P278 Type IV collagen formation/degradation ratio predicts response to infliximab induction therapy in patients with Ulcerative Colitis

Abstract

Abstract Background Up to 50% of patients with ulcerative colitis (UC) do not respond to biological therapies, including TNF-α-antagonists (e.g., infliximab, IFX) and migration inhibitors (e.g., the α4β7-integrin inhibitor vedolizumab, VEDO). Therefore, serological biomarkers that predict therapy response are urgently needed. In this respect, biomarkers of extracellular matrix (ECM) remodeling, including basement membrane (BM) remodeling, and intestinal inflammation may hold promise as these processes play an important role in the pathophysiology of UC. In this study, we aimed to assess the predictive ability of serological biomarkers of ECM turnover and intestinal inflammation with regard to clinical response to IFX and VEDO therapy in patients with UC. Methods Serological biomarkers of collagen formation (PRO-C3, PRO-C4), matrix metalloproteinase (MMP)-mediated collagen degradation (reC1M, C3M, C4M, C4G, C6Ma3) and intestinal inflammation (macrophage activity marker [VICM] and calprotectin degradation fragments [CPa9-HNE]) were measured using neo-epitope solid-phase competitive enzyme-linked immunosorbent assay (ELISA) technology in 80 patients with UC who received either infliximab (IFX, n=26) or vedolizumab (VEDO, n=54) induction therapy. Clinical response to therapy was defined by composite assessment of available Simple Clinical Colitis Activity Index (SCCAI) scores and physician’s global assessments (PGA) derived from clinical records. Logistic regression modelling and receiver operating characteristics (ROC) statistics were used to assess discriminative capacity regarding response to IFX and VEDO. Results Only baseline ratios of type IV collagen formation and degradation (PRO-C4/C4M) were significantly elevated in patients with UC who clinically responded to IFX therapy (P<0.05). In addition, PRO-C4/C4M ratios accurately discriminated responders and non-responders to IFX therapy (area under the curve [AUC]=0.77, P<0.05), also after adjustment for confounding factors (AUC=0.88, P<0.01). Serum VICM levels were elevated in patients not responding to IFX therapy, but did not reach statistical significance (AUC=0.76, P=0.06). None of the biomarkers predicted response to VEDO therapy. Conclusion The baseline ratio of type IV collagen formation/degradation demonstrated high predictive accuracy with regard to response to IFX therapy in patients with UC, which may imply that patients with increased BM formation vs. degradation are more likely to respond. Furthermore, VICM may hold promise as a predictor for non-response to IFX therapy. Serological biomarkers of ECM turnover and intestinal inflammation may aid in predicting response to biological therapy and deserve validation in larger, prospective cohort studies.