We performed a retrospective analysis of data collected from 110 patients with UC who participated in a phase 2 placebo-controlled trial of etrolizumab, as well as from 21 patients with UC or without inflammatory bowel disease (controls) enrolled in an observational study at a separate site. Colon biopsies were collected from patients in both studies and analyzed by immunohistochemistry and gene expression profiling. Mononuclear cells were isolated and analyzed by flow cytometry. We identified biomarkers associated with response to etrolizumab. In the placebo-controlled trial, clinical remission was defined as total Mayo Clinic Score ≤2, with no individual subscore >1, and mucosal healing was defined as endoscopic score ≤1.
Colon tissues collected at baseline from patients who had a clinical response to etrolizumab expressed higher levels of T-cell−associated genes than patients who did not respond (P < .05). Colonic CD4+ integrin αE+ cells from patients with UC expressed higher levels of granzyme A messenger RNA (GZMA mRNA) than CD4+ αE− cells (P < .0001); granzyme A and integrin αE protein were detected in the same cells. Of patients receiving 100 mg etrolizumab, a higher proportion of those with high levels of GZMA mRNA (41%) or ITGAE mRNA (38%) than those with low levels of GZMA (6%) or ITGAE mRNA (13%) achieved clinical remission (P < .05) and mucosal healing (41% GZMAhigh vs 19% GZMAlow and 44% ITGAEhigh vs 19% ITGAElow). Compared with ITGAElow and GZMAlow patients, patients with ITGAEhigh and GZMAhigh had higher baseline numbers of epithelial crypt-associated integrin αE+ cells (P < .01 for both), but a smaller number of crypt-associated integrin αE+ cells after etrolizumab treatment (P < .05 for both). After 10 weeks of etrolizumab treatment, expression of genes associated with T-cell activation and genes encoding inflammatory cytokines decreased by 40%−80% from baseline (P < .05) in patients with colon tissues expressing high levels of GZMA at baseline.
Levels of GZMA and ITGAE mRNAs in colon tissues can identify patients with UC who are most likely to benefit from etrolizumab; expression levels decrease with etrolizumab administration in biomarkerhigh patients. Larger, prospective studies of markers are needed to assess their clinical value.