Search in the Abstract Database

Abstracts Search 2019

P477 Clinical remission by legacy vs. FDA definitions: definition justification and results from UNIFI Study

W. J. Sandborn*1, R. Strauss2, H. Zhang2, J. Johanns2, P. Szapary2, C. Marano2, S. Danese3

1University of California San Diego, La Jolla, USA, 2Janssen Research and Development, LLC, Spring House, USA, 3Humanitas Research Hospital, Milan, Italy


Ustekinumab (UST), an interleukin-12/23 blocker, was evaluated as induction and maintenance for moderate–severe ulcerative colitis (UC). Clinical remission was analysed using a US-specific definition (FDA) excluding the PGA (Physician’s Global Assessment) and the legacy definition which includes the PGA to accommodate regional regulatory preference.


Patients (pts) were randomised to receive a UST intravenous (IV) induction dose (either 130 mg [n = 320] or approximating 6 mg/kg [n = 322]), or PBO (n = 319). Responders to UST IV induction were randomised to SC maintenance of 90 mg UST (either every 12 weeks [n = 172] or every 8 weeks [n = 176]), or PBO (n = 175). The primary endpoint for induction (Week 8) and maintenance (Week 44) was clinical remission. In a prior UC induction study with golimumab, 87.5% had >3 stools per day at baseline and stool number ≤3 aligned with what approximately 98% of patients reported as normal. FDA clinical remission definition was developed after FDA requested PGA removal (ie, absolute stool number ≤3 [aligned with upper limit of normal stool number in the general population], Mayo rectal bleeding subscore 0, and Mayo endoscopy subscore 0/1). This differed from the legacy definition (total Mayo score ≤2 points, with no individual subscore >1). Using golimumab and infliximab UC study data, FDA definition was assessed for agreement with legacy definition, treatment effect, and clinical meaningfulness using the Inflammatory Bowel Disease Questionnaire and the 36-item short form health survey as anchor variables. UNIFI remission was analysed using both definitions.


The FDA definition demonstrated high concordance, specificity and sensitivity with the legacy definition with a similar treatment effect, and defined patients who had clinically meaningful benefit. In the UNIFI study, Week 8 clinical remission rates among patients receiving IV UST at either 130 mg or ~6 mg/kg were significantly higher than PBO patients by both legacy (15.6%, 15.5%, and 5.3%, respectively p < 0.001 for both doses) and FDA definitions (16.6%, 18.9%, and 6.3%, respectively; p < 0.001 for both doses). Week 44 clinical remission rates among patients randomised to q12wk or q8wk UST were significantly higher than PBO patients by both legacy (38.4%, 43.8%, and 24.0%, respectively; p = 0.002 q12wk and p < 0.001 q8wk) and FDA definitions (39.5%, 42.6%, and 24.6% respectively; p = 0.002 q12wk and p < 0.001 q8wk).


Using either legacy or FDA remission definition, IV UST induced remission and SC UST maintained remission in UST induction responders with moderately-to-severely active UC. Importantly, the FDA definition with absolute stool number ≤3 is clinically meaningful, easily understood by physicians and patients, and is not based on patients’ distant recall of normal stool pattern.