Search

Background There is a growing need for biomarkers of inflammation to monitor and predict therapeutic outcome in Crohn’s disease (CD) patients. We aimed to evaluate whether the level of circulating blood cells, expressed as ratios (NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; ELR, eosinophil-to-lymphocyte ratio and ENLR, eosinophil*neutrophil-to-lymphocyte ratio), could be used as early prognostic biomarker of endoscopic response (ER) in patients starting biological therapy with infliximab, adalimumab, vedolizumab and ustekinumab. Association with steroid-free clinical remission at week 54 and endoscopic disease activity at baseline, as well as other variables related with ER were also evaluated as secondary outcomes. Methods Consecutive patients with CD who started biological therapy between June 2016 and July 2019 in two Italian tertiary centers were enrolled. We used multivariate analysis to evaluate whether NLR, PLR, ELR and ENLR at baseline and after 12 weeks of treatment could predict ER (SES-CD ≤ 2 or SES-CD ≤ 2 and Rutgeerts score i0-i1) after 54 weeks of treatment. Receiver operating characteristic (ROC) curves were generated to identify the area under the curve (AUC) and find the best cut-off values of NLR, PLR, ELR and ENLR for the prediction of ER. To enhance the diagnostic performance, these ratios were combined with baseline Harvey Bradshaw Index (HBI) in a prognostic model. Results 107 patients were included. Patients who achieved ER had significantly lower baseline NLR (p = 0.025), ELR (p = 0.013) and ENLR (p = 0.021) compared with those without ER; similar results were confirmed after 12 weeks of treatment for ELR (p = 0.006) and ENLR (p = 0.026). AUC was 0.64 (p = 0.003), 0.67 (p = 0.006) and 0.65 (p = 0.014) for NLR, ELR and ENLR, respectively. The best cut-offs values were included in the HBI-based prognostic model: 81.8% of patients with HBI ≥ 5 and higher ratios did not achieve ER, while all patients with HBI < 5 and lower ratios reached ER. Interestingly, ENLR was found to be linearly associated with SES-CD at baseline (p = 0.035). Conclusion To the best of our knowledge, this is the first study assessing NLR, ELR and ENLR as predictive markers of therapeutic response in CD patients. Low levels of these ratios can predict endoscopic response and could be used in clinical practice for a better management of CD patients needing biological therapy. Further prospective studies are needed to confirm these results in larger cohorts.

Background The role of Vedolizumab (VDZ) as therapeutic option for the postoperative recurrence of Crohn’s disease (CD) following ileocolonic resection is currently unknown. We aimed to assess the effectiveness of VDZ in this setting. Methods All consecutive CD patients with an available baseline colonoscopy at 6-12 months from the ileocolonic resection and treated with VDZ for the postoperative recurrence after the baseline colonoscopy were extracted from the cohort of the Sicilian Network for Inflammatory Bowel Diseases (SN-IBD). The primary outcome was endoscopic success, assessed at the first colonoscopy following initiation of VDZ. In patients with Rutgeerts score i0 or i1 at baseline, endoscopic success was defined by maintenance of Rutgeerts score i0 or i1; in patients with Rutgeerts score ≥ i2 at baseline, it was defined as reduction of at least one point of Rutgeerts score. The secondary outcome was clinical failure, assessed at one year and at the end of follow-up. Results Seventy patients were included (median follow-up: 23.5 months). All 9 patients without endoscopic recurrence at baseline (Rutgeerts score i0 or i1) and available post-treatment colonoscopy maintained a Rutgeerts score i0 or i1 (treatment success: 100%). In patients with endoscopic recurrence (Rutgeerts score ≥ i2 at baseline), a reduction of at least one point in the Rutgeerts score was obtained in 20 out of 42 patients (47.6%). By combining the two subgroups, the overall endoscopic success was achieved in 29 out of 51 patients (56.9%). Furthermore, 14 out of 42 patients (33.3%) with endoscopic recurrence at baseline achieved a Rutgeerts score i0 or i1 at the subsequent colonoscopy. Clinical failure was reported in 13/70 patients (18.6%) at one year, and in 23/70 patients (32.9%) at the end of follow-up. A new resection was required in 7/70 patients (10.0%). Conclusion VDZ may be a therapeutic option for the management of postoperative recurrence of CD. Further studies are needed to confirm these results.

Background In a 6-year, multicenter, prospective nested case–control study, we aimed to evaluate risk factors for incident cancer in inflammatory bowel disease (IBD), when considering clinical characteristics of IBD and immunomodulator use. The secondary end point was to provide characterization of incident cancer types. Methods All incident cases of cancer occurring in IBD patients from December 2011–2017 were prospectively recorded in 16 Italian Group for the Study of Inflammatory Bowel Disease units. Each of the IBD patients with a new diagnosis of cancer was matched with 2 IBD patients without cancer, according to IBD phenotype (ulcerative colitis [UC] vs Crohn’s disease [CD]), age (±5 years), sex. Risk factors were assessed by multivariate logistic regression analysis. Results Cancer occurred in 403 IBD patients: 204 CD (CD cases), 199 UC (UC cases). The study population included 1209 patients (403 IBD cases, 806 IBD controls). Cancer (n = 403) more frequently involved the digestive system (DS; 32%), followed by skin (14.9%), urinary tract (9.7%), lung (6.9%), genital tract (6.5%), breast (5.5%), thyroid (1.9%), lymphoma (2.7%, only in CD), adenocarcinoma of the small bowel (SBA; 3.9%, 15 CD, 1 pouch in UC), other cancers (15.9%). Among cancers of the DS, colorectal cancer (CRC) more frequently occurred in UC (29% vs 17%; P < 0.005), whereas SBA more frequently occurred in CD (13% vs 6.3% P = 0.039). In CD, perforating (B3) vs non-stricturing non-perforating (B1) behavior represented the only risk factor for any cancer (odds ratio [OR], 2.33; 95% confidence interval [CI], 1.33–4.11). In CD, risk factors for extracolonic cancer (ECC) were a B3 vs B1 and a stricturing (B2) vs B1 behavior (OR, 2.95; 95% CI, 1.62–5.43; OR, 1.79; 95% CI, 1.09–2.98). In UC, risk factors for ECC and for overall cancer were abdominal surgery for UC (OR, 4.63; 95% CI, 2.62–8.42; OR, 3.34; 95% CI, 1.88–5.92) and extensive vs distal UC (OR, 1.73; 95% CI, 1.10–2.75; OR, 1.99; 95% CI, 1.16–3.47). Another risk factor for ECC was left-sided vs distal UC (OR, 1.68; 95% CI, 1.00–2.86). Inflammatory bowel disease duration was a risk factor for skin and urinary tract cancers. Conclusions Perforating CD, extensive UC, and abdominal surgery for UC were identified as risk factors for overall incident cancer and for ECC. The clinical characteristics associated with severe IBD may increase cancer risk.

Background No real-life study aiming at comparing at the same time the effectiveness of vedolizumab (VDZ), adalimumab (ADA), and golimumab (GOL) in Ulcerative colitis (UC) is currently available. Methods Data of consecutive patients with UC treated with VDZ, ADA, and GOL from June 2015 to December 2018 were extracted from the cohort of the Sicilian Network for Inflammatory Bowel Disease (SN-IBD). A three-arms propensity score-adjusted analysis was performed to reduce bias caused by imbalanced covariates at baseline, including the proportion of TNF-α inhibitor naïve and non-naïve patients, using the Inverse Probability of Treatment Weighting (IPTW) method. The effectiveness was evaluated at 8 weeks, 52 weeks, and as treatment persistence at the end of follow-up. The clinical endpoints were steroid-free clinical remission (partial Mayo score Results A total of 463 treatments (VDZ: n = 187; ADA: n = 168; GOL: n = 108) were included, with a median follow-up of 47.6 weeks (IQR 20.0–85.9). At 8 weeks, a clinical benefit was achieved in 70.6% patients treated with VDZ, in 68.5% patients treated with ADA, and in 67.6% patients treated with GOL (p = n.s. for all comparisons). After 52 weeks, VDZ showed better rates of clinical benefit compared with both ADA (71.6% vs. 47.5; OR: 2.79, 95% CI 1.63–4.79, p < 0.001) and GOL (71.6% vs. 40.2%; OR: 3.77, 95% CI 2.08–6.80, p < 0.001), while the difference between ADA and GOL was not significant. Cox survival analysis demonstrated that patients treated with VDZ had a reduced probability of treatment discontinuation compared with those treated with ADA (HR: 0.42, 95% CI 0.28–0.64, p < 0.001) and GOL (HR: 0.30, 95% CI 0.19–0.46, p < 0.001), while patients treated with ADA had a reduced risk of treatment discontinuation compared with those treated with GOL (HR: 0.71, 95% CI 0.50–1.00, p = 0.048). Post-treatment mucosal healing rates showed a numerical but non-significant difference in favour of VDZ (48.1%) compared with ADA and GOL (38.0% and 34.6%, respectively). Conclusion In the first study comparing at the same time the clinical effectiveness of VDZ, ADA, and GOL in UC patients via propensity score-adjusted analysis, VDZ was superior to both subcutaneous agents at 52 weeks and as treatment persistence, while ADA showed a superior treatment persistence compared with GOL.

Background Currently, there are no randomised controlled trials focussing on direct comparisons between biologics in Crohn’s disease (CD), while there is an unmet need to better understand and compare the effectiveness of these drugs. We performed a multicentre, real-life, comparison of the effectiveness of vedolizumab (VDZ) and adalimumab (ADA) in CD. Methods Data of consecutive patients with CD treated with VDZ and ADA from January 2016 to April 2019 were extracted from the cohort of the Sicilian Network for Inflammatory Bowel Disease (SN-IBD). A propensity score analysis was performed using the Inverse Probability of Treatment Weighting (IPTW) method. The effectiveness was evaluated at 12 weeks, 52 weeks, and as treatment persistence at the end of follow-up. The clinical endpoints were steroid-free clinical remission (Harvey–Bradshaw Index < 5 without steroid use) and clinical response (reduction of the Harvey–Bradshaw Index ≥ 3 points with a concomitant decrease of steroid dosage compared with baseline). The sum of the two outcomes was defined as a clinical benefit. The achievement of endoscopic response (a reduction of Simple Endoscopic Score for CD ≥ 50% compared with baseline) or mucosal healing (Simple Endoscopic Score for CD ≤ 2) was assessed after at least 6 months of treatment. Results A total of 585 treatments (VDZ: n = 277; ADA: n = 308) were included, with a median follow-up of 56.0 weeks (IQR 24.0–104.0). After 12 weeks, a clinical benefit was achieved in 64.3% patients treated with VDZ and in 83.1% patients treated with ADA (p = 0.11 in propensity score analysis), while at 52 weeks a clinical benefit was observed in 54.0% patients treated with VDZ and in 69.1% patients treated with ADA (p = 0.33 in propensity score analysis). The median treatment persistences for VDZ and ADA were 52 weeks and 64 weeks, respectively. Cox survival analysis weighted for IPTW showed no significant difference in the probability of treatment discontinuation between the two drugs (HR for VDZ=1.20; p = 0.34). Post-treatment endoscopic response and mucosal healing rates were similar between the two drugs (endoscopic response: 35.3% for VDZ and 25.5% for ADA, p = 0.15; mucosal healing: 31.8% for VDZ and 33.8% for ADA, p = 0.85). Conclusion This is the first real-life study comparing VDZ and ADA in CD patients via propensity score-adjusted analysis. Even if crude rates of all clinical outcomes were higher for ADA compared with VDZ, such differences were not significant when patients’ characteristics at baseline were weighted by propensity score.

No data on mortality in celiac disease are currently available in southern Europe. Our aim was to evaluate mortality and the cause of death in adult celiac disease in a Mediterranean area. In all, 228 adults with celiac disease were histologically diagnosed in our department from 1980 to 1997. Full information on their state of health was obtained in 216 of 228 patients. A tabulation of patient-years at risk was constructed in terms of age at diagnosis and the interval from diagnosis. Standardized mortality ratio was calculated by dividing the number of observed deaths by the number of expected deaths. Twelve deaths were observed, whereas 3.12 deaths were expected (SMR = 3.8; 95% CI 2-7). The increased mortality was mainly observed within four years from diagnosis (8 observed; 1.4 expected) (SMR = 5.8; 95% CI 2.5-11.5). Twelve tumors were observed (six lymphomas). In conclusion, mortality from adult celiac disease in our geographical area is increased compared with the general population, and this increased risk seems due to non-Hodgkin's lymphoma.

Concordance in the extent of disease among the family members of patients with Crohn's disease has not been widely investigated. Furthermore, the relationship between the site of the disease and familial occurrence has never been studied. Our aim was to evaluate the familial occurrence of Crohn's disease in the various sites. Nine hundred thirty-four patients with Crohn's disease, observed consecutively in two gastrointestinal departments, were investigated to determine first-degree familial incidence (in both Crohn's disease and ulcerative colitis). Whenever two or more members were attending the same clinic, only one was regarded as a propositus. The analysis, therefore, was carried out on 882 patients. The exact site of the disease was determined in all patients either at diagnosis or during the follow-up by colonoscopy and by small bowel enema. The rate of concordance in the extent of disease and familial occurrence in the various sites was evaluated and the difference was calculated by chi-square test. Sixty-one propositi were identified among all the patients. Forty-nine had familial occurrence for the same disease (concordant patients), whereas 12 had at least one relative with ulcerative colitis (discordant patients). In 44 propositi with only one relative affected, the rates of concordance in the extent of the disease were 84, 68, 18, and 0% respectively, for the ileum, the ileum-right colon, the ileum-total colon, and the colon. The number of propositi in the various sites was as follows: 4 of 162 (2.4%) patients with the disease located in the colon, 1 of 9 (11%) with the jejunum site, 24 of 380 (6.3%) with the ileum site, 16 of 165 (9.7%) with the ileum and right colon site, and 16 of 164 (9.7%) with the ileum and total colon site. The chi-square values of propositi distribution among other sites and the colon was, respectively, as follows: jejunum, 2.2 (N.S.); ileum, 3.4 (P = 0.06); ileum and right colon, 7.4 (P = 0.006); and ileum and total colon, 7.4 (P = 0.006). This study shows a pronounced concordance in the site of the disease for family members with Crohn's disease and suggests that familial occurrence in Crohn's disease is less frequent when the disease is located in the colon rather than elsewhere.

Background and aim: Intraoperatory radiofrequency ablation (RFA) of pancreas cancer has