Updated

Some people may be genetically programmed to feel better after taking placebo pills, while others may only heal with real drugs, suggests a new review of existing research.

The study team looked at evidence that some people’s genes may make them more prone to experience the placebo effect. If true, and a genetic profile of such “placebo responders” could be identified, it might change the way medications are prescribed and the way clinical trials are designed, the authors say.

"Our findings strongly support the idea that genetic signatures for placebo responses exist, but our findings are preliminary," said lead author Kathryn Hall, a researcher at Harvard Medical School and Beth Israel Deaconess Medical Center in Boston. "Something is definitely there, but more needs to be known."

In the journal Trends in Molecular Biology, Hall and her colleagues note that past research suggests people’s response to placebos may be influenced by the way certain signaling molecules in the brain and body respond to pain and reward, or the expectation of those experiences.

Among the important signaling systems identified are those involved in responses to opiate drugs and others related to mood, like the serotonin and dopamine systems. Differences between people in how these systems function can be linked to variations in their genes.

"We are still in the early stages of using genetic screening for the placebo response in clinical trials, and as our knowledge of personalized medicine evolves it makes sense that we also consider how the placebo effect fits into treatment response," Hall told Reuters Health in an email.

Evidence that the placebo effect is real was first publicized in 1978, after an experiment done on patients having molar teeth extracted found that some people experienced pain relief with a placebo pill instead of a narcotic painkiller.

More recently, researchers looked at the gene COMT, which regulates the amount of dopamine in the brain and is connected to feelings of pain and pleasure.

In one experiment, researchers offered three types of placebo treatment to patients with irritable bowel syndrome. Patients were either put on a waiting list for treatment, given fake acupuncture by an unfriendly provider, or given fake acupuncture by an affectionate provider.

Then they tested patients to see which version of COMT they had. People with a high-dopamine variant of the gene were the ones most likely to report that the fake treatment had actually relieved their pain.

While the work is intriguing, "finding a few correlations between gene mutations and placebo responses to specific drugs does not nail down the genetic basis of the placebo response," said Dr. Tim Lahey, chair of clinical ethics at Dartmouth's Geisel School of Medicine in Hanover, New Hampshire.

"Another unanswered question is whether the genetic traits that drive the placebo response to one drug are the same as would drive the placebo response to another drug – they could be entirely different," Lahey, who wasn't involved in the research review, said by email.

There are also ethical challenges in using a potential genetic profile to prescribe treatments, Hall and colleagues note. It might make sense for physicians to do genetic testing before prescribing some drugs, but there’s a question of what might happen if patients refuse this screening.

The study authors also raise the possibility of changing the gold standard of clinical trials for new medicines - currently done with one group receiving placebo and another group getting the experimental drug - to add an extra group that receives no treatment. This, Hall says, might help measure the magnitude of any placebo response and help to better gauge the effectiveness of the drug.

"Most people tend to think first and foremost of a placebo not evoking a biological response but a psychological response," said Arthur Caplan, an ethics researcher at New York University Langone Medical Center. "If there is indeed a biological element and you can actually find a genetic basis for it, you would absolutely be causing a revolution in clinical trial design."

Even if it gets more accurate trial results to include an extra group that receives no treatment, it might be difficult to get people to volunteer for this type of experiment, added Caplan, who wasn't involved in the study. "Americans are completely unwilling to believe that doing nothing is better than doing something."