Abstract 3388: Genetic Prediction of VEGF-A Plasma Levels in Cancer Patients
Federico Innocenti,Chen Jiang,Alexander Sibley,Amy Etheridge,Yoichi Furukawa,Michiaki Kubo,Hedy L. Kindler,Alan P. Venook,Herber I. Hurwitz,Andrew B. Nixon,Kouros Owzar
DOI: https://doi.org/10.1158/1538-7445.am2016-3388
IF: 11.2
2016-01-01
Cancer Research
Abstract:Abstract ABSTRACT Background: Angiogenesis is an essential event in tumor growth, progression and metastasis and is strongly regulated by multiple VEGF ligands and receptors. We sought to discover genetic variants that could predict levels of circulating angiogenic proteins in cancer patients prior to receiving therapy. Methods: EDTA plasma was collected at baseline (before treatment) in 216 treatment naïve pancreatic cancer patients (CALGB 80303, discovery) and 149 treatment naïve colorectal cancer patients (CALGB 80203, validation). Thirty-one angiogenic factors were measured by a multiplexed ELISA assay. Genetic variants associated with levels of each of the 31 proteins were selected from a genome-wide genotyping of 484,523 common variants in CALGB 80303. Using a cut off of p<10−8 for association, variants were selected from CALGB 80303, genotyped in CALGB 80203 and associations tested with the same protein levels. The Jonckheere-Terpstra test was used. Additionally, mRNA was extracted from formalin-fixed, paraffin-embedded primary tumors from patients in CALGB 80203 and tissue levels of VEGF-A analyzed by RT-PCR. Results: In CALGB 80303, three genetic variants passed the p<10−8 cut off for significance: rs2284284 for MCP1, rs7504372 for VEGFC, and rs7767396 for VEFG-A levels. Of these, only rs7767396 (A>G) for VEFG-A levels was validated in CALGB 80203 (p = 1.23×10−5). Patients with the AA genotype exhibited 2.2-fold higher VEGF-A levels than AG patients, who had 1.2-fold higher levels than GG patients. rs7767396 was not associated with VEGF-A mRNA levels from the primary tumors of patients in CALGB 80203 (p>0.05). rs7767396 is a common variant (frequency of 49% in Europeans), is located about 150 Kb 3’ to the VEGFA gene, and is predicted by HaploReg to disrupt the binding motifs of two transcription factors, NF-AT1 and ZBRK1. Conclusions: A common genetic variant predicts the levels of circulating VEGF-A in cancer patients. A similar effect has been also shown in non-cancerous individuals (Debette S et al., Circ Res, 2011). Due to the central role of VEFG-A in the pathophysiology of many conditions, genetic testing could predict patients who have high versus low levels, potentially helping to guide the use of anti-angiogenesis therapies. Citation Format: Federico Innocenti, Chen Jiang, Alexander Sibley, Amy Etheridge, Yoichi Furukawa, Michiaki Kubo, Hedy L. Kindler, Alan P. Venook, Herber I. Hurwitz, Andrew B. Nixon, Kouros Owzar. Genetic prediction of VEGF-A plasma levels in cancer patients. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3388.