ORIGINAL ARTICLES ALIMENTARY TRACT

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2012;10: ORIGINAL ARTICLES ALIMENTARY TRACT Regurgitation Is Less Responsive to Acid Suppression Than Heartburn in Patients With Gastroesophageal Reflux Disease PETER J. KAHRILAS,* ANDREAS JONSSON, HANS DENISON, BÖRJE WERNERSSON, NESTA HUGHES, and COLIN W. HOWDEN* *Division of Gastroenterology and Hepatology, Northwestern University Feinberg School of Medicine, Chicago, Illinois; AstraZeneca Research & Development, Mölndal, Sweden; and Research Evaluation Unit, Oxford PharmaGenesis Ltd, Oxford, United Kingdom BACKGROUND & AIMS: Although most patients with gastroesophageal reflux disease (GERD) achieve substantial symptom relief with acid suppression, many have some residual symptoms. We evaluated the responsiveness of regurgitation, characterized by the reflux disease questionnaire (RDQ) to potent acid suppression. METHODS: We analyzed data from 2 randomized controlled trials of AZD0865 (a potassium-competitive acid blocker) mg/day vs esomeprazole mg/day for the treatment of nonerosive reflux disease (NERD, n 1460) or reflux esophagitis (RE, n 1514). Inclusion criteria for both studies were high-severity substernal burning ( 4 days per week of at least moderate intensity) during the week before enrollment. Pooled data from all treatment arms were used to ascertain the response of the reflux disease questionnaire regurgitation items to potent acid suppression during the fourth week of treatment. RESULTS: When the study began, 93% of patients with NERD or RE had either acid taste in the mouth (regurgitation-taste) or unpleasant movement of material upwards from the stomach (regurgitation-movement). Either or both symptoms were present and severe in 53% of NERD (n 717) and 54% of RE patients (n 751) for the main study outcome. During week 4 of therapy, patients with severe regurgitation-taste and regurgitation-movement responded significantly less well than patients with NERD and high severity substernal burning (34% and 26% vs 49%) or those with RE (44% and 33% vs 55%). There were no differences in symptom response between patients with healed and nonhealed RE. CONCLUSIONS: Regurgitation was less responsive to acid suppression than heartburn in patients with gastroesophageal reflux disease, indicating that persistent regurgitation is a common cause of incomplete treatment response. Keywords: PPI; Esophagus; Pain; Stomach; Drug. Acid suppression, particularly with proton pump inhibitors (PPIs), is the mainstay of gastroesophageal reflux disease (GERD) therapy. 1,2 The use of PPIs has accelerated steadily since the 1990s with the repeated demonstration of the remarkable capacity of these drugs to heal reflux esophagitis (RE), and the correlation between the intensity of acid suppression and clinical efficacy, in both healing and symptom control therapeutic end points. Enthusiasm for pharmacologic acid suppression has led to research into increasingly potent molecules in the hope of eclipsing the existing generation of PPIs. One such agent, AZD0865, a potassium-competitive acid blocker, was tested up to phase 2b clinical trials, at which point it was found to be no more efficacious than esomeprazole in healing esophagitis or in resolving heartburn in patients with RE- or nonerosive reflux disease (NERD). 3,4 Further development of this drug was therefore discontinued. The conclusion from these trials was that a ceiling effect existed, such that beyond a critical level of acid suppression, no further increment in efficacy was observed. In the AZD0865 phase 2b clinical trials, both the eligibility of patients for inclusion and symptom responsiveness to therapy was gauged using the Reflux Disease Questionnaire (RDQ), an extensively used and validated instrument developed as an aid to the diagnosis of GERD. 5,6 The RDQ comprises 6 items (Table 1), with 2 items relating to each of the 3 symptom domains (heartburn, regurgitation, dyspepsia), each scored for frequency and intensity. Patients with qualifying RDQ items underwent endoscopy and were then randomized into either the RE 3 or the NERD trial 4 based on the presence or absence of esophagitis. The entry criteria for the 2 trials were identical with regard to symptoms. Four RDQ items refer to heartburn and regurgitation. However, only a fraction of the data collected was used as end points in the published AZD0865 trials. The primary outcome of the RE trial was the healing rate of esophagitis at 4 weeks. Secondary outcomes were the rates of substernal burning relief at 4 weeks and esophagitis healing at 2 and 8 weeks. 3 In the NERD trial, the primary outcome was the time to sustained absence (for 7 consecutive days) of substernal burning. 4 Data on the response of all RDQ items to therapy were collected meticulously using patient-reported e-diaries for 4 weeks in both trials. The aim of our study was to use this wealth of clinical data to conduct a critical evaluation of the responsiveness of the RDQ regurgitation items to potent acid suppression in well characterized patients with RE or NERD. Methods We retrospectively reviewed data from the 2 parallelgroup, double-blind, randomized trials introduced above: (1) AZD , 50, or 75 mg/day vs esomeprazole 20 mg/day for NERD, 4 clinicaltrials.gov identifier: NCT ; and (2) Abbreviations used in this paper: CI, confidence interval; GERD, gastroesophageal reflux disease; NERD, nonerosive reflux disease; PPI, proton pump inhibitor; RDQ, Reflux Disease Questionnaire; RE, reflux esophagitis; SF-36, Short Form by the AGA Institute /$

2 June 2012 REGURGITATION RESPONSE TO ACID SUPPRESSION 613 Table 1. RDQ Items and Scoring System Used in This Study Item Symptom domain Terms used in text Burning feeling behind the breastbone Heartburn Substernal burning Pain behind the breastbone Heartburn Substernal pain An acid taste in the mouth Regurgitation Regurgitation-taste Unpleasant movement of material upwards from the stomach Regurgitation Regurgitation-movement Burning feeling in the center of the upper stomach Dyspepsia Dyspepsia-burning Pain in the center of the upper stomach Dyspepsia Dyspepsia-pain Item frequency levels Definition 0 None on any day of the week 1 Present on 1 3 days of the week 2 Present on 4 6 days of the week 3 Present daily Item intensity levels Definition 0 Absent 1 Mild: awareness of symptoms but easily tolerated 2 Moderate: discomfort sufficient to cause interference with normal activities 3 Severe: incapacitating with inability to perform normal activities Item severity levels Low High Definition Symptom being present but to an insufficient degree to qualify as high severity Having the symptom with at least moderate intensity on at least 4 days of the qualifying week AZD , 50, or 75 mg/day vs esomeprazole 40 mg/day for RE, 3 clinicaltrials.gov identifier: NCT There were no placebo controls in these trials. At the screening visit to assess eligibility, baseline symptoms were recorded using 7-day patient recall of the RDQ. Each baseline RDQ item was scored by patients on a 4-point scale for frequency and intensity (Table 1). Eligibility for enrollment required that patients identified substernal burning as their main GERD symptom, that they had been experiencing this for at least 6 months, and that they recalled experiencing substernal burning of at least moderate intensity on at least 4 days in the week before randomization (in this study defined as high severity substernal burning ). Patients who met these inclusion criteria underwent endoscopy and were then entered into the NERD or RE trials, as appropriate. Patients reporting the dyspepsia pain or the dyspepsia burning item as their dominant symptom were excluded. In order to facilitate analysis of the responsiveness of the baseline symptoms substernal pain, regurgitation-taste, and regurgitation-movement, these items were subcategorized as low or high severity to mirror the high severity of substernal burning that was a prerequisite for entry into the trials. High severity was defined as having the symptom with at least moderate intensity on at least 4 days of the qualifying week. The main outcome measure of the current analyses was complete resolution of high severity RDQ regurgitation and heartburn items during week 4 of potent acid suppression. This outcome was based on patient reports of RDQ symptom intensity recorded twice daily in an e-diary during 4 weeks of treatment. Assessment of the Potential to Pool Data From the NERD and RE Trials Neither trial showed a significant difference between esomeprazole and AZD0865 for the relief of substernal burning. In addition, the RE trial did not show any significant differences in esophagitis healing rates (Table 2). Given the results in Table 2, pooled data from all treatment arms of each study (NERD, n 1460; RE, n 1514) were reviewed to gauge Table 2. Equivalence of RE Healing and Complete Relief of Substernal Burning Among the 3 AZD0865 Doses and Esomeprazole After 4 Weeks of Treatment RE Healed NERD Complete resolution of substernal burning Treatment arm n % (95% CI) n % (95% CI) n % (95% CI) AZD mg ( ) ( ) ( ) AZD mg ( ) ( ) ( ) AZD mg ( ) ( ) ( ) Esomeprazole 20 mg (NERD) or 40 mg (RE) ( ) ( ) ( )

3 614 KAHRILAS ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 6 Figure 1. Baseline symptom profile in the RDQ heartburn and regurgitation domains collected at the initial screening visit. In each panel, the number and corresponding percentage of NERD and RE patients with evaluable data for that specific item are indicated. All patients were required to have substernal burning of at least moderate intensity on at least 4 days per week to qualify for the studies. Most patients also reported experiencing at least 1 of the regurgitation items. the therapeutic effect of 4 weeks of potent acid suppression on the RDQ heartburn and regurgitation items. For each item, complete response was defined as the absence of that symptom during the final week of therapy. With an incomplete response, the mean intensity of each RDQ heartburn and regurgitation item was calculated during the fourth week of therapy as the average of each patient s worst daily scores that week (potential range of 0 to 3 points based on the 4-grade intensity rating from none to severe). For example, the mean intensity necessary to achieve the entry criterion of at least moderate (score of 2) substernal burning on 4 days in a week would be (intensity 2 4 days/7 days mean intensity 1.14). These values were then used to construct bubble plots of the mean intensity of each symptom item during the fourth week of therapy to illustrate both complete response and residual symptom severity. Statistics Statistical analysis was performed using SAS 8.2 (SAS Institute Inc, Cary, NC). 2 tests were used to assess the difference between RDQ item resolution at 4 weeks in patients in the RE trial whose esophagitis healed vs those whose esophagitis did not heal. P values.05 were considered to be statistically significant. All confidence intervals (CIs) are 95% throughout this report. Results Baseline Symptoms Figure 1 illustrates the baseline symptom profile for the RDQ heartburn and regurgitation domains in both trials. There were no substantive differences in the baseline symptom profile between patients with RE and those with NERD. Overall, regurgitation was reported by approximately 93% of patients both in the NERD and RE trials. In 53% of patients with NERD (n 717) and 54% of patients with RE (n 751), either or both regurgitation item was of high severity at entry.

4 June 2012 REGURGITATION RESPONSE TO ACID SUPPRESSION 615 Figure 2. Magnitude of residual heartburn and regurgitation symptoms after 4 weeks of acid suppressive therapy in healed vs nonhealed RE patients. Note that if RE patients were healed at 2 weeks they completed the study at that point, but their symptom data were carried forward for the purpose of this analysis. No consistent differences were seen between groups. Patient Population Included in the Symptom Response Analysis The proportion of patients with heartburn and regurgitation relief at 4 weeks in the RE trial was similar in the 1056 patients whose esophagitis healed and in the 221 whose esophagitis did not heal (Figure 2). There were no statistically significant differences between healed and nonhealed patients at week 4 concerning the symptom frequency distribution ( 2, all P.05). Consequently, the data were pooled in the subsequent analysis. Relative Response of Heartburn and Regurgitation to Treatment The treatment responses for each of the RDQ heartburn and regurgitation items that were of high severity at baseline are shown in Figure 3. Both regurgitation items were significantly less responsive to potent acid suppression than was substernal burning in patients with NERD ( substernal burning 48.5% [CI, ]; regurgitation-taste 34.3% [CI, ]; regurgitation-movement 26.0% [CI, ]) and those with RE ( substernal burning 54.9 [CI, ]; regurgitation-taste 43.9% [CI, ]; regurgitation-movement 32.8% [CI, ]). Between the regurgitation items, the proportion of patients relieved of high severity regurgitation-movement was significantly lower than regurgitation-taste in RE (32.8% [CI, ] vs 43.9% [CI, ]). In the patients with NERD, the same trend was observed but there was slight overlap in the CI ( regurgitationmovement 26.0% [CI, ] vs regurgitation-taste 34.3% [CI, ]). High severity baseline substernal pain achieved a similar degree of relief after 4 weeks (42.8% [CI, ] in NERD and 54.1% [CI, ] in RE) as did high severity substernal burning (48.5% [CI, ] in NERD and 54.9% [CI, ] in RE). The magnitude of relief for low severity regurgitation-taste was also similar (51.4% [CI, ] in NERD and 54.4% [CI, ] in RE). However, although low severity regurgitationmovement showed similar responsiveness to high severity substernal burning in the patients with NERD (42.1% [CI, ] vs 48.5% [CI, ]), this was significantly less responsive in the patients with RE (45.2% [CI, ] vs 54.9% [CI, ]). Concordance Between Regurgitation Items During Week 4 of Treatment Figure 4 is a bubble plot of the concordance of residual regurgitation item intensity after acid suppression. All patients

5 616 KAHRILAS ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 6 Figure 3. Responsiveness of high severity regurgitation and heartburn items on the RDQ to 4 weeks of acid suppressive therapy. Complete resolution indicates no report of the item for all 7 days. For each item, the numbers given next to the symbols indicate the number of evaluable patients reporting each RDQ item at high severity at baseline. The RE numbers are somewhat lower than the NERD numbers because patients in the RE study who healed at 2 weeks completed the study at that point and did not have data for analysis at 4 weeks. included in this analysis (617 with NERD and 449 with RE) had either high severity regurgitation-taste or high severity regurgitation-movement at entry, continued in the study for at least 4 weeks (not patients healed of RE at 2 weeks), and had a complete RDQ data set for week 4. Figure 4 plots the mean daily intensity of each item of the regurgitation domain during the fourth week of therapy. For point of comparison, the minimal mean score necessary to achieve the criterion of high severity was set to 1.14 (corresponding to 4 days at moderate intensity and 3 days without the symptom; the lowest reported symptom intensity and frequency qualifying as high severity at baseline). Seventy-one percent of patients with NERD and 75% of patients with RE had improvement to either low severity regurgitation or complete resolution of both RDQ regurgitation items (bubbles in the lower left-hand quadrant of Figure 4A and B). Among patients with persistent high severity regurgitation, approximately half had either regurgitation-taste or regurgitation-movement of high severity (54.1% [n 98/181] for patients with NERD and 54.0% [n 61/113] for patients with RE), and the other half had both items of high severity (45.9% [n 83/181] for patients with NERD and 46% [n 52/113] for patients with RE). Discussion The objectives of GERD treatment are to eliminate esophageal mucosal damage and reflux-induced symptoms. Much has been written regarding the utility of acid-suppressive medications, particularly PPIs, in resolving mucosal injury. 2 This study aimed to quantify the effectiveness of potent acid suppression in relieving the cardinal symptoms of GERD: heartburn and regurgitation. Data from 2 clinical trials that used a well-validated questionnaire, both to screen patients for study enrollment and to quantify the responsiveness of heartburn ( substernal burning and substernal pain ) and regurgitation concepts ( regurgitation-taste and regurgitation-movement ) were analyzed. This provided a database comprising a combined total of 2974 patients. The major findings were that: (1) regurgitation was substantially less responsive to acid suppression than was substernal burning, particularly when conceptualized as regurgitation-movement ; (2) heartburn conceptualized as substernal pain demonstrated similar responsiveness to substernal burning when both were present (approximately 50% of patients had both of these symptoms at baseline); and (3) for all questionnaire items in both domains, symptom responsiveness was, in absolute terms, about 7% less for patients with NERD than for those with RE. Regurgitation is commonly accepted as a typical symptom of GERD. However, in an international panel convened to develop the Montreal GERD definition, experts in esophageal disorders were divided in their views of this symptom. About half of the experts required there to be pharyngeal reflux with accompanying acid taste, while half accepted the perception of movement of reflux within the chest as indicative of regurgitation. 7 A similar outcome emerged from patient focus groups assembled for the development of the RDQ, leading to the inclusion of 2 questionnaire items to encompass the concept of regurgitation: regurgitation-taste and regurgitation-movement (Table 1). The trials evaluated here analyzed responsiveness of regurgitation in these terms with potent acid suppression being achieved with either esomeprazole or AZD0865. Ninety-three percent of the patients enrolled in these trials reported 1 or the other concept of regurgitation during the screening week of any frequency or intensity; in approximately 50% of patients, the symptom was of high severity at entry. A major finding of this study was that the responsiveness of both high severity regurgitation items was substantially less than that of high severity substernal burning : regurgitationtaste was over 10% less responsive (34.3% vs 48.5% NERD; 43.9% vs 54.9% RE) and regurgitation-movement was more than 20% less responsive (26.0% vs 48.5% NERD; 32.8% vs 54.9% RE). Furthermore, severe persistent regurgitation was manifest by only 1 of the 2 RDQ regurgitation items in about half of the cases, further validating the structure of this questionnaire (Figure 4). These observations strongly reinforce the conclusions of a recent systematic review of the responsiveness of regurgitation to PPI therapy, which found that regurgitation was approximately 20% less responsive than heartburn (the therapeutic gain with PPIs relative to placebo was 17% vs 41%, respectively), and concluded that persistent regurgitation was a significant cause of incomplete response to PPI therapy in GERD. 8 Consequently, it is important that clinicians distinguish ongoing regurgitation from ongoing

6 June 2012 REGURGITATION RESPONSE TO ACID SUPPRESSION 617 Figure 4. Concordance between residual RDQ regurgitation items during the fourth week of therapy. (A) Patients with NERD; (B) patients with RE. All included patients reported either high severity regurgitation-taste or regurgitation-movement at baseline. Severity on each axis is the average intensity with which the symptom was reported during the fourth week of treatment with a potential range of 0 3; high severity is indicated by a score of at least The size of the bubbles is proportional to the number of patients represented at specific coordinates. Colors indicate concordance or discordance between RDQ items: red, residual high severity regurgitation-taste and regurgitation-movement ; blue, residual high severity regurgitation-taste only; green, residual high severity regurgitationmovement only; yellow, residual low severity regurgitation-taste and regurgitation-movement ; white, complete resolution of both regurgitation items.

7 618 KAHRILAS ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 6 heartburn in patients with an incomplete response to PPI therapy and consider alternative therapies such as treatment with transient lower esophageal sphincter relaxation inhibitors or fundoplication. Few studies have investigated the impact of severe regurgitation on patient quality of life. The majority of published studies in this area focus either on a combined heartburn and regurgitation symptom severity score, or on heartburn severity only. Two studies have reported separately on the impact of regurgitation by severity. 9,10 However, both studies have shortcomings. One study, by Eloubeidi et al, reported that regurgitation severity correlated significantly with social functioning assessed with the Short Form-36 (SF-36) 9 ; however, actual values and data from other SF-36 domains were not reported. Eslick et al reported that SF-36 domain scores decreased with increasing regurgitation severity but statistical significance and actual values were not reported. 10 The separation of the heartburn experience into substernal burning or substernal pain as characterized by the RDQ items was new in this study. The responsiveness of the 2 heartburn items was not very different, but the important caveat to that was that high severity substernal burning was a prerequisite for inclusion in both trials. Hence, these data support the finding of a recent systematic review and metaanalysis of chest pain response to PPI therapy. 11 Only patients with coexisting objective evidence of GERD (in this case, as assessed by a validated questionnaire with or without coexisting RE) have a high probability of chest pain response to PPI therapy. Consequently, substernal pain can be a typical GERD symptom when substernal burning is also present. Other interesting observations regarding the response of heartburn to therapy in this analysis were: (1) the degree of heartburn relief was very similar between RE patients with and without mucosal healing; and (2) there was no demonstrable dose-response relationship between acid suppression and the relief of substernal burning in either patients with NERD or those with RE. These data, together with the finding that the resolution of all heartburn and regurgitation items was approximately 7% less for patients with NERD than for patients with RE, emphasize the independence of at least some of the pathophysiological determinants behind symptom generation as opposed to mucosal injury. These observations also support the concept of a ceiling effect, such that beyond a certain potency (exceeded by all treatment arms in this case), there was no additional therapeutic gain achieved by increasing the level of acid suppression further. This study has some potential limitations. These relate mainly to methodological issues regarding the retrospective use of data from studies with different primary objectives and that the studies analyzed had no placebo control arm. With respect to enrollment, patients were included only on the basis of having high severity substernal burning during the 1-week recall of screening. Ideally, the same e-diary methodology used to assess 4-week efficacy would have been used during the screening period to provide a more robust baseline symptom assessment. Furthermore, it was assumed that the threshold for defining high severity regurgitation in terms of the frequency and intensity experienced was the same as for heartburn. The Montreal definition stipulates that a symptom should be troublesome for GERD to be diagnosed. 7 Clearly, all of the patients in our study population had troublesome heartburn, but not necessarily troublesome regurgitation. Ideally, we would have validation from patient focus groups of criteria for defining troublesome regurgitation. However, in the absence of any such data providing a clear definition, it seemed reasonable to accept the relatively high threshold of frequency and intensity already in place for substernal burning. Finally, the intensity of acid-suppressive therapy used in these studies was not consistently or appropriately tested with ph-metry or gastric acid analysis. Instead, the drug doses used were based on either the Food and Drug Administration s approved doses for esomeprazole, or pharmacokinetic and pharmacodynamic data for AZD In neither case was twice-daily therapy tested. However, if a dose-response relationship existed, the range of doses (AZD , 50, and 75 mg) of acid-suppressive therapy tested should have demonstrated this. In conclusion, we used data from 2 large clinical trials of potent acid suppressors in patients with GERD to perform a post hoc analysis of the relative responses of heartburn and regurgitation. High severity regurgitation, particularly when perceived as unpleasant movement of material upwards from the stomach, was much less responsive to potent acid suppression than was heartburn in both NERD and RE patients. These observations reinforce previous reports 8 that persistent regurgitation is a significant contributor to partial response to PPI therapy in GERD. References 1. Kahrilas PJ, Shaheen NJ, Vaezi MF, et al. American Gastroenterological Association Institute technical review on the management of gastroesophageal reflux disease. Gastroenterology 2008;135: Kahrilas PJ, Shaheen NJ, Vaezi MF, et al. American Gastroenterological Association medical position statement on the management of gastroesophageal reflux disease. Gastroenterology 2008;135: Kahrilas PJ, Dent J, Lauritsen K, et al. A randomized, comparative study of three doses of AZD0865 and esomeprazole for healing of reflux esophagitis. Clin Gastroenterol Hepatol 2007;5: Dent J, Kahrilas PJ, Hatlebakk J, et al. A randomized, comparative trial of a potassium-competitive acid blocker (AZD0865) and esomeprazole for the treatment of patients with nonerosive reflux disease. Am J Gastroenterol 2008;103: Shaw MJ, Talley NJ, Beebe TJ, et al. Initial validation of a diagnostic questionnaire for gastroesophageal reflux disease. Am J Gastroenterol 2001;96: Dent J, Vakil N, Jones R, et al. Accuracy of the diagnosis of GORD by questionnaire, physicians and a trial of proton pump inhibitor treatment: the Diamond study. Gut 2010;59: Vakil N, van Zanten SV, Kahrilas P, et al. The Montreal definition and classification of gastroesophageal reflux disease: a global evidence-based consensus. Am J Gastroenterol 2006;101: Kahrilas PJ, Howden CW, Hughes N. Response of regurgitation to proton pump inhibitor therapy in clinical trials of gastroesophageal reflux disease. Am J Gastroenterol 2011;106: Eloubeidi MA, Provenzale D. Health-related quality of life and severity of symptoms in patients with Barrett s esophagus and gastroesophageal reflux disease patients without Barrett s esophagus. Am J Gastroenterol 2000;95: Eslick GD, Talley NJ. Gastroesophageal reflux disease (GERD): risk factors, and impact on quality of life-a population-based study. J Clin Gastroenterol 2009;43: Kahrilas PJ, Hughes N, Howden CW. Response of unexplained chest pain to proton pump inhibitor treatment in patients with

8 June 2012 REGURGITATION RESPONSE TO ACID SUPPRESSION 619 and without objective evidence of gastro-oesophageal reflux disease. Gut 2011;60: Holstein B, Holmberg A, Florentzson M, et al. Pharmacokinetic and pharmacodynamic profiles of AZD0865, a novel, potassiumcompetitive acid blocker. Gastroenterology 2004;126:Abstract M1440. Reprint requests Address requests for reprints to: Peter J. Kahrilas, MD, Division of Gastroenterology and Hepatology, 676 N. Saint Clair Street, Suite 1400, Chicago, Illinois p-kahrilas@northwestern. edu; fax: (312) Conflicts of interest The authors disclose the following: Peter Kahrilas has acted as a consultant for AstraZeneca, Eisai, EndoGastric Solutions, Ironwood, Novartis, and XenoPort. Colin Howden has acted as a consultant for Boehringer Ingelheim, Novartis Consumer Health, Novartis Oncology, Otsuka, Takeda, and XenoPort, and as a speaker for Novartis, Otsuka, Takeda and GlaxoSmithKline. Andreas Jonsson, Hans Denison, and Börje Wernersson are employees of AstraZeneca R&D, Mölndal, Sweden. Nesta Hughes is an employee of Oxford Pharma- Genesis Ltd, Oxford, United Kingdom, and was funded by AstraZeneca R&D, Mölndal, Sweden. Funding This study was supported by AstraZeneca R&D, Mölndal, Sweden.