Skip to main content
Premium Trial:

Request an Annual Quote

Meta-Analysis Links Five New Variants to Brain Cancer Risk

NEW YORK (GenomeWeb) – An Institute of Cancer Research-led team has uncovered five new genetic variants linked to risk of developing brain cancer.

The team led by ICR's Richard Houlston conducted a meta-analysis that drew upon more than 4,000 cases and nearly 7,500 controls, and after genotyping an additional 1,500 cases and 1,700 controls, they linked five new variants to brain cancer risk, bringing the total number of risk variants to 12, as reported in Nature Communications today.

All together, these 12 risk variants explain about 12 percent of the variation in glioma risk. One of these new variants, the research team added, increased glioblastoma susceptibility by some 23 percent.

"One of the risk factors we identified is linked to quite a dramatically increased chance of developing glioblastoma, a particularly aggressive kind of brain cancer," Houlston said in a statement. "Our study sheds fresh light on the biology of gliomas, and could provide clues to why the disease develops, and how it could be treated or prevented."

Houlston and his colleagues pooled together data from four genome-wide association studies that were conducted in populations of European ancestry. After filtering, they had genotyping data on 4,147 cases and 7,435 controls. They further imputed unobserved genotypes in that cohort using a panel of 1,092 people from the 1,000 Genomes Project and 3,781 people from the UK10K project.

In this pooled dataset, they noted that seven of the known glioma risk loci had associations with glioma in a consistent direction of effect.

After filtering, Houlston and his colleagues selected 10 SNPs for replication genotyping in a separate cohort of 1,490 glioma cases and 1,723 controls. They then folded data from this new cohort into a larger meta-analysis of 5,637 cases and 9,158 controls — and some 8 million SNPs in all — through which they uncovered five new SNPs associated with glioma risk.

One signal, rs3851634 at 12q23.3 was specific for glioblastomasusceptibility, the researchers reported. They traced this signal to intron 12 of POLR3b, which encodes polymerase III, RNA, subunit b. POLR3B is involved in the transcription of small noncoding RNAs, short interspersed nuclear RNAs, and transfer RNAs, they noted.

The other signals were specific to non-glioblastoma glioma. One of these signals was located in intron 7 of the VTI1A, which encodes a protein involved in regulating insulin-stimulated trafficking of secretory vesicles and is important in neuronal development and in selectively maintaining spontaneous neurotransmitter release. VTI1A, the researchers added, has been implicated in other cancers like lung and colorectal cancer.

Another signal mapped to an intron of ZBTB16, another to a lincRNA, and the final signal to a substitution within ETFA, which is also in linkage disequilibrium with the genes ISL2, TYRO3P, and SCAPPER.

A functional analysis of these SNPs indicated that a few SNPs, such the ETFA p.Thr171Ile substitution-linked rs1801591, were located in highly conserved regions of the human genome. The signal tracing to VTI1A, meanwhile, had a low conservation score.

Through a pathway analysis, the researchers noted an enrichment of 14 pathways in glioma overall, eight in glioblastoma, and nine in non-glioblastoma tumors. For glioblastoma, implicated pathways included DNA repair and Notch signaling, while non-glioblastoma tumors were enriched for cell cycle progression and energy metabolism pathways.

Using mutation data from TCGA, the researchers also gauged the frequency of recurrent mutations in the genes marked by these new SNPs and found they are altered in 8 percent of low-grade gliomas and 3 percent of glioblastomas. This, they added, indicates a role for them in glioma tumorigenesis.

In addition, Houlston and his colleagues estimated the variance in glioma risk that can be linked to all 12 SNPs to be 26 percent, 27 percent, and 43 percent for all glioma, glioblastoma, and non-glioblastoma, respectively.

In particular, they said the SNP near the RNA polymerase III-encoding gene increased the risk of developing glioblastoma by 23 percent. They added, though, that the risk to any one person remains low as glioblastomas are rare.

"[I]dentifying genetic mistakes that increase the risk of glioma could be a vital first step towards developing new treatments against the disease," added Aine McCarthy from Cancer Research UK, one of the funders of the study. "Building on these findings and determining the exact role these genetic changes play in the development of glioma could help doctors personalize treatment for the disease in the future and save more lives."