The Gastro & Hep Report: Spring 2013

Comprehensive Reports on the Latest Advances in Gastroenterology and Hepatology from:

• The 2012 Annual Scientific Meeting of the American College of Gastroenterology
October 19–24, 2012 • Las Vegas, Nevada

• The 63rd Annual Meeting of the American Association for the Study of Liver Diseases
November 9–13, 2012 • Boston, Massachusetts

• The 2012 Advances in Inflammatory Bowel Diseases
December 13–15, 2012 • Miami, Florida

Indexed through the National Library of Medicine (PubMed/Medline), PubMed Central (PMC), and EMBASE

Disclaimer: Opinions expressed in articles, case reports, and other editorial content are those of the individual authors and do not necessarily reflect the opinions of the editors or the publishers of Gastroenterology & Hepatology. General comments in this journal are based on the experience of those involved in the composition and editing of the individual piece, and should not be taken as a formal consultation or recommendation for a patient, disease state, or treatment approach.

The authors, editors, and publishers make every effort to ensure that drugs and dosage recommendations are precise and accurate, and thatgeneric and trade names of drugs are correct. However, errors can occur and readers should confirm all dosage schedules against the manufacturer’s package information data. Some dosages and delivery methods may not reflect package insert information, due to the clinical experience of the authors. Please report any errors to the editor. Publication of an advertisement or other product mention in Gastroenterology & Hepatology is not an endorsement of either the product or the manufacturer’s claims and should not be construed as such. The editors and publishers do not assume any responsibility for any injury and/or damage to persons or property related to any use of the content contained herein.

Gastro-Hep Communications, Inc., is an independent publishing company and is not associated with any pharmaceutical or other agency.

©2013  Gastro-Hep Communications, Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means now or hereafter known, electronic or mechanical, without permission from the publisher, except for “fair use” purposes, as defined by US Copyright Law, Chapter 1, Section 107. Up to 5 copies of an article may be made for personal or internal use at no charge. Publishers: Steven H. Kurlander and Paul H. McDaniel. For questions concerning reproduction of content, please contact the publisher at (212) 995-2211. Printed in the United States of America.


Presentations in IBS

Mesalamine Granules Provide Relief of Symptoms in Patients with Irritable Bowel Syndrome with Diarrhea

Extended-release mesalamine granule capsules are currently indicated for the maintenance of remission in ulcerative colitis. Uncontrolled studies suggest that aminosalicylate therapy may be beneficial in patients with irritable bowel syndrome with diarrhea (IBS-D). Jeffrey Aron, from the California Pacific Medical Center in San Francisco, reported the results of a randomized, double-blind, placebo-controlled, multicenter phase II study at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #7). The study was designed to evaluate the efficacy of extended-release mesalamine granules for the treatment of IBS-D.

A total of 148 patients with IBS-D were randomly selected to receive either extended-release mesalamine granules at a dosage of 750 mg/day or 1,500 mg/day or else placebo for 12 weeks. Eligible patients had IBS-D as defined by Rome III criteria and had experienced diarrhea with no constipation in the 7–13 days before receiving the first dose of the study drug.

Patients described as weekly responders in relation to abdominal pain were defined as those patients who had a 30%-or-greater improvement from baseline in a weekly average abdominal pain score on a 10-point scale. Weekly responders in relation to stool consistency were defined as those patients who had a 50%-or-greater reduction in the number of days per week in which stool consistency was type 6 or 7 compared with baseline, according to the Bristol Stool Scale. Patients defined as being monthly responders were those who were weekly responders in relation to both abdominal pain and stool consistency for at least 2 of 4 weeks.

A significantly greater number of patients who were treated with 1,500 mg extended-release mesalamine granules were monthly responders compared with patients who were receiving placebo (47% vs 28%; P=.0432). Although more patients receiving 750 mg/day of extended-release mesalamine granules were monthly responders compared with those receiving placebo, the difference in the rate of response did not reach statistical significance (32% vs 28%; P=.6059).

The adverse event profiles were similar between the extended-release mesalamine granule dosage groups and the mesalamine therapy groups and the placebo group.


Linaclotide Acts Rapidly on Abdominal and Bowel Symptoms in Patients with Irritable Bowel Syndrome with Constipation

Linaclotide achieved rapid improvement in symptoms, compared with placebo, in patients with irritable bowel syndrome with constipation (IBS-C), according to Lin Chang, of the Division of Digestive Diseases at the University of California Los Angeles, who presented study results during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1559). Chang and colleagues evaluated the changes in abdominal and bowel symptoms from baseline through the first 7 days of treatment with linaclotide.

Patients qualified for each study if they had IBS-C and met modified Rome II criteria. In 2 phase III trials, patients were randomly selected to receive either 290 mg/day linaclotide (n=805) or placebo (n=797). Pooled results of the intent-to-treat population showed that patients treated with linaclotide achieved a significant mean percent improvement in abdominal bloating (8% vs 4%; P<.05) and fullness (9% vs 4%; P<.05) on Day 1, pain (12% vs 7%; P<0.05) and discomfort (11% vs 7%; P<.05) by Day 2, and cramping (15% vs 10%; P<.05) by Day 3, compared with patients receiving placebo.

Stool consistency and straining on Day 1 also were significantly improved in the linaclotide group compared with the placebo group (P<.0001). More patients receiving linaclotide than those receiving placebo reported 1 or more spontaneous bowel movements on each of the first 7 treatment days (Day 1, 49% vs 24%; Day 2, 57% vs 40%; Day 3, 54.3% vs 42%; Day 4, 55% vs 40%; Day 5, 55% vs 39%; Day 6, 54% vs 39%; Day 7, 50% vs 39%; P<.0001). The same was also true for patients who reported a complete spontaneous bowel movement.

Although the median time to the first spontaneous bowel movement was the same—2 days—for patients in both treatment groups, the median time to the first complete spontaneous bowel movement was significantly shortened in the linaclotide group compared with the placebo group (5 vs 20 days; P<.0001). During Week 1, patients treated with linaclotide had an average of 6.6 spontaneous bowel movements and 2.4 complete spontaneous bowel movements. These frequencies of spontaneous bowel movements were significantly higher than those reported by patients receiving placebo (3.5 and 0.9, respectively; P<.0001). The incidence of diarrhea during the first week of treatment was 10% among patients treated with linaclotide, which was higher than that reported in patients receiving placebo (0.4%).


Linaclotide May Increase Bowel Movements and Lessen Straining in Patients with Chronic Constipation 

A post-hoc analysis of pooled data from 2 phase III trials showed that linaclotide therapy can increase bowel movements and lessen straining in patients with chronic constipation. The findings were delivered by Satish S. Rao, of the Georgia Health Sciences University Medical Center, during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #455).

The goals of this study were 2-fold: 1) to evaluate the effects of linaclotide compared with placebo on the distribution of bowel movements occurring with the use of rescue medications, spontaneous bowel movements, and complete spontaneous bowel movements, and 2) to determine the effects of linaclotide compared with rescue medication on straining associated with spontaneous bowel movements.

All patients had an average of fewer than 3 complete spontaneous bowel movements per week and 6 or fewer spontaneous bowel movements per week during a 2-week pretreatment period. In this study, spontaneous bowel movements were defined as those occurring in the absence of rescue medication, and complete spontaneous bowel movements were accompanied by a patient-reported feeling of complete evacuation.

A total of 1,271 patients were included in the pooled intent-to-treat population, including 423 patients who received placebo and 430 patients who received linaclotide at a dosage of 145 mg/day and 418 patients who received linaclotide at a dosage of 290 mg/day. The rates of spontaneous and complete spontaneous bowel movements were significantly increased with linaclotide therapy compared with placebo (38% for the 145 mg/day dosage and 42% for the 290 mg/day dosage vs 22% for placebo).

During the treatment period, more patients treated with linaclotide reported decreased straining compared with patients receiving placebo. The proportion of patients who reported a little straining or none at all was 69% for both linaclotide cohorts versus 49% for the cohort receiving placebo who used rescue medication.


Benefit Was Demonstrated for Lubiprostone in Patients with Irritable Bowel Syndrome with Constipation 

Global response criteria for lubiprostone provided clinically meaningful measures of treatment benefit in relation to symptoms of irritable bowel syndrome with constipation (IBS-C) and were consistent with improvement in individual secondary symptoms, according to Raymond M. Panas and colleagues from Sucampo Pharma Americas, Inc. in Bethesda, Maryland, whose study results were presented during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1009). The researchers conducted a sensitivity analysis that compared previously established global response criteria for evaluating treatments for IBS-C with improvement in individual symptoms.

Data from 1,421 patients were pooled from 2 randomized, double-blind, 12-week studies that compared lubiprostone (8 mg twice daily) with placebo for the treatment of IBS-C. The primary endpoint—based on patient response to the question “How would you rate your relief of IBS symptoms [abdominal discomfort/pain, bowel habits, and other IBS symptoms] over the past week compared with how you felt before you entered the study?”— was rated on a 7-point balanced scale. The monthly response was defined by patient-reported symptom rating as either significantly relieved for 2 weeks or more or moderately relieved for 4 weeks without having an increase from baseline in the use of rescue medication, a discontinuation due to lack of efficacy, or any ratings of moderate or significant worsening. Individual secondary symptoms, which included abdominal discomfort/pain, abdominal bloating, and constipation severity, were rated on a 5-point scale, on which a 1-point shift was considered to be clinically meaningful.

Global response was significantly improved with lubiprostone treatment versus placebo (P<.0005). Global response appeared to be closely related to improvement in individual symptoms. Responding patients achieved a 1-point-or-greater increase in the mean change from baseline for each symptom. Patients who failed to respond to treatment showed little change (<1 point). The global response criteria demonstrated a high sensitivity for detecting treatment-dependent changes of individual IBS-C symptoms (0.923 for abdominal discomfort/pain; 0.936 for abdominal bloating; and 0.935 for constipation severity).


A Tri-Component Endpoint in Irritable Bowel Syndrome Shows Value in Response Assessment

A tri-component endpoint for symptoms of irritable bowel syndrome (IBS) without constipation is responsive and correlates with all daily symptoms of IBS with diarrhea (IBS-D), according to Anthony J. Lembo, of the Beth Israel Deaconess Medical Center in Boston, Massachusetts, who presented findings on the use of a tri-component endpoint for assessing response to treatment in patients with IBS-D during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1571A). Lembo and colleagues noted that current guidance from the US Food and Drug Administration recommends the use of a composite endpoint of 2 primary IBS-D symptoms: abdominal pain and stool consistency; however, this endpoint may ignore other gastrointestinal symptoms that patients consider to be important, such as urgency and bloating. Thus, the research team developed a tri-component endpoint that included a daily assessment of abdominal pain and stool consistency as well as global IBS symptoms.

Patient data were pooled from 2 phase III trials, TARGET 1 and TARGET 2. A total of 1,260 patients with IBS without constipation were randomly selected to receive either rifaximin (550 mg 3 times daily) or placebo for 2 weeks and then were evaluated for 10 weeks. IBS symptom severity for both global IBS symptoms and IBS-related abdominal pain was rated on a 7-point scale. Stool consistency was rated on a 5-point scale. The tri-component endpoint defined weekly responders as those patients who reported a 30%-or-greater improvement from baseline in abdominal pain, a 50%-or-greater improvement in the number of days in a week that they passed loose or watery stools, and a 1-point-or-greater improvement in the weekly average score of daily global IBS symptoms.

In both studies, as well as in pooled analysis, significantly increased improvements in all daily IBS symptom severity measures were observed at each study week in those responders defined by the tri-component endpoint compared with nonresponders. Results also suggested that the tri-component endpoint had convergent validity; it consistently correlated with all daily IBS symptom severity measures at each treatment week (Spearman correlation of ≥0.40 with other daily symptom severity measures).


Presentations in IBD

Results of GEMINI I: Vedolizumab Is Effective in Maintaining Clinical Remission in Patients with Ulcerative Colitis

Vedolizumab maintenance therapy showed efficacy in achieving clinical corticosteroid-free remission, with durable response, in patients with ulcerative colitis. These findings from GEMINI I, a phase III trial evaluating the efficacy and safety of the investigational anti-α4β7 integrin gut-selective monoclonal antibody vedolizumab, were reported by Brian G. Feagan, of the Robarts Research Institute in London, Ontario, Canada, at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #5).

A total of 895 adults with moderate to severely active ulcerative colitis who had failed at least 1 prior therapy received 2 induction doses of vedolizumab at Weeks 0 and 2. Patients with a clinical response at Week 6 (n=373) were then randomly selected to receive 46 weeks of maintenance therapy with vedolizumab, administered intravenously at a dosage of 300 mg every 4 or 8 weeks, or placebo. The primary outcome was clinical remission at Week 52.

Superior outcomes were achieved in significantly more patients treated with vedolizumab than placebo. The rate of clinical remission was 45% for patients receiving vedolizumab every 4 weeks, 42% for patients receiving vedolizumab every 8 weeks, and 16% for patients receiving placebo (P<.0001). The rate of steroid-free remission was 45% for the 4-week vedolizumab arm, 31% for the 8-week vedolizumab arm, and 14% for the placebo arm (P<.0001 and P=.0120 for the 4-week and 8-week vedolizumab arms, respectively). The rate of mucosal healing also was significantly improved with both 4-week (56%) and 8-week (52%) vedolizumab compared with placebo (20%; P<.0001). Clinical remission and durable clinical response were achieved regardless of prior exposure to tumor necrosis factor alpha (TNF-α) inhibitors.

The rates of clinical remission among patients with prior TNF-α inhibitor failure were 35% for vedolizumab every 4 weeks, 37% for vedolizumab every 8 weeks, and 5% for placebo. The comparable rates among patients with no prior TNF-α inhibitor exposure were 48%, 46%, and 19%, respectively.

There was no significant increase in the frequency of adverse events, including the rates of opportunistic or enteric infections, among patients treated with vedolizumab compared with patients receiving placebo.


GEMINI II Results: Vedolizumab Shows Good Response in Treatment-Resistant Crohn’s Disease 

Vedolizumab maintenance therapy was associated with higher rates of remission and an enhanced corticosteroid-free response compared with placebo in patients with Crohn’s disease (CD) that was resistant to immunsuppressive agents and tumor necrosis factor alpha (TNF-α) inhibitors. These findings from the GEMINI II trial were presented by Stephen Hanauer of the University of Chicago Medical Center at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #42).

The trial enrolled 1,115 patients with moderate-to-severe CD. A total of 461 patients who had received induction therapy, consisting of 2 intravenous 300-mg doses of vedolizumab, showed a response at Week 6 and went on to be randomly selected to receive vedolizumab maintenance therapy (300 mg intravenously) every 4 or 8 weeks or else placebo.

Significantly superior rates of clinical remission (36% for the 4-week arm and 39% for the 8-week arm) were achieved with vedolizumab compared with placebo (22%; P=.0042 and P=.0007 for the 4- and 8-week vedolizumab arms, respectively). Significantly superior rates of corticosteroid-free remission were achieved as well in patients receiving active treatment (29% in the 4-week and 32% in the 8-week vedolizumab arm vs 16% in the placebo arm; P=.0450 and P=.0154 for the 4- and 8-week vedolizumab arms, respectively). Further, the rate of remission was improved with vedolizumab regardless of whether patients had histories of failing TNF-α inhibitor therapy or had no prior exposure to TNF-α inhibitors.

The rate of enhanced clinical response, defined as a decrease in the CD Activity Index of 100 points, was also significantly improved in both vedolizumab arms compared with the placebo arm (P=.0132 and P=.0053 for the 4- and 8-week vedolizumab arms, respectively). The rate of durable remission also was improved with vedolizumab therapy; however, the difference compared with placebo did not reach statistical significance.

The exposure-adjusted rates of all adverse events, including serious infections, were comparable across the 3 treatment groups. However, the exposure-adjusted rate of serious adverse events was increased in the placebo arm versus either vedolizumab arm.


Antibody Production May Be Linked to Lack of Response to Adalimumab Therapy 

Up to a third of patients with Crohn’s disease (CD) will show a primary nonresponse to tumor necrosis factor alpha (TNF-α) inhibitor therapy, and an additional 30–40% of patients will lose response to initial TNF-α therapy during the first year of treatment. Dose escalation or a switch to another anti–TNF-α therapy is often required to maintain a response in patients in whom secondary resistance develops. Failure of anti–TNF-α therapy may, in part, be attributed to low serum drug levels and/or the development of antidrug antibodies. Monitoring of both of these parameters is emerging as an important strategy in the management of inflammatory bowel disease (IBD).

To analyze adalimumab levels in relation to production of antibodies to adalimumab, Shui Long Wang, from Prometheus Laboratories, Inc. in San Diego, California, and colleagues used the homogeneous mobility shift assay (HMSA) to measure adalimumab levels and antibodies to adalimumab in serum samples of 100 patients with IBD who had initially responded to adalimumab therapy for at least 3 months but were beginning to lose response to treatment. Findings were reported during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1499).

The HMSA demonstrated a lower limit of detection of 0.026 mg/mL for antibodies to adalimumab and 0.018 mg/mL for adalimumab. The coefficient of variation was less than 15%, and the accuracy was within 20% of detection for both assays. Tolerance for adalimumab in antibodies to adalimumab HMSA reached 40 μg/mL. Using serum samples from 100 drug-naïve healthy individuals, a cutoff threshold value of 0.55 U/mL was determined for antibodies to adalimumab and 0.66 μg/mL was calculated for serum adalimumab levels. Overall, antibodies to adalimumab were present in 40% of samples.

Approximately one third (36%) of the patients who were losing response to adalimumab had a serum adalimumab level of less than 3 mg/mL. Of these, 58% were positive for antibodies to adalimumab. However, antibodies to adalimumab were present in only 4 (18%)of 22 patients whose serum adalimumab level increased to greater than 20 mg/mL.


Adalimumab Shows Promise for Treatment of Moderately to Severely Active Ulcerative Colitis

Efficacy results suggest that adalimumab may hold promise in the treatment of severe ulcerative colitis (UC), according to Jean-Frederic Colombel, of the Centre Hospitalier Universitaire de Lille, France, who discussed interim results of the ongoing adalimumab UC development program extension study at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P387). The development program consists of the ULTRA 1 and ULTRA 2 clinical trials followed by an ongoing multicenter open-label extension study.

ULTRA 1 and ULTRA 2 both demonstrated that adalimumab effectively induced remission in patients with moderately to severely active UC. Patients enrolled in the extension study who were already receiving open-label adalimumab weekly continued with the same regimen. Patients who entered the extension study from a blinded cohort or who were receiving open-label adalimumab at a dosage of 40 mg every other week received adalimumab at this dosage during the extension study. In cases of disease flare or nonresponse, patients who entered the extension study from a blinded cohort were permitted to increase their adalimumab dosage to 40 mg weekly at Week 12 or thereafter. The same protocol was allowed for patients from open-label cohorts who were in clinical response when they entered the extension study. Those who had a flare or no response were permitted to increase their dosage at Week 2. The partial Mayo score, defined as the Mayo score without the endoscopy subscore, was determined at each study visit.

On the day of the first adalimumab dose in the extension study, the observed mean partial Mayo score was 5.9. This score decreased gradually over the extension study, reaching a mean of 4.2 at 4 weeks, 3.9 at 8 weeks, 2.6 at 52 weeks, and 1.8 at 112 weeks of treatment. Of the 588 patients enrolled in the extension study, 351 (60%) achieved clinical remission by Week 60. No new adverse events have been reported.


New Therapy Shows Little Impact on Colorectal Cancer Risk in Patients with Ulcerative Colitis 

No significant difference in colorectal cancer risk was seen between patients receiving adalimumab and standard therapy for treatment of ulcerative colitis (UC), reported Jean-Frederic Colombel, from the Centre Hospitalier Universitaire de Lille, France, at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P389). To assess whether adalimumab showed a benefit over standard therapy in relation to colorectal cancer risk in patients with UC, Colombel and colleagues identified 23,867 patients, age 18–64 years, from a patient database for years 2000–2010. Patients who had initiated adalimumab therapy for UC during this time were assigned to the adalimumab group (n=581), while patients who had initiated corticosteroid, aminosalicylate, or immunosuppressant therapy (without concomitant adalimumab, etanercept, or infliximab) were assigned to the standard therapy group (n=23,286).

For the 6 months prior to baseline, patients could have no evidence of colorectal cancer; benign neoplasm of the colon, rectum, or anal canal; or any other malignancies. The time to colorectal cancer diagnosis was calculated using Kaplan-Meier survival curves and Cox proportional-hazard models.

The risk of colorectal cancer at 1 and 2 years after initiating therapy was similar between the adalimumab- and standard-therapy groups before adjusting for baseline demographics (age, sex, and year of treatment initiation) and comorbidity factors (noninfective gastroenteritis, colitis, and gastrointestinal hemorrhage). At 1 year, the colorectal cancer–free rate was 100% in both groups. The comparable 2-year colorectal cancer–free rates were the same; however, after adjusting for baseline factors, colorectal cancer was approximately 1.36-fold less likely to develop in patients treated with adalimumab by Year 2 compared with patients receiving standard therapy. Statistical significance was not reached (hazard ratio, 0.735; 95% confidence interval, 0.101–5.373; P=.762).


Adalimumab Therapy Is Associated with Reduced Hospitalization and Colectomy Rates in Patients with Ulcerative Colitis 

Rates of hospitalizations and colectomies in patients with ulcerative colitis (UC) who had a first response to adalimumab therapy were reduced compared with rates in patients receiving placebo, reported Brian G. Feagan, of the Robarts Research Institute in London, Ontario, Canada, during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P424). Feagan and colleagues studied the effect of a 160/80/40-mg adalimumab regimen on

reduction of risk in all-cause and UC-related hospitalization and colectomy, specifically among patients with an initial response to adalimumab (n=939). The investigators used data from the ULTRA 1 and ULTRA 2 trials, both of which demonstrated significant reductions in hospitalization rates and nonsignificant reductions in colectomy rates in patients receiving adalimumab for treatment of moderately to severely active UC.

Two gastroenterologists blinded to the treatment arms reviewed hospitalization and colectomy rates based on the safety reports of the patients in the study cohort. Hospitalizations of patients treated with adalimumab who had an initial nonresponse through Week 8 were counted but were censored after Week 8 to reflect the clinical practice pattern of continuing treatment in initial adalimumab responders.

Compared with the placebo group, a 35% reduction in the number of patients hospitalized for any reason and a 36% reduction in the number of hospitalizations for any reason were seen in the adalimumab group. For the number of patients who had an all-cause hospitalization, the person-year–based incidence rate was 17% for adalimumab versus 26% for placebo (P=.035). The person-year–based incidence rate for the number of hospitalizations was 20% for adalimumab versus 31% for placebo (P=.015). Both the rate and number of UC-related hospitalizations also were significantly reduced (55% and 56%, respectively). The person-year–based incidence rate for the number of patients who underwent a UC-related hospitalization was 10% for adalimumab versus 22% for placebo (P=.001), and the person-year–based incidence rate for the number of UC-related hospitalizations was 11% versus 25% (P<0.001). The difference in the person-year–based incidence rate for colectomy did not reach statistical significance between the adalimumab and placebo groups (2% vs 4%; P=.099).


Presentations in GERD

Prevalence and Distinguishing Features of PPI-Responsive Esophageal Eosinophilia and Eosinophilic Esophagitis Are Defined 

The prevalence and distinguishing characteristics of proton pump inhibitor–responsive esophageal eosinophilia (PPI-REE) and eosinophilic esophagitis (EoE) were defined in a prospective study that aimed to determine the prevalence of PPI-REE and EoE in patients undergoing esophagogastroduodenoscopy (EGD). The study was presented by Evan S. Dellon, from the Division of Gastroenterology at the University of North Carolina Chapel Hill, at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #48).

Between 2009 and 2011, 173 patients with dysphagia revealed by EGD were enrolled in the study conducted by Dellon and colleagues. Esophageal biopsies were used to determine the maximum eosinophil count per high-power field (hpf). Patients with 15 or more eosinophils/hpf were treated twice daily with a PPI for 8 weeks, after which the EGD was repeated. Patients who exhibited persistent symptoms and 15 or more eosinophils/hpf received a diagnosis of EoE, but PPI-REE was diagnosed in cases of symptomatic and histologic response, defined as the finding of fewer than 15 eosinophils/hpf.

Of the 173 patients enrolled, 66 (38%) had an eosinophil count of 15 or more eosinophils/hpf. Following PPI treatment, a diagnosis of EoE was made in 40 (23%) patients, and 24 (14%) patients had a diagnosis of PPI-REE. EoE or eosinophilic gastroenteritis was diagnosed in 1 (2%) patient.

Among the 24 patients with PPI-REE, 9 (38%) had clinical features consistent with gastroesophageal reflux disease (eg, heartburn and hiatal hernia), whereas 12 (50%) had clinical features consistent with EoE, including dysphagia and food impaction. Three (12%) patients with PPI-REE had indeterminate clinical features. Compared with patients with EoE, those with PPI-REE were more likely (P<.05) to be older (age, 36 years vs 48 years), male (63% vs 88%), and have erosive esophagitis (5% vs 21%) or a Schatzki ring (2% vs 21%). Patients with PPI-REE were less likely (P<.05) to have esophageal rings (90% vs 63%), diffuse narrowing (42% vs 8%), linear furrows (88% vs 58%), or decreased vascularity (32% vs 0%). There was no difference in the maximum eosinophil count between patients with a diagnosis of PPI-REE and those with a diagosis of EoE (50 vs 64 eosinophils/hpf).


Patients with Type 2 Diabetes May Be at Increased Risk for Barrett Esophagus

Although a strong association exists between central obesity and an increased risk of Barrett esophagus and esophageal adenocarcinoma, the reason remains unclear. Findings suggesting a potential epidemiologic link between type 2 diabetes and Barrett esophagus were reported by Prasad G. Iyer of the Mayo Clinic College of Medicine in Rochester, Minnesota, at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #49). This was a population-based case-control study using patient data from the General Practice Research Database of the United Kingdom.

Cases of Barrett esophagus (n=14,245) were compared with controls (n=70,361), matched for age, sex, enrollment date, duration of follow-up, and practice region. The association of an initial diagnosis of type 2 diabetes prior to a diagnosis of Barrett esophagus was calculated after adjusting for known risk factors for Barrett esophagus, including cigarette smoking, alcohol consumption, body mass index, and gastroesophageal reflux disease.

Baseline characteristics showed that cases and controls were comparable for age and sex, but cases were more likely than controls to have ever smoked cigarettes or consumed alcohol. The prevalence of type 2 diabetes before a diagnosis of Barrett esophagus was higher in cases versus controls (6% vs 5%; P<.001). The same was also true for the mean body mass index (27.0 kg/m2 vs 26.8 kg/m2; P<.001).

A diagnosis of type 2 diabetes was associated with a 2-fold increased risk of Barrett esophagus, independent of other risk factors. This association was stronger in men (odds ratio [OR], 2.03; 95% confidence interval [CI], 1.01–4.04) compared with women (OR, 1.37; 95% CI, 0.63–2.97). Having a diagnosis of type 2 diabetes for more than 1 year was also associated with a higher risk of Barrett esophagus (OR, 7.7; P=.004). Being overweight or obese prior to the diagnosis of Barrett esophagus was significantly associated with the disease (P=.013 and P=.08, respectively).


Failure of Empiric PPI Therapy May Rule Out a Suspected GERD Diagnosis

Most patients referred for testing to confirm a suspected diagnosis of gastroesophageal reflux disease (GERD) after failing empiric proton pump inhibitor (PPI) therapy had normal 24-hour pH/impedance study results while off acid suppressive therapy, reported Fong-Kuei Cheng of the Walter Reed National Military Medical Center in Bethesda, Maryland, during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P3). The study investigated whether a final diagnosis of GERD was made in patients referred for a 24-hour esophageal pH test and impedance testing.

A total of 348 patients who had undergone 24-hour esophageal pH and impedance testing while off of acid suppressive therapy between years 2006 and 2011 were identified from a clinical database. The mean age of the patient set was 47±13 years, and just over half (55%) were men. The majority of patients had received empiric PPI treatment before testing, and, of these, most (68%) were assigned to a daily pharmacotherapy regimen. Most patients (72%) had a normal Johnson-DeMeester score on 24-hour esophageal pH testing.

Among the 97 (28%) patients with abnormal scores (≥22), 56 (58%) had typical GERD symptoms, including heartburn or acid regurgitation, and 41 (42%) had atypical symptoms, including chest pain, chronic cough, hoarseness, dysphagia, or dyspepsia (P=.023). Although there were no significant differences in relation to race or age of the patients with abnormal 24-hour esophageal pH testing, a significantly greater proportion of men were found to have an abnormal Johnson-DeMeester score compared with women (34% vs 20%; P=.004).

Seventy-five (22%) patients had abnormal findings on impedance testing. The majority of patients (87%) who had normal 24-hour esophageal pH tests also had a normal impedance tests. However, fewer than half (44%) of the patients with an abnormal 24-hour esophageal pH test had a normal impedance test (kappa=.331; P<.001).


Esophageal Body Hypomotility and Acid Exposure Are Independent Predictors of Barrett Esophagus

The associations between Barrett esophagus and esophageal dysmotility and reflux were described by Wai-Kit Lo of the Brigham and Women’s Hospital at Harvard Medical School in Boston, Massachusetts, during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P5). Lo and colleagues used new technologies, such as high-resolution manometry and combined multichannel intraluminal impedance and pH, to assess esophageal motor characteristics and reflux profiles, such as acid and bolus exposure time. All patients included in the study had undergone esophagogastroduodenoscopy (EGD) within 2 years of the esophageal physiologic testing and had a diagnosis of Barrett esophagus based on EGD and histopathologic findings.

A total of 23 cases of Barrett esophagus and 69 controls (matched by age at endoscopy, gender, and race) who had been treated over a 5-year period at a tertiary care center were identified. Patients with a history of gastroesophageal surgery were excluded from the study.

A univariate analysis showed that a history of cigarette smoking, esophageal body hypomotility (defined as the average of esophageal body contraction amplitudes <30 mmHg or >30% failed sequences), and increased acid exposure time were significantly associated with Barrett esophagus. In a multivariate analysis that was adjusted for smoking history, each of these factors remained significant independent predictors of Barrett esophagus: increased acid exposure time (odds ratio [OR], 6.22; P=.008), esophageal body hypomotility (OR, 7.78; P=.01), and history of cigarette smoking (OR, 4.82; P=.04).


PPIs and Laparoscopic Antireflux Surgery Are Comparable in Long-Term Symptom Control of GERD

Proton pump inhibitors (PPIs) and laparoscopic anti-reflux surgery (LARS) were shown to have similar efficacy in long-term symptom control of gastroesophageal reflux disease (GERD) in a study by Jyothsna Talluri and colleagues from the McLaren Regional Medical Center in Flint, Michigan. The team presented study results of an evidence-based review during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P8).

Two randomized trials published between 1980 and 2011 that met the inclusion criteria for the review, including a minimum follow-up of 3 years, were identified. The primary outcome in the first study was a 5-year remission rate, which was significantly higher among patients treated with the PPI esomeprazole than LARS (92% vs 85%; P=.05). However, this comparison lost significance following best-case scenario modeling of the effects of study dropout.

The 5-year prevalence of some symptoms was different—and, in some cases, significantly different—in the esomeprazole group compared with the LARS group. Symptoms included heartburn (16% vs 8%; P=.14), acid regurgitation (13% vs 2%; P<.001), dysphagia (5% vs 11%; P<.001), bloating (28% vs 40%; P<.001), and flatulence (40% vs 57%; P<.001).

In the second study, the efficacy of PPI therapy and LARS at 3 years was compared using the GERD symptom scale (GERSS), which was the primary treatment outcome in this trial. At 3 years, GERSS scores were statistically similar between the groups receiving PPI therapy and LARS (9.05 vs 6.21; P=.17). However, LARS was associated with a greater frequency of heartburn-free days compared with PPI therapy (6.81 vs 5.98; P=.007). Acid reflux symptoms, measured by 24-hour esophageal pH testing, improved from a GERSS score of 9.46 at baseline to a score of 4.29 at 3 years in patients receiving PPI therapy. Acid reflux symptoms improved from a GERSS score of 10.26 at baseline to a score of 2.11 at 3 years in patients who underwent LARS. The change from baseline to 3 years did not statistically differ between the PPI and LARS groups (P=.13).


Electrical Stimulation Therapy of the Lower Esophageal Sphincter May Be Useful in Treatment-Refractory GERD

Preliminary results of an international multicenter trial evaluating the efficacy and safety of lower esophageal sphincter electrical stimulation therapy (LES-EST) in patients with refractory gastroesophageal reflux disease (GERD) show promise and were presented by Albert J. Bredenoord of the Academic Medical Center in Amsterdam, The Netherlands, during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P581).

Patients with GERD who were partially responsive to PPI therapy were included if they had the following characteristics: an off–proton pump inhibitor (PPI) GERD health-related quality-of-life score of more than 20, an improvement in score of more than 5 points while on PPI therapy, a LES end-expiratory pressure of greater than 5 mmHg, a 24-hour esophageal pH of less than 4 for more than 5% of the time interval, and a hiatal hernia. At the time of presentation, 11 patients had been enrolled and had undergone implantation. The mean age of the patients was 54 years, and more than half (n=7) were men.

During the implantation procedure, a small bowel trocar perforation, which was successfully repaired, occurred in 1 patient. LES-EST is continuing in the remaining 10 patients. Of these 10, 6 (60%) have completed a 3-month evaluation and 3 (30%) have completed a 6-month evaluation.

The median off-PPI GERD health-related quality-of- life score at baseline was 32 (interquartile range: 25–38), which was found to be improved to 9 at the 3-month evaluation following LES-EST (P<.001). This improvement remained stable at the 6-month evaluation (P=.05). The median esophageal acid exposure at baseline was 12% (interquartile range: 8.8–15), which improved to 8% (interquartile range: 2.4–12) at the 3-month evaluation and 7% (interquartile range: 0.2–15.3) at the 6-month evaluation. Ten (91%) of the 11 patients enrolled were able to discontinue PPI therapy.

A total of 13 adverse events, including 1 serious adverse event, were reported in 4 patients. Of these, 9 (69%) events were related to the device or the procedure, 7 (54%) events were pain at the implant site, and 1 event was postoperative nausea.


Transoral Incisionless Endoscopic Fundoplication May Be an Option in Select Patients with Chronic GERD 

Transoral incisionless fundoplication (TIF) may be a safe and effective therapeutic option in carefully selected patients with chronic gastroesophageal reflux disease (GERD) who have unsatisfactory outcomes with medical management or who choose to discontinue proton pump inhibitor (PPI) therapy, reported Peter G. Mavrelis of Internal Medicine Associates in Merrillville, Indiana, during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1049). The presentation related results of a 12-month follow-up study of patients with chronic GERD who underwent treatment with TIF performed with an EsophyX device. The study included 100 consecutive patients with GERD who had failed medical management and/or had extraesophageal manifestations of GERD. The mean patient age at baseline was 52 years, the mean body mass index was 26.5 kg/m2, and approximately one third (35%) of the study population were men. The average duration of GERD was 10 years, and the average duration of PPI use was 8 years.

No complications were reported either during or after the procedure. The procedure failed in 6 patients who underwent revision either with laparoscopic Nissen fundoplication (n=5) or TIF (n=1). Whereas 92 (92%)patients were on daily PPI therapy before the TIF procedure, 75 (75%) were able to discontinue PPI therapy after the TIF  procedure (P<.001).

Slightly more than half (53%) of patients with an available 12-month pH test demonstrated either a normalization or a 50%-or-greater improvement in esophageal acid exposure. A significant improvement was observed in the mean GERD health-related quality-of-life score from when patients were receiving PPI therapy to after the TIF procedure (23.4 vs 6.6; P<.001). A significant reduction in the average heartburn score (16.1 vs 4.3; P<.001) and regurgitation score (13.6 vs 3.2; P<.001) were also seen.

Most patients experienced an elimination of daily bothersome heartburn (65%) and regurgitation (86%). Atypical symptoms also were improved, demonstrated by a reduction in the mean Reflux Symptom Index score from 19.9 to 7.9 (P<.001).

At baseline, 80% of patients were dissatisfied with their current health condition; this dissatifacction rate decreased to 15% after TIF. De novo dysphagia was reported in 2 patients and bloating in 1 patient.


Presentations in Endoscopy

Peroral Endoscopic Myotomy Demonstrates Efficacy and Safety 

Peroral endoscopic myotomy (POEM) improves on natural orifice transluminal endoscopic surgery (NOTES), allowing for the treatment of achalasia. Stavros Stavropoulos and colleagues from the Winthrop University Hospital in Mineola, New York, presented their experience with POEM at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #6). Their findings reflect the first time that POEM has been performed outside of Japan and the first time worldwide that POEM has been performed by a gastroenterologist.

A total of 31 patients with achalasia (mean age, 51.8 years) underwent POEM with NOTES between years 2009 and 2012 and were included in the study. Presurgical evaluation included an office visit, upper endoscopy, barium swallow, computed tomography imaging of the chest and abdomen, and esophageal manometry. The primary outcome of this analysis was symptom resolution, defined as a decrease in the Eckardt score to 3 or less. Secondary outcomes included adverse events, postprocedure lower esophageal sphincter (LES) pressure, length of stay, and reflux symptoms or use of antacids. The investigators noted that their technique differed from the technique published in the literature and used in Japan. For example, although balloon inflation was initially used for tunnel dissection, the technique used by Stavropoulos and colleagues involved a knife that allows simultaneous submucosal injection and dissection and is a method that may be more accessible to Western gastroenterologists who do not have extensive experience with endoscopic submucosal dissection.

Patients were observed overnight following the procedure and underwent a barium swallow at 24 hours postsurgery to assess for presence of a leak. A clear liquid diet was initiated if the barium swallow results were negative, and the patient was discharged with a 7-day antibiotic therapy regimen. The liquid diet was switched to a soft diet at 1 week postsurgery.

The success rate of POEM was 94%. Treatment failed in 2 patients who had recurrent symptoms at 3 months (both responded to pneumatic dilation). Significant reductions were observed in both Eckardt score (7.5 to 1.1; P<.0001) and LES pressure (49mmHg to 19 mmHg; P<.0001). No complications were reported, including the need for intensive care unit stay, the need for hospital stays over 5 days, or the need for surgical interventions or blood transfusions. No patients underwent a surgical conversion or POEM-related readmission, and most (87%) patients did not require any post-POEM analgesia.


Rectal NSAIDs Are More Effective Than Pancreatic Stents in Preventing Post-ERCP Pancreatitis

Pancreatic stent placement and rectal administration of nonsteroidal anti-inflammatory drugs (NSAIDs) have shown benefit in the prevention of postendoscopic retrograde cholangiopancreatography pancreatitis (PEP) in high-risk patients. However, a gap exists in the literature regarding randomized controlled trials that directly compare pancreatic stent placement with rectal NSAIDs in this setting. In comparing these procedures, Venkata S. Akshintala and colleagues from the Johns Hopkins Hospital in Baltimore, Maryland, deduced that rectal NSAIDs are more effective than pancreatic stents in preventing PEP in at-risk patients. This study, reported at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #55), was a systematic review and network meta-analysis aimed to compare rectal NSAIDs with PEP.

MEDLINE, EMBASE, and Cochrane Library databases were searched for randomized controlled trials meeting the inclusion criterion, which was being a full-text publication that reported the incidence of PEP as the primary outcome in high-risk patients receiving either rectal NSAIDs or pancreatic stenting. The investigators identified a total of 7 randomized controlled trials (n=1,168) that met the inclusion criterion. Of these, 5 trials evaluated pancreatic stenting, 1 trial evaluated rectal NSAIDs, and 1 trial evaluated both procedures. All 7 trials were simultaneously analyzed with the mixed treatment comparisons method of network meta-analysis, using a Bayesian approach. Treatments were evaluated according to the relative predictive probability of being ranked as the most effective.

The network meta-analysis showed that rectal NSAID use was more effective for prevention of PEP than either pancreatic stenting or placebo (probability of 0.66 for ranking as the best procedure). Pancreatic stenting ranked as the second best (probability of 0.33). Rectal NSAIDs were predicted to decrease the risk of PEP by 47% compared with pancreatic stenting and were associated with a relative risk reduction of 61.7% compared with pancreatic stenting for the prevention of PEP.


Covered or Uncovered Metal Stents: Utility Is the Same in the Management of Malignant Biliary Strictures 

There were no differences in utility between covered and uncovered self-expanding metal stents for management of malignant biliary obstruction, according to Jeffrey H. Lee from the MD Anderson Cancer Center in Houston, Texas. Lee reported study results during a poster session at the 2012 Annual Scientifc Meeting of the American College of Gastroenterology (Abstract #P650). Study outcomes included measurement of stent patency rate, overall survival, and complications in this retrospective cohort study from a single tertiary cancer center.

The study included 749 patients seen between years 2000 and 2011 who met the following inclusion criteria: 1) presence of a malignant biliary obstruction and presentation for endoscopic retrograde cholangiopancreatography, biliary decompression, and first-time metal stent placement; and 2) a history of malignant biliary obstruction, prior placement of a plastic biliary stent, and presentation for first-time metal stent placement. Of these patients, 171 received a covered self-expanding metal stent and 578 received an uncovered self-expanding metal stent. No difference was observed in median overall survival between the covered and uncovered groups (10.4 months vs 11.7 months; P=.84). The median time to recurrent biliary obstruction was 6.2 months for the group with a covered stent and 4.05 months for the group with an uncovered stent (hazard ratio, 1.06; 95% confidence interval, 0.70–1.58).

The proportion of patients who had recurrent biliary obstruction did not differ significantly between groups (P=.61), but the type of obstruction differed significantly. Tumor ingrowth occurred in 76% of patients with an uncovered stent and 9% of patients with a covered stent (P<.001). Patients with covered stents were more likely than those with uncovered stents to have tumor overgrowth (15% vs 2%), sludge stone (18% vs 3%), food debris (12% vs 5%), and stent migration (36% vs 2%). Acute pancreatitis also was more likely to occur in patients with covered stents than in patients with uncovered stents (6% vs 1%; P<.001). A total of 109 patients underwent surgery with the self-expanding metal stents in place. No stent-related intraoperative or postoperative complications were reported.


A Novel Teaching Tool Aids Trainees in Gastroenterology in Histologic Characterization of Diminutive Colorectal Polyps 

Significant improvement in both accuracy of histologic characterization of polyps and the proportion of high-confidence diagnoses among trainees in gastroenterology was achieved via a novel computer-based teaching tool combined with short narrow-band imaging (NBI). The study, which was presented by Swati G. Patel from the University of Colorado at Denver School of Medicine during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1096), aimed to determine the performance characteristics and learning curve among gastroenterology trainees who were taught characterization of diminutive polyp histology using NBI. A computer-based training tool and real-time NBI video clips were used in training to simulate clinical practice. The teaching module included previously validated NBI criteria that could be used to differentiate adenomas from hyperplastic polyps.

A total of 80 randomly distributed short videos of polyps (both adenomas and hyperplastic polyps) under NBI with magnification were viewed, after which participants reported the predicted polyp histology and their degree of confidence. Following each video assessment, feedback was provided regarding the histology, incorporating NBI criteria that supported the diagnosis. A total of 12 gastroenterology trainees were included in the study. Three trainees were in their first year of training, 4 were in their second year, and 5 were in their third year. They had a wide range of colonoscopy experience (from 51 to >500 colonoscopies performed).

There was a significant improvement in accuracy rates and the proportion of high-confidence predictions with increasing views of video blocks (P<.001 for the trend). The overall accuracy rate was 90%, with the accuracy rate steadily increasing with the numbers of videos viewed (83%, 86%, 93%, and 96% for the 1–20, 21–40, 41–60, and 61–80 video blocks, respectively). The overall positive predictive value was 95%, and the overall negative predictive value was 82%. A high degree of confidence had a greater positive correlation with a high accuracy rate than a low degree of confidence (96% vs 72%; P<.001). A substantial overall interobserver agreement was seen (kappa=.71), and no significant differences were observed in relation to year of training or extent of colonoscopy experience.


Endoscopic Ultrasound-Guided Rendezvous May Facilitate Pancreatic Endotherapy after Pancreaticoduodenectomy

Endoscopic ultrasound (EUS)-guided rendezvous following pancreaticoduodenectomy (PD) or pylorus-preserving PD (PPPD) can facilitate pancreatic endotherapy in select cases, according to Ihab I. El Hajj and colleagues from the Indiana University in Indianapolis, whose finding were presented during a poster session at the 2012 Annual Scientific Meeting of the American College of Gastroenterology (Abstract #P1212). Their study sought to characterize the performance characteristics of EUS rendezvous following classic PD or PPPD.

This was a single-center case series of EUS-guided rendezvous procedures that were performed following PD or PPPD between years 2005 and 2012. In all cases, the technique involved an EUS transgastric puncture of the pancreatic duct, followed by wire passage into the jejunum and intraluminal retrieval of the wire by another endoscope for an attempted endotherapy via the pancreatojejunal anastomosis. A total of 26 patients, with a median age of 55.5 years, underwent 30 EUS procedures. The procedures occurred at a median of 933 days (range, 128–180) following either PD (n=9) or PPPD (n=17). Indications for the procedure included suspected anastomotic stricture (n=13), stricture with filling defects (n=3), stricture and stones (n=6), impacted main pancreatic duct stents (n=2), and acute recurrent pancreatitis (n=6). Endoscopic retrograde cholangiopancreatography had failed once in 16 patients and twice in 2 patients prior to EUS and had not been previously attempted in 12 patients.

The EUS-guided pancreatogram was successful in all 30 procedures, and wire passage across the pancreatojejunal anastomosis was successful in 16 (54%). The wire was successfully grasped in 9 (30%) procedures, with successful retrieval of the wire and pancreatic endotherapy.

Complications included a needle fracture and a peripancreatic abscess. In a univariate analysis, no differences were observed between cases that did or did not have successful wire passage into the jejunum. Short-term clinical success, defined as pain relief up to 6 months after the procedure, was achieved in all 9 procedures in which pancreatic endotherapy was successful. Of 21 patients who had an unsuccessful EUS-guided wire retrieval, 2 patients improved after endoscopic pancreatogastrostomies, 15 were referred for surgery, 2 had celiac plexus blocks, and 2 were lost to follow-up.


Single-Center Retrospective Study Sheds Light on Recurrence Rate of Previously Resected Large Polyps 

Endoscopic mucosal resection (EMR) appears to be a safe and generally effective procedure for removal of large (>2 cm) and difficult-to-remove colorectal polyps and was associated with a very low risk of local recurrence, according to Niket Sonpal from the Lenox Hill Hospital in Hauppauge, New York. Sonpal reported findings of a retrospective review of recurrence rates of colorectal polyps during a poster session at the 2012 Annual Meeting of the American College of Gastroenterology (Abstract #P1469). The review sought to determine the recurrence rate of colorectal polyps after removal of large polyps by advanced polypectomy and EMR.

A total of 262 patients whose mean age was 66 years were identified from an endoscopy database and patient records. It was determined that a failed attempt at polyp removal was made by the referring gastroenterologist in 47 (18%) of these patients. Polyps were successfully removed in the remaining 215 (82%) patients, with a recurrence rate of 5.4%.

In the 47 patients in whom polyp removal failed, 45 (95%) were successfully treated; the remaining 2 patients required subsequent surgical referral.

No immediate complications or hospitalizations were reported. A 2-stage procedure was required for completion in 3 patients. Surveillance colonoscopies were performed at a mean follow-up of 14.7 months. Of the 45 treated patients, 13 (29%) were referred for surgical intervention and resection.

At follow-up, histology revealed invasive cancer in 10 (77%) and tubular adenomas in 3 (23%) of the 13 patients referred for surgery. Eight (61%) of these patients underwent surgery. Six (75%) of these 8 patients had a segmental resection with anastomosis, and 2 (25%) had subtotal and total colectomies.


Bowel Wall Thickening on Radiologic Imaging Warrants Endoscopic Evaluation

The finding of bowel wall thickening on radiologic imaging warrants further endoscopic evaluation, according to Pierre Hindy, of the State University of New York Health Science Center at the University Hospital of Brooklyn. Hindy and colleagues assessed the rate of malignancy, clinically significant pathology, and risk factors in patients with gastrointestinal wall thickening who were undergoing endoscopic evaluation. The findings were reported during a poster session at the 2012 Annual Meeting of the American College of Gastroenterology (Abstract #P1475).

An interim analysis of retrospective imaging (computed tomography and magnetic resonance imaging) data from years 2001 to 2007 was performed. A total of 11,935 patients with the word “thickening” on imaging reports were included in the study. A total of 3,103 cases were reviewed. Gastrointestinal thickening was found in 489 (15.7%) cases (all in male veterans). Endoscopy was performed in 352 (72%) of these cases within 3 months of the imaging findings. Significant pathology was evident on endoscopy in 156 (32%) of these cases, and malignant lesions were observed in 127 (26%).

Using a stepwise approach, the investigators determined the odds of malignancy with adjustment for age, race, body mass index, incidence of diabetes mellitus, cigarette smoking, anemia, and ferritin levels. Patients with malignancy had a higher mean age compared with those without malignancy (70.9 years vs 64.8 years, respectively; P<.01). Among those with cancer, 45% of patients were racially white, 36% were black, and 20% were Hispanic. No significant difference was found when comparing patients with or without a cancer diagnosis with regard to either body mass index or diabetes mellitus, however; cigarette smoking carried a significant risk for malignancy (odds ratio [OR], 1.772; 95% confidence interval [CI], 1.116–2.815) as did anemia (OR, 1.598; 95% CI, 1.006–2.538; P<.05 for both risk factors). The risk of malignancy also was significantly increased with age (OR, 1.042; 95% CI, 1.022–1.064; P<.01).


Presentations in Hepatology

A Clinical Decision Tool May Help Predict Response in Patients Receiving Triple Therapy 

Triple therapy consisting of boceprevir plus peginterferon and ribavirin is an effective treatment option for many patients with hepatitis C virus (HCV) infection. However, a number of factors can influence patient response. To explore these factors, a multicenter team developed clinical decision tools to predict HCV undetectability at Week 8 of treatment and sustained virologic response (SVR). The process was described by Scott Devine of Merck in Whitehouse Station, New Jersey, during a poster session at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #1842).

Devine and colleagues built logistic regression models to predict HCV undetectability at Week 8 of treatment and SVR. The analysis used data from 1,227 patients from the SPRINT-2, RESPOND-2, and PROVIDE trials of boceprevir. Factors used for the development of the models included prior treatment with peginterferon and ribavirin, interleukin (IL)-28B genotype, HCV genotype 1 subtype, initial ribavirin dose, age, race, sex, HCV RNA level after 4 weeks of peginterferon and ribavirin therapy, log10 reduction in HCV RNA levels from baseline to Week 4, and baseline characteristics (weight, body mass index [BMI], hemoglobin level, fibrosis score, the ratio between alanine aminotransferase [ALT] level and the upper limit of normal, platelet count, statin use, steatosis score, and HCV RNA level). Final models that included baseline variables plus HCV RNA level at Week 4 were developed to predict response at Week 8 (n=856) and SVR (n=522). Both models included treatment-naïve patients, relapsers, and partial and prior nonresponders.

A step-down approach was used to reduce the final number of predictors. In the model to predict response at Week 8, the final variables were race, initial ribavirin dose, platelet count, log10 reduction in HCV RNA level from baseline to Week 4, and HCV RNA level at Week 4. In the SVR model, the final factors were sex, BMI, ribavirin use, platelet count, HCV genotype 1 subtype, and HCV RNA level at Week 4. The final model calibration curves had good discrimination for both the Week-8 response and SVR models (C-statistics, 0.89 and 0.83, respectively). In addition to successfully predicting response at Week 8 and SVR without invasive testing, these nomograms could also be useful for clinical decision-making about the initiation and maintenance of therapy.


Ritonavir’s PK Effect on Boceprevir in HCV/HIV Coinfection May Not Compromise Boceprevir Efficacy

Administration of ritonavir-boosted HIV protease inhibitors reduces boceprevir concentrations in healthy volunteers. To further explore this interaction, Larissa A. Wenning of Merck, Whitehouse Station, New Jersey, and colleagues assessed boceprevir pharmacokinetics (PK) in patients coinfected with hepatitis C virus (HCV) and HIV. The study also evaluated the relationships among boceprevir PK and pharmacodynamics (PD), sustained virologic response (SVR), and anemia. These data were presented during a poster session at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #770).

Data from a phase II HCV/HIV coinfection study, the phase III SPRINT-2 study, and the phase III RESPOND-2 study were analyzed. Data on boceprevir-related PK were available for 51 patients in the coinfection study, 105 patients in SPRINT-2, and 84 patients in RESPOND-2. A population PK model was used to estimate the PK parameters. For the study arms that contained boceprevir, the overall response was estimated using a linear regression model for SVR or anemia, in which boceprevir PK (area under the curve from 0–8 hours [AUC0-8hr] or concentration at 8 hours [C8hr]) were used as predictors.

The cross-study comparison of boceprevir PK found that the boceprevir AUC0-8hr was approximately 20% lower in the HCV/HIV coinfection study than in the studies of patients monoinfected with HCV. In addition, the C8hr was approximately 27% lower in the coinfection study compared with the monoinfection studies. The SVR and anemia results regarding PK and PD were similar for both boceprevir AUC0-8hr and C8hr. Thus, Wenning and colleagues were unable to determine whether AUC0-8hr or C8hr was a better predictor of efficacy or safety. The study also found no significant relationship between boceprevir PK and SVR rates. However, there was a lower probability of anemia (hemoglobin level of 8.5–10 g/dL) with decreasing boceprevir PK, although this result was not significant for the coinfection study data alone.

The investigators concluded that overall boceprevir exposure was reduced in patients coinfected with HCV/HIV compared with patients monoinfected with HCV. However, reduced boceprevir exposure is unlikely to adversely influence the efficacy of treatment, given that the relationship between boceprevir PK and SVR rates was not significant. Although reduced boceprevir exposure was associated with a reduced probability of anemia, data on ribavirin PK were not collected in the coinfection study, which showed no relationship between boceprevir PK and ribavirin dose; however, ribavirin cannot be eliminated as a confounding factor in the analysis.


The Second-Generation HCV NS3/4A Protease Inhibitor MK5172 Retains Potent   In Vitro Activity Against Boceprevir-Resistant Genotype 1 HCV Isolates

The second-generation hepatitis C virus (HCV) NS3/4A protease inhibitor MK5172 demonstrated activity against multiple HCV genotypes and has been shown to significantly reduce viral load in patients with genotype 1 HCV infection, according to Robert A. Ogert, from Merck Sharpe & Dohme in Kenilworth, New Jersey, who presented study findings during a poster session at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #1724). The study sought to confirm the activity of MK5172 against HCV isolates from patients who were clinically resistant to boceprevir. Ogert and coinvestigators amplified the NS3 gene of 13 different clinical isolates from patients who failed therapy with boceprevir plus peginterferon and ribavirin. The amplified NS3 genes were then tested against MK5172 using an in vitro, replicon-based phenotypic assay. The resistant isolates also were tested against boceprevir, telaprevir, and simeprevir.

Six genotype 1a isolates and 8 genotype 1b isolates with boceprevir-resistance–associated variants were grouped according to virologic response. There were 8 isolates from patients with incomplete virologic response (3 genotype 1a, 5 genotype 1b), 4 isolates from patients who experienced virologic breakthrough (1 genotype 1a, 3 genotype 1b), 1 relapser (genotype 1a), and 1 nonresponder (genotype 1a). Viral load plots that indicated the presence of resistance-associated variants were presented based on these groupings.

Among patients with incomplete virologic response, the resistance-associated variants in genotype 1a isolates were V36M, T54S, R155K, R155K/T, and A156S, while the variants in genotype 1b isolates were T54A/S, T54A, V170A, T54S, and R155K. Among patients in whom virologic breakthrough occurred, the resistance-associated variant in the genotype 1a isolate was R155T, and the resistance-associated variants in the genotype 1b isolates were V55A, T54A, V170A, and M175L. The variants present in the patient with genotype 1a who relapsed were T54S and R155K, while the patient who was a nonresponder had V36M and R155K variants.

The isolates from patients with genotype 1a who failed boceprevir-based therapy demonstrated in vitro resistance to boceprevir, telaprevir, and simeprevir. An 8–13-fold (boceprevir), 18–36-fold (telaprevir), and greater-than-10-fold (simeprevir) shift in the half maximal inhibitory concentration (IC50) from baseline was observed. The boceprevir failure genotype 1a isolates that were resistant to boceprevir, telaprevir, and simeprevir were responsive to MK5172 (IC50, 0.6–4.4 nM).

Compared with baseline isolates, the isolates from patients with genotype 1b who failed boceprevir were resistant to boceprevir (2.7-fold shift in IC50) and telaprevir (2.8-fold shift in IC50). In contrast to the boceprevir failure genotype 1a isolates, the majority of genotype 1b isolates remained sensitive to simeprevir. Similar to the boceprevir failure genotype 1a isolates, the boceprevir failure genotype 1b isolates were sensitive to MK5172 (IC50, 0.04 –0.25 nM) and had a greater-than-2-fold shift in IC50 from baseline. Ogert noted that further studies are underway, including a clonal sequence analysis and deep sequencing of select patient samples.


OPTIMIZE Results Show Noninferiority of Telaprevir Twice Daily Compared with 3 Times Daily 

The OPTIMIZE trial, the first phase III clinical trial comparing twice-daily administration of telaprevir with 8-hour administration, met its primary endpoint of showing noninferiority in sustained virologic response at Week 12 (SVR12) rates for twice-daily versus 8-hour dosing of telaprevir in combination with peginterferon and ribavirin.

The findings were presented by Maria Buti, of the Hospital General Universitari Vall d’Hebron and Ciberehd in Barcelona, Spain, in a late-breaker poster at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #LB-8).

A total of 740 treatment-naïve patients with genotype 1 hepatitis C virus (HCV) infection were randomly selected to receive 12 weeks of peginterferon (180 µg/week) and ribavirin (1,000–1,200 mg/day) plus telaprevir at either 1 of 2 dosages: 750 mg every 8 hours or 1,125 mg every 12 hours. All patients then received peginterferon and ribavirin without telaprevir for an additional 12 or 36 weeks. The total treatment duration was 24 or 48 weeks. Administration of telaprevir was halted if HCV RNA levels were greater than 1,000 IU/mL at Week 4 or if HCV RNA levels were at or above 25 IU/mL at Weeks 12, 24, 32, or 40. Patients were followed until Week 72.

Twice-daily telaprevir was found to be noninferior to telaprevir administered every 8 hours (SVR12, 74% vs 73%, respectively; 95% confidence interval, -4.9–12). A subgroup analysis based on liver fibrosis status and interleukin (IL)-28B genotype also demonstrated similar SVR12 rates for both dosing regimens. Among cirrhotic patients, SVR12 rates were 54% for patients who received telaprevir at a dosage of 1,125 mg twice daily versus 49% for those who received telaprevir at a dosage of 750 mg every 8 hours. In noncirrhotic patients, SVR12 rates were 78% and 77% for twice-daily versus every-8-hour dosing of telaprevir, respectively. In addition, rapid virologic response (RVR) rates were similar for both dosing regimens (69% and 67%, respectively). In patients who achieved RVR, SVR rates were 86% and 85% for twice-daily versus every-8-hour dosing of telaprevir; in patients who did not achieve RVR, the SVR rate was 47% with either dosing regimen. Relapse rates were 8% for patients who received 1,125 mg of telaprevir twice daily and 7% for patients who received 750 mg of telaprevir every 8 hours. Both dosing regimens had an on-treatment virologic failure rate of 10%.

The safety and tolerability of telaprevir were similar in patients receiving 1,125 mg twice daily and those receiving 750 mg every 8 hours. The most common adverse events in both groups were fatigue, pruritus, anemia, nausea, rash, and headache. Serious adverse events occurred in 8–9% of patients. Treatment discontinuation due to adverse events occurred in 15% of patients who received 1,125 mg of telaprevir twice daily and 19% of patients who received 750 mg of telaprevir every 8 hours. Because the safety profiles and SVR rates were similar for both treatment arms, Buti and her coinvestigators concluded that telaprevir given at a dosage of 1,125 mg twice daily plus peginterferon and ribavirin could offer a safe, effective, and simplified treatment option for patients with genotype 1 HCV infection.


Interim Study Results Show Promise for Telaprevir in Patients with HCV Infection and Severe Fibrosis or Compensated Cirrhosis

Interim results of HEP3002—an ongoing, international, early-access program for patients infected with genotype 1 hepatitis C virus (HCV) with severe fibrosis or compensated cirrhosis—suggest that telaprevir has value in this patient population. Results from the 609 patients of the more than 1,900 patients enrolled in the study were presented by Massimo Colombo of the Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico at the University of Milan, Italy, in a late-breaker poster at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #LB-15).

Enrollment criteria included genotype 1 HCV infection, severe fibrosis, or compensated cirrhosis (Metavir score of F3 or F4), and a platelet count of more than 90,000 cells/mm3. The mean age of the patients was 53.5 years, 67% of the patients were men, and 98% were white. In addition, 66% of patients had HCV RNA levels of at least 800,000 IU/mL, 45% of patients had severe fibrosis, 55% had cirrhosis, and 28% had genotype 1a HCV infection. At baseline, 20% of patients were treatment-naïve, 28% were prior relapsers, 15% were partial prior responders, 29% were prior null responders, 3% were nonresponders for unspecified reasons, and 5% had prior viral breakthrough.

Patients were treated with telaprevir (750 mg every 8 hours) plus peginterferon and ribavirin for 12 weeks. Peginterferon and ribavirin were then administered for an additional 12–36 weeks using a response-guided treatment paradigm. At Week 4, 329 (54%) patients had undetectable HCV RNA levels. By Week 12, 481 (79%) patients had undetectable HCV RNA levels. The percentage of patients who showed an HCV RNA response at Week 12 was lower for prior null responders (73%) than for prior relapsers or treatment-naïve patients (85% for both groups).

Grade 1–4 anemia developed in 359 (59%) patients, with severe anemia occurring in 31%. Grade 1–4 rash developed in 256 (42%) patients, with severe rash occurring in 4% of patients. Discontinuation due to adverse events occurred in 14% of patients (12% of patients with F3 fibrosis and 16% of patients with F4 fibrosis). Reasons for discontinuation included rash (5%), anemia (3%), asthenia (1%), abdominal pain (1%), nausea (1%), pruritus (1%), and vomiting (1%). The investigators noted that the rates of discontinuation for rash and anemia were similar to those observed in the phase III registration trials for telaprevir. Three cirrhotic patients (0.5%) died during the peginterferon and ribavirin phase of therapy due to hepatic failure/ischemic colitis and multiorgan failure; 1 of these deaths was deemed to be treatment-related, and 1 death was possibly treatment-related.


Long-Term Tenofovir DF for Chronic Hepatitis B Appears to be Safe, Well Tolerated, and Associated with Sustained Response

Six-year results from 2 ongoing 8-year studies demonstrate that tenofovir disoproxil fumarate (tenofovir DF) has good safety and tolerability profiles and is associated with sustained response, according to Patrick Marcellin, of the Hôpital Beaujon in Clichy, France. Marcellin reported these findings at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #374).

Patients were randomly selected for treatment with either tenofovir DF or adefovir dipivoxil for 48 weeks in a double-blind comparison, after which those who underwent a liver biopsy were permitted to continue with open-label tenofovir DF for 7 additional years. Monitoring for adverse events and hepatitis B virus (HBV) DNA occurred every 3 months, resistance surveillance was performed annually, and annual bone mineral density assessments of the spine and hip were added starting at Year 4.

Of the initial 641 patients treated, 585 (93%) entered into the tenofovir DF extension phase and 477 (73%) remained on study at Year 6.

In the long-term evaluation analysis set, in which missing patients were counted as failures, 281 (81%) of 345 patients who were hepatitis B e antigen (HBeAg)-negative and 157 (62%) of 251 patients who were HBeAg-positive had HBV DNA levels of less than 400 copies/mL. Comparable percentages for patients in the on-treatment analysis set, in whom HBV DNA levels of less than 400 copies/mL, were achieved in 283 (~100%) of the 284 patients who were HBeAg-negative and 167 (99%) of the 169 patients who were HBeAg-positive. In the on-treatment analysis set, 228 (86%) of 265 patients who were HBeAg-negative and 127 (78%) of 162 patients who were HBeAg-positive showed normalization of alanine aminotransferase levels. Half of the patients who were HBeAg-positive showed loss of HBeAg, and 61 (37%) of 163 patients who were HBeAg-positive had HBeAg seroconversion.

Over the 6-year follow-up, tenofovir DF proved to be well tolerated, with fewer than 2% of patients discontinuing due to an adverse event. A confirmed renal event occurred in 1.5% or fewer patients. Over 2 years, bone mineral density levels remained stable. Importantly, no tenofovir DF resistance was detected through Year 6.


Tenofovir DF Is Safe and Effective in Patients with Chronic Hepatitis B Virus Infection Resistant to Lamivudine 

Tenofovir disoproxil fumarate (tenofovir DF)  was shown to suppress hepatitis B virus (HBV) DNA without signs of emerging drug resistance in patients with documented resistance to lamivudine, reported Scott Fung of the Toronto General Hospital in Ontario, Canada, during a presentation at the 63rd Annual Meeting of the American Association for the Study of Liver Diseases (Abstract #20). Fung and colleagues conducted a randomized, double-blind, phase IIIb trial that compared tenofovir DF (n=141) with emtricitabine plus tenofovir DF (n=139; given as a fixed-dose combination tablet). All patients had chronic HBV and documented lamivudine resistance, with 103 or more HBV DNA copies/mL at the time of study screening despite receiving lamivudine. At the time of study entry, patients were stratified by alanine aminotransferase (ALT) levels and hepatitis B e antigen (HBeAg) status.

The majority of patients in both the tenofovir DF and emtricitabine/tenofovir arms completed the 96-week study period (94% and 90%, respectively). At Week 96, a similar proportion of patients in the tenofovir DF and the emtricitabine/tenofovir DF arms had less than 400 HBV DNA copies/mL (89% and 86%, respectively). Normalized ALT levels were present in 44 (62%) of 79 patients in the tenofovir DF and 52 (63%) of 83 patients in the emtricitabine/tenofovir DF arm, and normal ALT levels were achieved in 70% of both arms.

Among HBeAg-positive patients, HBeAg loss occurred in 10 (15%) of 65 patients treated with tenofovir DF and 9 (13%) of 68 patients treated with emtricitabine/tenofovir DF. HBeAg seroconversion occurred in 7 (11%) of 65 patients in the tenofovir DF monotherapy group and 7 (10%) of 68 patients in the emtricitabine/tenofovir DF group.

Both treatments were well tolerated, with only 1% of patients discontinuing therapy due to an adverse event. There were no confirmed cases of increased levels of serum creatinine (≥0.5 mg/dL from baseline). Serum phosphorous levels below 2 mg/dL occurred in 1% of patients, and reduced creatinine clearance (<50 mL/min) occurred in 3% of patients. No clinically relevant bone loss or nontraumatic bone fractures were observed. Over the 96-week study period, no tenofovir DF resistance was observed.

Millennium Medical Publishing, Inc