The Placebo Effect: What's Your Gut Feeling?

The scene has played itself out in thousands of homes across America for years: A young child falls and bumps her knee. She runs to her mother, tears cascading down her cheeks, and begs for a bandage. The mother complies, despite the absence of even a minute amount of blood, and smoothes the bandage over the "injury." Almost instantly, the child feels better. Her tears cease, her cries fade to a whimper, and before long she is back at play. And the cartoon characters on the bandage didn't hurt either, seeming to speed relief to the area.

It's called the placebo effect, and it's not something to be dismissed. The word placebo has its roots in Latin, meaning "I shall please." By definition, placebos are supposed to be inactive; in most cases, they're sugar pills or solutions containing no active drugs. But sometimes the mere expectation of relief from a placebo can be enough to make you feel better.

Placebos are key players in the assessment of new therapies in clinical trials. "In order to prove a therapy is effective, you have to compare it to something that is already proven to work or to something that should have no effect," explains William Tremaine, M.D., Professor of Medicine at the Mayo Clinic in Rochester, Minnesota. "When we use placebos, we also don't need to enroll as many patients in a clinical trial as we would if we were comparing a new drug to an established, effective therapy." This means it takes less time to complete a study, and fewer patients are exposed to a new tratment that may not be effective. Patients participating in such studies are not told what they are receiving—the active drug or the placebo—until the study is over.

In theory, then, if a new drug is effective against a disease, we should see more patients in the treatment group respond to the agent than in the placebo group. But that's not always the case in clinical trials of patients with inflammatory bowel disease (IBD), where it's not uncommon for as many as 30 to 40 percent of patients in the placebo group to report that they feel better. If the placebo response is similar to that of the patients who received the study drug, researchers might conclude that the drug was not effective— when, in fact, it may have actually helped some of the patients. "Situations like this may confuse doctors who do research on new therapies for IBD," adds Dr. Tremaine.

How can we explain this phenomenon? First, there is most likely a mind-body connection. The very nature of the way IBD behaves may also create the illusion of a placebo response. One thing we know for sure: the knowledge that investigators are gleaning regarding the placebo response in IBD is changing the design of clinical trials and clarifying new directions for future research.


The Power of Emotion

Explanations for the placebo effect have been debated in the medical literature for decades. In a 1955 article called "The Powerful Placebo," published in the Journal of the American Medical Association, the author wrote, "It is evident that placebos have a high degree of therapeutic effectiveness in treating subjective responses, decided improvement, interpreted under the known technique as a real therapeutic effect."

More recently, in a 2001 report, Danish researchers evaluated 114 studies that compared a placebo to no treatment at all and found that placebos exerted their most powerful influence on subjective measures, chronic conditions, and perception of pain. In a 2002 study, researchers at the University of California, Los Angeles used electroencephalograms to show differences in brain function between people with depression who responded to placebo and those who responded to antidepressant medication. And a 2004 investigation at the University of Michigan employed functional magnetic resonance imaging (MRI) to demonstrate that activity in certain parts of the brain decreased in response to placebo and increased in anticipation of pain.

Comparing the brain with the intestine is not as big a jump as one might think: There are an equal number of neurons in the intestine— a complex network of nerves making up the "enteric nervous system." Think about it: the need to vomit when you see something disturbingly graphic, a sudden attack of diarrhea in a particularly nerve-wracking or stressful situation, even "butterflies" in the stomach are all regulated by the enteric nervous system, the study of which is still in its infancy.

Emotions and response to stress may indeed play a role in some patients' perceptions of their IBD symptoms, at least partially explaining some of the placebo response found in clinical trials. In patients who've tried a variety of drugs and then enter a clinical trial of a new agent, they may have a strong urge to think, "This is going to be the one that makes me better."

"There may very well be a mind-body connection in IBD. We can't prove it, but we can't disprove it," asserts Dr. Tremaine. "Patients may feel a little shamefaced about responding to a placebo, and say, 'Oh, it was all in my head to begin with.' We don't think it's a psychological thing. It may be partly in your head, but it may also be a neurophysiologic response."

In Crohn's disease, part of the explanation lies in understanding the Crohn's Disease Activity Index (CDAI), a measure of disease activity developed during the 1970s. Many of the measures contributing to this index are subjective, including a patient's report of their general well-being. Just participating in a clinical trial may make a patient feel better. Indeed, the placebo response has been linked to the number of visits to the doctor that a patient makes as a clinical trial participant.

"Patients in clinical trials may feel they're not just doing something to help themselves, but they're also doing something to help everybody else with these diseases, whether or not the clinical trial is successful," says Scott Plevy, M.D., Associate Professor of Medicine at the University of Pittsburgh School of Medicine. Since clinical trial participants who feel better about themselves may report better overall well-being, their CDAI may improve, even though they may have received a placebo.

Patients in clinical trials are also asked to keep accurate diaries about how many medications they are taking each day. "Just being in a clinical trial is different from real-life experience," explains James Lewis, M.D., Assistant Professor of Medicine and Epidemiology at the University of Pennsylvania School of Medicine. "There are more visits to the doctor and more record-keeping in a very controlled setting." Such vigilant chronicling of daily drug use may make patients more compliant with their regular medications than they might have been otherwise — leading them to feel better, even if they didn't receive the study drug.

Medical evidence also supports a link between stress and the immune system. Dr. Plevy notes that there have been studies of patients with ulcerative colitis where 20 to 30 percent of the patients taking a placebo not only felt better, but showed evidence of healing of the intestinal lining when examined with endoscopy. "The mucosa was actually healing before our eyes," he says. "We don't know why. But there's a lot of literature about the release of neuro-psychiatric hormones that may have direct effects on the immune system and on inflammation. So I think the effects the mind can have on the immune system are absolutely real."

Many patients with IBD also have irritable bowel syndrome—a common disorder whose symptoms of diarrhea and pain may be exacerbated by stress. Some patients with IBD report that their symptoms flare during stressful periods, but their discomfort may be due more to increased intestinal movement (motility) from irritable bowel syndrome than an increase in inflammation. Resolution of irritable bowel symptoms in IBD patients taking a placebo could therefore skew the placebo response rate in a clinical trial.


IBD: A Disease with a "Mind" of its Own

Most doctors believe the placebo response in patients with IBD can be attributed to the natural course of the disease. "IBD waxes and wanes, even if you don't do anything differently," asserts Dr. Tremaine. "So in a clinical trial, if you wait long enough, somebody's going to get better."  In a recent clinical trial of the new drug certolizumab pegol in patients with Crohn's disease, for example, twice as many patients who received certolizumab experienced an improvement in their CDAI by the end of the second week ( 29.7 to 33.3 percent of patients) compared to patients who received a placebo (15.1 percent), with the effect being greatest among patients who received the highest dose of the drug. But by the end of week 12— the study's primary endpoint—the proportion of patients in the placebo group who reported improvement in CDAI increased to 35.6 percent, obscuring the statistical difference between the two groups that was observed earlier in the trial.

Does this mean certolizumab is not an effective treatment for Crohn's disease? More likely, some of the patients in the placebo group experienced an improvement in their symptoms that would have occurred regardless of their therapy, simply due to the natural behavior of IBD. Upon further analysis, the
researchers also found that the patients who were most likely to benefit from certolizumab were those with an elevated level of a marker called C-reactive protein (CRP), and the placebo response was greatest in those with low CRP levels (less than 10 mg/L). CRP, which has gained public recognition as a marker of heart disease risk, is also an indicator of inflammation in patients with IBD, especially those with Crohn's disease. CRP is made by the liver, and its production is driven by inflammatory proteins called cytokines.

If certolizumab, which was designed to reduce inflammation, really works, it will prove most effective in patients with true inflammation. Since high CRP levels indicate the presence of active inflammation, it would make sense that patients with high CRP were more likely to respond to the drug and less likely to experience a placebo response.

So what's going on in the patients with low CRP levels? Why would they be less likely to respond to certolizumab or similar drugs? The answer lies in the cause of their symptoms. Some of them may have inflammation that is caused by cytokines other than those related to CRP. In many cases, however, patients may have symptoms unrelated to inflammation. They may experience diarrhea due to complications of surgery, for example, or to strictures (narrowed areas in the intestine). They may not respond as well to anti-inflammatory drugs as patients with high CRP levels and true inflammation, but they may report relief during periods when their disease wanes in severity.


The Future of Clinical Trials for IBD

Placebo response data are not for naught, but may in fact advance IBD research by providing clues about the optimal design of clinical trials and the use of markers such as CRP. The screening of patients for clinical trials may become more rigorous as researchers measure the presence or absence of active inflammation. One method of monitoring inflammation is endoscopy (colonoscopy or sigmoidoscopy), but not all patients can undergo this procedure.

Measuring CRP is far easier and very cost-effective. Some investigators believe that a high CRP level could become a requirement for patients to enter clinical trials of new IBD drugs—a move that would minimize the placebo effect. But that approach would exclude the 50 percent of IBD patients who do not have elevated CRP and are in equally critical need of better therapies. Dr. Lewis explains that future trials will most likely include patients with a variety of CRP levels. Patients would be stratified by their CRP levels, and the effects of the drug (and placebo) would be assessed group by group.

Study endpoints may also need to be altered in clinical trials of new drugs to account for the ebb and flow of symptoms in IBD patients. "It may behoove investigators to evaluate the effectiveness of a new drug earlier in the clinical trial, and then continue to assess patients to determine the drug's long-term safety," says Dr. Lewis.

Clinical investigators may also need to revisit the measures they use to judge a drug's success. Many feel the CDAI has outgrown its usefulness. Analyzing the effect of a drug on objective measures such as CRP could provide a more accurate assessment of its effectiveness compared to a placebo, and would account for differences in levels of inflammation between patients.

Researchers are continuing to look for other markers, in addition to CRP, that could be used to predict a patient's response to therapy. "We're in dire need of better biomarkers and better indices of disease activity," emphasizes Dr. Plevy. "IBD is really many different diseases. At a minimum, there are several types of Crohn's and several types of colitis. Identifying new biomarkers will help define new clinically important subgroups of disease. This is going to be the wave of the next ten years.

"We haven't hit the home run yet where we can say if a test is going to predict or not predict how a patient responds to therapy," he continues. "The best we have for that at this point is CRP."


What This Means for Patients

Despite the debate over the meaning of the placebo response in IBD research, placebos will remain an important part of the clinical trial process. "If you go into a clinical trial and you wind up getting a placebo, that's not necessarily a bad thing," Dr. Lewis advises. "Some patients will improve even on placebo. Furthermore, there are times when being in the group receiving the active drug is not a good thing because of side effects. Patients should not let the fear of getting a placebo prevent them from participating in clinical trials."

Posted June 8, 2006