Chair of the Digestive Disease & Surgery Institute
Cleveland Clinic, Cleveland
This month’s edition of The Regueiro Report looks at minimizing placebo effects in clinical trials and at an encouraging finding that low-volume colonoscopy preparations are safe and efficacious for people with inflammatory bowel disease.
In the first study I discuss, the researchers found that people with Crohn’s disease who have low C-reactive protein (CRP) levels and high albumin levels are more likely to respond to placebo, possibly because inflammation is not driving their symptoms.
Placebo effects in IBD clinical trials arise when we focus on subjective measures (e.g., symptoms) to assess how someone in a clinical trial is responding to a medication. Symptoms are very important for understanding quality of life for people with IBD, and these subjective measures have their place in clinical practice. However, the challenge in a randomized controlled trial that compares a medication with a placebo head-to-head is that improvements in these subjective measures could be seen in those receiving the placebo (i.e., the “placebo effect”) and lead to outcomes that look similar to outcomes seen in those taking the new therapy. This could lead to the study drug failing to show a statistically significant advantage compared with the placebo, making it less likely to move further along toward development, even if it’s a promising therapy for IBD.
The solution to this is to use more objective measures of improvement in IBD, such as biomarkers (e.g., CRP and albumin), or grading mucosa on endoscopy. Similarly, when we enroll a patient with IBD into a trial, we want to ensure that they have an appropriate degree of active bowel inflammation, without symptoms that may be unrelated to inflammation (e.g., comorbid irritable bowel syndrome). Many patients with irritable bowel will have symptoms of diarrhea and abdominal pain that are also common in Crohn’s disease and ulcerative colitis. If symptoms alone were the inclusion criteria or outcome measure of success, it is conceivable that the placebo would show similar results to the active drug being studied. Disease duration also may play a role in outcomes, with the placebo effect being less likely in those with a shorter disease duration and new onset of disease.
The take-home message is to consider disease duration, CRP levels and albumin levels, as well as other objective measures, when designing clinical trials to reduce the effect of placebo. In addition, patients who are newly diagnosed with signs of active inflammation (e.g., high CRP levels) are less likely to have a placebo effect and will better indicate whether the study medication is effective for IBD.
The other study I selected addresses the long-standing idea that high-volume colonoscopy preparations do a better job of cleaning the bowel before the procedure. It turns out this is not true. Nobody likes these preparations, but they can be especially hard on people with IBD. The study found that three different low-volume preps were safe and effective in people with IBD who were about to undergo a colonoscopy. It is helpful to have strong evidence that we can give IBD patients a more palatable, low-volume way to prepare for this procedure.
Reducing the Placebo Effect
Inflamm Bowel Dis 2023;29(9):1390-1398
This article is a post hoc analysis of three placebo-controlled clinical trials of treatments for moderate to severe Crohn’s disease evaluating predictors of clinical remission. The trials—GEMINI-2, UNITI-1/2 and CLASSIC-1—collectively enrolled 683 patients treated with placebo.
The following predictors were included in the multivariable model for clinical remission: CRP level lower than 5 mg/L (odds ratio [OR], 1.66; 95% CI, 1.04-2.67; P=0.035), albumin higher than 40 g/L (OR, 1.57; 95% CI, 1.05-2.93; P=0.023) and disease duration shorter than five years (OR, 1.70; 95% CI, 1.05-2.75; P=0.032).
The researchers found that of the three predictors in the multivariable analysis, shorter disease duration was the only variable that retained significance (adjusted OR, 1.67; 95% CI, 1.02-2.73; P=0.040).
The researchers conclude that strategies, such as exclusion of participants by disease duration and mild disease severity based on objective biomarkers, can minimize placebo response rates.
Low-Volume Prep in IBD
Scand J Gastroenterol 2023; 58(6):656-663
This study enrolled 92 outpatients with IBD who were preparing for a colonoscopy and received various low-volume colonoscopy preparations: 1 L of polyethylene glycol ascorbate (n=33), 2 L of polyethylene glycol ascorbate (n=28) or the oral laxative sodium picosulfate (n=31).
No serious adverse events occurred for any of the preparations. The Boston Bowel Preparation Scale showed that each preparation effectively cleansed the bowel. Patients were most likely to intake sodium picosulfate in full and were most likely to say they would use sodium picosulfate again.
This article is from the December 2023 print issue.