Development and Validation of Machine Learning Models in Prediction of Remission in Patients With Moderate to Severe Crohn Disease.
ABSTRACT: Importance:Biological therapies have revolutionized inflammatory bowel disease management, but many patients do not respond to biological monotherapy. Identification of likely responders could reduce costs and delays in remission. Objective:To identify patients with Crohn disease likely to be durable responders to ustekinumab before committing to long-term treatment. Design, Setting, and Participants:This cohort study analyzed data from 3 phase 3 randomized clinical trials (UNITI-1, UNITI-2, and IM-UNITI) conducted from 2011 to 2015. Participants (n?=?401) were individuals with active (C-reactive protein [CRP] measurement of ?5 mg/L at enrollment) Crohn disease who received ustekinumab therapy. Data analysis was performed from November 1, 2017, to June 1, 2018. Exposures:All included patients were exposed to 1 or more dose of ustekinumab for 8 weeks or more. Main Outcomes and Measures:Random forest methods were used in building 2 models for predicting Crohn disease remission, with a CRP level lower than 5 mg/dL as a proxy for biological remission, beyond week 42 of ustekinumab treatment. The first model used only baseline data, and the second used data through week 8. Results:In total, 401 participants, with a mean (SD) age of 36.3 (12.6) years and 170 male (42.4%), were included. The week-8 model had a mean area under the receiver operating characteristic curve (AUROC) of 0.78 (95% CI, 0.69-0.87). In the testing data set, 27 of 55 participants (49.1%) classified as likely to have treatment success achieved success with a CRP level lower than 5 mg/L after week 42, and 7 of 65 participants (10.8%) classified as likely to have treatment failure achieved this outcome. In the full cohort, 87 patients (21.7%) attained remission after week 42. A prediction model using the week-6 albumin to CRP ratio had an AUROC of 0.76 (95% CI, 0.71-0.82). Baseline ustekinumab serum levels did not improve the model's prediction performance. Conclusions and Relevance:In patients with active Crohn disease, demographic and laboratory data before week 8 of treatment appeared to allow the prompt identification of likely nonresponders to ustekinumab without the need for costly drug-level monitoring.
Project description:Background/Aims:Efficacy and safety of ustekinumab were evaluated in a Japanese subpopulation with moderately to severely active Crohn's disease (CD) in UNITI-1, UNITI-2 and IM-UNITI studies and results were compared with the overall population. Methods:Overall, patients in UNITI-1 (Japan, n=56; failed response to tumor necrosis factor antagonist) and UNITI-2 (Japan, n=26; failed response to prior conventional therapy) were randomized to placebo or ustekinumab intravenous induction (130 mg or ~6 mg/kg) at week 0. Responders to ustekinumab induction therapy (Japan, n=21) were randomized to placebo or ustekinumab (90 mg, subcutaneous) maintenance (every 12 weeks [q12w] or 8 weeks [q8w]) in IM-UNITI. The primary endpoint was clinical response at week 6 for induction studies and clinical remission at week 44 for maintenance study. Results:Percentage of patients achieving clinical response at week 6 was greater in ustekinumab 130 mg and ~6 mg/kg groups than in the placebo group (UNITI-1: 36.8% and 31.6% vs. 27.8%, respectively, for Japanese; 34.3% and 33.7% vs. 21.5%, respectively, for overall; UNITI-2: 37.5% and 55.6% vs. 11.1%, respectively, for Japanese; 51.7% and 55.5% vs. 28.7%, respectively, for overall). Clinical remission rate at week 44 during maintenance was greater in the ustekinumab 90 mg SC q12w and q8w groups than in the placebo group (50.0% and 55.6% vs. 25.0%, respectively, for Japanese; 48.8% and 53.1% vs. 35.9%, respectively, for overall). Efficacy and safety results observed in the Japanese subpopulation were generally consistent with those in the overall population. Conclusions:Ustekinumab could be considered as a new therapeutic option for moderately to severely active CD in Japanese patients. Both ustekinumab induction and maintenance treatments were generally well tolerated (Clinical Trial Registration: NCT01369329, NCT01369342, NCT01369355).
Project description:BACKGROUND:In Phase 3 studies of ustekinumab, a fully human monoclonal IL-12/23p40 antibody approved for moderate-to-severe Crohn's disease, patients entered a long-term extension after completing 8 weeks of induction and 44 weeks of maintenance treatment. Efficacy through 92 weeks and safety through 96 weeks of IM-UNITI maintenance are reported. METHODS:UNITI-1 (TNF-antagonist failures) and UNITI-2 (conventional therapy failures) patients (N = 1281) entered IM-UNITI, including 397 ustekinumab intravenous induction responders randomised to subcutaneous ustekinumab 90 mg every 12 weeks, every 8 weeks, or placebo and 884 nonrandomised patients. Dose-adjustment to 90 mg every 8 weeks occurred in patients randomised to 90 mg every 12 weeks and placebo patients with loss of response (Weeks 8-32). All Week 44 completers could enter the long-term extension without further dose adjustment. Placebo patients discontinued following study unblinding. RESULTS:A total of 718 patients (all treated) entered the long-term extension (298 randomised and 420 not randomised). Overall, 86.5% (621/718) completed Week 96. The proportions of randomised patients in clinical remission were generally maintained from Week 44 through 92 in ustekinumab 90 mg every 12 weeks (77.4% to 72.6%), every 8 weeks (84.1% to 74.4%), and prior dose adjustment groups (63.4% to 53.5%). At Week 92, the proportions of patients in clinical remission were similar in the ustekinumab 90 mg every 12 weeks and every 8 weeks groups and lower in patients with prior dose adjustment. Proportions of patients in clinical remission at Week 92 for all treated every 8 weeks (64.4%) and every 12 weeks (64.3%) groups were lower than randomised every 8 weeks (74.4%) and every 12 weeks (72.6%) groups, but similarly maintained. Safety events (per hundred patient-years) were similar among all placebo and ustekinumab patients (Week 0-96), including adverse events (484.39 vs 447.76), serious adverse events (19.24 vs 18.82), and serious infections (4.09 vs 4.02). No dose effect was observed. CONCLUSIONS:Subcutaneous ustekinumab maintained clinical response and remission through Week 92. No new safety signals were observed. ClinicalTrials.gov number NCT01369355.
Project description:Crohn's disease is characterized by a dysregulation of both innate and adaptive immunity responses. Interleukin-12/23 (IL-12/23) pathway has been found to be a major driver of inflammation in adaptive immune responses. Ustekinumab is a fully human immunoglobulin G1 kappa monoclonal antibody that blocks the p40 subunit of IL-12 and IL-23 and prevents their interaction with their cell surface receptor and further cytokine activation. It is currently approved in the management of plaque psoriasis and psoriatic arthritis. Very promising data have emerged through phase II and phase III trials (UNITI-1, UNITI-2, and IM-UNITI) for both induction and maintenance of clinical response and remission in moderate-to-severe Crohn's disease, resulting in approval by the Food and Drug Administration for this condition. This article reviews the immunology of the IL-12/23 pathway, available data regarding the initial designing of ustekinumab, drug development through clinical trials including pharmacokinetics, efficacy, and safety, and its potential place in the treatment of Crohn's disease.
Project description:BACKGROUND:Ustekinumab is an effective therapy for Crohn disease currently approved for adults. Off-label use in the pediatric population is increasing, but its effectiveness in this age group has not been reported. AIMS:The aim of the study was to describe real-world experience with ustekinumab at a tertiary care pediatric inflammatory bowel disease (IBD) center. METHODS:As part of an ongoing observational cohort study of biologic-treated pediatric IBD patients initiated in October 2014, data on demographics, disease behavior, location and activity, treatment, and surgical history were collected for all patients receiving ustekinumab. Disease activity was assessed using the Harvey Bradshaw index or partial Mayo score. Primary outcome was steroid-free remission at 52 weeks. Descriptive statistics summarized the safety and efficacy outcomes, and univariate analyses were performed to examine associations of clinical characteristics with efficacy. RESULTS:Fifty-two children and young adults initiating ustekinumab were analyzed; 81% Crohn Disease, 8% ulcerative colitis, and 11% IBD-unspecified. Median [IQR] age at induction was 16.8 [14-18] years. Patients were followed for a minimum of 12 months. Most patients (81%) failed >1 anti-TNF, and 37% failed anti-TNF and vedolizumab; 10 patients were biologic-naïve. At week 52, 75% were still on ustekinumab, and 50% (bio-exposed) and 90% (bio-naïve) were in steroid-free remission. Two infusion reactions and neither serious adverse events nor serious infections were observed. CONCLUSIONS:Our results suggest that ustekinumab is efficacious and safe in pediatric patients with IBD. Controlled clinical trial data are needed to confirm these observations.
Project description:<h4>Background and aims</h4>Ustekinumab is approved for the treatment of Crohn's disease [CD]. Systematically registered prospective real-world data are scarce. We therefore aimed to study the effectiveness, safety and usage of ustekinumab for CD in everyday practice.<h4>Methods</h4>We prospectively enrolled CD patients initiating ustekinumab in regular care between December 2016 and January 2019. Clinical (Harvey Bradshaw Index [HBI]), biochemical (C-reactive protein [CRP] and faecal calprotectin [FCP]), extra-intestinal manifestations and, peri-anal fistula activity, ustekinumab dosage, concomitant medication use, and adverse events were documented at weeks 0, 12, 24, and 52. The primary outcome was corticosteroid-free clinical remission.<h4>Results</h4>In total, 221 CD patients were included (98.6% anti-tumour necrosis factor [TNF] and 46.6% vedolizumab exposed) with a median follow-up of 52.0 weeks [interquartile range 49.3-58.4]. Corticosteroid-free clinical remission rates at weeks 24 and 52 were 38.2% and 37.1%, respectively. An initial dosing schedule of 8 weeks, compared to 12 weeks, correlated with a lower discontinuation rate [20.0% vs 42.6%, p = 0.01], but comparable corticosteroid-free clinical remission at week 52 (46.3% [q8w] vs 34.6% [q12w], p = 0.20). There was no clinical benefit of combination therapy after 52 weeks when compared to ustekinumab monotherapy [combi 40.6% vs mono 36.0%, p = 0.64]. At baseline, 28 patients had active peri-anal fistula, of whom 35.7% showed complete clinical resolution after 24 weeks. During follow-up we encountered six severe infections [3.5 per 100 patient-years], with all patients being on concomitant immunosuppressant therapies. Ustekinumab treatment discontinuation was observed in 75 [33.9%] patients mainly due to lack of response.<h4>Conclusion</h4>Ustekinumab is a relatively safe and effective treatment option for CD patients with prior failure of anti-TNF and anti-integrin therapies.
Project description:Psoriatic arthritis (PsA) is an inflammatory arthritis that commonly occurs with psoriasis and is attributed to genetic, immunologic and environmental factors. The T-helper (Th)-17 pathway and the interleukin (IL)-23/IL-17 axis have become prominent players in PsA and considerably increased our understanding of disease pathogenesis. In this review article, we will focus on the emerging role of IL-12/23 and its blockade, in the pathogenesis and management of PsA as well as of psoriasis and inflammatory bowel disease. Ustekinumab, is a fully human monoclonal immunoglobulin (Ig)G1 antibody that binds specifically to the p40 subunit of IL-12 and IL-23, primarily inhibiting downstream Th-17 signalling pathways. Ustekinumab produced consistent and sustained clinical efficacy in two phase III clinical trials in PsA, PSUMMIT-1 and PSUMMIT-2, with data out to 52 weeks, and no new safety signals. PSUMMIT-1 included patients with active PsA despite conventional therapy who were all naïve to anti-tumour necrosis factor (TNF) agents, whereas PSUMMIT-2 also included anti-TNF experienced patients. Similarly, ustekinumab produced consistent clinical efficacy in two phase III clinical trials in psoriasis, PHOENIX-1 and PHOENIX-2, and in both induction and maintenance of moderate-to-severe Crohn's disease, UNITI-1, UNITI-2 and IM-UNITI, without an increased safety signal. Currently, ustekinumab is used in the treatment of PsA following the failure of nonsteroidal anti-inflammatory drugs (NSAIDs) and conventional disease-modifying antirheumatic drugs (DMARDs), and as an alternative to, or after failure of an anti-TNF agent.
Project description:To evaluate the efficacy and safety of ustekinumab through 2 years in adult patients with active psoriatic arthritis (PsA).A total of 615 adult patients with active PsA were randomized to placebo, ustekinumab 45 mg, or ustekinumab 90 mg, at weeks 0, 4, and every 12 weeks through week 88 (last dose). At week 16, patients with <5% improvement in both tender and swollen joint counts entered blinded early escape (placebo to 45 mg, 45 mg to 90 mg, and 90 mg to 90 mg). All remaining placebo patients crossed over to ustekinumab 45 mg at week 24. Clinical efficacy measures included American College of Rheumatology criteria for 20% improvement (ACR20), Disease Activity Score in 28 joints using the C-reactive protein level (DAS28-CRP), and ≥75% improvement in the Psoriasis Area and Severity Index (PASI75). Radiographic progression was evaluated using the modified Sharp/van der Heijde score (SHS).At week 100, ACR20, DAS28-CRP moderate/good response, and PASI75 rates ranged from 56.7-63.6%, 71.9-76.7%, and 63.9-72.5%, respectively, across the 3 treatment groups. In both ustekinumab groups, the median percent improvement in dactylitis and enthesitis was 100% at week 100. The mean changes in SHS score from week 52 to week 100 were similar to those observed from week 0 to week 52 in the ustekinumab groups. Through week 108, 70.7% and 9.7% of patients had an adverse event (AE) or serious AE, respectively. The rates and type of AEs were similar between the dose groups.Clinical and radiographic benefits from ustekinumab treatment were maintained through week 100 in the PSUMMIT 1 study. No unexpected safety events were observed; the safety profile of ustekinumab in this population was similar to that previously observed in psoriasis patients treated with ustekinumab.
Project description:Biopsy samples (n=362) were collected from terminal ileum at baseline, 8 weeks after induction (Ustekinumab or placebo), and 44 weeks after maintenance (Ustekinumab 90 mg SC q12w, Ustekinumab 90 mg SC q8w, or placebo) therapies during endoscopy for RNA extraction and microarray analysis from patients with moderate-to-severe CD who participated in stelara CD phase 3 studies (UNITI-2 and IM-UNITI). These patients failed conventional therapies previously and largely naive to anti-TNF therpy. Ileum biopsies (n=26) collected from non-IBD subjects were also analyzed to serve as control. Enrichment of a core microvilli gene set was found to be reduced in UNITI-2 CD samples relative to controls, correlated with microvilli length and endoscopy score, and associated with response to treatment. Overall design: A gene expression profiling study was conducted in which ileum biopsy samples were collected for RNA extraction and hybridization to microarrays from patients (n=362) with moderate-to-severe Crohn's disease and from non-IBD subjects (n=26).
Project description:BACKGROUND:Reliable assessment of remission is important for the optimal management of rheumatoid arthritis (RA) patients. In this study, we used the multi-biomarker disease activity (MBDA) test to explore the role of biomarkers in predicting point remission and sustained remission. METHODS:RA patients on >?6?months stable therapy in stable low disease activity (DAS28-ESR ??3.2) were assessed every 3?months for 1?year. Baseline, intermittent (IR) and sustained (SR) remission were defined by DAS28-ESR, DAS28-CRP, simple disease activity index (SDAI), clinical disease activity index (CDAI) and ACR/EULAR Boolean criteria. Patients not fulfilling any remission criteria at baseline were classified as 'low disease activity state' (LDAS). Patients not fulfilling any remission criteria over 1?year were classified as 'persistent disease activity' (PDA). MBDA score was measured at baseline/3/6?months. The baseline MBDA score, the 6-month time-integrated MBDA score and MBDA biomarkers were used for analyses. The area under the receiver operating characteristic curve (AUROC) assessed the ability of the MBDA score to discriminate between remission and non-remission. Biomarkers were analysed at baseline using the Mann-Whitney test and over time using the Jonckheere-Terpstra trend test. RESULTS:Of 148 patients, 27% were in the LDAS, 65% DAS28-ESR remission, 51% DAS28-CRP remission, 40% SDAI remission, 43% CDAI remission and 25% ACR/EULAR Boolean remission at baseline. Over 1?year, 9% of patients were classified as PDA. IR and SR were achieved in 42%/47% by DAS28-ESR, 46%/29% by DAS28-CRP, 45%/20% by SDAI, 44%/21% by CDAI and 35%/9% by ACR/EULAR Boolean criteria, respectively. By all remission criteria, baseline MBDA score discriminated baseline remission (AUROCs 0.68-0.75) and IR/SR (AUROCs 0.65-0.74). The 6-month time-integrated MBDA score discriminated IR/SR (AUROCs 0.65-0.79). Baseline MBDA score and concentrations of IL-6, leptin, SAA and CRP were significantly lower in all baseline remission criteria groups vs LDAS. They and the 6-month time-integrated values were lower among patients who achieved IR/SR vs PDA over 1?year. CONCLUSIONS:This study demonstrated that the MBDA score and its biomarkers IL-6, leptin, SAA and CRP differentiated between small differences in disease activity (i.e. between low disease activity and remission states). They were also predictors of remission over 1 year.
Project description:BACKGROUND & AIMS:Crohn disease (CD) presents as chronic and often progressive intestinal inflammation, but the contributing pathogenic mechanisms are unclear. We aimed to identify alterations in intestinal cells that could contribute to the chronic and progressive course of CD. METHODS:We took an unbiased system-wide approach by performing sequence analysis of RNA extracted from formalin-fixed paraffin-embedded ileal tissue sections from patients with CD (n = 36) and without CD (controls; n = 32). We selected relatively uninflamed samples, based on histology, before gene expression profiling; validation studies were performed using adjacent serial tissue sections. A separate set of samples (3 control and 4 CD samples) was analyzed by transmission electron microscopy. We developed methods to visualize an overlapping modular network of genes dysregulated in the CD samples. We validated our findings using biopsy samples (110 CD samples for gene expression analysis and 54 for histologic analysis) from the UNITI-2 phase 3 trial of ustekinumab for patients with CD and healthy individuals (26 samples used in gene expression analysis). RESULTS:We identified gene clusters that were altered in nearly all CD samples. One cluster encoded genes associated with the enterocyte brush border, leading us to investigate microvilli. In ileal tissues from patients with CD, the microvilli were of decreased length and had ultrastructural defects compared with tissues from controls. Microvilli length correlated with expression of genes that regulate microvilli structure and function. Network analysis linked the microvilli cluster to several other down-regulated clusters associated with altered intracellular trafficking and cellular metabolism. Enrichment of a core microvilli gene set also was lower in the UNITI-2 trial CD samples compared with controls; expression of microvilli genes was correlated with microvilli length and endoscopy score and was associated with response to treatment. CONCLUSIONS:In a transcriptome analysis of formalin-fixed and paraffin-embedded ileal tissues from patients with CD and controls, we associated transcriptional alterations with histologic alterations, such as differences in microvilli length. Decreased microvilli length and decreased expression of the microvilli gene set might contribute to epithelial malfunction and the chronic and progressive disease course in patients with CD.