Search in the Abstract Database

Abstracts Search 2019

P317 Characterisation of patients with delayed response to ustekinumab for Crohn’s disease

B. E. Sands*1, A. Oortwijn2, N. Rijnders2, J. Izanec3, C. Gasink3, D. Jacobstein4, O. J. Adedokun4, T. Ma4, L-L. Gao4, J-F. Colombel5, S. Targan6, S. Ghosh7, W. J. Sandborn8

1Icahn School of Medicine at Mount Sinai, New York, USA, 2Janssen Biologics BV, Leiden, The Netherlands, 3Janssen Scientific Affairs, LLC, Horsham, USA, 4Janssen Research and Development, LLC, Spring House, USA, 5Icahn School of Medicine at Mt Sinai, New York, USA, 6Cedars-Sinai Medical Center, Los Angeles, USA, 7University of Birmingham, Birmingham, UK, 8University of California San Diego, La Jolla, USA

Background

In UNITI-1 and 2, pivotal induction studies of ustekinumab (UST) in patients with CD, 467 non-responders to UST 130 mg or ~6 mg/kg IV received UST 90 mg SC at Week 8. Overall, 50.5% achieved response at Week 16 (delayed responders, DR).1 We sought to characterise DR population induced with ~6 mg/kg IV and identify predictors for delayed response.

Methods

UNITI-1 and 2 patients who were induced with UST~6 mg/kg IV and had a UST SC dose at Week 8 were included in this post hoc analysis and classified as Week 8-responders (ER), DR (no response at Week 8/response at Week 16) or non-responders (NR, no response at Weeks 8 and 16). Levels of UST and CRP at Weeks 8 and 16 and FeCal at Weeks 6 and 16 were described. Pearson and Spearman correlations between drug exposure, changes in FeCal and CRP and change from baseline (BL) in CDAI at Weeks 6, 8, and 16 were analysed. Univariate logistic regression modelling was performed on BL variables, including concomitant medications, UST, FeCal, and CRP levels and changes from BL. Factors from the univariate model with p < 0.15 were included in a multivariate logistic regression model and significant predictors (p < 0.15) were selected by backward method.

Results

Among 387 patients induced with UST~6 mg/kg IV, 38.7% were ER, 23.8% were DR, 37.5% were NR. Serum UST, CRP and FeCal levels, correlation results, and multi-variate logistic regression are in Table 1. Drug levels were similar in ER and DR and slightly lower in NR. UST levels weakly correlated with CDAI changes from Wk0–16 in ER, but not in delayed responders. Among DR, Wk16 response rates were equally distributed in different quartiles from Wk8 exposure. No or weak correlation was found between changes from BL in FeCal CRP and CDAI at the same visit. Univariate logistic regression model identified age, BMI, corticosteroid (CS), active fistula (AF) at BL and history (hx) of extra intestinal manifestation (EIM) and colonic disease as potential predictors for delayed response vs. Wk8 response. The final model identified patients with younger age, non-CS at BL, hx of EIM, pure ileal disease hx and AF as more likely to have delayed response vs. Wk8 response.

Table 1. Main results

Conclusion

Delayed response to UST induction is observed in about 24% of patients with CD induced with the ~6 mg/kg IV dose and one additional dose SC at Wk8. Among tested variables, neither Wk8 drug levels nor biomarker response or previous anti-TNF failure was predictive of delayed response. Younger age, no CS or active fistula at BL, hx of extra intestinal manifestation and pure ileal disease were associated with increased risk for delayed response vs. Wk8 response.

References

1. Feagan BG, Sandborn, WJ, Gasink C, et al. Ustekinumab as induction and maintenance therapy for Crohn's disease. N Engl J Med 2016;375:1946–60.