Such testing could become even more valuable as new smoking-cessation drugs are developed. Varenicline, a new smoking-cessation drug that works via a different mechanism than bupropion, went on the market last year. "Ideally, we would like to be able to say, 'If you have these variants, you'll do better on bupropion, but if you have these variants, you'll do better on varenicline,'" says Rutter.

No gene tests are commercially available for smoking cessation, and scientists caution that individual findings must be replicated before such tests move into standard use. (The current study focused on smokers of European descent, so additional studies are needed to find out if the variation has a similar effect in those of African and Asian descent.)

Not everyone is sure that such tests would help smokers more than existing strategies, such as public-service announcements and government-sponsored booklets on quitting. "We need to develop ways to compare it to existing public-health measures," says Chris Carlsten, a scientist and pulmonary specialist at the University of British Columbia, in Vancouver. "Intuitively, it's reasonable to have some doubt that these new approaches can trump the power of the existing, proven, broad-based health approaches."

If gene tests for smoking cessation do prove useful, they are likelyto come up against some of the issues that have plagued personalized medicine in general: uncertainty over willingness on the part of doctors to prescribe the tests and on the part of smokers to take them, says Nancy Rigotti, thedirector of the Tobacco Research and Treatment Center at Massachusetts General Hospital (MGH), in Boston.

Doctors say that they are less likely to prescribe genomics tests than other tests to patients interested in quitting smoking, according to surveys conducted by Alexandra Shields, the director of the Harvard / MGH Center on Genomics, Vulnerable Populations and Health Disparities. Shields has just finished a new study surveying smokers in rural Alabama and inner-city Baltimore that suggests that they are apprehensive about undergoing genetic testing to help doctors tailor smoking-cessation treatments. "They don't understand genetics and are fearful of what it might mean for them and their family," says Shields. "But if we can make the resources available to educate physicians and patients, then I think the potential to increase the quit rate and reduce the public-health burden of smoking is substantial."