THE GASTRO & HEP REPORT
Comprehensive Reports on the Latest Advances in Gastroenterology and Hepatology From:
• The 76th American College of Gastroenterology Annual Scientific Meeting and Postgraduate Course
October 28–November 2, 2011
Washington, DC
• The 62nd Annual Meeting of the American Association for the Study of Liver Diseases
November 4–8, 2011
San Francisco, California
• 2011 Advances in Inflammatory Bowel Diseases/Crohn’s & Colitis Foundation’s Clinical and Research Conference
December 1–3, 2011
Hollywood, Florida
Presentations in GERD
Comparison of Impedance-pH Monitoring and a Symptom-Based Reflux Disease Questionnaire for the Identification of GERD
The GerdQ is a 6-item, symptom-based reflux disease questionnaire that has been validated to help identify patients with gastroesophageal reflux disease (GERD). At the 2011 American College of Gastroenterology (ACG) Annual Scientific Meeting and Postgraduate Course, Chehade and colleagues presented results from a prospective study that aimed to evaluate the association between the GerdQ and parameters of impedance-pH monitoring.
A total of 85 consecutive patients who were referred for impedance-pH monitoring over an 11-month period were included. The majority of patients were white (99%), half were female (52%), the average age was 53±15 years, and the mean body mass index (BMI) was 27.2±6.2 kg/m2. Patients had a mean duration of symptoms of 100±108 months, and 59% of patients completed the study while receiving proton pump inhibitor (PPI) therapy.
Abnormal acid exposure occurred in 8% and 18% of patients on and off PPI therapy, respectively (P=.19). Abnormal acid and nonacid events, combined, also occurred in a similar number of patients on and off treatment (20% and 21%, respectively; P=.99). After controlling for age, sex, and BMI, no significant differences were observed between patients on and off PPI therapy in terms of likelihood of an abnormal impedance-pH monitoring result.
Among patients on and off PPI therapy, mean GerdQ scores were 9.2±2.9 and 8.8±2.6, respectively (P=.64). For patients who were on treatment, a GerdQ score of 8 or higher had 75% sensitivity (range, 19–99%) for identifying patients with abnormal acid exposure time. However, the specificity of the GerdQ at this threshold was only 26% (range, 14–41%). In patients off treatment, the sensitivity and specificity were 100% (range, 54–100%) and 37% (range, 19–58%), respectively. The investigators concluded that the GerdQ may be an important tool for differentiating acid versus nonacid reflux disease. Further study in a larger patient population is required for confirmation.
Prevalence, Association, and Effect of Gender on Cough and GERD in Untreated Patients Over Age 65
Cough is a reported symptom of GERD; like GERD, cough is also known to increase with patient age. At the 2011 ACG meeting, DeVault and colleagues presented a study that assessed the effect of gender on cough and GERD in older patients.
A total of 300 consecutive patients older than 65 years (mean age, 72 years; range, 65–88 years) were identified after being referred for screening or surveillance colonoscopy. Patients had not previously undergone an upper endoscopic procedure and were not on antisecretory therapy. Nearly half (46%) of the patients were female.
Approximately one fourth (27%) of the 300 patients reported cough as a symptom, with frequency of cough ranging from 1 to 75 times weekly. Of these 82 patients, 20 were not distressed by their cough, 46 were somewhat distressed, 12 were quite a bit distressed, and 4 were very much distressed. Mucosal disease or symptoms consistent with GERD were identified in 52% of patients, including 19% with Barrett esophagus (BE), 14% with erosive esophagitis (EE), and 19% with other significant reflux symptoms.
Cough was significantly more prevalent among GERD patients versus non-GERD patients (31.2% vs 23.1%; P=.006). Cough also occurred more frequently in females than males (28.9% vs 24.2%; P=.09). GERD was more common in males than females (57.8% vs 43.0%; P<.0001). Among GERD patients, mucosal disease was more common in males than females (76.3% vs 45.3%; P<.0001), while cough was more frequent in females (35.9% vs 28.0%; P<.0001). The investigators concluded that GERD and cough were indeed associated in older patients, but further study is required to determine the cause of this link.
Increasing Body Mass Index Does Not Diminish Heartburn Relief in Patients Treated for Nonerosive GERD or Erosive Esophagitis
As presented by Peura and colleagues at the 2011 ACG meeting, a post–hoc analysis was performed using data from patients with nonerosive GERD (NERD) or EE who were enrolled in phase III trials. The goal of this analysis was to determine if BMI affected the severity and/or frequency of heartburn symptoms.
A total of 3,369 patients were included in this post–hoc analysis. NERD patients were enrolled in a randomized, double-blind study comparing 4 weeks of dexlansoprazole (Dexilant, Takeda; N=315) versus placebo (N=317) for 24-hour heartburn relief. EE patients (N=2,737) were enrolled in a double-blind, randomized study comparing 8 weeks of dexlansoprazole versus lansoprazole for EE healing.
When patients were distributed by baseline BMI into 1 of 3 groups (<25 kg/m2, 25 kg/m2 to <30 kg/m2, and ≥30 kg/m2), there were 190, 193, and 238 NERD patients in each group, respectively. The corresponding numbers for EE patients were 547, 1,013, and 1,132, respectively. The severity of heartburn symptoms at baseline rose with increasing BMI in both the NERD and EE patient cohorts. Among NERD patients, heartburn symptom severity (as rated on a 5-point scale from 0=none to 4=very severe) was 1.14–1.32, 1.21–1.42, and 1.36–1.50 for patients with BMIs below 25 kg/m2, 25 kg/m2 to less-than-30 kg/m2, and 30 kg/m2 or higher, respectively. In EE patients, heartburn symptom severity was 1.00–1.07, 1.17–1.25, and 1.43 for these 3 BMI groups, respectively.
During treatment with either dexlansoprazole or lansoprazole, patients with higher BMIs had lower severity of heartburn (NERD patients: 0.50, 0.33, and 0.38; EE patients: 0.19–0.20, 0.13–0.15, and 0.12–0.14; for BMI <25 kg/m2, 25 kg/m2 to <30 kg/m2, and ≥30 kg/m2, respectively). Patients with higher BMIs also exhibited a higher percentage of heartburn-free 24-hour days with treatment (NERD patients: 51.9%, 57.7%, and 56.0%; EE patients: 73.6–77.0%, 80.0–82.2%, and 81.8–84.8%; for BMI <25 kg/m2, 25 kg/m2 to <30 kg/m2, and ≥30 kg/m2, respectively). The investigators concluded that NERD and EE patients with higher BMIs had increased severity and frequency of heartburn at baseline but that these patients may actually derive a greater benefit during dexlansoprazole treatment.
Diagnostic Utility of Major Basic Protein and Eotaxin-3 Staining to Differentiate Eosinophilic Esophagitis From GERD
At the 2011 ACG meeting, Dellon and colleagues presented a report on the potential utility of 2 biomarkers, major basic protein (MBP) and eotaxin-3, that might help to differentiate eosinophilic esophagitis (EoE) from GERD. A total of 51 patients with EoE were enrolled based on a diagnosis of EoE as defined by the criteria in current consensus guidelines. The 55 GERD patients enrolled in this study all had signs of inflammation (including the presence of eosinophils on biopsy) and thus represented a patient group in which diagnostic difficulty is likely. MBP and eotaxin-3 expression in the esophageal epithelium was quantified by immunohistochemistry staining density (positive cells/mm2); the number of eosinophils per high-power field (hpf) was also determined.
The characteristics of the EoE and GERD patient groups were relatively similar, including mean age (24 years and 34 years, respectively) and sex (69% male and 61% male, respectively). At baseline, patients with EoE had a much higher mean eosinophil count compared to GERD patients (143 eosinophils/hpf vs 20 eosinophils/hpf, respectively). MBP maximum staining density was also significantly higher in patients with EoE versus GERD (1,479±1,290 cells/mm2 vs 59±103 cells/mm2; P<.001), as was the maximum staining density of eotaxin-3 (2,219±1,782 cells/mm2 vs 479±777 cells/mm2, respectively; P<.001).
Construction of a receiver operator characteristic (ROC) curve revealed a strong association between MBP expression and eosinophil count (R=.81; P<.001) and a weaker association between eotaxin-3 expression and eosinophil count (R=.25; P=.01). The area under the ROC curve for diagnosis of EoE based on a combination of MBP and eotaxin-3 staining plus eosinophil count was .99, compared to .96 for MBP staining alone, .87 for eotaxin-3 staining alone, and .96 for MBP plus eotaxin-3 staining. The investigators concluded that eotaxin-3 and MBP expression may have greater diagnostic utility
than eosinophil count alone for the differentiation of
EoE versus GERD.
Comparison of Aerosolized Swallowed Fluticasone to Esomeprazole for Treatment of Eosinophilic Esophagitis
In a prospective, single-blinded, randomized, controlled trial presented by Moawad and colleagues at the 2011 ACG meeting, aerosolized swallowed fluticasone steroid therapy was compared to the oral PPI esomeprazole for the treatment of EoE. A total of 42 patients with EoE were enrolled; all were diagnosed based on both clinical symptoms (including dysphagia and food impaction) and histologic criteria (>15 eosinophils/hpf). Patients were randomized to 8 weeks of treatment with aerosolized, swallowed, twice-daily fluticasone or oral daily esomeprazole. The vast majority of patients were male (90%) and white (81%), with a mean age of 38+10 years. At randomization, patients with coexisting GERD at baseline (n=8) were equally stratified into each treatment arm.
No significant difference was observed between the 2 treatment groups in terms of resolution of EoE (defined as <7 eosinophils/hpf). The proportion of patients in the fluticasone and esomeprazole treatment groups who achieved resolution of EoE was 19% and 35%, respectively (P=.247). Clinical assessment using the Mayo Dysphagia Questionnaire revealed a significant improvement in the esomeprazole group (mean score, 19+21 before therapy vs 1+5 after therapy; P<.001) but not in the fluticasone group (mean score, 17+18 before therapy vs 12+16 after therapy; P=.162). Neither of the treatment groups exhibited significant changes in endoscopic findings, eosinophil counts, or other histologic markers.
Of the EoE patients who had coexisting GERD, 0% and 100% of patients in the fluticasone and esomeprazole arms, respectively, achieved resolution of EoE (P=.029). Among patients with coexisting GERD, neither treatment led to an improvement in the Mayo Dysphagia Questionnaire. The investigators concluded that there seemed to be no overall differences between fluticasone and esomeprazole for resolution of EoE, although patients with coexisting GERD may benefit more from esomeprazole.
Can Proton Pump Inhibitors Prevent Progression to Dysplasia in Barrett Esophagus?
Chemoprevention to limit the progression of BE precursor lesions to esophageal adenocarcinoma is a major topic of interest. PPIs have shown some promise as a chemopreventative agent in this setting; based on a lack of definitive studies, however, current guidelines do not recommend their use for this purpose. In a retrospective review presented at the 2011 ACG meeting, Altawil and colleagues identified several characteristics in patients with BE and used these characteristics to examine the association between PPI therapy and progression to dysplasia or esophageal adenocarcinoma.
A total of 77 patients with pathologically confirmed BE were included; most of these patients were white (75%) and male (96%), with an average age of 60 years. Of the 77 patients in this study, over half (64%) were on PPI therapy. Significantly fewer patients on PPI therapy developed dysplasia or esophageal adenocarcinoma compared to patients who were not on PPI therapy (14% vs 36%; P=.03).
Two characteristics were identified as being significantly different in patients on and off PPI therapy: BMI (28.9 kg/m2 vs 26 kg/m2; P=.03) and use of histamine-2 receptor antagonists (2% vs 29%; P=.006). There were no significant differences between these 2 groups in terms of patient age (58.3 vs 61.7 years; P=.27), race (78% white vs 71% white; P=.53), sex (96% male vs 96% male; P=.91), or use of other medications (including aspirin [10% vs 11%; P=.95], nonsteroidal anti-inflammatory drugs [NSAIDs; 16% vs 14%; P=.82], and statins [37% vs 32%; P=.69]). From these retrospective data, the investigators concluded that PPI therapy was indeed a potentially valuable chemopreventative tool in patients with BE, one that should be studied further in larger prospective studies.
Helicobacter pylori Gastritis Is Inversely Correlated with Dysplasia in Patients with Barrett Esophagus
A large national study recently demonstrated an inverse association between Helicobacter pylori infection and BE.1 At the 2011 ACG meeting, Trapasso and colleagues presented a follow-up study in which they investigated the role of H. pylori on the progression of BE to dysplasia and neoplastic lesions.
A total of 172,329 patients (median age, 56 years; 43.5% male) were identified via electronic medical records from a gastrointestinal pathology laboratory’s large nationwide patient population. These individuals had simultaneous esophageal and gastric biopsies. Control patients (N=78,505) had no BE or evidence of cancer; this group included only patients with no history of GERD, in order to avoid a false low rate of H. pylori positivity. The control group was used as a reference comparator group in this analysis; the prevalence of H. pylori infection in these patients was 7.2%.
BE with no evidence of dysplasia was diagnosed in 13,836 patients, while BE with evidence of low-grade or high-grade dysplasia was diagnosed in 283 and 150 patients, respectively. The prevalence of H. pylori infection was 3.7% in the group with BE and no dysplasia (odds ratio [OR], 0.49; 95% confidence interval [CI], 0.45–0.54; P<.0001), compared to 3.2% (OR, 0.42; 95% CI, 0.21–0.82; P<.01) and 2.7% (OR, 0.35; 95% CI, 0.13–0.95; P<.05) in the low-grade and high-grade dysplasia groups, respectively.
BE with adenocarcinoma was diagnosed in 104 patients; the prevalence of H. pylori infection was 6.7% in these patients (OR, 0.93; 95% CI, 0.42–2.00; P=not significant). Esophageal squamous-cell carcinoma was diagnosed in 83 patients; this group had an H. pylori prevalence of 9.6% (OR, 1.36; 95% CI, 0.66–2.85; P=not significant).
The investigators concluded that H. pylori infection is inversely related with both low-grade and high-grade dysplasia in patients with BE. In contrast, the investigators found no such relationship in patients with either adenocarcinoma or squamous-cell carcinoma, possibly due to the inclusion of junctional adenocarcinomas of gastric origin in the former group and similar risk factors (such as lower socioeconomic status, ethnic background, and smoking) in the latter group.
Reference:
1. Sonnenberg A, Lash RH, Genta RM. A national study of Helicobactor pylori infection in gastric biopsy specimens. Gastroenterology. 2010;139:1894-1901.
Presentations in IBS
Efficacy and Safety of Once-Daily Linaclotide in Patients with IBS with Constipation
At the 2011 ACG meeting, Chey and colleagues reported on pooled results from 2 randomized, double-blind, placebo-controlled, phase III trials that evaluated the investigational agent linaclotide for the treatment of irritable bowel syndrome associated with constipation (IBS-C).
In both trials, linaclotide was administered at a dose of 290 µg once daily for 12 weeks. Patients were randomized to treatment with either linaclotide or placebo following a 2-week baseline period.
A total of 1,602 patients were included in the pooled intent-to-treat population (median age, 44 years; 90% female). During baseline assessment, the majority of patients (87%) experienced abdominal pain on a daily basis. No complete spontaneous bowel movements (CSBMs) were reported in 76% of patients.
Following treatment, linaclotide was associated with a significant improvement in abdominal and bowel symptoms, even when controlling for multiplicity (nominal P<.0001 for all 14 primary and secondary endpoints). Specifically, linaclotide was associated with significant improvements in the following primary endpoints: 30% or greater reduction in abdominal pain, at least 3 CSBMs, and an increase of at least 1 CSBM from baseline, all within the same week for at least 9 of 12 weeks (12.4% vs 4.0% for linaclotide vs placebo); at least 3 CSBMs and an increase of at least 1 CSBM from baseline, both within the same week for at least 9 of 12 weeks (18.8% vs 5.6% for linaclotide vs placebo); 30% or greater reduction in abdominal pain for at least 9 of 12 weeks (36.6% vs 23.3% for linaclotide vs placebo); and 30% or greater reduction in abdominal pain and an increase of at least 1 CSBM from baseline, both within the same week for at least 6 of 12 weeks (33.7% vs 17.4% for linaclotide vs placebo).
The most common adverse event reported was diarrhea, which led to discontinuation in 5.3% and 0.4% of the linaclotide and placebo groups, respectively. The investigators concluded that linaclotide was associated with sustained and clinically meaningful and significant improvements in abdominal and bowel symptoms in patients with IBS-C.
Subanalysis of the TARGET 1 and TARGET 2 Studies to Evaluate the Efficacy of Rifaximin by Baseline Disease Severity
In a presentation at the 2011 ACG meeting, Pimentel and colleagues evaluated the efficacy of rifaximin (Xifaxan, Salix Pharmaceuticals) for the treatment of nonconstipating IBS (non–C-IBS) in 2 double-blind, placebo-controlled trials (TARGET 1 and TARGET 2). A total of 1,260 non–C-IBS patients from these trials were assessed over a 10-week post-treatment follow-up period. In both TARGET trials, patients were randomized to 2 weeks of treatment with rifaximin (550 mg 3 times daily) or placebo.
No interaction was found between rifaximin and non–C-IBS disease severity, suggesting that patients in all severity categories benefit equally from treatment. Improvement in IBS symptoms, the primary study endpoint, was noted with rifaximin versus placebo in mild non–C-IBS (39.8% vs 30.2%; P=.0261), moderate non–C-IBS (39.5% vs 31.6%; P=.1112), and severe non–C-IBS (43.0% vs 33.7%; P=.0643). When all severity groups were considered together, improvement in symptoms was observed in 40.7% of rifaximin-treated patients versus 31.7% of placebo-treated patients (P=.0008); rifaximin also improved several secondary endpoints compared to placebo, including bloating symptoms (40.2% vs 30.3%; P=.0002), IBS symptoms from daily data (40.2% vs 29.5%; P<.0001), and abdominal pain and stool consistency (46.6% vs 37.4%; P=.0009). The investigators concluded that rifaximin was associated with improved symptom relief in the weeks following treatment, and this benefit was apparent in patients with non–C-IBS of all severities.
Probiotic Bifidobacterium infantis 35624 in a Nonpatient Population with a History of Abdominal Discomfort and Bloating
Ringel and colleagues presented results from a double-blind, randomized, placebo-controlled, parallel study at the 2011 ACG meeting. In this trial, the efficacy of the probiotic Bifidobacterium infantis 35624 was investigated in nonpatient individuals with IBS.
A total of 275 evaluable individuals who had experienced abdominal discomfort and bloating more than 2 times per week, on average, for at least 3 months were recruited by advertisement from the general population. All subjects had not seen a physician for their symptoms and had not received medication for their symptoms within the prior 12 months. After a 2-week placebo run-in phase, B. infantis 35624 was administered in a 4-week intervention phase at 10 clinical centers.
At baseline, the mean severity score was 2.3 for abdominal discomfort and 2.5 for bloating. Over the 4-week intervention period, both the B. infantis 35624 and placebo groups demonstrated significant improvements in abdominal discomfort and bloating scores (P<.05). However, neither score showed a significant difference between the B. infantis 35624 and placebo groups.
Although previous studies had suggested a significant benefit with B. infantis 35624 in IBS patients, the authors of this study concluded that a population of nonpatients with IBS did not experience the same trend. Reasons for this finding may include a high placebo effect in IBS, as well as the potential for lower severity, frequency, and impact of IBS symptoms among nonpatients.
Interleukin-10 in the Susceptibility and Pathogenesis of IBS
At the 2011 ACG meeting, Jiang and colleagues presented results of an investigation in which they evaluated a particular single nucleotide polymorphism (SNP) present in the –1082 interleukin-10 allele and assessed its association with particular types of IBS. In their previous studies, these investigators had demonstrated an association between the AA SNP and reduced frequency of IBS.
Fifty patients with postinfectious IBS, 50 patients with idiopathic IBS, and 52 healthy control subjects were included in this study. DNA was extracted from blood samples, amplified by polymerase chain reaction, and analyzed by pyrosequencing. The level of interleukin-10 in the fecal matter was measured by enzyme-linked immunosorbent assay.
The incidence of the AA SNP was 40%, 32%, and 52% for postinfectious IBS, idiopathic IBS, and control subjects, respectively. When the incidence in both IBS groups was combined (35%), it was significantly lower than the incidence of the AA SNP in the control group (P=.037). The median levels of fecal interleukin-10 were also assessed and were found to be 2.78 pg/mL, 2.59 pg/mL, 1.85 pg/mL, and 5.79 pg/mL for the postinfectious IBS, idiopathic IBS, combined IBS, and control groups, respectively (P=.023 for combined IBS vs control). The investigators concluded that both postinfectious and idiopathic IBS were associated with a reduced incidence of this AA SNP in the –1082 interleukin-10 allele, as well as reduced interleukin-10 protein expression in fecal matter.
Questionnaire Evaluation of IBS Patients’ Outlook on Diagnosis and Treatment
In a presentation at the 2011 ACG meeting, Lurix and colleagues presented results of a questionnaire they administered to 108 IBS patients to help determine patients’ outlooks on diagnosis and treatment. This anonymous questionnaire was distributed to IBS patients who presented to an outpatient tertiary referral gastroenterology clinic.
The vast majority of respondents were female (96%), with a mean age of 42 years. Most patients (81%) self-classified their symptoms as moderate to severe, and patients reported that their symptoms had been present for an average of 40 months. One third of patients (33%) reported a prior diagnosis of IBS, but fewer than half (44%) of these patients agreed with this diagnosis. Almost half (44%) of respondents had seen multiple healthcare providers (2–5) for the same complaint; 50% of these patients had seen another gastroenterologist. Colonoscopy, endoscopy, and/or imaging studies were reported in 58–60% of patients; however, nearly all respondents (95%) felt they required more testing. Of the 84% of patients who reported prior treatment, most (64%) had tried lifestyle changes. A diagnosis was expected either that day or within 1 month by 69% of patients, and nearly three quarters of patients (72%) desired some type of treatment at their current visit. Fewer than half (41%) were willing to see a psychiatrist or psychologist. The investigators concluded from the questionnaire responses that IBS patients utilize a significant amount of resources in gastroenterology clinics, and they recommended that physicians be aware of the expectations of their IBS patients.
A Palpable Bowel Loop Is a Highly Specific Physical Sign in IBS and Other Functional Gastrointestinal Disorders
At the 2011 ACG meeting, McWilliams and colleagues presented a study in which they assessed the diagnostic utility of a palpable bowel loop for diagnosis of functional gastrointestinal disorders and IBS. Among a referral population of 2,115 patients, 947 patients (72% female) had an ultimate diagnosis of a functional gastrointestinal disorder. Patients in this group had a significantly younger mean age at first presentation compared to patients without a functional gastrointestinal disorder (38.5 years vs 44.2 years; P<.001).
Of the 2,115 patients in the referral population, palpable bowel loops occurred with an incidence of 15.8%. The sensitivity and specificity of a palpable bowel loop for diagnosis of a functional gastrointestinal disorder were 23.9% and 90.8%, respectively. Similar sensitivity and specificity were observed for the diagnosis of IBS (25% and 88.8%, respectively). Among the 334 patients with a palpable bowel loop, a tender palpable bowel loop (present in 211 patients) had a sensitivity of 69.5% for diagnosis of any functional gastrointestinal disorder and a sensitivity of 76.1% for the diagnosis of IBS. The investigators concluded that a palpable bowel loop was most useful as a diagnostic tool when absent, although the presence of a tender palpable bowel loop may have utility for aiding in the positive diagnosis of IBS.
Presentations in Hepatology
The PROVIDE Study Demonstrates Efficacy of Boceprevir in Prior Null Responders to Peginterferon and Ribavirin
The ongoing, multicenter, single-arm rollover PROVIDE study aims to evaluate the efficacy of boceprevir (Victrelis, Merck), peginterferon, and ribavirin in patients who previously failed to respond to peginterferon and ribavirin. The PROVIDE treatment regimen included boceprevir (800 mg 3 times daily), peginterferon α-2b (1.5 µg/kg weekly), and weight-based ribavirin (600–1,400 mg/day in 2 divided doses). During the 2011 Annual Meeting of the American Association for the Study of Liver Diseases (AASLD), Vierling and coworkers presented the results of a subset analysis from the PROVIDE study in which they assessed the efficacy of boceprevir, peginterferon, and ribavirin in 48 hepatitis C virus (HCV)-infected patients classified as prior null responders (defined as a reduction in HCV RNA level <2 log10 from baseline to Week 12 during previous peginterferon and ribavirin treatment). All patients in this subanalysis received 4 weeks of peginterferon and ribavirin, followed by boceprevir plus peginterferon and ribavirin for up to 44 weeks.
The primary study endpoint was sustained virologic response (SVR), defined as an undetectable HCV RNA level at 24 weeks post-therapy. SVR was achieved in 16 of 42 evaluable patients (38%). Twenty of 43 patients (47%) achieved end-of-treatment responses, and relapses following end-of-treatment response were detected in 3 of 19 patients (16%). An association was detected between declines in HCV RNA level after the 4-week lead-in period and the likelihood of subsequent SVR. SVR was achieved in 50% of patients with a decline in HCV RNA level of at least 1 log10 at Week 4 versus an SVR rate of 34% among patients whose HCV RNA decline at Week 4 was less than 1 log10. These results suggest the effectiveness of a treatment regimen consisting of boceprevir, peginterferon, and ribavirin in a group of well-documented prior null responders.
Interim Analysis of the ZENITH Study Demonstrates Efficacy of VX-222 and Telaprevir in Combination with Peginterferon and Ribavirin in Treatment-Naïve Patients
The ZENITH trial is an ongoing, phase II study evaluating 12-week response-guided treatment with the HCV polymerase inhibitor VX-222 plus telaprevir (Incivek, Vertex), with or without peginterferon and/or ribavirin, in treatment-naïve patients with genotype 1 HCV infection. In a presentation at the 2011 AASLD meeting, Nelson and coworkers presented a Week 24 interim analysis of the 59 patients who received the 4-drug regimen: VX-222 (100 mg [n=29] or 400 mg [n=30] twice daily), telaprevir (1,125 mg twice daily), peginterferon (180 µg/week), and ribavirin (1,000–1,200 mg/day). Patients received all 4 drugs for 12 weeks and were allowed to stop treatment at Week 12 if they achieved undetectable levels of HCV RNA at Weeks 2 and 8; patients whose HCV RNA levels were detectable at either Week 2 or Week 8 received additional peginterferon and ribavirin for a total of 24 weeks.
Among patients who received VX-222 at a dose of 400 mg, 50% (15 of 30) were eligible to stop treatment at Week 12; SVR was achieved in 93% (14 of 15) of those patients. Of the 15 patients receiving the 400-mg dose of VX-222 who were assigned to 24 weeks of treatment, 87% (n=13) had undetectable levels of HCV RNA at 12 weeks post-treatment. Among patients receiving the 100-mg dose of VX-222, 38% (11 of 29) were eligible to stop treatment at Week 12; 82% of those patients (9/11) attained SVR. Of the 18 patients in the 100-mg VX-222 arm who were assigned to 24 weeks of treatment, 83% (n=15) had undetectable levels of HCV RNA at 12 weeks post-treatment. An intent-to-treat analysis of all patients revealed undetectable HCV RNA levels at Week 24 in 90% of patients who received the 400-mg dose of VX-222 and 83% of patients who received the 100-mg dose of VX-222. A total of 3 patients experienced relapses: 2 in the 100-mg arm and 1 in the 400-mg arm. Fatigue (56%), nausea (49%), diarrhea (48%), anemia (37%), pruritus (34%), and rash (31%) were among the most commonly reported adverse events. Severe adverse events occurring in more than 1 patient included neutropenia (5.1%), hypomagnesemia (3.4%), and anemia (3.4%).
HCV SPRINT-2 Study Addresses Use of Boceprevir Combined with Peginterferon and Ribavirin in Treatment-Naïve Black Patients
Because black patients typically show lower response rates to therapy, the SPRINT-2 study enrolled and analyzed black patients separately from nonblack patients. At the 2011 AASLD meeting, McCone and colleagues presented data on the efficacy and safety of boceprevir-based therapy in black patients enrolled in this study. Of the 159 treatment-naïve black patients who were enrolled in this study, 52 patients were assigned to receive 48 weeks of peginterferon and ribavirin;
52 patients were assigned to receive response-guided therapy with boceprevir, peginterferon, and ribavirin; and 55 patients were assigned to receive boceprevir-based triple therapy for 48 weeks. All patients received peginterferon and ribavirin alone for 4 weeks before the addition of boceprevir or placebo.
The addition of boceprevir to peginterferon and ribavirin was associated with a significant improvement in SVR rates. Among the 154 patients without cirrhosis, SVR rates were 54% for patients treated with boceprevir plus peginterferon and ribavirin for 48 weeks, 50% for patients treated with response-guided boceprevir-based therapy, and 26% for patients treated with peginterferon and ribavirin alone. Among patients without cirrhosis who had an undetectable HCV RNA level at Weeks 8–24, boceprevir-containing treatment yielded SVR rates of 92–95%. Among patients without cirrhosis who were late responders—patients whose HCV RNA level was detectable at Week 8 but undetectable at Week 24—SVR rates were 58–86%.
In terms of adverse events, patients who received 48 weeks of boceprevir-based treatment or response-guided boceprevir-based therapy had higher rates of anemia than patients in the control group (71%, 63%, and 31%, respectively). Dysgeusia was also more common among patients who received boceprevir-containing regimens: 40% among patients who received the 48-week boceprevir-based regimen, 29% among patients who received the response-guided boceprevir-based regimen, and 12% in the control group. Rates of neutropenia were similar across all 3 arms (31%, 35%, and 35%, respectively).
24-Week Interim Analysis of Telaprevir in Combination with Peginterferon and Ribavirin in HCV/HIV Co-Infected Patients
In a late-breaking abstract at the 2011 AASLD meeting, Sherman and colleagues presented interim results of a phase II study evaluating telaprevir, peginterferon α-2a, and ribavirin in genotype 1 HCV treatment-naïve patients who were co-infected with HIV. The study was divided into 2 parts; in both parts, patients were randomly assigned to 1 of 2 antiviral treatment regimens: telaprevir (750 mg every 8 hours), peginterferon α-2a (180 µg/week), and ribavirin (800 mg/day) for 12 weeks, followed by 36 weeks of peginterferon and ribavirin; or 48 weeks of placebo plus peginterferon and ribavirin. In Part A, patients received no concurrent antiretroviral therapy (ART). In Part B, patients received their assigned antiviral therapy plus a stable, predefined ART: either an efavirenz (EFV)-based regimen or an atazanavir/ritonavir (ATV/r)-based regimen. Patients assigned to telaprevir and an EFV-based regimen received telaprevir at a dose of 1,125 mg every 8 hours.
Of the 62 randomized patients, 60 received at least 1 dose of study medication, including 13 patients in Part A (no ART) and 47 patients in Part B (with ART). On-treatment virologic responses were higher among patients receiving the telaprevir-based antiviral regimen than among patients receiving peginterferon and ribavirin alone; early virologic response, defined as an undetectable level of HCV RNA at Weeks 4 and 12, was observed in 63% of patients receiving the telaprevir-based regimen compared to 4.5% of patients receiving peginterferon and ribavirin alone. Of the 60 patients evaluable at
Week 24, undetectable HCV RNA levels were observed in 74% of telaprevir-treated patients and 55% of patients who received peginterferon and ribavirin alone.
Two HCV breakthroughs were detected in patients receiving telaprevir-based therapy and ART, including 1 patient receiving an EFV-based regimen and 1 patient receiving an ATV/r-based regimen. No HIV viral
breakthroughs were detected. The absolute number of CD4+ T cells declined while patients were on therapy, but the relative proportion of CD4+ T cells remained stable. In regard to toxicity, the safety profile of telaprevir in this study was similar to that seen in HCV monoinfected patients.
Comparing the 2 ART regimens, the incidence of bilirubin adverse events was higher among patients who received the ATV/r-based regimen than among patients who received the EFV-based regimen (27% vs 0%); the incidence of indirect hyperbilirubinemia was also higher in patients who received the ATV/r-based regimen. Among patients who received ART and telaprevir, 3 patients discontinued at least 1 study drug due to an adverse event. Finally, the ART regimen did not appear to affect the pharmacokinetics of telaprevir; the effects of telaprevir on ART pharmacokinetics were consistent with previous reports in healthy individuals.
5 Years of Treatment with Tenofovir for Chronic HBV Infection Is Associated with Sustained Viral Suppression and Significant Regression of Fibrosis and Cirrhosis
Marcellin and colleagues presented 5-year on-treatment virologic and paired histologic assessment data from Study 102 and Study 103 during the 2011 AASLD meeting. These multicenter, randomized, double-blind, phase III trials compared tenofovir disoproxil fumarate (Viread, Gilead Sciences) and adefovir dipivoxil (Hepsera, Gilead Sciences) in chronic hepatitis B virus (HBV)–infected patients with compensated liver disease who were hepatitis B e antigen (HBeAg)-negative (Study 102; n=375) or HBeAg-positive (Study 103; n=266). The majority of patients were treatment-naïve. In both studies, patients who were originally randomized to adefovir dipivoxil were rolled over to open-label tenofovir disoproxil fumarate (n=196) at Week 48, and patients who were originally randomized to tenofovir disoproxil fumarate continued on open-label treatment (n=389).
Of the 641 patients who were initially randomized and treated in these studies, 91% (n=585) entered the open-label extension phase of the trials. At Year 5, 76% (n=490) remained on study. Normalization of alanine aminotransferase (ALT) levels at Week 240 was achieved in 72% of patients in Study 102 and 50% of patients in Study 103. Across both studies, only 2.1% of patients who received tenofovir disoproxil fumarate for 5 years discontinued treatment due to an adverse event; 0.9% of patients experienced a confirmed increase in serum creatinine level of at least 0.5 mg/dL or a calculated creatinine clearance less than 50 mL/min. Resistance to tenofovir disoproxil fumarate over a 5-year treatment period was not detected. Overall histologic improvement was observed in 88% of the 331 patients who underwent biopsies before therapy and again at 5 years. Of the 94 patients who were cirrhotic at the start of therapy, 73% experienced regression of histologic cirrhosis at 5 years.
BE-LOW Study Shows That Entecavir Monotherapy Is Comparable to Entecavir Plus Tenofovir in Nucleos(t)ide-Naïve Patients with Chronic HBV Infection
In a presentation at the 2011 AASLD meeting, Lok and colleagues presented data from the open-label, multicenter, phase IIIb BE-LOW study, in which treatment-naïve adult patients with HBeAg-positive and HBeAg-negative chronic HBV infection and compensated liver disease were randomized to receive either entecavir monotherapy (Baraclude, Bristol-Myers Squibb) 0.5 mg once daily (n=182) or entecavir 0.5 mg plus tenofovir disoproxil fumarate 300 mg once daily (n=197) for 100 weeks.
Prior to Week 96, 6.5% (n=12) of patients in the entecavir monotherapy arm discontinued treatment, compared to 11.6% (n=23) of patients in the entecavir plus tenofovir arm. A comparable proportion of patients in both treatment arms achieved HBV DNA levels below 50 IU/mL at Week 96 (76.4% in the entecavir monotherapy arm vs 83.2% in the entecavir plus tenofovir arm; 95% CI, –1.0, 14.9; P=.0882). Fewer patients in the combination therapy arm experienced ALT normalization or HBeAg seroconversion compared to the entecavir monotherapy arm. Virologic breakthrough was similar in both arms (4% in the entecavir plus tenofovir arm vs 1% in the entecavir monotherapy arm). Both study arms had similar safety profiles, with serious adverse events reported in 6.6% of patients in the entecavir monotherapy arm and 7.1% of patients in the entecavir plus tenofovir arm.
Cirrhotic Patients on Rifaximin Have a Lower Incidence of Clostridium difficile Infection and Improved Outcomes
In a study presented at the 2011 AASLD meeting, Zuchelli and coworkers sought to determine the incidence of Clostridium difficile–associated diarrhea (CDAD) in patients with cirrhosis who were receiving rifaximin and/or lactulose and to establish outcomes and confounding factors among cirrhotic patients with CDAD. A total of 144 patient charts were retrospectively reviewed, including 69 with CDAD and 75 without CDAD. Among patients with CDAD, 26% (n=18) were receiving lactulose and 9% (n=6) were receiving rifaximin. Of the patients without CDAD, 80% (n=60) were receiving lactulose and rifaximin and 20% (n=15) were receiving rifaximin alone. Although there were no significant differences among patients in regard to gender, age, etiology of cirrhosis, or PPI and/or antibiotic use, cirrhotic patients with CDAD were found to have significantly more chronic kidney disease (P<.0001), hypertension (P<.03), and cardiac disease (atrial fibrillation, congestive heart failure; P<.014) than cirrhotic patients without CDAD.
Among all patients, those treated at home with rifaximin had a significantly lower incidence of CDAD than those treated at home with lactulose (P<.007). The incidence of CDAD did not differ significantly among patients receiving rifaximin alone compared to rifaximin and lactulose combined. Cirrhotic patients with CDAD had a longer average length of hospitalization (35 days vs 13 days; P<.00004) and higher mortality rate (26% vs 13%; P<.19) compared to cirrhotic patients without CDAD. Of the 15 patients with CDAD who died, 33% (n=5) were receiving lactulose and 7% (n=1) were receiving rifaximin at the time of death. The study results suggest the need for future prospective studies to reconfirm that rifaximin is protective against C. difficile infection in cirrhotic patients.
Presentations in Endoscopy
Resect-and-Discard Strategy in Real-Time Colonoscopy Using a Validated Classification System Provides Accurate Surveillance Interval Recommendations
At the 2011 ACG meeting, Kaltenbach and colleagues evaluated the use of a resect-and-discard strategy for real-time evaluation of polyp histology during colonoscopy. This strategy has been proposed as a means of improving the cost-effectiveness of colorectal cancer screening; it is based on the fact that histopathologic review of polyps makes up a large proportion of related costs, but most small polyps do not exhibit evidence of dysplasia or neoplasia. In this prospective analysis, the resect-and-discard strategy was applied using the recently validated Narrow-Band Imaging International Colorectal Endoscopic (NICE) classification system; this approach was compared to traditional histopathologic assessment.
The NICE classification system is used to differentiate adenomatous and hyperplastic polyps in real time without magnification during endoscopy. This classification scheme differentiates type I (hyperplastic) polyps from type II (adenomatous) polyps based on 3 criteria: color, vessels, and surface pathology.
In this study, endoscopists used the NICE classification system to differentiate polyps less than 1 cm in diameter as type I or type II; they also assigned either a high or low level of confidence to this differentiation.
A total of 220 patients were enrolled in the study; 49% had at least 1 polyp that was less than 1 cm in diameter. A high-confidence assignment was made in 75% of the 236 consecutive polyps that were less than 1 cm in diameter. These high-confidence assignments were correlated with 89% accuracy, 98% sensitivity, and a 95% negative predictive value.
In the vast majority of cases (98%), the endoscopists made accurate recommendations for follow-up surveillance colonoscopy intervals. Importantly, most of the incorrect recommendations erred on the side of a recommendation for a shorter surveillance interval (3–5 years vs 10 years). The investigators concluded that use of the NICE classification scheme for evaluation of polyps in real time during colonoscopy could provide a high-confidence prediction of colorectal polyp histology in most cases, resulting in accurate surveillance interval recommendations.
Retrograde Visualization Improves Adenoma Detection Rate Among Individuals Undergoing Surveillance or Diagnostic Colonoscopy
Two thirds of the adenomas missed during colonoscopy are located behind haustral folds. Thus, the TERRACE trial evaluated the ability of the Third Eye Retroscope to provide a retrograde view of areas behind these folds.1 In the TERRACE study, an additional 23.2% of adenomas were detected using this system, compared to standard colonoscopy; the relative risk (RR) of missing adenomas with standard colonoscopy versus the Third Eye Retroscope was significant (1.92; P=.029).1 Siersema and colleagues presented a subanalysis of the TERRACE trial at the 2011 ACG meeting. In this report, the TERRACE study results were reviewed to determine if the indication for colonoscopy was correlated with adenoma detection rate using the Third Eye Retroscope.
A total of 448 individuals were enrolled in the prospective, multicenter, randomized TERRACE trial; 349 subjects were included in the per-protocol population. These individuals were randomized to same-day, tandem examinations with either standard colonoscopy followed by the Third Eye Retroscope (group A) or the Third Eye Retroscope followed by standard colonoscopy (group B). Because tandem colonoscopy studies generally result in the discovery of additional lesions during the second procedure, the adenoma detection rate of group B was subtracted from the adenoma detection rate in group A, resulting in a net additional detection rate that could be attributed to the Third Eye Retroscope.
Among the study population, indications for colonoscopy were screening (51.0%), surveillance following prior polypectomy (25.2%), and diagnostic work-up (23.8%). The additional adenoma detection rates for the Third Eye Retroscope compared to standard colonoscopy were 4.4%, 35.7%, and 55.4% for the screening, surveillance, and diagnostic groups, respectively. The RRs for missing adenomas with standard colonoscopy versus the Third Eye Retroscope were 1.11 (P=.81), 3.15 (P<.05), and 8.64 (P<.05), respectively. When the surveillance and diagnostic groups were combined, an additional 17.5% of patients were determined to have at least 1 adenoma due to the Third Eye Retroscope. Further, an additional 27.3% of patients in the combined surveillance and diagnostic group were advised (per guidelines) to return for a 3-year follow-up colonoscopy due to the use of the Third Eye Retroscope. The investigators concluded that individuals undergoing colonoscopy for an indication of either surveillance or diagnostic work-up benefited most from the use of the Third Eye Retroscope.
Reference:
1. Leufkens AM, DeMarco DC, Rastogi A, et al; Third Eye Retroscope Randomized Clinical Evaluation [TERRACE] Study Group. Effect of a retrograde-viewing device on adenoma detection rate during colonoscopy: the TERRACE study. Gastrointest Endosc. 2011;73:480-489.
Indigocarmine Added to the Water Method Enhances Adenoma Detection Rate in Screening Colonoscopy
In a late-breaking abstract presented at the 2011 ACG meeting, Leung and colleagues reported the results of a randomized controlled trial in which they evaluated whether the addition of indigocarmine to the water method could improve adenoma detection rates during screening colonoscopy.
After receiving intravenous (IV) conscious sedation, 168 patients underwent screening colonoscopy with the water method, either with or without 0.008% indigocarmine. Overall, the adenoma detection rate was 62% in the indigocarmine group versus 44% in the group evaluated with the water method alone (P=.0302). One incidence of cancer was detected in each group, and the number of patients with normal biopsies was comparable (12 in the water method group vs 18 in the indigocarmine group; P=.3139). The investigators concluded that detection
of adenomas could be significantly improved by the addition of indigocarmine to the water method for screening colonoscopy.
Correlation Between Endoscopic Ultrasound and Histopathology for Diagnosis of Minimal Change Chronic Pancreatitis
Although endoscopic ultrasound (EUS) has been widely used to diagnose minimal change (noncalcific) chronic pancreatitis (MCCP), the 9 criteria used for diagnosis of MCCP via EUS are derived from patients undergoing resection for non–MCCP-related conditions. Therefore, Vega-Peralta and colleagues presented a retrospective study at the 2011 ACG meeting in which they assessed the relationship between EUS findings and histopathology in patients with MCCP.
A total of 50 adults and children with MCCP were identified from a total of 141 patients who underwent total pancreatectomy with islet autotransplantation at a single center between 2008 and 2010. EUS findings obtained within 1 year prior to surgery and histopathology records were examined.
MCCP was identified via histopathology in 90% of cases using a composite of any of 3 abnormalities: fibrosis, atrophy, or inflammation. All of the patients identified via histopathology had fibrosis. The 10% of patients with normal histology had a documented history of acute recurrent pancreatitis in addition to disabling pain. More than half (60%) of the patients with MCCP identified via histopathology fulfilled 4 or more of the 9 EUS criteria. A minority (7%) of patients with 4 or more of 9 EUS criteria had normal histology. A total of 85% of patients with 3 or fewer EUS criteria and 80% of patients with no EUS criteria had fibrosis and histologically identified MCCP.
The negative predictive value of a normal EUS was 38%, and the positive predictive value of an abnormal EUS only exceeded 72% with a threshold of 6 or more EUS criteria. The investigators concluded that there was little correlation between EUS and histopathology for MCCP diagnosis in patients undergoing total pancreatectomy with islet autotransplantation.
Does Timing of Endoscopic Retrograde Cholangiopancreatography Affect Outcomes in Patients with Cholangitis?
While cholangitis is a serious medical issue, patients can improve with antibiotics, and some patients do not undergo endoscopic retrograde cholangiopancreatography (ERCP) until several days after presentation. There is a question of whether ERCP should be performed within 24 hours or whether it is more prudent to wait for sicker patients to recover and/or to wait for weekdays when supportive care is more routine. A presentation at the 2011 ACG meeting by Craft and colleagues addressed these questions by retrospectively reviewing patient outcomes with either immediate versus conservative implementation of ERCP.
Patients from a single institution were identified; cholangitis was either the preliminary diagnosis or the indication following ERCP. Only patients with clinical cholangitis were included, manifested as fevers, elevated white blood cell counts, and evidence of biliary obstruction. Patients were stratified into 4 groups according to when they underwent ERCP: 0-day delay; 1–2-day delay; 2–5-day delay; and more-than-5-day delay. A total of 53 patients were included; 43% had stones, and 28% had blocked stents. All patients received IV fluid resuscitation and antibiotic therapy during the interval prior to ERCP. Most patients (89%) had a fever, and 81% showed improvement in fever with antibiotic therapy.
The average time from presentation to ERCP was 2.61 days (range, 0–17 days). Organ failure was identified in 6 patients prior to ERCP and in an additional 4 patients after ERCP. One patient died, 6 patients required management in an intensive care unit, and 2 adverse events occurred post-ERCP (bleeding and pneumonia). ERCP was unsuccessful in 2 patients, who then required percutaneous transhepatic cholangiography; 5 patients required repeat ERCP during their hospital admission. Death, organ failure, and length of hospital stay were not found to differ among the time stratifications for ERCP implementation. The investigators concluded that there were no significant differences in patient outcomes when comparing immediate versus conservative (delayed) use of ERCP for cholangitis.
Is Ablation Needed After Endoscopic Mucosal Resection in Patients with Short-Segment Barrett Esophagus?
At the 2011 ACG meeting, Tian and colleagues investigated the role of ablative therapies following endoscopic mucosal resection (EMR) in patients with short-segment BE. While EMR followed by ablation is widely used to treat BE patients with either high-grade dysplasia or early esophageal adenocarcinoma, the benefit of this strategy is less clear in the setting of limited residual intestinal metaplasia, as seen in short-segment BE.
In this study, EMR followed by ablation was compared to EMR alone in patients with short-segment BE. Ablative therapies included radiofrequency ablation, photodynamic therapy, multipolar/bipolar electrocautery, cryotherapy, and argon plasma coagulation. All short-segment BE patients with high-grade dysplasia or early esophageal adenocarinoma who underwent EMR at a single center between 2006 and 2011 were included. A total of 213 patients met these criteria and were identified as having a complete response for intestinal metaplasia or dysplasia. Most of these patients were male (86.9%), with a mean age of 65.4±10.3 years. EMR was used alone in 43.7% of patients and in combination with ablation in 56.3% of patients.
All-cause mortality was 16.4%. Recurrence, defined as a finding of either dysplasia or esophageal adenocarcinoma after 2 consecutive, negative esophagogastroduodenoscopy examinations with complete response for intestinal metaplasia or dysplasia, occurred in 10.8% of patients during a mean follow-up period of 39.1 months. No differences were found between EMR alone and EMR followed by ablation in regard to age, sex, Charlson comorbidity index, recurrence rate, or mortality rate. After adjusting for age, sex, comorbidity index, and complete response, the use of ablative therapies post-EMR did not result in differences in recurrence or mortality rates. These results led the investigators to conclude that short-segment BE patients with high-grade dysplasia or early esophageal adenocarcinoma did not benefit from the addition of ablative therapies following EMR if they had achieved a complete response.
Utility of Biomarkers in Predicting Response to Radiofrequency Ablation Among Patients with Barrett Esophagus
Although largely considered to be effective for the treatment of BE with dysplasia, radiofrequency ablation does not yield a response in up to 30% of patients. In a presentation at the 2011 ACG meeting, Prasad and colleagues noted that biomarkers detected by fluorescence in situ hybridization (FISH) were able to predict response to photodynamic therapy. Therefore, they sought to determine if FISH-detected biomarkers could serve a similar function in predicting response to radiofrequency ablation.
A total of 99 patients who underwent radiofrequency ablation for dysplastic BE at a single center between 2007 and 2010 were identified. The mean patient age was 63.7±10.7 years, and 86% of patients were male. At baseline, 29% and 62% of patients had low-grade or high-grade dysplasia, respectively; intramucosal adenocarcinoma was identified in 8% of patients. Prior to radiofrequency ablation, the mean BE segment length was 6±2.5 cm. Approximately two thirds (69%) of patients underwent EMR prior to radiofrequency ablation. Most patients (n=81) were treated with Halo360 followed by Halo90; the remaining 18 patients received only Halo90 ablation. Total elimination of dysplasia was achieved in 61% of patients, and an additional 30% of patients achieved elimination of intestinal metaplasia.
FISH was performed on archived, paraffin-embedded biopsies taken prior to radiofrequency ablation. FISH was performed using a validated set of fluorescently labeled probes directed against the p16 tumor suppressor gene (chromosome 9p21) and the proto-oncogenes HER2 (chromosome 17q12), Myc (chromosome 8q24), and ZNF217 (chromosome 20q13.2). FISH signal patterns were evaluated in 50 consecutive cells. Loss of p16 was hemizygous or homozygous in 14% and 16% of patients, respectively. Gains in HER2, Myc, and ZNF217 occurred in 13%, 4%, and 9% of patients, respectively. Polysomy was observed in about half (53%) of cases.
In a multivariate analysis that adjusted for age, sex, and length of BE segment, only polysomy was significant for predicting the elimination of dysplasia (OR, 0.4; 95% CI, 0.2–0.9; P=.03) and the elimination of intestinal metaplasia (OR, 0.5; 95% CI, 0.2–1.1; P=.08) with radiofrequency ablation. The investigators concluded that the presence of polysomy, detected by FISH, could independently predict a lack of response (elimination of dysplasia) following radiofrequency ablation in patients with BE.
Presentations in IBD
Nonsteroidal Anti-Inflammatory Drugs Are Associated with Risk of Crohn’s Disease and Ulcerative Colitis
At the 2011 ACG meeting, Ananthakrishnan and colleagues presented a study in which they prospectively evaluated whether the risk of inflammatory bowel disease (IBD) was associated with use of aspirin or NSAIDs.
A total of 76,814 women enrolled in the Nurses’ Health Study were prospectively evaluated in this report. The mean age of these women in 1990 was 57 years. These women have provided data on their use of aspirin and NSAIDs twice per year since 1990. Medical records were used to determine a diagnosis of either Crohn’s disease (CD) or ulcerative colitis (UC), which was confirmed by 2 gastroenterologists. A total of 1,295,317 person-years of follow-up over a total of 18 years were included in
this assessment.
A total of 240 IBD cases were identified and confirmed, including 123 cases of CD and 117 cases of UC. At baseline, regular aspirin use and regular NSAID use were reported by 44% and 37% of women, respectively. Women who reported using NSAIDs for more than 15 days per month had an increased risk of both CD (RR, 1.59; 95% CI, 0.99–2.56) and UC (RR, 1.87; 95% CI, 1.16–2.99). Risks of CD (RR, 1.71; 95% CI, 1.05–2.77) and UC (RR, 1.78; 95% CI, 1.10–2.89) were also heightened among women who used more than 5 NSAID tablets per week. A longer duration (>6 years) of NSAID use was also associated with a greater risk of CD (RR, 2.83; 95% CI, 1.65–4.85) and UC (RR, 2.00; 95% CI, 1.15–3.49). In contrast, neither dose, duration, nor frequency of aspirin use was associated with a risk of developing CD or UC. The investigators concluded that increased frequency, longer duration, and higher doses of NSAIDs, but not aspirin, were associated with a greater risk of both CD and UC in women.
Over 5 Years of Follow-Up Data From the TREAT Registry Confirms No Increased Risk of Malignancy in Crohn’s Disease Patients Treated with Infliximab
The TREAT Registry is an ongoing, large-scale, observational effort aimed at determining long-term outcomes of various treatments for CD; it includes both community-based and academic clinical practices throughout North America. At the 2011 ACG meeting, Lichtenstein and colleagues provided an updated analysis of data from the TREAT Registry that focused on the incidence of malignancy in CD patients treated with anti–tumor necrosis factor α (anti-TNFα) therapy.
This prospective evaluation with over 5 years of follow-up data included 6,273 patients, of whom 3,764 received infliximab (Remicade, Janssen Biotech) and 2,509 received other treatments. A standardized incidence ratio (observed/expected) was calculated by dividing the number of patients in the TREAT Registry in whom malignancies were observed by the expected number of malignancies for the general US population as estimated by the Surveillance, Epidemiology, and End Results 2009 database.
The incidences of malignancies were found to be comparable between patients who received infliximab and those who received other treatments. Several baseline factors were significantly associated with risk of malignancy, including increased age, disease duration at enrollment, smoking status, and use of immunomodulatory therapy. However, infliximab use was not found to be a significant predictor of malignancy (hazard ratio, 0.79; 95% CI, 0.56–1.10; P=.17). The investigators concluded that the risk of malignancy was not significantly increased with infliximab use, based on this longer follow-up analysis of data from the TREAT Registry. However, the researchers did note an increased risk of lymphoma irrespective of infliximab treatment, confirming reports that CD patients have a heightened risk of lymphoma compared to the general population.
US Food and Drug Administration Review of Primary Efficacy Endpoints in Ulcerative Colitis Registration Trials
In an abstract presented at the 2011 Advances in IBD meeting, Johnson and colleagues discussed the US Food and Drug Administration (FDA) review of the various primary efficacy endpoints used in registration trials of recently approved UC treatments. The goal of this review was to evaluate these primary efficacy endpoints and to determine the consistency and use of drug development approaches.
Six recent registration trials were included in this analysis; all of these trials had led to the FDA approval of a therapy for adult or pediatric UC. The agents evaluated in these trials were delayed-release mesalamine (Asacol, Warner Chilcott), high-dose delayed-released mesalamine (Asacol HD, Warner Chilcott), balsalazide (Colazal, Salix Pharmaceuticals), once-daily delayed-release mesalamine (Lialda, Shire), infliximab, and controlled-release mesalamine caplets (Pentasa, Shire).
Primary efficacy was most frequently assessed at Week 8 of therapy in these studies. Studies of delayed-release mesalamine and high-dose delayed-released mesalamine used treatment success as a primary endpoint, defined as improvement in the Physician’s Global Assessment score. Balsalazide studies used reduction in rectal bleeding and improvement in at least 1 other symptom as the primary efficacy endpoint in adult trials, while clinical improvement as assessed by the modified UC Disease Activity Index (UCDAI) was the primary efficacy endpoint in pediatric trials. Once-daily delayed-release mesalamine was evaluated in a registration trial that used remission as measured by the UCDAI as the primary efficacy endpoint, while the infliximab adult registration trials used clinical response as defined by the Mayo score as the primary efficacy endpoint. In the registration trials of controlled-release mesalamine caplets, the primary efficacy endpoint was comprised of the Physician’s Global Assessment score, treatment failure rate, and sigmoidoscopic index score.
The primary efficacy endpoints in these 6 registration trials differed, and the investigators attributed this difference to the availability of numerous disease activity indices, historical precedence, and evolving science. The authors concluded by stating the FDA “seeks consensus on the definition of successful treatment outcome (disease response and/or remission) and development of a standardized UC activity index that defines the outcome.”
Induction of Clinical and Endoscopic Remission with 9-mg Budesonide MMX in Patients with Mild-to-Moderate Ulcerative Colitis
In a presentation at the 2011 ACG meeting, Sandborn and colleagues reported an analysis of pooled data from 2 phase III trials that evaluated the efficacy and safety of Budesonide MMX (Cosmo Pharmaceuticals) in patients with mild-to-moderate UC. A total of 672 patients were pooled from the modified intent-to-treat populations; all patients received at least 1 dose of the study agent and were randomized to receive placebo, 6-mg Budesonide MMX, or 9-mg Budesonide MMX for 8 weeks.
The rate of clinical and endoscopic remission at Week 8 was higher in patients treated with 9-mg Budesonide MMX and 6-mg Budesonide MMX versus placebo (17.7% and 10.9% vs 6.2%; P=.0002 and P=.0809 for comparison of 9-mg and 6-mg Budesonide MMX vs placebo, respectively). Accordingly, more patients in the 9-mg Budesonide MMX and 6-mg Budesonide MMX groups experienced symptom resolution compared to the placebo group (26.3% and 21.7% vs 14.3%; P=.0018 and P=.0429 for comparison of 9-mg and 6-mg Budesonide MMX vs placebo, respectively). Patients treated with 9-mg Budesonide MMX showed clinical improvements compared to patients treated with placebo (37.5% vs 28.6%; P=.0466); the 9-mg Budesonide MMX group also showed endoscopic improvements compared to the control group (41.8% vs 32.4%; P=.0407). In contrast, patients treated with 6-mg Budesonide MMX were not significantly different from placebo-treated patients in terms of either clinical improvement (28.3% vs 28.6%; P=.9425) or endoscopic improvement (30.9% vs 32.4%; P=.7334). Similar incidences of treatment-related adverse events were reported across the study groups. From these pooled data, the investigators concluded that a 9-mg dose of Budesonide MMX is both safe and effective in patients with mild-to-moderate UC, resulting in both clinical and endoscopic remission and symptom resolution.
Long-Term Clinical Experience with Vedolizumab in Patients with Mild-to-Moderate Ulcerative Colitis
The integrin molecule α4β7 binds the mucosal addressin cell adhesion molecule 1 and mediates leukocyte migration into the gastrointestinal mucosa and lymphoid tissue. Vedolizumab is an investigational monoclonal antibody directed against α4β7. In an abstract presented at the 2011 ACG meeting, Parikh and colleagues reported on the long-term clinical experience of 53 UC patients treated with vedolizumab.
All patients had mild-to-moderate UC and had received vedolizumab for up to 2.5 years. A total of 38 patients received vedolizumab in a randomized dose-ranging study in which vedolizumab was given at doses of 2 mg/kg, 6 mg/kg, or 10 mg/kg on Days 1, 15, 29, and 85. After completion of the dose-ranging portion of the study (Day 253), all of these patients then continued to receive vedolizumab (2 mg/kg) every 8 weeks for up to 547 additional days as part of an open-label extension study. Fifteen treatment-naïve patients who had not participated in the dose-ranging study were also enrolled into the extension study, resulting in a total of 53 patients for this long-term analysis. Following completion of the extension study, patients were eligible to continue to receive vedolizumab in an ongoing, phase III safety extension trial; 55% of patients elected to continue in the phase III trial.
A total of 81% of patients completed the extension study, with a mean vedolizumab exposure of 580.3 days. Discontinuation was due to lack of efficacy (5 patients), adverse events (3 patients), and withdrawal of consent (2 patients). The mean Partial Mayo Score, used to measure response to therapy, was 3.7±1.43 at the initiation of the dose-ranging study and 5.4±1.88 at the initiation of the extension study. At Day 491 of the extension study (prior to enrollment in the phase III safety trial), the mean Partial Mayo Score had dropped to 0.8±1.15. At that time point, 88% of patients were in clinical remission.
Adverse events were reported in 70% of patients during the extension study; 9% reported a serious adverse event. The most frequently observed adverse events included nasopharyngitis (15%), headache (8%), cough (8%), arthralgia (4%), and upper respiratory infection (4%). One serious infection was reported (self-limited viral gastroenteritis requiring hospitalization). A serious infusion reaction occurred in 1 patient who had previously been exposed to vedolizumab in another clinical trial. Anti–human vedolizumab antibodies were found in 3 patients. The investigators concluded that long-term administration of vedolizumab was well tolerated and associated with durable decreases in UC disease activity.
Phase IIb Study of Ustekinumab in Patients with Moderate-to-Severe Crohn’s Disease: Results Through Week 36 From the CERTIFI Trial
Ustekinumab is a novel human monoclonal antibody directed against interleukins 12 and 23. At the 2011 ACG meeting, Feagan and colleagues provided Week 36 results of ustekinumab therapy in the CERTIFI trial. This trial was a multicenter, randomized, double-blind, placebo-controlled, phase IIb study that investigated the safety and efficacy of ustekinumab in patients with moderate-to-severe CD who had previously failed anti-TNFα therapy.
Patients were randomized to receive either IV placebo or IV ustekinumab (1 mg/kg, 3 mg/kg, or 6 mg/kg) at Week 0. At Week 8, patients in the ustekinumab arm were separately rerandomized to receive maintenance therapy with ustekinumab (90 mg) or placebo (regardless of Week 6 response); maintenance therapy was administered subcutaneously at Weeks 8 and 16. Steroid tapering was required during maintenance therapy. Patients were followed through Week 36. The primary study endpoint was clinical response rate at Week 6.
All 526 patients included in this study had a Crohn’s Disease Activity Index (CDAI) score of 220–450 (baseline mean CDAI score=324). Nearly half of patients (49.6%) were on steroid therapy (including budesonide), and 48.8% had failed more than 2 prior anti-TNFα therapies.
The clinical response rates at Week 6 were 39.7% for 6-mg/kg ustekinumab versus 23.5% for placebo (P<.05). The clinical response rate at Week 6 for all ustekinumab dosage groups combined was also significantly higher than the response rate in the placebo group (36.8% vs 23.5%; P<.05). During the maintenance phase of the study, 41.7% of the patients who had achieved a response with ustekinumab at Week 6 and had continued with ustekinumab maintenance therapy were in remission at Week 22, compared to 27.4% of patients who had continued with placebo during the maintenance phase of the study (P=.029). The median dose of systemic steroids at baseline was 20 mg; at Week 36, it was reduced to 5 mg in ustekinumab-treated patients and 15 mg in placebo-treated patients (P<.05).
No major adverse events were reported in this study, and the toxicity profiles of the ustekinumab and placebo arms remained similar during both the induction and maintenance phases of the study. The investigators concluded that ustekinumab therapy was well tolerated and resulted in induction and maintenance of clinical response in patients with moderate-to-severe CD who had previously failed anti-TNFα therapy.
Narrow-Band Imaging and Chromoendoscopy for the Detection of Colonic Dysplasia in Patients with IBD
In a presentation at the 2011 Advances in IBD meeting, Feitosa and colleagues presented data from a study that evaluated the use of narrow-band imaging and chromoendoscopy for the detection of colonic dysplasia in patients with IBD. This prospective, randomized study compared narrow-band imaging with chromoendoscopy as a diagnostic technique in 29 patients who had been diagnosed with IBD at least 8 years previously. Baseline patient characteristics were well balanced between the intervention groups, including sex (61.5% female vs 68.75% female) and mean age (50.3 years vs 49.5 years) for the chromoendoscopy and narrow-band imaging groups, respectively. The mean duration of disease was 15.9 years for the narrow-band imaging group and 15.3 years for the chromoendoscopy group. Patients had a diagnosis of CD in 56.3% and 53.8% of the narrow-band imaging and chromoendoscopy groups, respectively. No significant differences were noted in terms of patient demographics, disease behavior, medication use, endoscopic grade of disease activity, or symptoms at the time of examination.
Thirty percent of patients in the chromoendoscopy group exhibited evidence of dysplastic lesions on histologic examination (biopsies of mucosal lesions). In contrast, no patients in the narrow-band imaging group showed signs of dysplastic lesions (chi-square=4.13, ∑critical >3.841, considering a 5% error). Two adenomas and 1 dysplastic lesion were observed, both of which are typical of IBD. While these results are preliminary, the investigators concluded that there was a strong statistical tendency for superiority of chromoendoscopy versus narrow-band imaging. This study was limited by a small sample size, and increasing the number of patients would help to confirm this suspected superiority.