IBD: Optimal Endpoints of Therapy

In the realm of inflammatory bowel disease (IBD), endpoints can be based on many things. We commonly use subjective assessments and complement these with more objective evaluations, including laboratory, endoscopic and radiologic studies. A great deal depends on the context in which these are considered: are they in reference to an individual patient or to assess response in a clinical trial of a new biologic agent?

Research studies and drug development demand scores allowing for consistent patient selection, quantification of the proportion of patients responding based on the rules of the study, and comparisons across patients and between studies. By contrast, metrics for the individual patient can be and should be far more nuanced. A patient-specific response should take into account the patient’s overall well-being, most bothersome symptom, individualized treatment goals and objective assessments of inflammation.
Historically, subjective scales have dominated assessments in clinical trials. These include the Crohn’s disease activity index for Crohn’s disease and the Mayo score for ulcerative colitis. However, it is worth noting that both have objective parameters: the Crohn’s disease activity index includes weight and anemia, and the Mayo score includes endoscopy. Finally, the IBD Questionnaire is a validated instrument measuring quality of life. Most studies include it as a secondary endpoint and generate a graph of the data, but generally it is not used to gauge drug efficacy.

Under scrutiny because of two competing issues in the field. The first is FDA and patient-led efforts to focus on patient-reported outcomes, which has led to a distillation of all IBD endpoints into primarily two things: number of bowel movements and abdominal pain. Given that these outcomes were derived from trials, they are not truly independent of the old scales. Now that we are left with only two things to assess, we all miss the Crohn’s disease activity index.

On the other extreme, infliximab ushered in the era of mucosal healing. So what about mucosal healing as an endpoint? In particular, what does mucosal healing mean in trials versus practice? Several instruments have been validated for assessment of endoscopy in IBD. In Crohn’s disease, the Crohn’s disease Endoscopic Index of Severity is the most detailed and has a shortened sister version, the Simple Endoscopic Score. The abridged version does not capture the depth of the ulcerations, which we know to be important for prognostication. There is also the Rutgeert’s score, which measures neoterminal ileal disease recurrence after ileocolic resection. The Mayo endoscopic score looks at the severity of the inflammation in ulcerative colitis. It has been criticized because of inter-observer variability and improved upon by recent studies that tested and validated the ulcerative colitis endoscopic index of severity.1

Now comes the devil in the details: how much mucosal healing does one need in order to say that a patient is mucosally healed? In recent clinical trials of biologics for IBD, it is surprising to see how the definition varies” some allow persistence of mild inflammation and some require no inflammation. As clinicians and investigators, we would like to see all studies providing data reflecting complete mucosal healing and those with a substantial improvement compared to their baseline. One can understand that a patient with very limited mucosal inflammation is more likely to have complete healing with a therapy whereas one with deep ulcerating disease may have substantial improvement in their ulcerations that is actually more meaningful and more difficult to achieve.

In addition to endoscopy, there are other important ways to assess disease activity and therapeutic endpoints less invasively. C-reactive protein (CRP) is a reasonably specific marker of inflammation in patients with IBD without other co-morbidities, though flawed by the fact not all individuals generate CRP (because of genetic polymorphisms) and because small bowel disease is less likely to induce c-reactive protein. Likewise, fecal calprotectin has emerged as a sensitive and specific marker in IBD, but again is of limited use in those with small bowel disease. Each provides complementary data. It is best to think of these tests like an ammonia level, important to know a baseline and follow the course as encephalopathy improves.

Finally, radiologic tests are not only useful but, for a given patient, may be the only way to follow small bowel disease activity or look at perianal fistula tracts. For magnetic resonance enterography, the MaRIA score (not named for the author) tracks well with severity of disease and response to therapy.2

In terms of individual endpoints, several things must be considered concomitantly including patient symptoms, disease severity, and risk of progression or complications of disease. Clinicians seeing an IBD patient should have a standard way to query symptoms and identify those that are most bothersome for that patient. A simple how do you feel? is insufficient. Although number of bowel movements and abdominal pain are reasonable places to start, some patients are felled by fatigue, by urgency and fear of incontinence, and others by anal pain. Therefore, establishing what they most want to accomplish with therapy is essential. This becomes especially important for a seemingly asymptomatic patient, who has either become accustomed to a lower quality of life and does not remember normal or is truly asymptomatic and the clinician must therefore have a clear reason to treat the patient (e.g., deep ulcerations, shortened small intestine) since medications cannot make them feel any better. In Atul Gawande’s new book, Being Mortal, he highlights a physician’s task of establishing a patient’s therapeutic goals and how much they are willing to risk to achieve them. One would be surprised by what patients really want.

The ultimate endpoint of therapy would be a cure. Unfortunately, we are not there yet. But many have witnessed firsthand the transformation possible with current therapies. About 30 percent of patients on biologics experience complete mucosal healing.3 Not surprisingly, ample studies demonstrate those lucky patients who do achieve mucosal healing do better, with fewer hospitalizations and surgeries. Based on these data, the aspirational goal should be complete mucosal healing. Indeed, histologic remission, especially in ulcerative colitis, is emerging as even better than mucosal healing alone. In a recent study, we found patients may have adequate serum levels of anti-tumor necrosis factor (TNF) agents and yet the intensity of the inflammation and local TNF production may outstrip the available anti- TNF in the tissue.4 Thus, operationally it is not always possible to give sufficient anti-TNF to achieve mucosal healing.

The authors suggest establishing individual treatment endpoints prior to starting new medications. First, have a clear understanding of the patient’s most bothersome symptom and the degree and extent of inflammatory disease using the tools described above (see figure on the next page). The choice of therapy is beyond what can be described here but should be appropriate for the degree of inflammation and patient-specific factors like age and co-morbidities. When and how to assess response to therapy should be based on a medication’s onset of action and the best test for that particular patient. For example, anti-TNF agents should be expected to work by 12 weeks, whereas vedolizumab may need slightly longer.

If the goals of therapy are not achieved, the clinician should try to optimize the patient’s current therapy. Depending on these goals and patient-specific risk factors, pushing to mucosal healing may be ideal but has to be tempered by the risk of doing so. Along the way, keep in mind adjunctive agents like anti- diarrheals that may have dramatic benefits on the endpoints for the patient. In the sum, the endpoints of IBD therapy should be a balance focusing on patient goals and also patient- specific characteristics.

Dr. Perlini has no conflicts to disclose.

Dr. Abreu serves as a consultant for AbbVie, Prometheus Labs, Sano Aventis, Takeda, UCB, Pfizer, Janssen, Mucosal Health Board, GSK, Hospira, Shire, Eli Lilly and Ferring, She serves on the scienti c advisory boards of Asana Medical, Celgene and Shire. She has lectured or trained on behalf of Imedex, Pri-Med Institute, ECCO, Cornerstones Health, Focus Medical Communications, Prova Education and WebMD Health. She serves on the board of directors of the GI Health Foundation.


1. Travis SP et. al, Reliability and initial validation of the ulcerative colitis endoscopic index of severity. Gastroenterology. 2013 Nov;145(5):987-95.
2. Ordás I et al Accuracy of magnetic resonance enterography in assessing response to therapy and mucosal healing in patients with Crohn’s disease.Gastroenterology. 2014 Feb;146(2):374-82.e1.
3. Rutgeerts P et. al, Adalimumab induces and maintains mucosal healing in patients with Crohn’s disease: data from the EXTEND trial. Gastroenterology 2012 May;142(5):1102-1111.e2.
4. Yarur AJ1 et al. Feb 10. The association of tissue anti- TNF drug levels with serological and endoscopic disease activityinin ammatoryboweldisease:theATLASstudy. [Epub ahead of print] / Gut 2015 PMID: 25670812

One comment

Join the discussion

Your email address will not be published. Required fields are marked *