Skip to main content

Non-smokers' lung cancer gene found

Monday 22 March 2010

“People who contract lung cancer and who have never smoked appear to have a genetic variation in common that predisposes them to the disease,” The Times reported. It said researchers have found that a gene called GPC5 is more common in non-smokers with lung cancer than non-smokers without the disease.

The news story reports a genome-wide association study that found an association between a particular gene variant (on chromosome 13q31.3) and the functioning of a gene called GPC5. A reduced expression (activity) of this gene appears to contribute to the development of lung cancer in people who have never smoked.

This well-conducted study increases our understanding of why people who never smoke may develop lung cancer. It has some shortcomings, such as small sample sizes (because of the difficulty of finding enough people with lung cancer who have never smoked), but it is a sound starting point for research into this area. Screening programmes or drugs to prevent or treat the condition based on these findings are a long way off. Smoking remains the most important known risk factor for lung cancer, causing 90% of all cases.

Where did the story come from?

The research was directed by Dr Xifeng Wu from the University of Texas MD Anderson Cancer Center and Dr Ping Yang from the Mayo Clinic College of Medicine, with colleagues from other academic and medical institutions across the USA. The research was funded by the US National Institutes of Health and the Mayo Foundation along with funding to individual researchers. The research paper was published in the peer-reviewed medical journal The Lancet.

This was well-conducted genetic research that sought to replicate its initial findings from one population in several different populations. The authors declare no conflicts of interest.

What kind of research was this?

This was a genome-wide association study of an initial population of 754 people who had never smoked or ‘never smokers’, 377 people with lung cancer and 377 matched healthy controls. The researchers examined the DNA of these individuals for variants (single nucleotide pairs or SNPs) that were more or less common in the cases and that could therefore be associated with the risk of lung cancer. Further analysis was carried out on those variants that were most strongly associated with disease.

What did the research involve?

Genome-wide association studies look at how often particular DNA variants occur in people with a disease compared to how often they occur in people without the disease. They are a form of case-control study. This study investigated whether any particular genetic variants were associated with lung cancer in people who have never smoked. It is widely known that smoking is the single greatest risk factor for lung cancer and nine out of ten cases are caused by smoking. However, some people who never smoke do get the disease.

The researchers initially analysed the DNA from a population of 377 never smokers with lung cancer and 377 healthy adults, who were matched in terms of their age, gender and ethnic origin. Never smokers were defined as people who had smoked fewer than 100 cigarettes in their lifetime.

This analysis resulted in 44 ‘top’ single nucleotide pairs (SNPs) that were associated with lung cancer. These SNPs were assessed further by replicating the DNA analyses in another two independent samples, the first consisting of 328 lung cancer cases and 407 controls, and the second of 92 cases and 161 controls. The top two SNPs (i.e. the variants most associated with disease) from these two samples were further replicated in another independent sample of 91 cases and 439 controls.

In their analysis to determine how strongly the gene variants were associated with disease risk, the researchers took into account some factors that might influence risk (confounders). These included exposure to second-hand smoke in adulthood, childhood or over the lifetime; family history of lung and other cancers; and the occurrence of chronic obstructive lung disease.

A separate analysis looked at the genes that were in close proximity to the SNPs that had been identified. This allowed the researchers to determine whether the SNPs had some functional effect, for example, whether they were reducing or increasing the expression of some genes.

What were the basic results?

The first analysis found 44 top SNPs that might alter the risk of lung cancer in people who never smoke. One particular SNP was associated with the disease in four independent samples. Analysis showed that the presence of this variant significantly increased the risk of a never smoker having lung cancer by 1.46 times (odds ratio 1.46, 95% confidence interval 1.26 to 1.70).

Further analysis showed a strong inverse link between this variant and levels of a gene called GPC5, which occurs in normal lung tissue. This means that the presence of the variant was associated with lower levels of the gene’s activity. Further experiments indicated that there was double the concentration of GPC5 in normal lung tissue than in tissue from lungs with adenocarcinoma (a type of lung cancer).

How did the researchers interpret the results?

The researchers say that genetic variants at a particular region (rs2352028) on chromosome 13q31.3, in a gene called GPC5, altered the expression of this gene in people who had never smoked, and increased their susceptibility to lung cancer.


This study reported reliable, scientific research and will be of interest to the genetic research community. The study has identified a variant in the DNA sequence linked with the expression of a particular gene and which affects the risk of lung cancer in people who have never smoked.

By ruling out smoking, the single greatest risk factor for lung cancer, the researchers have increased our understanding of the complex link between genetics and lung cancer. They further strengthened their study by attempting to take into account the participants’ exposure to second-hand smoke, although accurately quantifying this exposure would be difficult.

There are some important factors to consider when interpreting the results of this study. The researchers acknowledge that they had difficulties recruiting ‘pure’ never smokers, and particularly people with lung cancer who had never smoked, given that 90% of all lung cancers are caused by smoking. This resulted in a small sample of people to analyse. Though the researchers succeeded in identifying new genetic variants associated with lung cancer, they say that the small size may have meant a “lack of power” to detect other significant associations. In addition, they say they were not able to adjust fully for confounders, as some of the studies analysed did not have complete data to make this possible.

The genetics of lung cancer are likely to be complex, and further studies are needed to determine whether there are other associations and how genetics and environment interact and contribute to risk. More information is needed before genetic screening programmes for susceptibility to lung cancer become a reality.

Smoking remains the most important known risk factor for lung cancer, and stopping smoking is the best way of reducing your risk.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Lung cancer gene found in non-smokers.

BBC News, 22 March 2010

Scientists identify common gene linked to lung cancer in non-smokers.

The Times, 22 March 2010

Gene can lead to lung cancer in non-smokers.

The Daily Telegraph, 22 March 2010

Links to the science

Li Y, Sheu C-C, Ye Y, et al.

Genetic variants and risk of lung cancer in never smokers: a genome-wide association study.

The Lancet Oncology 2010; Early Online Publication, March 22

Further reading

Manser R, Irving LB, Stone C, Byrnes G, Abramson MJ, Campbell D.

Screening for lung cancer.

Cochrane Database of Systematic Reviews 2004, Issue 1

There are a number of Cochrane reviews relating to treatments for lung cancer, including: Caraballoso M, Sacristan M, Serra C, Bonfill Cosp X.

Drugs for preventing lung cancer in healthy people.

Cochrane Database of Systematic Reviews 2003, Issue 2