And ever since the 1960s, the U.S. government has required inert placebos as a comparison in clinical trials to determine whether a potential drug is effective or not. Since the act of taking a pill can make many patients feel better, a drug must perform better than a placebo to be allowed on the market.
But as I found while writing a new book on the suggestibility of the human mind, this so-called placebo control has become increasingly difficult to manage. For certain ailments, so many people respond so strongly to placebos that it’s impossible to tell if the drug being tested is working or not. And this can be disastrous for people desperate for new therapies.
Take pain. In 2011, Clinicaltrials.gov registered 4,152 clinical trials to investigate new pain treatments. Yet over the next few years, the Food and Drug Administration approved just five new treatments.
“We’ve all been there. If you are in this field, you know that weekly there will be a clinical trial that doesn’t work and one of the interpretations was that the placebo effect was too high,” says Neal Farber, CEO of Neurohealing Pharmaceuticals, a company that, among other things, tries to resuscitate drugs that failed previous clinical trials.
But there may be a way to run drug trials that could bypass the strongest placebo effects. People may differ in how susceptible they are to the placebo effect. If we could detect who would respond to placebos and exclude them from a trial, it would be much easier to tell if a therapy worked.
A former biotech executive believes he may have the key to do just that. Gunther Winkler first developed a dislike of the placebo effect in the late 1990s. As head of clinical operations for the pharmaceutical giant Biogen, it was his job to help choose which new drugs to invest in and shepherd them through the trials that would prove their effectiveness and clear them with the Food and Drug Administration for approval.
The company had developed a drug, Amevive, to treat psoriasis, which causes irritating rashes on patients, especially in times of high stress. The Phase 2 trials — the “proof of concept” tests that show a drug is safe and effective in small groups of patients — had gone well, and the company was optimistic about its prospects.
But the next step would be the hardest — the drug would need to be tested against a placebo control in a larger Phase 3 trial that every drug must surmount to be sold in the United States. A Phase 3 trial generally compares a drug to a placebo control, with those administering it ignorant of whether they are giving patients the placebo or the drug. Winkler’s research told him that between 5 and 40 percent of psoriasis patients feel their symptoms are alleviated when taking placebo pills. The only way to prove that Amevive was more effective than placebos would be to recruit hundreds of subjects and hope they were enough to provide the statistical power to tease out the effect of the drug over placebos.
In the end, the placebo response was well below 40 percent and the drug passed easily, but at a much higher cost than it could have. (He estimates each subject in a study like this costs a company about $30,000.)
“If you can halve the costs, let’s say, suddenly you can afford running twice as many drug candidates,” he says.
A few years ago, he stumbled across an academic paper on a gene called COMT that codes for an enzyme in our brain with the same name. That enzyme helps control levels of the neurotransmitter dopamine. Few brain chemicals are as important to our health and well-being as dopamine, which influences dozens of functions from digestion to bodily movement to how we process rewards. And COMT regulates dopamine.
Think of COMT as a vacuum cleaner that cleans up dopamine (in reality, by partly neutralizing it). Those people with shiny, powerful vacuum cleaners have very little excess dopamine in their brains while those with cheaper, slower machines have plenty of it scattered about.
The team found that those with slick vacuums (about a quarter of the human population) respond poorly to placebos while those with hinky vacuums (another quarter of the population) respond extremely well and those with a mix (half the population) are in the middle.
A light went off in Winkler’s head. For decades, he realized, the placebo effect had been a sort of fuzzy phenomenon.
“But now we have an understanding that actually it is mechanistically based. And that is the real breakthrough,” says Winkler. “The mechanistic part of it turned out to be genetically measurable. And that creates this huge opportunity.”
Winkler quickly set up a company, called Biometheus, which would offer to screen out people with cut-rate vacuums from pharmaceutical trials — the people most likely to respond to the placebo. He then got access to large data sets from drug trials and retrospectively screened out these theoretically placebo-prone subjects and found that indeed companies could have saved millions by excluding them.
“It’s a really interesting technology. I think if it works, it will be transformational,” says Jeanie Chu, 20-year veteran of the pharma industry involved in drug development. “This would just change the way we think about proof-of-concept studies.”
Former biotech scientist Kathryn Hall, first author on the COMT study, says Winkler is on an exciting track. After decades of placebos being an annoying curiosity, she says, it’s hard for some pharma researchers to wrap their minds around this concept.
“People are not thinking this way. They’re not ready yet. Placebo is more of a kind of novelty — a gee whiz kind of thing,” she says.
Since that first paper, she and Joseph Loscalzo of the Brigham and Women’s Hospital in Boston have helped identify more than 30 genes related to placebo effects for various conditions. Loscalzo works in a branch of science called network medicine that looks not just at genes but at relationships between the molecules they code for. If a given gene is like a large company, network medicine examines what its employees do in their off hours and what other workers they associate with.
Interestingly, Loscalzo says, the genes related to placebos seem to be connected to each other through their “employee” protein associations. Almost as if all placebos — be they for pain, irritable bowels or depression — follow a similar set of rules.
“They really do physically stick together. The pathways do, if not the genes,” he says.
Not everyone is so buoyant about Winkler’s approach to placebo screening. Tor Wager, a neuroscientist at the University of Colorado, was among the first to use brain imaging to show the placebo effect in action and has published widely on the phenomenon. He’s in favor of putting basic research to work but worries that Winkler might be getting ahead of the science.
“Complex phenomena like placebo effects are much more complicated than simple traits,” he says.
TransCelerate is a joint organization run by 12 pharmaceutical companies to solve shared problems. Ed Bowen, a Pfizer executive who works on the implications of placebos on drug development, says that the problem of high placebo response is exactly the kind of thing that the organization was built to address and all their member have put in resources and experts to understand it, focusing mostly on statistical or design solutions rather than genetic ones. From his perspective, a truly usable screen is still a few years off.
One of the biggest problems, he says, is that drug trials limited to people with, say, specific COMT genes might mean that the drug won’t be approved for the population at large. If you happen to be a person with slow vacuum cleaners, you may find that many new drugs are not certified for you simply because some other people who share some of your genes respond too readily to placebo pills. Ironically, you would also be the most likely to respond to the drug — and any drug — in the first place.
But if COMT can reduce the cost of trials, Winkler says, it can be a powerful tool to bringing hundreds of new drugs to the market. The tool can always get sharper in the future, but for now, he says, it works fine. He says he didn’t create Biometheus to get rich (he did pretty well for himself with a long career in biotech) but to revolutionize biotechnology across multiple diseases. Asked if this is the way he envisioned his retirement, he smiles.
“I truly feel that this will change how the industry will develop drugs. And hopefully pass the advantages of that through to society,” he says.
Erik Vance will be reading from his new book “Suggestible You” at Politics and Prose on Dec. 3. This story was supported in part by the Pulitzer Center on Crisis Reporting.