Multimodal profiling of peripheral blood identifies proliferating circulating effector CD4+ T cells as predictors for response to integrin α4β7-blocking therapy in inflammatory bowel disease

Circulating proliferating CD4+ effector T cells as predictors of non-response to vedolizumab in inflammatory bowel disease

In recent years, the number of advanced treatment options for inflammatory bowel disease (IBD) has increased significantly. However, many patients still experience primary or secondary drug failure and there is no cure for the disease. To date, the choice of therapy is patient-centered, taking into account age, concomitant diseases, family planning, the affected parts of the gastrointestinal tract, and disease severity. Head-to-head studies between approved substances are scarce. The current guidelines therefore do not include a recommendation on the treatment sequence. For the individual patient, this often means trying out different mechanisms of action until the right medication is found. This in turn leads to prolonged periods of inadequate disease control, with reduced quality of life, as well as possible organ damage and complications. In addition, this approach is unfavorable from a health economics point of view. The development of reliable biomarkers to predict treatment response to advanced therapies would significantly improve the management of IBD. 
Horn et al. used a multimodal approach to investigate predictive factors for treatment response to vedolizumab. Vedolizumab is an antibody directed against the integrin α4β7 that prevents the interaction with its ligand MadCAM-1, an adhesion molecule expressed by intestinal endothelial cells. This limits the intestinal homing of α4β7-expressing immune cells and reduces their number in the mucosa. Peripheral blood is easy to collect in everyday clinical practice and is therefore well suited for determining potential biomarkers. In this study, peripheral blood was prospectively sampled from 2 independent cohorts before and during treatment with vedolizumab (at week 2, 6, 30 and 50 after initiation of treatment). Analysis of proteome, transcriptome, and clonality of immune cells and serum protein levels was then performed. For patients with Crohn’s disease, treatment response was defined as a reduction in the Harvey-Bradshaw Index by at least 3 points and for ulcerative colitis patients as a reduction in the partial Mayo score by at least 2 points after 30 weeks of treatment. The co-expression of α4 and β7 could be detected on various other immune cells in addition to T cells and monocytes before therapy. Frequencies of α4⁺β7⁺ cells were lower in the peripheral blood of IBD patients than in healthy donors before treatment was initiated. During treatment, there was an increase in α4⁺β7⁺ cells over time, an increased expression of α4β7, particularly by myeloid populations, and an increase in the clonal diversity of circulating CD4⁺ memory T cells (but not CD8⁺ T cells or B cells). Furthermore, there was an increase in proinflammatory serum proteins during treatment. Proliferating CD4⁺ memory T cells in the peripheral blood prior to initiation of treatment were identified as a predictive factor for non-response to vedolizumab by using machine-learning algorithms. The increased proportion of cells positive for the proliferation marker Ki67 was specific to CD4⁺ T cells, and was not seen in CD8⁺ memory T cells or regulatory T cells. The proliferating CD4⁺ memory cells showed characteristics of activated T helper 1 (Th1) and Th17 cells, which are known to be key players in the pathogenesis of IBD. Moreover, circulating proliferating CD4⁺ memory T cells showed an increased expression of homing molecules and the integrin α4β1, whose interaction with its ligand is not affected by vedolizumab.
In summary, this multimodal study shows that treatment with vedolizumab leads to the retention of myeloid and lymphoid immune cells in the peripheral blood and that consecutively increased proinflammatory cytokines and chemokines can also be measured in the peripheral blood. An increased proportion of proliferating activated CD4⁺ T cells was identified as a predictive marker for non-response to vedolizumab. Indeed, a significant proportion of these cells expressed the integrin α4β1, which is not a target molecule of vedolizumab. In clinical trials, the blocking of α4β1 and α4β7 by an antibody directed against α4 (natalizumab) showed anti-inflammatory effects, but also an increased risk of fatal progressive multifocal leukoencephalopathy (PML). An orally administered integrin α4 antagonist (AJM300) demonstrated clinical response in a phase 3 study involving patients with moderate ulcerative colitis, without an increased risk of PML. These findings indicate that such a treatment could be beneficial in cases where proliferating activated CD4⁺ T cells are detected. Studies such as the one presented here not only help identify predictors of therapeutic response, but may also point toward alternative treatment mechanisms. However, the study here also has several limitations. Although the biomarker was identified in 2 independent cohorts, both consisted of a relatively small number of participants. Not all analytical modalities could be applied to every patient. Analyses using paired biopsies to investigate the immunological changes in the intestine are lacking, and functional analyses exploring the pathogenic role of the relevant immune cell populations were not performed. A multicenter, prospective study for biomarker validation is therefore needed. Within the context of precision medicine, a key ongoing challenge remains: biomarkers must be determined and validated separately for each therapeutic class to ultimately determine an optimal, individualized treatment sequence.