Abstract 877: Genetic variation in podoplanin in relation to glioma risk and outcome

Abstract

Abstract Gliomas are highly invasive tumors that invariably disseminate throughout the brain. The mechanisms of glioma cell migration are still poorly understood. Human PDPN (eg. podoplanin) is a novel transmembrane glycoprotein and a lymphatic endothelial marker thought to play a role in wound healing and frequently upregulated in human neoplasia. Although its physiologic function is still unknown, the recent observation of restricted expression at the leading invasive edge of solid tumors in experimental systems has raised the possibility that PDPN plays a role in tumor cell migration and invasion. Microarray-based gene expression analysis has recently identified PDPN as a novel candidate gene in glioma pathogenesis (Clin Cancer Res. 2009 15:6541-50). In this study, we examined whether polymorphisms in the gene encoding PDPN (located at 1p36.21) are associated with glioma risk and outcome in a clinic-based case-control study encompassing 563 newly diagnosed (eg. nonrecurrent) glioma cases (including 324 WHO grade IV glioblastomas (GBM); 145 WHO grade II or III astrocytomas and 94 oligoastrocytomas and oligodendrogliomas) and 629 healthy controls with no history of brain cancer. DNA was isolated from saliva samples. A total of 22 candidate and haplotype tagging SNPs in PDPN including the 5’ and 3’ UTR region of PDPN were genotyped using the Illumina Goldengate assay. Logistic regression was used to estimate odds ratios (OR) and 95% confidence intervals (CI) for glioma risk in relation to individual SNPs adjusting for age and gender. Proportional hazards regression was used to estimate age- and gender-adjusted hazard ratios (HR) for GBM-related death (203 deaths; median follow up: 11.2 months) according to PDPN genotypes. No consistent associations were observed for any examined PDPN variant with glioma risk, overall, or according to glioma histological subtype. In contrast, 3 intronic SNPs (rs92406, rs2487643 and rs4391657) in moderately high linkage disequilibrium (r2: 0.57-0.96) residing in a haplotype block spanning 8kb were significantly associated with GBM mortality with the strongest signal observed in rs4391657 (NT_004610.19:g.617369T>C) (per variant “T” allele HR: 1.62; 95% CI: 1.28-2.05; P=0.00006; allele frequency: 0.20). The magnitude of association was similar for deaths occurring within 12 months (per variant allele HR: 1.51; 95% CI: 1.13-2.02; P=0.005) and greater than 12 months (per variant allele HR: 1.84; 95% CI: 1.21-2.80; P=0.004) following GBM diagnosis. The median survival times were 14.4, 11.4 and 8.9 months in those carrying 0, 1, or 2 at-risk “T” alleles in PDPN rs4391657 (log rank P value 0.0018), irrespective of therapy. To our knowledge this is the first report of an association of a PDPN genetic variant predicting survival times in malignancy. Further studies are needed to confirm this observation and to identify the putative causal variant. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 877. doi:10.1158/1538-7445.AM2011-877