Risk-benefit Assessment of Adalimumab and Certolizumab Pegol as Induction Treatment for Crohnʼs Disease

Abstract

Purpose: Benefit and risk of a treatment can be quantified by calculating number needed to treat (NNT) and number needed to harm (NNH)1. Relative to a comparator, treatments with smaller NNTs are more effective and those with smaller NNHs are more toxic; negative NNHs indicate reduced risk of an adverse event. We quantified the benefit-risk profile of adalimumab (ADA) and certolizumab pegol (CZP) by calculating NNT for clinical remission and NNH for serious adverse events (SAE) during induction treatment of patients (pts) with moderate to severe Crohn's disease (CD). Methods: Data were evaluated from induction studies of ADA (CLASSIC-I [bionaïve pts]2 and GAIN [infliximab-exposed pts]3) and CZP (bionaïve pts)4. Efficacy and safety data (Table 1) from randomization through the study's primary endpoint (ADA: 4 weeks; CZP: 6 weeks) were evaluated for intent-to-treat pts receiving ADA (160 mg/80 mg at Weeks 0 and 2), CZP (400 mg at Weeks 0, 2, 4) and corresponding placebo (pbo) to calculate NNT for clinical remission (CDAI<150) and NNH for SAE. Point estimates and two-sided 95% confidence intervals (CI) were derived for the treatment effect (NNT) and risk difference (NNH) between drug and pbo.Table 1: Data from ADA and CZP induction trialsResults: Point estimates for NNT and NNH are shown in Table 2. The NNT for ADA was 5 for bionaïve pts and 8 for infliximab-exposed pts (both values statistically significant); the NNT for CZP was not significantly different from placebo. The 95% CI values for NNHs for both compounds included ∞, indicating no statistically significant difference between drug and pbo for SAE. Because of differences in trial designs, patient populations, and study endpoints, comparisons of efficacy and safety from different clinical trials is not appropriate.Table 2: Point estimates for NNT and NNHConclusion: For adalimumab, the NNT value for remission induction was 5 for bionaïve patients and 8 for infliximab exposed patients, and NNH values did not differ significantly from placebo, indicating a favorable risk-benefit profile.