GENE POLYMORPHISMS AND GLIOMA RISK

July 9, 2009 — An international study of genetic variations involved in glioma, the most frequently occurring type of brain tumor, has identified 14 common gene variants in 5 genes associated with the occurrence of glioma.

Published online July 5 in Nature Genetics, the international investigation, led by researchers at the M.D. Anderson Cancer Center in Houston, Texas, analyzed data from 2 genomewide association studies from the United States and the United Kingdom. Together, the studies included 1878 patients with glioma and 3670 control individuals whose genotypes were obtained.

"[I]t's the first time we've had a large enough sample to understand the genetic risk factors related to glioma," said one of the senior authors, Melissa Bondy, PhD, professor, Department of Epidemiology, M.D. Anderson Cancer Center, University of Texas, in an M.D. Anderson press release.

The study contributes to understanding the possible cause of this common type of brain tumor, which constitutes approximately 80% of all primary brain malignancies. The only contributing factor identified previously was ionizing radiation. However, an increased familial risk had been recognized, leading to the US and UK genomewide association studies of glioma in populations with Western European heritage.

Beginning with the genotyping of more than a half-million single nucleotide polymorphisms (SNPs), the researchers found 34 SNPs demonstrating strong evidence of association with glioma (P <>−5). Replication studies were carried out in 3 case-control groups from French, German, and Swedish populations, totaling 2545 patients with glioma and 2953 control individuals.

Combined results from genomewide association and replication studies pointed to 14 SNPs in 5 genes that qualified for genomewide significance (P <>−7). Among these 14 SNPs, the 95% odds ratios (ORs) ranged from 1.18 to 1.36, with respective P values ranging from 1.07 × 10⁻⁸ to 2.34 × 10^-18. The 5 genes implicated were TERT, CCDC26, CDKN2A/B, PHLDB1, and RTEL1.

The strongest association was found for a SNP in CCDC26, a gene on chromosome 8 that influences retinoic acid, thus reducing telomerase activity and increasing programmed cell death. A SNP in TERT showed the next strongest association with glioma (P = 1.50 × 10⁻¹⁷). TERT is necessary for telomerase activity and maintaining telomeres.

The third strongest association (P = 7.24 × 10⁻¹⁵) was for a SNP in CDKN2B. The CDKN2A/B region activates p53, a well-known tumor suppressor. Interestingly, half of all brain tumors have lost at least 1 copy of this gene — a deficit that usually indicates poor prognosis. The fourth-ranked SNP (P = 2.52 × 10⁻¹²) was located in RTEL1, a gene on chromosome 20 that normally functions to stabilize the genome. The fifth strongest signal (P = 1.07 × 10⁻⁸) was demonstrated for a SNP in PHLDB1 — a gene previously implicated in neuroblastoma, but not in glioma.

The only interactive effects detected in the study were between SNPs in CDCC26, and analysis of other pairs showed that their effects on glioma were independent. As rank-ordered here by their strength of association with glioma, the 5 genes increased the risk for glioma by 36%, 27%, 24%, 28%, and 18%, respectively.

Because their effects are largely independent, the variants' contributions to glioma risk are additive: "Individuals with eight or more risk alleles have an approximately fourfold increase in glioma risk compared to those with a median number of risk alleles," the report stated. However, this may be a low estimate because the model weighted the SNPs equally, which is unlikely in reality.

At present, the authors caution against using the results of their study for individual screenings. Not only is more research needed on the function of the individual genes, but factors such as demographics, environmental factors, and lifestyle should be included when identifying at-risk individuals.

The investigators will begin an even more comprehensive study next year, enrolling 6000 patients with glioma and 6000 control individuals. "We will be able to look at all of the potential risk and protective factors we've identified in much smaller studies over the years," said Dr. Bondy, "such as exposure to ionizing radiation, allergies, infections, and use of nonsteroidal anti-inflammatory drugs."

Dr. Bondy is also optimistic about the future: "Our findings give reasons for hope for those who might be affected and an incentive for a more comprehensive investigation of what has been a mysterious disorder."

Nat Genet. Published online July 5, 2009.