Genetic Link to Lung Cancer Found on Chromosome 15

By Michael Smith
HOUSTON, April 3 -- A genetic variation that is strongly linked to the risk of developing lung cancer has been independently pinpointed by labs here, in Iceland, and in France.
Explain that all three studies found an increased risk of lung cancer for carriers of the genetic variation, but disagree on whether it causes cancer directly or by causing people to smoke more.
But the three groups disagree on whether the variant -- found on the long arm of chromosome 15 -- causes the disease directly or by increasing addiction to cigarettes.
All three studies show "remarkable consistency" in what region they pinpoint and how much increase in risk they find, said Christopher Amos, Ph.D., of M.D. Anderson Cancer Center, the first author of a paper published online by Nature Genetics.
But his group and researchers led by Paul Brennan, Ph.D., of the International Agency for Research on Cancer in Lyons, France, found at best a modest association with smoking behavior.
The key finding, Dr. Brennan said, is that the direct association with lung cancer development is "relatively large." In contrast, the conventional wisdom is that lung cancer is primarily environmentally caused, by smoking.
But Kari Stefansson, M.D., of Iceland's DeCODE Genetics, said he and colleagues found a clear link with increased smoking and difficulty quitting, which in turn contributed to smoking-related diseases like cancer and peripheral artery disease.
"The association is very solid, the P-values are vanishingly small, and the sample size that we are looking at was very large," Dr. Stefansson said.
His and Dr. Brennan's studies appear in the April 3 issue of Nature. All three studies use genome-wide association scanning, in which cases and controls are compared for differences called single nucleotide polymorphisms (SNPs).
In two of the studies, however, the cases were people with lung cancer, while the third -- by Dr. Stefansson and colleagues -- looked at smokers, in an attempt to discern genes that make people susceptible to smoking in the first place.
"There's still some discussion as whether this is an addictive effect or possibly an effect that works on a cellular level in driving forward lung cancer," Dr. Brennan said.
What they all agree on is that a region on chromosome 15 contains genetic variants that are relatively common in the population and that are linked, directly or indirectly, to lung cancer.
Having one copy of the variant increases the risk of disease by about 30% and having two copies increases the risk by about 80%, the researchers said.
Dr. Stefansson added that a smoker with one copy will tend to have one more cigarette a day than a person with no copies, while having two copies means an extra two cigarettes a day.
His study also calculated the risk of peripheral artery disease, and found that one copy of the variant increases the risk 20%, while two increases it 40%, compared with smokers with no copies of the variant.
The researchers said about 40% to 50% of the population have one copy of the variant, while about 10% to 16% have two.
All three studies found that a SNP dubbed rs1051730 appears to play a key role, but the two lung cancer studies also found a significant association with another, rs8034191. Both are in a region on chromosome 15 that contains several genes, including three that encode nicotinic acetylcholine receptor subunits.
In 13,945 smokers, Dr. Stefansson and colleagues found that the "T" version of SNP rs1051730 was more common in heavy smokers than in those less addicted. It also made it more difficult for smokers to quit, they found.
That finding led them to test the SNP for its relevance in lung cancer and peripheral artery disease directly, in two case-control cohorts totaling 62,208 volunteers.
The associations with lung cancer and peripheral artery disease were significant at P=1.5x10-8 and P=1.4x10-7, respectively, the researchers found.
Dr. Stefansson said it's possible that the variation works directly to cause the two very different diseases in some way, but he favors the idea that it first causes addiction, which then leads to disease.
The study led by Dr. Amos looked at 5,778 persons with lung cancer and 4,831 controls, who were matched by smoking status. They found that the two SNPs were significantly associated with lung cancer, at P=3.15x10-18 for rs1051730 and P=7.00x10-18 for rs8034191.
The study by Dr. Brennan and colleagues, involving 4,502 cases and 7,377 controls, found similar results -- significant associations for the two SNPs, at P=5x10-9 and P=9x10-10, respectively.
The studies cry out for further research to "clear the smoke clouding the differing conclusions of these papers" and establish a biological rationale for the "robust" associations, said Stephen Chanock, M.D., of the National Cancer Institute and David Hunter, MBBS, Sc.D., of Harvard.
But "it is at least reassuring that all three groups point to the same region on chromosome 15," they said in an accompanying comment article in Nature.
The study by Dr. Amos and colleagues was supported by the NIH, M.D. Anderson Cancer Center, the Flight Attendants Medical Research Institute, the Center for Inherited Disease Research, the Kelsey Research Foundation, Cancer Research UK, the National Cancer Research Network, HEAL, sanofi-aventis, and the Allan J. Lerner Fund. The researchers did not report any conflicts.
The study by Dr. Brennan and colleagues was supported by the French Institut National Du Cancer, the NCI, and the Ontario Institute for Cancer Research. The researchers did not report any conflicts.
The study by Dr. Stefansson and colleagues was supported by the National Institute of Drug Abuse and the European Commission. The researchers reported competing interests; several of the authors are employees or hold equity in deCODE Genetics.
Drs. Chanock and Hunter did not report any support or conflicts.
Primary source: NatureSource reference:Thorgeirsson TE, et al "A variant associated with nicotine dependence, lung cancer and peripheral arterial disease" Nature 2008; 452: 638-42. Additional source: NatureSource reference: Hung RJ, et al "A susceptibility locus for lung cancer maps to nicotinic acetylcholine receptor subunit genes on 15q25" Nature 2008; 452: 632-37. Additional source: NatureSource reference: Chanock SJ, Hunter DJ "When the smoke clears …" Nature 2008; 452: 537-38.