P237. Correlation between fecal calprotectin levels and clinical activity in ulcerative colitis in a phase I study with an anti-MAdCAM‑1 monoclonal antibody
S. Vermeire1, J. Panes2, F. Cataldi3, G. Yuan3, G. Comer4
1University Hospital Gasthuisberg, Gastroenterology, Leuven, Belgium; 2Hospital Clinic Barcelona, Inflammatory Bowel Disease Unit, Barcelona, Spain; 3Pfizer, GI, Biotherapeutics Clinical Programs, Boston, United States; 4Pfizer Inc., Bio-therapeutics, Collegeville, United States
Background: The fully human anti-MAdCAM 1 monoclonal antibody PF00547659 (PF) recently evaluated in ulcerative colitis (UC) represents a potential new treatment of UC (Vermeire Gut 2011). Outcome measures in UC clinical trials often combine findings from patient-reported outcome and invasive endoscopic tests. However, there is emerging evidence that fecal calprotectin may serve as a biomarker for the evaluation of efficacy. We evaluated the correlation between the decrease in fecal calprotectin and disease activity measured by the Total Mayo score in this first in human (FIH) study in UC with PF.
Methods: In a double blind dose-escalating study 80 subjects (60 active and 20 placebo (P)) were assigned to 1 of 6 single-dose (SD) or 5 multiple-dose (MD) cohorts (PF (0.03 to 10 mg/kg SC or IV) or P). In this post-hoc analysis, we compared clinical response based on Total Mayo score at week 4 and 12 to fecal calprotectin. Pearson correlation analysis and regression models were used to assess the relationship between the biomarker and the efficacy measurements. Due to the small sample size in each dose level, subjects were categorized into 3 dose groups for the statistical analysis: Placebo, low dose group (0.03 mg/kg iv (SD), 0.1 mg/kg iv (SD or MD), 0.3 mg/kg iv (SD or MD) and 0.3 mg/kg sc (MD)), and high dose group (1.0 mg/kg iv (SD), 1.0 mg/kg sc (MD), 3.0 mg/kg sc (SD), 3.0 mg/kg iv (MD) and 10 mg/kg iv (SD)).
Results: Fecal calprotectin levels decreased to a greater extent with PF than P (week 4: 63% vs 18%). Pearson correlation coefficient between the change from baseline in natural log transformed fecal calprotectin and change from baseline in Total Mayo score is 0.48 (P < 0.0001). Linear regression was used to model the relationship between the change from baseline in natural log transformed fecal calprotectin and change from baseline in Total Mayo score, which showed a regression coefficient of 0.95 (0.18). The statistical testing on the coefficient showed high significance (P < 0.0001), suggesting evidence of the correlation and predictive capabilities of fecal calprotectin as efficacy measure in UC.
Conclusions: In this FIH study with PF00547659, anti-MAdCAM 1 monoclonal antibody, efficacy was observed for both the decrease in fecal calprotectin levels and the correlation between fecal calprotectin and clinical response measured by the Total Mayo score. These findings suggest that fecal calprotectin could play a role in the future as an efficacy or pharmacodynamic endpoint in UC clinical trials.