Search

Background Post-marketing data are required to confirm the durability and the long-term benefit and safety of UST in CD in clinical practice. Our aims were: (1) to evaluate the retention rate of UST in CD patients and to identify predictive factors of UST discontinuation; (2) to assess UST short-term effectiveness; (3) to analyse the durability of the response to UST in the long-term; and (4) to evaluate the safety of UST in clinical practice. Methods Retrospective, multicentre study (>60 centres). Patients with active CD [(Harvey–Bradshaw (HBI) >4)] that received at least one dose of UST intravenously before July 2018 were included. Clinical activity plus biochemical parameters were assessed at every UST administration. Clinical remission was defined as HBI score ≤4, and clinical response as a decrease in HBI ≥3 points. Loss of efficacy was defined as reappearance of symptoms that led to intensify the treatment dose, add another medication to control CD, switching or surgery in patients with short-term remission. The retention rate of UST treatment and the cumulative incidence of loss of efficacy were evaluated by survival curves, and predictive factors were assessed by Cox-regression. The short-term response was evaluated at week 8 and after the induction (week 16). Factors associated with short-term remission were assessed by multivariate analysis. Adverse events were recorded. Data quality was assured by remote monitoring. Results 331 CD patients have been included up to date (Table 1). The incidence rate of UST discontinuation was 15% per patient-year of follow-up: 8%, 13% and 20% at 6, 12 and 18 months (Figure 1). Previous surgery was the only factor associated with a higher risk of UST discontinuation [Hazard ratio (HR) = 2.03, 95% confidence interval (CI) = 1.1–3.6]. Short-term efficacy is shown in Figure 2. Previous surgery (OR = 0.3, 95% CI = 0.2–0.6) and higher HBI score at baseline (OR = 0.8, 95% CI=0.8–0.9) were associated with an impaired response to UST at week 16. The cumulative incidence of loss of response was 32% per-patient-year of follow-up (Figure 3); A higher HBI score at baseline was associated with a higher risk of losing response (HR = 1.2, 95% CI = 1.1–1.3). Neither the concomitant treatment with immunosuppressants nor the number of previous biologics were associated with UST short- and long-term benefit. Thirty adverse events were reported in 25 (7%) patients (Table 2). Conclusion Sustain is the largest real clinical practice study of UST to treat CD patients with the longest follow-up reported to date. UST was demonstrated to be effective in real-world use in the short and long run. Safety was consistent with the known profile of UST.

Background It has been suggested that biologic therapy may increase the risk of postoperative complications in inflammatory bowel disease (IBD), but the evidence is scarce. Our aim was to evaluate whether the treatment with anti-TNF agents, ustekinumab or vedolizumab increase the risk of complications after surgery. Methods IBD patients undergoing intra-abdominal surgery between 1st January 2009 and 31st December of 2019 were retrospectively selected. Data collection included clinical characteristic of IBD, biochemical parameters and surgical aspects. Postoperative complications (PC) were defined as those occurring within 30 days after surgery. Exposed cohort (EC): Patients who received the last dose of the biologic within 3 months before surgery. Non-exposed cohort (NEC): Patients who did not receive biologic treatment within 3 moths prior to surgery. Predictive factors for PC and for infections were identified by logistic regression analyses. A genetic matching score was performed to balance the clinical characteristics of both groups. Results A total of 1,535 surgeries performed in 37 centres were included: 81% in Crohn’s disease, 18% in ulcerative colitis and 1% in unclassified-IBD patients. A total of 711 surgeries (46.3%) had been exposed to biologics (583 under anti-TNF therapy, 58 under vedolizumab and 69 under ustekinumab) and 824 surgeries (53.7%) the NEC. PC were reported in 38% (n=267) of patients in the exposed cohort and in 34% (n=280) of patients in the non-exposed one (p=0.15), including dehiscence, infection, obstruction, ileus, bleeding, thrombosis, fistula and evisceration. The most frequent complications were infections (48% of all the cases). A 30-day hospital readmission was needed in 7% (n=110) of the patients, and 2% (n=29) required a new surgery with no differences (p>0.05). Multivariate analysis for PC and infections is presented in table 1. The frequencies of PC for each biologic in the univariate analysis are represented in figure 1. No specific treatment was associated to PC or infections in multivariate analysis. Conclusion Preoperative administration of biological therapy does not seem to be a risk factor for overall PC, although it may be for postoperative infections.

Background The urinary metabolome of patients with Crohn’s disease (CD) differs significantly from healthy subjects and, among other features, reflects the specific gut dysbiosis affecting these patients. However, most of the studies included established and treated CD patients. Our aim was to characterize the urinary metabolome of onset and treatment-naïve CD patients and to identify the metabolic profile related to the different CD clinical classifications. Methods Patients newly diagnosed with CD (n=131) were prospectively included. Control healthy subjects (HC, n=338) were recruited among the general population and matched for sex, age and BMI to the IBD subjects. Fasting urine was obtained before starting any treatment. Metabolomic analysis was performed by proton nuclear magnetic resonance (1H NMR). We performed a comparative assessment of the urinary metabolome profile using a linear regression model for each metabolite, including sex, age, BMI, and smoking habit as covariates to control for confounding. The different subgroup comparisons within CD were made as follows: (1) CD; (2) CD location (Montreal Classification): L1 (ileal) + L4 (ileal and upper-intestinal), L2 (colonic) and L3 (ileocolonic); (3) endoscopic CD activity: 0, 1, 2 and 3; and (4) CD phenotype: B1 (inflammatory), B2 (stricturing) and B3 (penetrating), versus HC. In addition, data analysis was carried out using partial least squares-discriminate analysis (PLS-DA) to determine class membership based on distinct metabolomic profile. Results The primary characteristics of the CD patients and HC are shown in Table 1. Several metabolites were identified to be differently abundant in each group (Table 2). These metabolites are involved in relevant processes related to energy and aminoacids metabolism, and also include gut-derived metabolites. The PLS-DA model separated patients within the different clinical subgroups (Figures 1–4). Figures 1–4(b) show the main metabolites involved in each group separation. Many of these metabolites are in accordance with the differential metabolites obtained using the univariate analysis (Table 2), showing the potential of this approach to group CD patients and to identify potential biomarkers. Conclusion Analysis of urinary metabolites can help to understand the etiopathological mechanisms in CD. It has the potential to provide a non-invasive means of diagnosing CD, and can differentiate between CD clinical expressions.

Background Lack of adherence in patients with inflammatory bowel disease (IBD) is a relevant problem in our clinical practice. Non-adherence to anti-TNF increases healthcare costs. The aim of this study was both to measure adherence and also to study the factors and consequences related to non-adherence in patients with IBD under maintenance treatment with infliximab (IFX). Methods A prospective, observational cohort study was designed and patients with IBD under treatment with IFX were consecutively included. Adherence was measured with pharmacy refills and Morisky Medication Adherence Scale-8 (MMAS-8). Patients were systematically asked about adherence and the mean displacement days, the accumulated delay, the mean delay and the medication possession ratio (MPR) were calculated. MPR was calculated by dividing the number of days supplied within the refill interval by the number of days in the actual refill interval over 24 months. Potencial risk factors for non-adherence were evaluated: age, gender, disease duration, site of disease, behaviour of Crohn’s disease (CD), smoking status, educational level, marital status, type of housing, extraintestinal manifestations, previous surgery, concomitant treatments, anxiety and depression. Relapse was defined as a Harvey Bradshaw score > 4 in CD and a partial Mayo > 2 in ulcerative colitis (UC). The Mann-Whitney Wilcoxon U Test was used to distinguish the intergroup differences. Correlations were evaluated with Spearman rank correlation coefficients. Results Ninety patients with mean age 46 years (range: 22–85) were included. 49 (54.4%) were women and 63 (70%) had CD. Anxiety occurred in 38.9% of patients and depression in 78.9%. Three quarters of the patients were in clinical remission at inclusion. After 24 months of follow-up, 82 (91.1%) had delayed some dose of treatment, 35 (38.9%) had delayed on at least 7 days some dose of treatment, and 11 (12.2%) had not received some of the scheduled doses. The MPR was 87% (range 46–100). Lack of adherence was related to loss of response to IFX in CD (p = 0.035), but not in UC (p=0.078). In UC, lack of adherence was related with anxiety (p=0.046). The Spearman’s correlation between older age and non-adherence was 0.53 (p=0.006) in UC and 0.1 (p=0.308) in CD. Conclusion Lack of adherence is related to loss of response to IFX in CD. Non-adherence to IFX is high and strongly associated with age in UC. Older patients with UC are more prone to lack of adherence.

Background Biological drugs are being increasingly used for the treatment of inflammatory bowel diseases (IBD) in elderly patients. Despite the particular characteristics of this population subgroup, the efficacy and safety of these treatments in real clinical practice is poorly evaluated. Methods Retrospective and multicenter study of GETECCU, carried out in 28 Spanish hospitals. Patients with IBD who started biological treatment (Infliximab, Adalimumab, Golimumab, Ustekinumab or Vedolizumab) aged 65 years or older were included. Efficacy (clinical- at the criteria of the responsible physician-, biochemical and endoscopic) was assessed at 12-14 weeks and at 52 weeks of treatment. Adverse effects such as tumors or serious infections were also recorded. Results A total of 570 patients were included, baseline characteristics are shown in Table 1. Biologics used were: Infliximab (214, 37.5%), Adalimumab (167, 29.3%), Golimumab (16, 2.8%), Ustekinumab (73, 12.8%) and Vedolizumab (100, 17.5%). After 12-14 weeks of treatment, in 38.7% (220) of the cases clinical remission had been achieved and in 47.7% (270) there was clinical response without remission. However, 80 patients (13.9%) had no response, resulting in treatment discontinuation due to primary failure. At week 52, only 379 patients (66.5%) continued on biological treatment: 216 (57%) were in clinical remission (216, 57.0%) while 129 (34%) had response without remission and 34 (9%)had no response. In addition, 119 patients (21%) had an endoscopic study performed: 47 (39,5%) presented with endoscopic remission, 38 (31,9%) with mild activity, 28 (23,5%) with moderate activity and 6, (5.1%) with severe activity. At the end of the follow-up, only 60% of the patients continued on biological treatment, being the reason for withdraw lack of efficacy or due to the report of adverse side effects. Regarding treatment safety in this population, 12.1% (68 patients) suffered an infectious complication with a microbiological diagnosis, requiring hospitalization in 62.1% of the cases. In addition, 39 patients (6.9%) were diagnosed with a tumor until the end of the follow-up, noting that 34.2% of the cases continued on biological therapy after the diagnosis. Likewise, in 25 patients (36.8%) this infection forced discontinuation of biological treatment. Finally, 10 patients stopped biological treatment due to a serious adverse reaction to it Conclusion Response rates to biological treatment in elderly patients are similar to those described in the general population, with approximately one third of failures happening during the first year. However, a remarkable proportion of patients developed a serious adverse effect that could be related to treatment

Background The lack of interaction between immune system cells and microbiota is considered a key driver of the pathogenesis of inflammatory bowel disease (IBD). The association of age-associated B cells (ABC) from antigen-experienced B cells and autoimmunity is now well established. The increased demands in lipid metabolism during immune response and cell activation upon autoimmune signals leads to lipid tag modifications of proteins. In IBD patients, we observed an impairment of B cell development, identifying a transitional B cell subset expressing CD38Hi, CD24Hi and CD19+. We hypothesize that metabolic alterations of proteins and lipids during immune cell activation might be a relevant mechanism of autoimmune disease. Methods An IBD cohort of patients in clinical remission under anti-TNF treatment was included in order to test metabolic and proteomic changes in peripheral blood mononuclear cells (PBMCs). Proteomic analysis was performed by Mass Spectrometry using a label free quantitative method (SWATH-MS analysis) and lead to identify a number of proteins underregulated (< 0.6) and overexpressed (> 1.5) fold change respect to control. Lipidomics of serum samples analyzed by proton Nuclear Magnetic Resonance spectroscopy (NMR)2 and multivariant statistics. Results Serum samples were obtained in Crohn disease (CD) n = 18, ulcerative colitis (UC) n = 9, before biological infusion and healthy controls (CTRL) n = 10. Proteins related with lipid metabolism like APOC2 for CD and APOB, APOA1 for UC, were differentially modulated. Significant differences in non-polar metabolites and lipids that phenotypically define CD and UC groups respect to CTRL were observed (Fig. 1) Fig. 1 Orthogonal-Partial Least Square-Discriminant Analysis (OPLS-DA) of the global NMR data of the serum samples of groups CD and CTRL. Indeed, proteins associated with complement activation like Complement Factor 1 and Complement C1q subcomponent subunit B were differentially expressed in CD group. These results were confirmed by lipidomic analysis, in which palmitic (Fig. 2) as well as other candidate lipids were more abundant in IBD groups than CTRL. Fig 2. Differences of palmitic acid concentration in CD and UC patients vs CTRL (Bonferroni test showed statistical difference between CTRL vs. CD and CTRL vs. UC with p = 0.02 and p = 0.004 respectively). Conclusion Our study detected high levels of proteins regulating lipid metabolism and palmitic in IBD patients. This observation can be interpreted as part of deleterious alterations of proteins through palmitoylation during B-cell activation and might be a relevant mechanism of disease. Further exploration of palmitate catabolism might shed light to the mechanisms that result in immunometabolic disorders.

Background Ulcerative colitis (UC) is a chronic, progressive and disabling disease with a complex pathology of unknown aetiology influenced by genetic, environmental and microbiota factors that lead to an immunological and inflammatory response in the colon. Janus Activated Kinase (JAK) family plays a key role in modulating the adaptive and innate inflammatory response. The JAK/STAT pathway involvement in UC has been demonstrated in both animal models and human studies. Thus, overexpressed JAK-3 has been detected in the intestine of patients with UC, suggesting a key role in their pathophysiology and the inhibition of TYK-2 in animal models resulted in an improvement of the disease, which would explain its implication in the inflammatory process. We hypothesise here that there could be an activation of JAK-3 and TYK-2 signalling pathways in UC patients. Thus, we aimed to detect the activation of both signalling pathways by means of western-blot studies in UC patient samples Methods A prospective, observational single-centre study was designed. Inclusion criteria were adult patients with endoscopic active UC (more than Mayo-0) confirmed in a programmed colonoscopy. All patients signed informed consent. Samples were obtained from overstock of routine biopsies in the more severe segment affected of the large bowel. Tissues were homogenised and processed in order to obtain cell lysates by employing RIPA buffer and ultrasounds. The degree of activation of the JAK-3 and TYK-2 pathways was measured by detecting the phosphorylation of both targets as well as of STAT1, STAT3, STAT4, STAT5 and STAT6 through western blot by employing specific antibodies for total and phosphorylated proteins. Results 19 UC patients were consecutively included. Mean age was 46 years old. 53% were female, 47% were extensive colitis (E3) and 53% left-side colitis (E2). Regarding endoscopic activity, 26% had Mayo-1, 53% Mayo-2, and 21% Mayo-3. Immunoreactive bands for both phosphorylated JAK-3 and TYK-2 were detected in the biopsies from UC patients, evidencing that colonic inflammation leads to an activation of both targets. The study of STATs phosphorylation showed immunoreactive bands for phosphorylated forms of STAT1, STAT3, STAT4, STAT5 and STAT6 confirming the activation of both signalling-pathways in these patients (Figure 1). Conclusion The developed translational workflows involving basic/clinical research confirm the activation of both JAK-3 and TYK-2-dependent signalling pathways in UC patients, validating both kinases as targets for treating UC. The developed methodology allows studying the target engagement for future JAK-3/ TYK-2 inhibitors employed in clinical trials.

Background Inflammatory Bowel Disease (IBD) is a multifactorial pathology with an increasing incidence. There are environmental factors, many unknown, that participate in its development. There is no study having assessed a possible relationship with residential radon exposure, which is very high in the study area. The aim of this study was to analyze if residential radon concentration measured at a municipal level is associated with a higher incidence of IBD and its characteristics (clinical or demographic). Methods We performed an ecological study where we included all incident cases of IBD in the area of Santiago de Compostela between January and December 2017 in order to estimate municipal incidence rates. Radon levels at a municipal level were obtained from the Galician Radon Map and correlated with demographic factors and type of IBD. We used the Spearman’s correlation coefficient to test the existence of any association. Results 96 patients were included, 63 (65.6%) with Ulcerative Colitis, 29 (30.25) with Crohn’s Disease and 4 (4.2%) with indeterminate colitis. Median age was 41 (IQR: 33.5 to 56 years), and 50.0% were women. The incidence rate per 100.000 inhabitants-year in the study area was 21.6 cases. The median radon concentration was 104.9Bq/m3 (IQR: 91.0 to 154.6), without statistically significant differences in function of the location of the house (rural vs. urban) nor the type of edification (flat vs house), p >0.05. There were no statistically significant differences on the type of IBD developed (ulcerative colitis, Crohn’s disease or indeterminate colitis) regarding radon levels either (p>0.05). There were no statistically significant differences (p>0.05) between radon and sex of IBD cases. No correlation between radon levels and age of the individuals was observed (Spearman’s rho = -0.13, p-value 0.2), nor radon levels variation by age groups (p>0.05). There was no correlation between radon concentration and cumulative incidence of IBD at municipal level (Spearman’s rho = 0.13, p-value 0.5), as it is shown in figure 1. Conclusion In the area of Santiago de Compostela there is a higher incidence of IBD in comparison with previous studies taking western countries as reference. It is possible that some environmental risk factors, could be responsible of this difference. In this study we have not found any correlation with municipal average radon concentration and incidence of IBD or any of its types.