Influence of VEGF-A gene variation and protein levels in breast cancer susceptibility and severity Sabapathy P. Balasubramanian 1 , Angelo Cox 2 , Simon S. Cross 3 , Sue E. Higham 1 , Nicola J. Brown 1 and Malcolm W. Reed 1 * 1 Academic Surgical Oncology Unit, University of Sheffield, Royal Hallamshire Hospital, Sheffield S10 2JF, England 2 Institute for Cancer Studies, University of Sheffield Medical School, Beech Hill Road, Sheffield S102RX England 3 Academic Unit of Pathology, University of Sheffield, Royal Hallamshire Hospital, Sheffield S10 2JF, England Vascular endothelial growth factor-A (VEGF-A) plays an impor- tant role in tumour angiogenesis and cancer progression. VEGF gene variation may influence VEGF levels and therefore cancer susceptibility and progression. We studied the role of VEGF single nucleotide polymorphisms and haplotypes in breast cancer suscep- tibility and severity. We also studied the relationships of VEGF SNPs with circulating VEGF levels in healthy volunteers and pro- tein expression in breast cancers. Single nucleotide polymor- phisms (SNPs) in the regulatory regions of the VEGF gene were genotyped by high throughput methods in 500 breast cancer cases and 500 appropriate controls. Haplotype frequencies were inferred using methods based on the Expectation Maximisation algorithm. The effect of VEGF genotypes on serum and plasma VEGF levels were studied in another cohort of healthy individu- als. A semi-quantitative assessment of VEGF protein expression on tissue micro arrays (TMA) constructed from 300 breast can- cer samples was performed and compared with VEGF genotypes and with histopathological parameters and survival in breast can- cer. The 2460T/1405C/27C/936C haplotype in the VEGF gene was found to be associated with decreased breast cancer risk (p 5 0.029). The 27C>T polymorphism may influence overall breast cancer survival (p 5 0.027). Individual polymorphisms however did not affect breast cancer susceptibility. There was no associa- tion between the individual polymorphisms and circulating VEGF levels in healthy volunteers and VEGF expression on the breast cancer micro array. VEGF expression in breast cancers was how- ever associated with high grade (p 5 0.002) and ER negative tumours (p 5 0.03). ' 2007 Wiley-Liss, Inc. Key words: VEGF; breast cancer; polymorphisms; haplotypes Vascular endothelial growth factor-A (VEGF-A) belongs to a ligand family comprising 6 related proteins and plays an important role in tumour angiogenesis. 1 VEGF is strongly expressed in human solid tumours 2 and plays an important role in their patho- genesis. 3 There is evidence that tumours can promote VEGF pro- duction in adjacent stroma, which in turn predisposes to further tumour growth. 4 The prognostic value of VEGF in breast cancer has been shown by immuno-histochemical analysis and ELISA in a number of studies. 5 Cytosolic VEGF concentrations in breast tumours have been shown to be associated with tumour size and predict both relapse-free and overall survival of node-negative breast cancer patients. 6 The VEGF gene contains eight exons and seven introns over a 14-kb region 7 and has been mapped to chromosome 6p12 by fluo- rescence in situ hybridization. 8 Several polymorphisms have been described within the promoter and 5 0 UTR of the VEGF gene, 9,10 some of which have been associated with VEGF protein produc- tion 10 and circulating levels. 11 Some recent studies have not identified any association between breast cancer and some VEGF polymorphisms, which include 21154 A/G, 12,13 22578C/A, 2634G/C and 1936C/T. 13 In contrast however, other studies have demonstrated that the 2460 C/T and 1405 G/C polymor- phisms are associated with breast cancer survival 14 and 2634G/C, 22578C/A and 21154 A/G are associated with breast cancer prognosis. 12,13 The current study aims to evaluate the role of regu- latory VEGF polymorphisms (2460C > T, 1405G > C, 27C > T in the 5 0 UTR and 936C > T in 3 0 UTR) and haplotypes in influ- encing breast cancer susceptibility and severity, blood VEGF lev- els and VEGF protein expression in breast cancer. Material and methods Case-control study We 15,16 have previously described the design and methodology of this case control study. Briefly, recruitment commenced in November 1998 and is ongoing. The cases include women diag- nosed with breast cancer and being followed up at the Royal Hal- lamshire Hospital in Sheffield and Rotherham District General Hospital and controls recruited from women attending the Sheffield Breast Screening Service. The study was restricted to white Cauca- sians, as there were insufficient individuals from other ethnic groups for meaningful analysis. The South Sheffield Research Ethics Committee approved the study [Ref. no. SS98/137] and informed written consent was obtained from all subjects. Demo- graphic, environmental risk factors and family history data were recorded for all breast cancer cases and mammography screening controls, using a standard questionnaire. Pathological data (including tumour grade, lymph node status and presence of vascu- lar invasion) were obtained from medical records and validated by an experienced histopathologist (SSC). Data on disease recurrence and overall survival were obtained from the hospital records and the Trent Cancer Registry. The data on the recurrence and survival was limited to patients with Ôinvasive breast cancerÕ in whom sur- gery was the first line of treatment and where records were avail- able from either the hospital records or the Trent Cancer Registry. This group of patients had a median (inter-quartile range) follow up of 72 (51 and 99) months. The data was entered by trained per- sonnel and stored in a Microsoft Access database which was main- tained by a dedicated database administrator. The data was vali- dated for all the records (by SPB and database manager). Identification of potential candidate SNPs The Medline literature and the following SNP databases were searched for previously identified polymorphisms in the VEGF-A gene: dbSNP, Ensembl SNP, HGBASE, SNP consortium, HGMD and GDB. Of the 33 potential and confirmed SNPs in the regula- tory regions of the VEGF gene, we identified 14 that were located at potential transcription factor binding sites using a web tool, Matinspector Professional. 17 Sequences in the 5 0 and 3 0 ends of Grant sponsor: Yorkshire Cancer Research, Sheffield Hospitals Charita- ble Trust, The Sheffield Breast Cancer Support Group. *Correspondence to: Academic Surgical Oncology Unit, University of Sheffield, K Floor, Royal Hallamshire Hospital, Glossop Road, Sheffield S10 2JF, England. Fax: 10114-271-3314. E-mail: m.w.reed@sheffield.ac.uk Received 1 November 2006; Accepted after revision 1 March 2007 DOI 10.1002/ijc.22772 Published online 30 April 2007 in Wiley InterScience (www.interscience. wiley.com). Abbreviations: A, adenine; C, cytosine; ELISA, enzyme linked immuno- Sorbant assay; ER, oestrogen receptor; G, guanine; IQR, inter-qaurtile range; LD, linkage disequilibrium; PCR, polymerase chain reaction; PR, progesterone receptor; SNP, single nucleotide polymorphism; T, thymi- dine; TMA, tissue micro array; UTR, un-translated region; VEGF, vascular endothelial growth factor. Int. J. Cancer: 121, 1009–1016 (2007) ' 2007 Wiley-Liss, Inc. Publication of the International Union Against Cancer