Background Previous retrospective studies of paediatric ulcerative colitis have had limited ability to describe disease progression and identify predictors of treatment response. In this study, we aimed to identify characteristics associated with outcomes following standardised therapy after initial diagnosis. Methods The PROTECT multicentre inception cohort study was based at 29 centres in the USA and Canada and included paediatric patients aged 4–17 years who were newly diagnosed with ulcerative colitis. Guided by the Pediatric Ulcerative Colitis Activity Index (PUCAI), patients received initial standardised treatment with mesalazine (PUCAI 10–30) oral corticosteroids (PUCAI 35–60), or intravenous corticosteroids (PUCAI ≥65). The key outcomes for this analysis were week 12 corticosteroid-free remission, defined as PUCAI less than 10 and taking only mesalazine, and treatment escalation during the 12 study weeks to anti-tumour necrosis factor α (TNFα) agents, immunomodulators, or colectomy among those initially treated with intravenous corticosteroids. We identified independent predictors of outcome through multivariable logistic regression using a per-protocol approach. This study is registered with ClinicalTrials.gov, number NCT01536535. Findings Patients were recruited between July 10, 2012, and April 21, 2015. 428 children initiated mesalazine (n=136), oral corticosteroids (n=144), or intravenous corticosteroids (n=148). Initial mean PUCAI was 31·1 (SD 13·3) in children initiating with mesalazine, 50·4 (13·8) in those initiating oral corticosteroids, and 66·9 (13·7) in those initiating intravenous corticosteroids (p<0·0001 for between-group comparison). Week 12 outcome data were available for 132 patients who initiated with mesalazine, 141 with oral corticosteroids, and 143 with intravenous corticosteroids. Corticosteroid-free remission with the patient receiving mesalazine treatment only at 12 weeks was achieved by 64 (48%) patients in the mesalazine group, 47 (33%) in the oral corticosteroid group, and 30 (21%) in the intravenous corticosteroid group (p<0·0001). Treatment escalation was required by nine (7%) patients in the mesalazine group, 21 (15%) in the oral corticosteroid group, and 52 (36%) in the intravenous corticosteroid group (p<0·0001). Eight patients, all of whom were initially treated with intravenous corticosteroids, underwent colectomy. Predictors of week 12 corticosteroid-free remission were baseline PUCAI less than 35 (odds ratio 2·44, 95% CI 1·41–4·22; p=0·0015), higher baseline albumin by 1 g/dL increments among children younger than 12 years (4·05, 1·90–8·64; p=0·00030), and week 4 remission (6·26, 3·79–10·35; p<0·0001). Predictors of treatment escalation by week 12 in patients initially treated with intravenous corticosteroids included baseline total Mayo score of 11 or higher (2·59, 0·93–7·21; p=0·068 [retained in model due to clinical relevance]), rectal biopsy eosinophil count less than or equal to 32 cells per high power field (4·55, 1·62–12·78; p=0·0040), rectal biopsy surface villiform changes (3·05, 1·09–8·56; p=0·034), and not achieving week 4 remission (30·28, 6·36–144·20; p<0·0001). Interpretation Our findings provide guidelines to assess the response of children newly diagnosed with ulcerative colitis to standardised initial therapy and identify predictors of treatment response and failure. These data suggest that additional therapeutic interventions might be warranted to improve early outcomes, especially in patients presenting with severe disease and requiring intravenous corticosteroids. Funding National Institutes of Health.
Bibliographical noteFunding Information:
Support for this study was provided by the NIH via the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) 5U01DK095745 and P30 DK078392, Integrative Morphology Core. Further details about the study are available online. We thank Frank Hamilton and Stephen James from NIDDK for their guidance, and William Faubion for his role as safety monitor. The study investigators are deeply indebted to Shire Pharmaceuticals for providing Pentasa (mesalazine) for this study, to the research coordinators at the investigative sites for their tireless attention, and to the patients and families who agreed to participate in this important study.
JSH has served on an advisory board for Janssen and is acting as a consultant for Abbvie, Takeda, Lilly, Boehringer Ingelheim, Allergan, and Astra Zeneca. SD has served on an independent data monitoring committee for Lycera Corporation. LAD has received grant support from AbbVie and Janssen. DRM has served on advisory boards for Abbvie and Janssen, acted as a consultant for UVB, and is an owner and holds shares in Biotagenics. NSL has acted as a consultant for Abbvie. AP has participated in speakers bureaus for Abbvie and Janssen. JM has acted as a consultant for Janssen, UCB, and Lilly. AMG declares research support from Abbvie, has acted as a consultant for Abbvie, Janssen, Merck, and Takeda, and has been a speaker for Abbvie and Janssen. JR has acted as a consultant for Abbvie, Janssen, Luitpold, and UCB, and has received grant funding from Janssen and Abbive. AO has served on advisory boards for Janssen and Abbvie, and has received research support from Abbvie, Janssen, Shire, and Astellas. MDK has acted as a consultant for Abbvie, Janssen, GlaxoSmithKline, and Pfizer. MD has acted as a consultant for Prometheus Laboratories. PR has acted as a consultant for Shire and Leutpold, has been a speaker for Abbvie, and has received research support from TechLab. CGS has acted as a consultant for Abbvie. SK has acted as a consultant for Janssen and UCB. JSt has acted as a consultant and speaker for Abbvie. MH reports research grants from Genentech, Abbvie, Sucampo, and Janssen. All other authors declare no competing interests.
© 2017 Elsevier Ltd