• Welcome to Crohn's Forum, a support group for people with all forms of IBD. While this community is not a substitute for doctor's advice and we cannot treat or diagnose, we find being able to communicate with others who have IBD is invaluable as we navigate our struggles and celebrate our successes. We invite you to join us.

Using Genetic Markers to Predict Response to Anti-TNF Therapy in IBD

Genome Wide Association (GWA) Predictors of Anti-TNFalpha Therapeutic Responsiveness in Pediatric Inflammatory Bowel Disease
Dubinsky MC, Mei L, Friedman M, et al
Inflamm Bowel Dis. [Epub ahead of print]

Mucosal Gene Signatures to Predict Response to Infliximab in Patients With Ulcerative Colitis
Arijs I, Li K, Toedter G, et al
Gut. 2009;58:1612-1619. Epub 2009 Aug 20.

Two recent studies assessed genetic predictors of response to the antitumor necrosis factor (anti-TNF) therapy infliximab in patients with inflammatory bowel disease (IBD). Arijs and colleagues analyzed RNA from mucosal biopsies of 46 patients in 2 cohorts with ulcerative colitis, just prior to receiving infliximab for the first time, and then again 4-6 weeks after the first dose. They compared expression profiles in responders, defined as those with endoscopic mucosal healing, as well as in nonresponders, defined as all remaining participants not meeting the response criteria. The investigators found that the top 5 differentially expressed genes encoded proteins involved in adaptive immune signaling pathways and TNF pathways, and, using these markers, they were able to separate responders from nonresponders with an overall accuracy of 89%.

In the second study, Dubinsky and colleagues studied a cohort of 94 children with either Crohn's disease or ulcerative colitis who were treated with infliximab for the first time. Genetic associations with primary nonresponse to treatment defined by nonendoscopic disease activity indices were assessed with both a targeted approach for IBD-susceptibility genes as well as a genome-wide association study. In addition to the genetic findings, IBD-associated serologic markers and patient demographics were also assessed in a multivariable model. Twenty-two of the 94 children met the criteria for primary nonresponse. The most predictive model for nonresponse involved a combination of 3 novel genetic loci, a previously identified susceptibility gene (BRWD1), perinuclear antineutrophil cytoplasmic antibody (pANCA) status, and a diagnosis of ulcerative colitis (area under the curve, 98%).

The treatment of both Crohn's disease and ulcerative colitis has been revolutionized by anti-TNF therapy, with demonstrated reductions in hospitalizations, surgery, and improved quality of life. However, in clinical trials, one third of patents show no response to these therapies, and fewer than one third of patients enrolled in these trials are in remission and off of steroids at the end of 1 year. The costs of anti-TNF therapies are enormous, and concerns for serious infections and lymphoma underscore the need to ideally target therapy toward those most likely to benefit.

Previous attempts to assess for response to infliximab have suggested that increased pANCA levels may be associated with nonresponse,[1,2] although previously identified genetic predictors have been difficult to replicate.

These 2 studies used novel approaches in identifying genetic predictors. Arijs and colleagues assessed mucosal gene signatures and used rigorous endoscopic endpoints to separate responders from nonresponders, whereas Dubinsky and colleagues assessed both known susceptibility loci and the unbiased genome-wide association study, and incorporated serologic and phenotypic predictors into a final, highly predictive model. The 2 studies looked at different populations and endpoints for defining response, so it is not unexpected that the findings are not more similar. Also, both studies suffer from limited sample sizes, and both require replication in larger cohorts to confirm the findings.

Ultimately, a clinically relevant predictive panel will incorporate elements from both of these studies, -- ie, using both genotypic and phenotypic information from a defined patient population stratified by rigorous definitions of response -- because it is unlikely that any single gene or predictor will suffice. This comprehensive approach will help clinicians target the use of these highly potent, but expensive and potentially toxic medications to those who stand to gain the most. Similarly, these studies will help to identify those patients who are unlikely to benefit from therapy, and thus allow for alternative management strategies to be prioritized.