Search

Many patients with Crohn's disease (CD) require fecal diversion. To understand the long-term outcomes, we performed a multicenter review of the experience with retained excluded rectums.We reviewed the medical records of all CD patients between 1990 and 2014 who had undergone diversionary surgery with retention of the excluded rectum for at least 6 months and who had at least 2 years of postoperative follow-up.From all the CD patients in the institutions' databases, there were 197 who met all our inclusion criteria. A total of 92 (46.7%) of 197 patients ultimately underwent subsequent proctectomy, while 105 (53.3%) still had retained rectums at time of last follow-up. Among these 105 patients with retained rectums, 50 (47.6%) underwent reanastomosis, while the other 55 (52.4%) retained excluded rectums. Of these 55 patients whose rectums remained excluded, 20 (36.4%) were symptom-free, but the other 35 (63.6%) were symptomatic. Among the 50 patients who had been reconnected, 28 (56%) were symptom-free, while 22(44%) were symptomatic. From our entire cohort of 197 cases, 149 (75.6%) either ultimately lost their rectums or remained symptomatic with retained rectums, while only 28 (14.2%) of 197, and only 4 (5.9%) of 66 with initial perianal disease, were able to achieve reanastomosis without further problems. Four patients developed anorectal dysplasia or cancer.In this multicenter cohort of patients with CD who had fecal diversion, fewer than 15%, and only 6% with perianal disease, achieved reanastomosis without experiencing disease persistence.Patients with distal Crohn’s disease often undergo colon resection with a stoma to divert the intestinal stream from the rectum in hopes of achieving sufficient healing to allow ultimate re-establishment of intestinal continuity. Patients and practitioners alike should be aware of the long-term success rates of this procedure. Our retrospective study of 197 patients found that half required later proctectomy and an additional one-quarter remained symptomatic with excluded rectums. Only 14% remained symptom-free after reanastomosis, and only 6% if perianal disease was the initial surgical indication. These data provide estimation of long-term surgical outcomes.

International audience; Background and AimsThe identification of Crohn's disease (CD)-associated adherent and invasive Escherichia coli (AIEC) is time-consuming and requires ileal biopsies. We aimed to identify a faster and less invasive methods to detect ileal colonization by AIEC in CD patients.MethodsCD patients requiring ileo-colonoscopy were consecutively enrolled in this prospective multicenter study. Samples from saliva, serum, stools, and ileal biopsies of CD patients were collected.ResultsAmong 102 CD patients, the prevalence of AIEC on ileal biopsies was 24.5%. The abundance and global invasive ability of ileal-associated total E. coli were respectively ten-fold (p = 0.0065) and two-fold (p = 0.0007) higher in AIEC-positive (vs. AIEC-negative), while abundance of total E. coli in the feces was not correlated with AIEC status in the ileum. The best threshold of ileal total E. coli was 60 cfu/biopsy to detect AIEC-positive patients, with high negative predictive value (NPV) (94.1%[80.3–99.3]), while the global invasive ability (>9000 internalized bacteria) was able to detect the presence of AIEC with high positive predictive value (80.0% [55.2–100.0]). Overall, 78.1% of the AIEC + patients were colonized by two or less different AIEC strains. The level of serum anti-total E. coli antibodies (AEcAb) was higher in AIEC-positive patients (p = 0.038) with a very high negative predictive value (96.6% [89.9–100.0]) (p = 0.038) for a cut-off value > 1.9 × 10−3.ConclusionsMore than two thirds of AIEC-positive CD patients were colonized by two or less AIEC strains. While stools samples are not accurate to screen AIEC status, the AEcAb level appears to be an attractive, rapid and easier biomarker to identify patients with Crohn's disease harboring AIEC.

No abstract available

International audience; Background Phase III trials have demonstrated the efficacy and safety of ustekinumab in ulcerative colitis (UC), but few real-life long-term data are currently available. Aims To assess the real-world effectiveness and safety of ustekinumab in patients with UC. Methods From January to September 2019, all consecutive patients with active UC treated with ustekinumab in a GETAID centre were included. Patients were evaluated at week 52. Remission was defined as a partial Mayo Clinic score

We assessed the effectiveness of switching from intravenous to subcutaneous infliximab in patients with inflammatory bowel diseases (IBDs) treated with or without intensified intravenous regimen.In this multicenter observational study, IBD patients in clinical remission (partial Mayo score ≤2 or Harvey-Bradshaw index ≤4) were switched to a unique dose of subcutaneous infliximab (120 mg every other week). Pharmacological and biological data were collected at baseline, visit 1 (4-8 weeks postswitch), visit 2 (8-16 weeks postswitch), and visit 3 (16-24 weeks postswitch). Relapse was defined as clinical relapse or fecal calprotectin increase ≥150 μg/g compared with baseline.Among 184 eligible patients, 72.3% (n = 133 of 184) agreed to switch to subcutaneous infliximab. At visit 3, a relapse occurred in 10.2% (n = 6 of 59), 7.3% (n = 3 of 38), 16.7% (n = 3 of 18), and 66.7% (n = 10 of 15) (P.001) of patients receiving 5 mg/kg every 8 weeks, 10 mg/kg every 8 weeks, 10 mg/kg every 6 weeks, and 10 mg/kg every 4 weeks, respectively. Dose escalation to 240 mg every other week led to recapture clinical remission in 93.3% (n = 14 of 15). Infliximab serum levels increased after the switch (P.0001) except for patients receiving 10 mg/kg every 4 weeks. In multivariable analysis, 10 mg/kg every 4 weeks regimen (odds ratio, 12.4; 95% confidence interval, 1.6-98.4; P = .017) and fecal calprotectin250 μg/g at baseline (odds ratio, 5.4; 95% confidence interval, 1.1-27.6; P = .042) had a higher risk of relapse as well as reduced (41.7%) or stable (36.8%) infliximab serum levels between baseline and visit 1 compared with increased serum levels (12.7%) (P = .020 and P = .019, respectively). Patients' acceptability (10-point scale) was improved by the switch (6.9 ± 1.6 vs 8.6 ± 1.4; P.0001). No severe adverse event was reported.Switching from intravenous to subcutaneous infliximab 120 mg every other week is safe and well accepted, leading to a low risk of relapse in IBD patients except for those receiving 10 mg/kg every 4 weeks requiring 240 mg every other week.

The management of colorectal stricture complicating inflammatory bowel disease [IBD] remains a challenging condition. Stricture raises concern about neoplastic complications, which cannot be fully ruled out by negative endoscopic biopsies. Also, impassable strictures restrict the endoscopic monitoring of upstream disease activity and dysplasia. Surgery remains the ‘gold standard’ treatment for colonic strictures but is associated with high morbidity. Over the past few decades, our therapeutic arsenal for IBD has been reinforced by biologics and therapeutic endoscopy. Few studies have focused on colonic strictures, and so current therapeutic strategies are based on a low level of evidence and applied by analogy with the treatment of ileal strictures. With a view to facilitating the decision making process in clinical practice, we reviewed the literature on the epidemiology, natural history, and management of colonic strictures in IBD.

Summary Introduction Most anal fistulas are crypto-glandular. Nevertheless, anal fistulas can reveal Crohn's disease (CD). The aim of our study was to evaluate the risk of developing CD in patients undergoing surgery for anal fistula. Patients and methods All patients undergoing surgery for anal fistula in our center between January 1, 2008 and January 31, 2017 were identified through a prospective administrative database. Demographic, clinical, and laboratory data were retrospectively collected. Results Ninety-three patients underwent anal exploration under general anesthesia. The median age at diagnosis of fistula was 43 years (IQR, 34-56) and 27% (n = 29) were women. Twenty-seven percent (n = 16) had had at least one previous fistula episode. After a median follow-up of 16.8 months (IQR, 7.2–42.0), seven (7.4%) patients were diagnosed with CD. The median time between the diagnosis of fistula and that of CD was 7.6 months (IQR, 2.7, 26.1). Chronic diarrhea (P = 0.0003), weight loss (P = 0.001), and chronic abdominal pain (P = 0.002) were associated with the diagnosis of CD. Characteristics of the fistulas (number, simple/complex, abscess), smoking, extra-digestive manifestations of CD, or a family history of IBD were not associated with the diagnosis of CD. Conclusion A medical history of anal fistula surgery resulted in the diagnosis of CD in 7% of cases. Weight loss and the presence of digestive symptoms were associated with the diagnosis of CD. These elements could be used to select patients requiring endoscopic exploration after anal fistula.

International audience

Background & Aims Mucosal healing (MH) has been associated with good outcomes of patients with Crohn's disease (CD). It is not clear what levels of endoscopic healing, based on CD endoscopic index score (CDEIS), associate with different courses of disease progression. We assessed long-term outcomes of patients with CD according to different levels of MH. Methods We performed a retrospective study of 84 patients with CD and MH who received biologic therapy (80% with infliximab) from 2008 through 2015 at 2 university hospitals in France and compared outcomes of patients with CD endoscopic index scores (CDEISs) of 0 vs CDEISs greater than 0 but less than 4. Patients were followed until treatment failure or through June 2016. The primary outcome measure was treatment failure, defined by the need for biologic optimization, initiation of corticosteroids, or a Harvey-Bradshaw score above 4 associated with change in treatment, CD-related hospitalization, and/or intestinal resection. Results After a median follow-up time of 4.8 years (interquartile range, 2.1–7.2), 27 patients (32%) had treatment failure and 3 patients (3.6%) underwent an intestinal resection. Rates of treatment failure were 25% in patients with a CDEIS of 0 and 48% in patients with CDEISs greater than 0 but less than 4 (P = .045). Median times to treatment failure were 21 months (interquartile range, 5–43 months) in patients with a CDEIS of 0 and 13 months (interquartile range, 3.6–35 months) in patients with CDEISs greater than 0 but less than 4 (P = .047). None of the patients with a CEDIS of 0 underwent intestinal resection whereas 11% patients with CDEISs greater than 0 but less than 4 required intestinal resection (P = .031). Patients with a CDEIS of 0 also had a significant lower rate of CD-related hospitalizations than patients with CDEISs greater than 0 but less than 4 (3.5% vs 18%; P = .013). In multivariate analysis, CDEISs greater than 0 but less than 4 (vs CDEIS = 0) was the only factor associated with treatment failure (hazard ratio, 2.6; 95% CI, 1.2–5.8; P = .02). Conclusions Complete endoscopic healing (CDEIS = 0) is associated with better long-term outcomes than partial endoscopic healing (CDEIS = 1–4) in patients with CD, as well as fewer surgeries and hospitalizations and an overall decreased risk of treatment failure.

Epstein‐Barr virus‐related mucocutaneous ulcer lymphoma is a rare entity promoted by immunosuppression. It is less described in inflammatory bowel diseases, and mostly these are refractory diseases. CD30 acts to Epstein‐Barr virus (EBV) local proliferation and thus could be an interesting target. Brentuximab vedotin could become a new helpful tool.

Background The optimal endoscopic target in early Crohn’s disease (CD) that limits long-term disease complications is unknown. Methods We analysed medical records from patients who had follow-up data since the end of CALM. Patients with Crohn’s disease endoscopic index of severity (CDEIS) scores at the end of CALM were included. The primary outcome was a composite of major adverse outcomes reflecting CD progression: new internal fistula/abscess, stricture, perianal fistula/abscess, CD hospitalisation, or CD surgery since the end of CALM. We compared median CDEIS and per cent improvement from baseline CDEIS. Youden index analysis was used to identify optimal CDEIS cut-off score associated with CD progression. Kaplan–Meier and Cox regression methods were used to compare rates of progression by different CDEIS targets. Multivariable models were adjusted for age, prior surgery, and stricturing behaviour. Results 110 patients with median age 28 (IQR 22–38) years, disease duration 0.2 (0.1–0.5) years, and median follow up of 3.1 (1.9–4.4) years were included. Eleven per cent had a history of stricture, 5.5% history of surgery, and 52% were originally in the tight control arm of the CALM study. Median CDEIS score at end of CALM was 3 (0–5.4) and 32 (29%) patients had disease progression. Baseline median CDEIS score was similar between those with and without progression [10.9 (7.5–15.5) vs. 11.9 (8–17.5)]. Median CDEIS score at the end of CALM was higher among those with progression [1.3 (0–5.1) vs. 4.9 (3–9.1), p < 0.001)]. Patients within higher quartiles of CDEIS score had higher rates of progression over time (Figure 1). Patients without disease progression had a greater median decrease in CDEIS score from baseline to end of CALM [90% (60–100%) vs. 50% (30–80%), p < 0.001]. The optimal CDEIS score cut-off was 2 with sensitivity 84%, specificity 60% and NPV 90% for progression. Patients with CDEIS ≤ 2 had less progression over time compared with patients with > 50% improvement from baseline CDEIS (not reaching CDEIS ≤ 2) and those not meeting either endpoint (Figure 2). On adjusted analysis, CDEIS score ≤ 2 was associated with a decreased risk of progression (aHR 0.23, 95% CI 0.09–0.56). Conclusion In early CD, a CDEIS score ≤ 2 is optimal cut-off associated with a lower risk of disease progression.

Introduction Up to 25% of patients treated with infliximab experience hypersensitivity reactions. Prophylactic premedication prior to infliximab infusion, comprising corticosteroids and/or antihistamines, is widely used in clinical practice but its efficacy has recently been called into question due to the lack of pathophysiological rationale and validation by controlled trials. Materials and methods We conducted a comprehensive literature search of multiple electronic databases from inception to June 2017 to identify studies reporting the impact of corticosteroid and/or antihistamine premedication on the risk of acute ( Results Ten studies, eight observational studies and two randomized control trials, were identified including a total of 3892 patients with IMIDs, and 1,385 patients with IBD. Corticosteroid premedication was not associated with a decreased risk of hypersensitivity reaction in either IMIDs (7 studies; OR, 1.07, 95%CI, 0.64–1.78; I2 = 57.5%) or IBD (3 studies; OR, 1.04, 95% CI, 0.52–2.07; I2 = 57%). Antihistamine premedication was not associated with a decreased risk of hypersensitivity reaction in IMIDs (3 studies: OR, 1.39, 95% CI, 0.70–2.73; I2 = 85%). The combination of corticosteroids and antihistamines did not decrease the risk of acute infliximab infusion reaction in IMIDs (6 studies; OR, 2.12, 95% CI, 0.61–7.35; I2 = 94%), but was associated with an increased risk in IBD (4 studies, OR, 4.17, 95% CI, 1.61–10.78; I2 = 77%). Conclusion Corticosteroid and/or antihistamine premedication is not associated with a decreased risk of acute hypersensitivity reactions to infliximab in patients with IMIDs. We believe that these premedications should no longer be part of standard protocols.

International audience

Background Crohn’s disease (CD) has a significant impact on health status and quality of life, by affecting physical and emotional well-being and by impairment of social and functional abilities. In the era of treat to target concept, endoscopic healing has emerged to be the goal to achieve to prevent intestinal damage and disability. It is not clear what level of endoscopic healing is associated with CD disability. We, therefore, aimed to compare disability associated to complete endoscopic healing or partial endoscopic healing in patients with CD. Methods We conducted a multicentric prospective study, between September 2019 and November 2020, in one University Hospital, one general hospital and one private practice center including consecutive patients with CD in clinical remission and either complete mucosal healing (CDEIS =0) or partial mucosal healing (CDEIS > 0 and ≤ 4). The 10-item self-questionnaire IBD-disk was used to assess disability. Moderate-to-severe disability was defined as an overall IBD-disk score ≥ 40. Results A total of 82 patients were included. Forty-four (53%) were women, the median age and disease duration were respectively 35.3 years (Interquartile range (IQR 28.6–45.2) and 8.0 years (IQR, 3.0–17.0). The median overall IBD-disk score was 26.5 (IQR, 9 -45.0) and 30 (36.6%) had moderate-to-severe disability. Complete endoscopic healing was observed in 48 patients (57.3%). Median IBD-Disk score was respectively 24 (IQR, 9.0–40.5) and 34 (IQR, 9.5–51.5) for patients with complete and partial endoscopic healing (p=0.068). Respectively, 13/48 (27%) and 17/34 (50%) of patients with complete and partial endoscopic healing had moderate-to-severe disability (p=0.039). In multivariate analysis, partial mucosal healing (OR=3.04, CI95% [1.11, 8.94], p=0.03), female gender (OR=3.93, CI95% [1.43, 11.94], p=0.008) and smoking (OR= CI95% 5.88 [1.72, 25.0] p=0.039) were significantly associated with moderate-to-severe disability. Among the IBD-disk subscores, defecation score (median, IQR) (0.0 [0.0–3.0] vs 4.0 [0.0–7.5], p=0.028) and energy score (4.0 [0.0–6.0] vs 6.0 [2.5–8.0], p=0.023) were significantly lower in complete endoscopic healing. Conclusion One-third of patient with endoscopic healing reported moderate-to-severe disability. Complete endoscopic healing (CDEIS = 0) was associated with lower disability than partial endoscopic healing (CDEIS > 0 and ≤ 4) in CD. Deeper endoscopic healing may be needed to reduce the risk of disability in CD.

BACKGROUND: The significance of endoscopic activity in asymptomatic ulcerative colitis (UC) patients with an ileal pouch is unknown. AIM: We aimed to understand the association of endoscopic pouch activity in asymptomatic patients with the development of pouchitis. METHODS: We analyzed a retrospective cohort of patients with UC or IBD-Unspecified (IBD-U) who underwent a total proctocolectomy with ileal pouch anal anastomosis (IPAA). Asymptomatic patients with a Pouchitis Disease Activity Index (PDAI) symptom sub-score of zero who underwent an index surveillance pouchoscopy were included. Endoscopic pouch body activity was graded as 0: normal, 1: mucosal inflammation, or 2: mucosal breaks (ulcers and/or erosions). The primary outcome was primary acute idiopathic pouchitis defined as PDAI score ≥ 7 with symptoms lasting less than four weeks and responsive to standard antibiotics, not otherwise meeting criteria for secondary pouchitis. The secondary outcome was chronic idiopathic pouchitis defined as PDAI score ≥ 7 with symptoms lasting greater than four weeks despite standard antibiotics. Predictors of pouchitis were analyzed using Kaplan-Meier and Cox regression methods with hazard ratios (HR) and 95% confidence intervals (CI) reported. RESULTS: A total of 143 asymptomatic pouch patients were included. Index endoscopic pouch body activity was graded as 0 in 86 (60.1%) patients, 1 in 26 (18.2%) patients and 2 in 31 (21.7%) patients. The median length of follow-up after index surveillance pouchoscopy was 3.03 [IQR 1.24–4.60] years. Primary acute idiopathic pouchitis occurred in 44 (31%) patients and chronic idiopathic pouchitis in 12 (8.4%) patients. Grade 2 endoscopic pouch activity was associated with the development of acute pouchitis (HR 2.39, 95% CI 1.23–4.67), however not chronic pouchitis (HR 1.76, 95% CI 0.53–5.87). Histologic inflammation in endoscopically normal pouch mucosa was not associated with acute or chronic pouchitis. CONCLUSIONS: Mucosal breaks are present in nearly a quarter of asymptomatic pouch patients and are associated with an increased risk of acute pouchitis.

International audience; BACKGROUND & AIMS: There is debate over whether patients with inflammatory bowel diseases (IBD) treated with biologics that are not tumor necrosis factor antagonists (such as vedolizumab or ustekinumab) should receive concomitant treatment with immunomodulators. We conducted a meta-analysis to compare the efficacy and safety of concomitant immunomodulator therapy vs vedolizumab or ustekinumab monotherapy. METHODS: In a systematic search of publications, through July 31, 2019, we identified 33 studies (6 randomized controlled trials and 27 cohort studies) of patients with IBD treated with vedolizumab or ustekinumab. The primary outcome was clinical benefit, including clinical remission, clinical response, or physician global assessment in patients who did vs did not receive combination therapy with an immunomodulator. Secondary outcomes were endoscopic improvement and safety. We performed random-effects meta-analysis and estimated odds ratio (OR) and 95% CIs. RESULTS: Overall, combination therapy was not associated with better clinical outcomes in patients receiving vedolizumab (16 studies: OR, 0.84; 95% CI, 0.68-1.05; I2 = 13.9%; Q test P = .17) or ustekinumab (15 studies: OR, 1.1; 95% CI, 0.87-1.38; I2 = 11%; Q test P = .28). Results were consistent in subgroup analyses, with no difference in clinical remission or response in induction vs maintenance studies or in patients with Crohn's disease vs ulcerative colitis in studies of vedolizumab. Combination therapy was not associated with better endoscopic outcomes in patients receiving vedolizumab (3 studies: OR, 1.13; 95% CI, 0.48-2.68; I2 = 0; Q test P = .96) or ustekinumab (2 studies: OR, 0.58; 95% CI, 0.21-1.16; I2 = 47%; Q test P = .17). Combination therapy was not associated with an increase in adverse events during vedolizumab therapy (4 studies: OR, 1.17; 95% CI, 0.75-1.84; I2 = 0; Q test P = .110). CONCLUSIONS: In a meta-analysis of data from studies of patients with IBD, we found that combining vedolizumab or ustekinumab with an immunomodulator is no more effective than monotherapy in induction or maintenance of remission.

Objectives Data on long-term natural history of microscopic colitis (MC), including collagenous (CC) and lymphocytic colitis (LC), are lacking. Methods All new cases of MC diagnosed in the Somme area, France, between January 1, 2005, and December 31, 2007, were prospectively included. Colonic biopsies from all patients were reviewed by a group of 4 gastrointestinal pathologist experts to assess the diagnosis of CC or LC. Demographic and clinical data were retrospectively collected from diagnosis to February 28, 2017. Results One hundred thirty cases of MC, 87 CC and 43 LC, were included (median age at diagnosis: 70 [interquartile range, 61-77] and 48 [IQR, 40-61] years, respectively). The median follow-up was 9.6 years (7.6; 10.6). By the end of the follow-up, 37 patients (28%) relapsed after a median time of 3.9 years (1.2; 5.0) since diagnosis, without significant difference between CC and LC (30% vs 26%; P = 0.47). Twenty patients (15%) were hospitalized for a disease flare, and 32 patients (25%) presented another autoimmune disease. Budesonide was the most widely used treatment (n = 74, 59%), followed by 5-aminosalicylic acid (n = 31, 25%). The median duration of budesonide treatment was 92 days (70; 168), and no adverse event to budesonide was reported. Sixteen patients (22%) developed steroid dependency and 4 (5%) were corticoresistant. No difference in the risk of digestive and extradigestive cancer was observed compared with the general population. None of the death (n = 25) observed during the follow-up were linked to MC. In multivariate analysis, age at diagnosis (HR, 1.03; 95% confidence interval, 1.00-1.06; P = 0.02) and budesonide exposure (HR, 2.50; 95% confidence interval, 1.11-5.55; P = 0.03) were significantly associated with relapse. Discussion This population-based study showed that after diagnosis, two-third of the patients with MC observed long-term clinical remission. Age at diagnosis and budesonide exposure were associated with a risk of relapse.

Background and Aims Recurrence of Crohn’s disease (CD) after surgery is a major concern. Curcumin has anti-inflammatory properties and induces endoscopic remission in patients with ulcerative colitis. We investigated the efficacy of curcumin vs placebo in preventing post-operative recurrence of CD, based on endoscopic and clinical indices, in patients receiving concomitant thiopurine therapy. Methods We conducted a double-blind randomized controlled trial at 8 referral centers in France, from October 2014 through January 2018, of 62 consecutive patients with CD undergoing bowel resection. Patients received azathioprine (2.5 mg/kg) and were randomly assigned to groups given oral curcumin (3 g/day; n = 31) or an identical placebo (n = 31) for 6 months, and were then evaluated by colonoscopy. We also collected data on CD activity index, results from laboratory tests, and answers to quality of life questionnaires during this 6-month period. The primary endpoint was postoperative recurrence of CD in each group (Rutgeerts’ index score ≥i2) at month 6 (determined by central reading). An interim analysis (intent to treat) was scheduled after 50% of the patients were enrolled. Results At month 6, postoperative recurrence (Rutgeerts’ index score ≥i2) occurred in 18 patients (58%) receiving curcumin and 21 patients (68%) receiving placebo (P = .60). A significantly higher proportion of patients receiving curcumin (55%) had a severe recurrence of CD (Rutgeerts’ index score ≥i3) than patients receiving placebo (26%) (P = .034). We observed a clinical recurrence of CD (CD activity index score >150) at month 6 in 45% of patients receiving placebo and 30% of patients receiving curcumin (P = .80). Quality of life scores at month 6 did not differ significantly between groups (P = .80). Severe adverse events developed in 6% of patients receiving placebo and 16% of patients receiving curcumin (P = .42). Conclusions In a randomized controlled trial of patients who underwent surgery for CD and received thiopurine treatment, we found that curcumin was no more effective than placebo in preventing CD recurrence. There were no significant differences between groups in quality of life or severe adverse events. The study was discontinued after interim analysis due to futility. ClinicalTrials.gov no: NCT 02255370.