An innovative biomarker could predict response to IBD therapy

Not everyone responds equally well to treatments for inflammatory bowel disease (IBD). Determining what works for individual patients requires trial-and-error during their therapy journey. Now, researchers from Charite-Universitatsmedizin and colleagues in Berlin and Bonn have successfully developed a biomarker which accurately indicates whether immunomodulator medications will be successful treatments. Researchers writing in Gastroenterology note this will allow more focused use of therapy.Inflammatory Bowel Disease (IBD), including Crohn’s disease and ulcerative colitis, are caused by an overactive immune response in the digestive tract that leads to an inflammatory process in its entirety. People living with Crohn’s experience abdominal cramping, diarrhea and fatigue. There is currently no cure – therefore treatments have focused on relieving symptoms and managing inflammation. Prof. Ahmed Hegazy of Charite’s Department of Gastroenterology, Infectious Diseases and Rheumatology agrees. As an active patient care participant himself, Prof. Hegazy takes great pride in actively contributing to patient care as a clinical scientist. “Since this disease involves flares that can arise unexpectedly, treatment must constantly be adjusted accordingly. Unfortunately, no one can predict how an individual will progress with disease or respond to various treatment options; thus making treatment so complex. Integrin-blocking therapy has proved highly successful while only producing minor side effects.” Vedolizumab acts to stop certain immune cells from invading the GI tract and initiating inflammation there, including T helper cells that would otherwise enter it via their migration routes. By binding to T helper cells it acts like an effective blocking effect: no T helpers enter. “Integrin-blocking therapy has proven highly successful for approximately two-thirds of patients; however, for one-third it simply doesn’t work at all. Determining who would respond was once determined solely through trial-and-error methods.” Hegazy notes: “Treating patients requires lengthy and costly examination processes that often prove frustrating for patients themselves,” and suggests having an advance indicator to know whether treatment plans might prove worthwhile or not – an aim our study set out to accomplish.

Machine learning assists in pattern recognition.

Researchers conducted comprehensive studies with 47 patients suffering from chronic IBD. Blood samples were taken prior to and six weeks post treatment with Vedolizumab. Researchers used sophisticated analytical techniques such as mass cytometry, single-cell RNA sequencing and serum proteomics to examine samples. “We identified various immune cells and proteins affected by treatment as potential targets for change.” Hegazy explained. “This research produced vast amounts of data that we then analyzed with machine learning – an area of artificial intelligence which employs algorithms and statistical models to enable computers to autonomously learn from data by recognising patterns without needing to be explicitly programmed beforehand. Through this research, we were able to recognize patterns which allow us to predict which patients would respond best to this form of therapy. Researchers representing medicine, bioinformatics, mathematics and biology from institutions including Berlin Institute of Health at Charite (BIH), German Rheumatism Research Centre Berlin (DRFZ) and University of Bonn identified similar patterns among studies from another group of patients; 26 individuals participated in their analysis to help validate results of study.

With high biomarker levels and ineffective treatment results.

Ki67, which is produced at increased levels during T helper cell division, proved particularly valuable as an indicator. Patients who had high concentrations of these cells prior to starting on treatment did not respond positively to vedolizumab therapy: we identified its molecular basis: these T helper cells do not possess binding sites for vedolizumab so can pass unimpeded into the digestive tract and continue contributing inflammation,” Hegazy explained. Because different trafficking molecules enable T helper cells to move throughout it unimpeded into this route allowing T helper cells resistant to its influence which makes Ki67 an indicator that indicates when treatment options fail compared with T helper cells resistant to treatment.”

Making the Switch into clinical practice

Researchers plan to validate their findings through large multicenter studies and thoroughly explore the reliability of biomarkers that they have identified. Furthermore, plans call for further developing detection and measurement methods so they can be integrated into routine clinical practice. According to Prof. Britta Siegmund from the department’s directorship: “Reliable biomarkers are key for providing patients with chronic inflammatory bowel disease with personalized therapy more quickly and with greater accuracy; that step toward personalized medicine will bring clarity early for them”.
Source:Charite – Universitatsmedizin BerlinJournal Reference:Horn V et al 2024) Multimodal profiling of peripheral blood identified circulating effector CD4+ T cells proliferating rapidly as predictors for response to an Integrin A4B7 blocker therapy treatment of IBD Gastroenterology DOI 10.1053/j.gastro.2024.09.021