Development and Validation of Machine Learning Models in Prediction of Remission in Patients With Moderate to Severe Crohn Disease.

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.

Adalimumab persistence for inflammatory bowel disease in veteran and insured cohorts.

OBJECTIVES: Identify predictors of persistence with adalimumab (ADA) among veterans and privately insured patients with inflammatory bowel disease (IBD) in the United States. STUDY DESIGN: Retrospective cohort study. METHODS: Patients with IBD taking ADA as their first biologic were identified from the Veterans Health Administration (VHA) database from 2009 to 2013 and the Truven Health MarketScan database from 2009 to 2012 with a 12-month follow-up. Persistence was defined as continued use 1 year after initiation. Adherence was assessed by calculating a medication possession ratio, which was dichotomized as greater than 0.86 or less than or equal to 0.86. Multivariable logistic regression was used to evaluate predictors of persistence. RESULTS: There were 1030 patients in the VHA population compared with 3264 patients in the privately insured (MarketScan) cohort. In MarketScan, 1800 patients (55%) remained on ADA compared with 755 (73%) in the VHA cohort. In multivariable analysis, male sex (odds ratio [OR], 1.38; 95% CI, 1.16-1.63; P <.01), Crohn disease (OR, 1.27; 95% CI, 1.02-1.57; P = .03), greater adherence (OR, 1.83; 95% CI, 1.45-2.30; P <.01), and dose escalation (OR, 1.82; 95% CI, 1.42-2.33; P <.01) were associated with higher ADA persistence in the MarketScan cohort; narcotic use (OR, 0.71; 95% CI, 0.58-0.88; P <.01) and hospitalization or new steroid use after initiation (OR, 0.04; 95% CI, 0.03-0.05; P <.01) were associated with lower persistence. In the VHA cohort, only a hospitalization or new steroid use (OR, 0.50; 95% CI, 0.36-0.70; P <.01) was associated with lower persistence. CONCLUSIONS: Despite being older and having more comorbidities, patients in the VHA, which is an integrated healthcare system, appear to be more likely to remain on ADA at 1 year than patients in the MarketScan database. Further studies of system differences are needed to understand the reasons behind this discrepancy.

Can Contrast-Enhanced Sonography Detect Bowel Wall Fibrosis in Mixed Inflammatory and Fibrotic Crohn Disease Lesions in an Animal Model?

OBJECTIVES: To determine whether contrast-enhanced sonographic quantitative perfusion parameters can detect bowel wall fibrosis in the setting of mixed inflammatory and fibrotic lesions in a Crohn disease animal model. METHODS: This study was approved by the institutional Committee on the Use and Care of Animals. Multiple (range, 1-5) 2,4,6-trinitrobenzenesulfonic acid-ethanol enemas were used to create intestinal inflammatory lesions with variable fibrosis in female Lewis rats. Low-mechanical index contrast-enhanced sonography was performed 3 days after the final enema using a 0.2-mL bolus of sulfur hexafluoride microbubbles injected through a tail vein. Contrast-enhanced sonographic data were analyzed with software that converts video data into echo-power (linearized) data. Colorectal lesions were scored for histopathologic inflammation and fibrosis; bowel wall collagen was quantified by Western blotting. The Spearman correlation was used to assess associations between contrast-enhanced sonographic quantitative parameters and bowel wall collagen; the Kruskal-Wallis test was used to compare continuous results between histopathologic groups. RESULTS: Thirty-one animals were included in our analysis. Animals were placed into 3 histopathologic cohorts: (1) severe bowel wall inflammation/minimal or no fibrosis (n = 11); (2) severe bowel wall inflammation/moderate fibrosis (n = 9); and (3) severe bowel wall inflammation/severe fibrosis (n = 11). Western blotting showed a significant difference in bowel wall collagen between histopathologic cohorts (P = .0001). There was no correlation between any contrast-enhanced sonographic quantitative parameter and bowel wall collagen (P > .05). There was no difference between histopathologic cohorts for any contrast-enhanced sonographic quantitative parameter (P > .05). CONCLUSIONS: Contrast-enhanced sonographic quantitative perfusion parameters failed to effectively detect bowel wall fibrosis in the setting of superimposed inflammation in a Crohn disease animal model.

Optimisation of Intestinal Fibrosis and Survival in the Mouse S. Typhimurium Model for Anti-fibrotic Drug Discovery and Preclinical Applications.

BACKGROUND AND AIMS: Intestinal fibrosis is a frequent complication in Crohn's disease [CD]. The mouse Salmonella typhimurium model, due to its simplicity, reproducibility, manipulability, and penetrance, is an established fibrosis model for drug discovery and preclinical trials. However, the severity of fibrosis and mortality are host- and bacterial strain-dependent, thus limiting the original model. We re-evaluated the S. typhimurium model to optimise fibrosis and survival, using commercially available mouse strains. METHODS: Fibrotic and inflammatory markers were evaluated across S. typhimurium ΔaroA:C57bl/6 studies performed in our laboratory. A model optimisation study was performed using three commercially available mouse strains [CBA/J, DBA/J, and 129S1/SvImJ] infected with either SL1344 or ΔaroA S. typhimurium. Fibrotic penetrance was determined by histopathology, gene expression, and αSMA protein expression. Fibrosis severity, penetrance, and survival were analysed across subsequent CBA studies. RESULTS: Fibrosis severity and survival are both host- and bacterial strain-dependent. Marked tissue fibrosis and 100% survival occurred in the CBA/J strain infected with SL1344. Subsequent experiments demonstrated that CBA/J mice develop extensive intestinal fibrosis, characterised by transmural tissue fibrosis, a Th1/Th17 cytokine response, and induction of pro-fibrotic genes and extracellular matrix proteins. A meta-analysis of subsequent SL1344:CBA/J studies demonstrated that intestinal fibrosis is consistent and highly penetrant across histological, protein, and gene expression markers. As proof-of-concept, we tested the utility of the SL1344:CBA/J fibrosis model to evaluate efficacy of CCG-203971, a novel anti-fibrotic drug. CONCLUSION: The S. typhimurium SL1344:CBA/J model is an optimised model for the study of intestinal fibrosis.

Increased risk of venous thromboembolic events with corticosteroid vs biologic therapy for inflammatory bowel disease.

BACKGROUND & AIMS: We investigated whether treatment of active inflammatory bowel disease with biologic agents is associated with a reduced risk of venous thromboembolic events (VTEs) compared with corticosteroid therapy. METHODS: We performed a retrospective analysis of 15,100 adults with inflammatory bowel disease who were identified from the Truven Health MarketScan databases. We analyzed data from patients who received 6 months of continuous medical and prescription coverage before and 12 months after their first diagnosis and had no VTE during the 6 months before they first received biologic or corticosteroid therapy. The outcome assessed was any VTE that occurred during the 12-month follow-up period. A multivariate logistic regression model was used to evaluate the effects of biologic, corticosteroid, and combination therapies (biologics and corticosteroids) on VTE risk. RESULTS: Three hundred twenty-five VTEs occurred during the study period (in 2.25% of patients receiving only corticosteroids, in 0.44% of patients receiving biologics, and in 2.49% of patients receiving combination therapy). Compared with patients receiving only corticosteroids, the odds ratio for VTE in patients receiving only biologics was 0.21 (95% confidence interval, 0.05-0.87) in the multivariate model, and the odds ratio for VTE in patients on combination therapy was 1.01. CONCLUSIONS: Compared with treatment with only a biologic agent, corticosteroid therapy is associated with a nearly 5-fold increase in risk for VTE. Combination therapy with corticosteroids and biologic agents was associated with the same risk for VTE as that of corticosteroids alone. Corticosteroids therefore appear to increase risk for VTE.

Comparison of noncontrast MRI magnetization transfer and T2 -Weighted signal intensity ratios for detection of bowel wall fibrosis in a Crohn's disease animal model.

PURPOSE: To compare the abilities of magnetization transfer magnetic resonance imaging (MT-MRI) and T2 -weighted signal intensity (T2 WSI) ratios to detect intestinal fibrosis in a Crohn's disease animal model. MATERIALS AND METHODS: Ten rats ("Group 1") received one trinitrobenzenesulfonic acid enema to induce acute colonic inflammation, while 10 additional animals ("Group 2") received multiple enemas to induce colonic inflammation and fibrosis. Gradient recalled-echo MT-MRI (5 and 10 kHz off-resonance) and T2 -weighted spin-echo imaging were performed 2 days after the last enema. MT ratios (MTR) and T2 WSI ratios were calculated in the area of greatest colonic thickening. Bowel wall MTR, bowel wall MTR normalized to paraspinous muscle MTR ("normalized MTR"), and T2 WSI ratios were compared between animal groups using Student's t-test. RESULTS: At 10 kHz off-resonance, mean bowel wall MTR for Group 1 was 24.8 ± 3.1% vs. 30.3 ± 3.2% for Group 2 (P = 0.001). Mean normalized MTR was 0.45 ± 0.05 for Group 1 and 0.58 ± 0.08 for Group 2 (P = 0.0003). At 5 kHz off-resonance, mean bowel wall MTR for Group 1 was 34.7 ± 5.2% vs. 40.3 ± 3.6% for Group 2 (P = 0.015). Mean normalized MTR was 0.53 ± 0.08 for Group 1 and 0.64 ± 0.07 for Group 2 (P = 0.003). Mean T2 WSI ratio was 5.32 ± 0.98 for Group 1 and 3.01 ± 0.66 for group 2 (P < 0.0001). Mean T2 WSI ratio/MTR (10 kHz off-resonance) was 12.06 ± 2.70 for Group 1 and 5.22 ± 1.29 for Group 2 (P < 0.0001), with an ROC c-statistic of 0.98. CONCLUSION: MTR and T2 WSI ratios detect bowel wall fibrosis in a Crohn's disease animal model.

US elastography-derived shear wave velocity helps distinguish acutely inflamed from fibrotic bowel in a Crohn disease animal model.

PURPOSE: To determine if acoustic radiation force impulse elastography-derived bowel wall shear wave velocity (SWV) allows distinction of acutely inflamed from fibrotic intestine in a Crohn disease animal model. MATERIALS AND METHODS: University Committee on the Use and Care of Animals approval was obtained. An acute inflammation Crohn disease model was produced by treating eight Lewis rats with a single administration of trinitrobenzenesulfonic acid (TNBS) enema, with imaging performed 2 days later in the surviving six rats. Colonic fibrosis in an additional eight Lewis rats was achieved by administering repeated TNBS enemas during 4 weeks, with imaging performed in the surviving seven rats 7 days later to allow acute inflammation resolution. Nine transcutaneous bowel wall SWV measurements were obtained from the colon in all rats without and with applied strain. Mean SWVs without and with applied strain were compared between animal cohorts by using the Student t test, and receiver operating characteristic (ROC) curves were created to assess diagnostic performance. RESULTS: Mean bowel wall SWVs were significantly higher for fibrotic versus acute inflammation cohort of rats at 0% (3.4 ± 1.1 vs 2.3 ± 0.5 m/sec; P = .047) and 30% (6.3 ± 2.2 vs 3.6 ± 0.9 m/sec; P = .02) applied strain. Both acute inflammation and fibrotic cohort of rats demonstrated linear increases in mean SWV with increasing applied strain, with significantly different mean slopes (P = .02) and y-intercepts (P = .02). The area under the ROC curve of the SWV ratio (mean SWV/applied strain) for differentiating histopathologically confirmed fibrotic from inflamed bowel was 0.971. CONCLUSION: Bowel wall SWV helps distinguish acutely inflamed from fibrotic intestine in a Crohn disease animal model.

Therapeutic high-density barium enema in a case of presumed diverticular hemorrhage.

Many patients with lower gastrointestinal bleeding do not have an identifiable source of bleeding at colonoscopy. A significant percentage of these patients will have recurrent bleeding. In many patients, the presence of multiple diverticula leads to a diagnosis of presumed diverticular bleeding. Current treatment options include therapeutic endoscopy, angiography, or surgical resection, all of which depend on the identification of the diverticular source of bleeding. This report describes a case of recurrent bleeding in an elderly patient with diverticula but no identifiable source treated successfully with barium impaction therapy. This therapeutic modality does not depend on the identification of the bleeding diverticular lesion and was well tolerated by our 86-year-old patient.