likely to facilitate HIF-1 a activity, therefore stimulating angiogenesis. These data would explain why our cases that are carriers of GSTM1-wt presented a higher density of microvessels than the cases with a homozygous deletion of GSTM1 (GSTM1-null). Individuals that are carriers of the GSTM1-null genotype, lack the enzyme and have
GSTM1 genotypes and angiogenesis a reduced removal of chemical substrates. In characterising the frequency of GSTM1 and GSTT1 polymorphisms in cancer-free women within the Portuguese healthy population, our results are in concordance with other reported frequencies for Caucasian populations [27]. A recent report has demonstrated that polymorphisms in glutathione S-transferases (GSTM1 and GSTT1) may have an important impact on disease recurrence and survival [27]. An effect of GSTM1-null and GSTT1-null on survival after treatment for breast cancer was evident [27]. Our results are consistent with the role of GSTs in breast cancer and may also provide a good clue to explain the results obtained by Ambrosone et al. [27]. If cases presenting a GSTM1-null genotype are associated with a lower angiogenic phenotype (therefore a better prognosis), we should expect that GSTM1-null should be associated with a better prognosis as is demonstrated by the report by Ambrosone et al. [27] where women with the null genotype for GSTM1 have reduced hazard of death. The inhibition of angiogenesis can prevent diseases with excessive vessel growth such as cancer [28]. The anti-angiogenesis therapy represents new promising approaches to cancer treatment [29]. Several anti-angiogenic drugs are currently undergoing clinical trials for the treatment of cancer [30], and the successful translation of angiogenesis inhibitors to clinical application depends partly on the knowledge of the biology of angiogenesis [24]. The scientific relevance of our preliminary study could be enhanced by adding the determination of VEGF and related receptors as well as thrombospondin-1 (TSP-1) by immunohistochemical, immunoenzymatic or RT-PCR methods [24, 31–33]. The analysis of the relationship between GSTM1 and VEGF/VEGF receptors and TSP-1 may allow us to obtain more solid results with factors directly involved in angiogenesis. The development of new molecular markers may help to provide an accurate measurement of tumour angiogenesis [34–39]. Genetic polymorphisms are being evaluated for their role in multifactorial conditions, including angiogenesis [40]. The determination of GSTM1 polymorphisms by PCR may be achieved with a single blood sample and may be useful in the selection of groups of cancer patients at a higher risk of having a high angiogenic tumour, even before the beginning of the treatment. Individuals that are carriers of the GSTM1-null genotype, lack the enzyme and have a reduced removal of chemical substracts [7, 8, 27]. Other genetic polymorphisms and their association with the angiogenic phenotype are currently under evaluation [40–44].
Conclusions In conclusion, we have demonstrated an association between GSTM1 genotypes and angiogenesis in early onset breast cancer cases. These results are representa-
57 tive of a nice example of how an adaptive genetic response could influence the development and biologic behaviour of one neoplasia through the angiogenesis phenomenon. Further studies on the enzymatic components of the glutathione biosynthetic pathway in other cancers could define a pharmacogenomic profile of human neoplasia, and help identify targets for the development of therapeutic strategies. Acknowledgements The authors would like to thank to Liga Portuguesa Contra o Cancro-Centro do Porto (Portuguese League Against Cancer), for their support. We thank Dr Daniela Pinto for critical editorial comments. We gratefully acknowledge funding of this work by the Ministry of Health of Portugal (Ministe´rio da Sau´de CFICS-226/2001). References 1. Folkman J. What is the evidence that tumours are angiogenesis dependent? J Natl Cancer Inst 1990; 82: 4–6. 2. Marson LP, Miller WR, Dixon JM. Angiogenesis and Breast Cancer. Breast 1998; 7: 299–307. 3. Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis (review). Cell 1996; 86: 353–64. 4. Folkman J. Seminars in Medicine of The Beth Israel Hospital, Boston. Clinical Applications of research on angiogenesis. N Engl J Med 1995; 333: 1757–63. 5. Mallery SR, Lantry LE, Laufman HB et al. Modulation of human microvascular endothelial cell bioenergetic status and glutathione levels during proliferative and differentiated growth. J Cell Biochem 1993; 53: 360–72. 6. Schwartz JL, Shklar G. Glutathione inhibits experimental oral carcinogenesis, p53 expression, and angiogenesis. Nutr Cancer 1996; 26: 229–31. 7. Chen C, Liu Q, Relling MV. Simultaneous characterization of glutathione S-transferase M1 and T1 polymorphisms by polymerase chain reaction in American whites and blacks. Pharmacogenetics 1996; 6: 187–90. 8. Bell D, Taylor J, Paulson D et al. Genetic risk and carcinogenic exposure: A common inherited defect of the carcinogen-metabolism gene glutathione S-transferase M1 (GSTM1) that increase susceptibility to bladder cancer. J Natl Cancer Inst 1996; 85: 1159– 63. 9. Marinho A, Soares R, Ferro J et al. Angiogenesis in breast cancer is related to age but not to other prognostic parameters. Path Res Pract 1997; 193: 267–73. 10. Miliaras D, Kamas A, Kalekou H. Angiogenesis in invasive breast carcinoma: Is it associated with parameters of prognostic significance? Histopathology 1995; 26: 165–9. 11. Bosari S, Lee AK, Delellis RA et al. Microvessel quantitation and prognosis in invasive breast carcinoma. Hum Pathol 1992; 23: 755–61. 12. Weidner N, Folkman J, Pozza F et al. Tumor angiogenesis: A new significant and independent prognostic indicator in early-stage breast carcinoma. J Natl Cancer Inst 1992; 84: 1875–87. 13. Weidner N, Semple JP, Welche WR, Folkman J. Tumor angiogenesis and metastasis correlation in invasive breast carcinoma. N Engl J Med 1991; 324: 1–8. 14. Miller AS, Dykes DD, Polesky HF. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988; 16: 1215.
58 15. Brien ML, Tew KD. Glutathione and related enzymes in multidrug resistance. Eur J Cancer 1996; 32A: 967-78. 16. Medeiros R, Pereira D, Afonso N, Lopes C. Susceptibility to breast and ovarian cancer: The role of glutathione S-transferase polymorphism. Eur J Cancer 1999; 35: 737. 17. Weidner N. Tumoural vascularity as a prognostic factor in cancer patients: The evidence continues to grow. J Pathol 1998; 184: 119– 22. 18. Guidi AJ, Berry DA, Broadwater G et al. Association of angiogenesis in lymph node metastases with outcome of breast cancer. J Natl Cancer Inst 2000; 92: 486–92. 19. Kato T, Kameoka S, Kimura T et al. Angiogenesis and blood vessel invasion as prognostic indicators for node-negative breast cancer. Breast Cancer Res Treat 2001; 65: 203–15. 20. Kato T, Kameoka S, Kimura T et al. Angiogenesis as a predictor of long-term survival for 377 Japanese patients with breast cancer. Breast Cancer Res Treat 2001; 70: 65–74. 21. Lee AHS. The relationship of vasculariy of the primary tumour to lymph node metastases and to distant haematogenous spread in carcinoma of the breast. Histopathology 2002; 41: 366–386. 22. Klauber-DeMore N, Zee KJ, Linkov I et al. Biological behavior of human breast cancer micrometastases. Clin Cancer Res 2001; 7: 2434–9. 23. Hlatky L, Hahnfeldt P, Folkman J. Clinical application of antiangiogenic therapy: Microvessel density, what it does and doesn’t tell us. J Natl Cancer Inst 2002; 94: 883–93. 24. Kerbel R, Folkman J. Clinical translation of angiogenesis inhibitors. Nature Rev Cancer 2002; 2: 727–39 25. Haddad J, Olver R, Land S. Antioxidant/pro-oxidant equilibrium regulates HIF-1a activity and NF-kB redox sensitivity. J Biol Chem 2000; 275: 21130–9. 26. Bicknell R. An update of new targets for cancer treatment: Vessels and matrix. Ann Oncol 2002; 13 (S4): 25–8. 27. Ambrosone CB, Sweeney C, Coles BF et al. Polymorphism in glutathione S-transferases (GSTM1 and GSTT1) and survival after treatment for breast cancer. Cancer Res 2001; 61: 7130–5. 28. Griffioen AW, Molema G. Angiogenesis: Potentials for pharmacologic intervention in the treatment of cancer, cardiovascular diseases, and chronic inflammation. Pharmacol Rev 2000; 52: 237– 68. 29. Oehler MK, Bicknell R. The promise of anti-angiogenic cancer therapy. Br J Cancer 2000; 82: 749–52. 30. Fan TP, Jaggar R, Bicknell R. Controlling the vasculature: Angiogenesis, anti-angiogenesis and vascular targeting of gene therapy. Trends Pharmacol Sci 1995; 16: 57–66. 31. Nor JE, Christensen J, Mooney DJ, Polverini PJ. Vascular endothelial growth factor (VEGF)-mediated angiogenesis is asso-
R. Medeiros et al.
32.
33.
34. 35. 36. 37.
38. 39. 40.
41.
42.
43.
44.
45.
ciated with enhanced endothelial cell survival and induction of Bcl2 expression. Am J Pathol 1999; 154: 375–84. Nor JE, Mitra RS, Sutorik MM et al. Thrombospondin-1 induces endothelial cell apoptosis and inhibits angiogenesis by activating the caspase death pathway. J Vasc Res 2000; 37: 209– 18. Coradini D, Pellizzaro C, Veneroni S et al. Infiltrating ductal and lobular breast carcinomas are characterised by different interrelationships among markers related to angiogenesis and hormone dependence. Br J Cancer 2002; 87: 1105–11. Byrne GJ, Bundred NJ. Surrogate markers of tumoral angiogenesis. Int J Biol Markers 2000; 15: 334–9. Thompson WD. Tumour versus patient: Vascular and tumour survival versus prognosis. J Pathol 2001; 193: 425–6. Gould VE, Wagner BM. Angiogenesis: An expanding universe. Hum Pathol 2002; 33: 1061–3. Costa C, Soares R, Reis-Filho JS et al. Cyclo-oxygenase 2 expression is associated with angiogenesis and lymph node metastasis in human breast cancer. J Clin Pathol 2002; 55 (6): 429–34. Reis-Filho JS, Milanezi F, Schmitt FC. Maspin is expressed in the nuclei of breast myoepithelial cells. J Pathol 2002; 197 (2): 272–3. Reis-Filho JS, Schmitt FC. Lymphangiogenesis in tumors: What do we know? Microsc Res Tech 2003; 60 (2): 171–80. Balasubramanian SP, Brown NJ, Reed MWR. Role of genetic polymorphisms in tumour angiogenesis. Br J Cancer 2002; 87: 1057–65. Medeiros R, Morais A, Vasconcelos A et al. Outcome in prostate cancer: Association with endothelial nitric oxide synthase glu-asp298 polymorphism at exon 7. Clin Cancer Res 2002; 8: 3433–7. Medeiros R, Morais A, Vasconcelos A et al. Linkage between polymorphisms in the prostate specific antigen ARE1 gene region, prostate cancer risk, and circulating tumor cells. Prostate 2002; 53: 88–94. Vasconcelos A, Medeiros R, Veiga I et al. Analysis of estrogen receptor polymorphism in codon 325 by PCR-SSCP in breast cancer: Association with lymph node metastasis. Breast J 2002; 8: 226–9. Medeiros R, Morais A, Vasconcelos A et al. Endothelial nitric oxide synthase gene polymorphisms and the shedding of circulating tumour cells in the blood of prostate cancer patients. Cancer Lett 2003; 189: 85–90. Srivastava A, Laidler P, Davies RP et al. The prognostic significance of tumour vascularity in intermediate thickness (0.76–4.0 mm thick) skin melanoma – a quantitative histologic study. Am J Pathol 1988; 133: 419–23.
