Abstract
Background/aim: The objective of this study is to identify clinical predictors of primary non-response (PNR) and secondary loss of response (LOR), in Crohn’s disease (CD) patients treated with anti-tumor necrosis factor α (anti-TNF) agents.
Methods: This retrospective, longitudinal, and observational cohort study included 283 patients with CD who received anti-TNF treatments from November 2006 to July 2017 at Samsung Medical Center, Seoul, Korea.
Results: A total of 212 patients with CD were eligible and based on clinical responses, divided into three groups: PNR, LOR, and responder groups. PNR occurred in 13 patients (6.1%). C-Reactive protein (CRP) level at initiation of anti-TNF (baseline CRP) was a possible predictor of PNR compared to the non-PNR group (baseline CRP >1 mg/dl, OR = 4.34, 95% CI = 1.06–17.83, p = .042). During maintenance therapy, incidence of LOR was 12.2% at 1-year, 23.6% at 2-years, 36.3% at 3-years, and 52.1% at 5-years. Combining baseline CRP level and CRP reduction rate [(CRP at 12–14 weeks–baseline CRP)/baseline CRP] was a possible predictor of 1-year LOR compared to the responder group (baseline CRP >1 mg/dl and CRP reduction rate > −70%, OR = 18.86, 95% CI = 3.40–104.55, p = .001). In the Cox hazard proportional model, a combination of baseline CRP level and CRP reduction rate was possible predictors of long-term LOR during maintenance therapy (baseline CRP >1 mg/dl and CRP reduction rate > −70%, HR = 5.84, 95% CI = 2.75–12.41, p < .001).
Conclusions: Baseline CRP level and CRP reduction rate might be clinical predictors for PNR or LOR to anti-TNF in patients with CD, and could guide proper therapeutic interventions in patients with CD.
Acknowledgements
The authors would like to thank all of the enrolled patients.
Disclosure statement
No potential conflicts of interest relevant to this article were reported.