Search in the Abstract Database

Search Abstracts 2013

* = Presenting author

P245. Evaluation of the relationship between fecal calprotectin concentrations and clinical and endoscopic outcome measures in a phase 2 study of tofacitinib, an oral janus kinase inhibitor, in active ulcerative colitis

J. Panés1, W.J. Sandborn2, H. Zhang3, D. Yu3, W. Niezychowski3, C. Su3, 1Hospital Clínic de Barcelona, CIBERehd, Department of Gastroenterology, Barcelona, Spain, 2University of California, Division of Gastroenterology, San Diego, La Jolla, CA, United States, 3Pfizer Inc, Collegeville, PA., United States

Background

The relationship between fecal calprotectin (FCP) and clinical/endoscopic outcomes based on Mayo score (MS) were assessed after 8 weeks (wk) of treatment with tofacitinib in moderate-to-severe UC.

Methods

In a multicenter, double-blind Phase 2 trial, 194 patients (pts) were randomized to tofacitinib 0.5, 3, 10, or 15 mg BID or placebo. Primary endpoint was clinical response (CR; MS decrease ≥3 points and ≥30%; rectal bleeding subscore decrease ≥1 point or absolute subscore ≤1) at Wk 8. Secondary endpoints included clinical remission (CRem; MS ≤ 2 and no subscore >1), endoscopic remission (ERem; endoscopic subscore of 0), and mucosal healing (MH; endoscopic subscore 0 or 1; post-hoc). Post-hoc analyses including receiver operating characteristic (ROC) analysis evaluated the relationship of FCP concentrations (conc; mg/kg) vs clinical/endoscopic outcomes at Wk 8.

Results

Median Wk 8 FCP conc were significantly lower (p < 0.001) in pts achieving clinical responses vs pts with no response: 156 vs 725 (CR), 64 vs 617 (CRem), 44 vs 489 (ER), and 127 vs 753 (MH) mg/kg. Pts with lower Wk 8 FCP conc were more likely to achieve Wk 8 efficacy endpoints (Table 1). Area under the curve for ROC models with FCP were: CRem 0.80; ERem 0.81; MH 0.78. An FCP cut off of approx. 150 achieved the highest summation of specificity (spec) and sensitivity (sens) for CRem (spec 0.79, sens 0.68, kappa 0.44) and ERem (spec 0.75, sens 0.79, kappa 0.38). Table 2 shows the proportion of pts achieving CRem and/or FCP < 150 at Wk 8 based on treatment assignment. Safety outcomes were consistent with the known profile of tofacitinib.

Table 1. Proportion [n (%)] of patients achieving efficacy endpoints by FCP cut off
 FCP <50 at Wk 8FCP ≥50 at Wk 8FCP <150 at Wk 8FCP ≥150 at Wk 8
 (n = 36)(n = 158)(n = 67)(n = 127)
CRem at Wk 824 (66.7)32 (20.3)38 (56.7)18 (14.2)
ERem at Wk 817 (47.2)16 (10.1)26 (38.8)7 (5.5)
MH at Wk 828 (77.8)68 (43.0)52 (77.6)44 (34.6)
Table 2. Proportion [n (%)] of patients achieving CRem and FCP < 150 at Wk 8
 PlaceboTofacitinib
 BID (N = 48)0.5 mg BID (N = 31)3 mg BID (N = 33)10 mg BID (N = 33)15 mg BID (N = 49)
CRem5 (10.4)4 (12.9)11 (33.3)16 (48.5)20 (40.8)
  p = 0.76p = 0.01p < 0.001p < 0.001
FCP <1509 (18.8)8 (25.8)9 (27.3)17 (51.5)24 (49.0)
  p = 0.48p = 0.37p = 0.002p = 0.002
CRem &FCP <1504 (8.3)3 (9.7)6 (18.2)11 (33.3)14 (28.6)
  p = 0.87p = 0.19p = 0.005p = 0.011
Pair-wise p values vs placebo based on CMH chi-square test stratified by prior TNF exposure.

Conclusion

Although a similar treatment effect was seen with an FCP cut off of 150 mg/kg, FCP levels had only fair to good accuracy in classifying patients' clinical or endoscopic outcomes. These data show limitations to the use of FCP in lieu of MS-based outcome measures in clinical trials.