O-002 Evaluation of the Operating Properties of Candidate Patient Reported Outcomes for Use in Randomized Controlled Trials in Crohn’s Disease

Abstract

The Crohn’s Disease Activity Index (CDAI), the conventional composite outcome measure used in randomized controlled trials (RCTs), consists of 8 items that evaluate symptoms, signs and laboratory tests. Recently the US FDA has indicated that patient reported outcomes (PROs) should be utilized as, or incorporated into, the primary outcome measure in Crohn’s disease (CD) registration trials in preference to the CDAI. However no validated PROs currently exist. Development and validation of novel PRO instruments is a lengthy process. Therefore, we determined whether the outcomes reported by patients from the CDAI diary card (abdominal pain, stool frequency, general well-being) could be adapted to derive a PRO with favorable operating properties for use in RCTs. Development consisted of 2 phases. In Phase I, data from an RCT1 of rifaximin extended intestinal release therapy were used to identify cut-points that yielded optimum sensitivity and specificity for identification of CDAI –defined remission (score <150). Two item (abdominal pain, stool frequency) and 3 item (2 item plus general well-being) PRO instruments were generated. In Phase II, these instruments were assessed using data from an RCT2 of methotrexate (MTX) induction therapy in active CD for which the primary outcome measure was a CDAI score <150 without prednisone after 16 weeks of therapy. The magnitude of the treatment effect of MTX was estimated using the 2- and 3-item instruments and the CDAI in patients with and without objective evidence of inflammation at baseline (serum orosomucoid concentration >88 mg per deciliter). Figure 1 shows the Receiver Operator Characteristic (ROC) curves for the optimal cut-points for each of the daily average item scores (mean daily stool frequency ≤1.5, mean daily abdominal pain score ≤1, mean daily general well-being score ≤1). Table 1 shows the estimates of effect size of MTX for multiple outcome measures in the 2 patient populations. The estimates of MTX treatment effect were similar for the 2 and 3 item PROs and the CDAI—based outcomes. The largest estimates of the treatment effect were derived when 1) complete withdrawal of prednisone was included in the definition of outcome and 2) patients had objective evidence of inflammation as defined by the presence of an elevated baseline serum concentration of orosomucoid. These data indicate that readily available data derived from patients’ CDAI diary cards may be appropriate for use as a PRO in clinical trials for CD. 1. Prantera C, Lochs H, Grimaldi M, et al. Rifaximin-extended intestinal release induces remission in patients with moderately active Crohn's disease. Gastroenterology 2012;142:473–81 e4. 2. Feagan BG, Rochon J, Fedorak RN, et al. Methotrexate for the treatment of Crohn's disease. The North American Crohn's Study Group Investigators. N Engl J Med 1995;332:292–7.