Current Challenges and Emerging Solutions in Upper GI Disorders: A Brief Report of the 2016 AGA Drug Development Conference

In October 2016, the Center for Diagnostics and Therapeutics (CDT) of the American Gastroenterological Association (AGA) held its first Drug Development Conference. Over a 2-day period, researchers, clinicians, Food and Drug Administration (FDA) personnel, and representatives of the pharmaceutical industry and patient advocacy groups convened in Washington, DC, to discuss current unmet needs and future strategies in 4 disorders of the upper gastrointestinal (GI) tract. The 4 disorders discussed were gastroesophageal reflux disease (GERD), eosinophilic esophagitis, gastroparesis, and functional dyspepsia. Experts in these various disorders presented information on current and potential therapeutic strategies, issues of clinical trial design, and possible therapeutic endpoints to be adopted in future clinical trials. Panel discussions after each session included representatives from the FDA and the pharmaceutical industry. Established in 2014, the CDT is the AGA’s third specialty center. It joins the AGA Center for GI Innovation and Technology and the AGA Center for Microbiome Research and Education. The mission statement of the CDT is “To support the development of therapies and diagnostic tests that will enhance human health and improve the lives of patients with digestive disorders.” The aim of the recent Drug Development Conference was to bring together the principal stakeholders (as outlined) to further the mission of the CDT as it pertains to upper GI disorders. In this article, we aim to present a summary of the recent Drug Development Conference and to highlight some of the issues that were discussed. Further, more detailed information will be made available in 4 individual white papers that have been prepared by the participating faculty members. These papers have been submitted for publication to Clinical Gastroenterology and Hepatology. Disorders of the upper GI tract are highly prevalent in specialized and primary care and constitute a large proportion of outpatient and hospital visits in the United States. Data from the AGA’s Burden of Disease survey1Peery A.F. Dellon E.S. Lund J. et al.Burden of gastrointestinal disease in the United States: 2012 update.Gastroenterology. 2012; 143: 1179-1187Abstract Full Text Full Text PDF PubMed Scopus (1415) Google Scholar found that, in 2009, there were >15 million outpatient visits for abdominal pain, almost 3 million each for problems with nausea or vomiting, almost 2 million for heartburn or “indigestion,” and >1 million for dysphagia. Among the top 15 physician-made diagnoses for GI disorders in 2009 were GERD (ranked first with >8.8 million diagnoses), “gastroenteritis and dyspepsia” (ranked third with >4 million diagnoses), nausea and vomiting (ranked tenth with >1.6 million diagnoses), and dysphagia (ranked 15th with >1 million diagnoses). Furthermore, the burden of upper GI tract disorders extends beyond the outpatient setting; during 2009, there were >65,000 hospital discharge diagnoses of reflux disease and >130,000 discharge diagnoses of functional/motility disorders. Discharge diagnoses of reflux disease had clearly diminished over the preceding 9 or 10 years. However, this decrease had been more than offset by a marked increase in the number of patients who had been hospitalized and found to have some functional GI or motility disorder. Among the latter group, there had been a 26% increase over the preceding 9–10 years. Those patients had a median duration of hospital stay of 4.0 days at an estimated cost of >$972 million. Among the patients with diagnoses of functional and/or motility disorders would have been many with gastroparesis and/or functional dyspepsia—or related functional GI disorders such as irritable bowel syndrome. Proton pump inhibitors (PPIs) have been available since the early 1990s. The extensive, cumulative clinical experience with them justifies their use for patients with troublesome heartburn owing to GERD, and in patients at high risk of upper GI tract ulcers because of long-term use of nonsteroidal anti-inflammatory drugs and/or aspirin. However, there has been a tendency to prescribe or recommend PPIs for patients with less specific symptoms or weaker indications. PPIs have been used frequently for patients with dyspeptic symptoms despite the observation from clinical trials and routine clinical practice that only a minority of patients show marked benefit. PPIs are also often recommended for a variety of symptoms thought to emanate from the upper GI tract, even when such symptoms (eg, upper abdominal fullness, bloating, early satiety, etc) would be unlikely to be acid-mediated or helped by the pharmacologic suppression of gastric acid secretion. It is clear that better therapeutic options are required for specific disorders of the upper GI tract, including those discussed at the recent conference. In the early 1990s, there was much optimism that this question might have been answered in the affirmative. At that time, patients with heartburn due to GERD often had a miserable time and were recommended to try a variety of (generally ineffective and non–evidence-based) lifestyle modifications along with liberal quantities of antacids and H2-receptor antagonists. Our approach to those patients was simple then; we offered them an upgrade to PPI treatment and felt generally satisfied with the outcomes. After all, early and subsequent randomized, controlled trials consistently showed superiority of any PPI over any H2-receptor antagonist in the healing of erosive esophagitis and in the control of heartburn in patients with erosive esophagitis. So, problem solved—right? Unfortunately, life is rarely that simple. For one thing, most patients with GERD do not have erosive esophagitis and—perhaps—our focus on endoscopic healing rates of erosive esophagitis was not appropriate. GERD patients often experience a variety of symptoms aside from heartburn and not all of these respond satisfactorily to PPI treatment. Many of the GERD patients whom we now see at our clinics are dissatisfied with PPI treatment and often complain of persistent heartburn and/or other symptoms such as regurgitation2Kahrilas P.J. Jonsson A. Denison H. et al.Regurgitation is less responsive to acid suppression than heartburn in patients with gastroesophageal reflux disease.Clin Gastroenterol Hepatol. 2012; 10: 612-619Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar, 3Kahrilas P.J. Howden C.W. Wernersson B. et al.Impact of persistent, frequent regurgitation on quality of life in heartburn responders treated with acid suppression: a multinational primary care study.Aliment Pharmacol Ther. 2013; 37: 1005-1010Crossref PubMed Scopus (16) Google Scholar and epigastric burning or discomfort. This in itself represents one of the recurring themes from the Drug Development Conference—namely, that of an overlap diagnosis. For example, patients with a firm diagnosis of GERD may have other symptoms that are troublesome to them. These symptoms may be sufficient to fulfill the diagnostic criteria for an additional disorder such as dyspepsia or irritable bowel syndrome. It is clear that patients’ own assessments of treatment outcomes may be very different from those of their health care providers—hence, the growing importance of properly designed and validated patient-reported outcomes (PRO) measures for use in future clinical trials. (The issue of PRO measures was another recurring theme of the conference impinging on all 4 topic areas discussed.) The forthcoming white paper on GERD will include future directions for the development of agents that can, perhaps, be added to PPI therapy and that might be suitable for GERD patients whose heartburn is well controlled by a PPI but whose other symptoms are not. Regurgitation is the second most frequent symptom of GERD. It is clear that PPIs are much less effective for this complaint than they are for heartburn and that regurgitation is frequently responsible for apparent “failure” of PPI treatment in GERD.2Kahrilas P.J. Jonsson A. Denison H. et al.Regurgitation is less responsive to acid suppression than heartburn in patients with gastroesophageal reflux disease.Clin Gastroenterol Hepatol. 2012; 10: 612-619Abstract Full Text Full Text PDF PubMed Scopus (80) Google Scholar, 3Kahrilas P.J. Howden C.W. Wernersson B. et al.Impact of persistent, frequent regurgitation on quality of life in heartburn responders treated with acid suppression: a multinational primary care study.Aliment Pharmacol Ther. 2013; 37: 1005-1010Crossref PubMed Scopus (16) Google Scholar Furthermore, persistent regurgitation has a major negative impact on patients’ quality of life. Future clinical trials of such agents might focus more on improvement in regurgitation as a predefined primary endpoint. Again, the importance of properly validated PRO measures and their incorporation into any assessment of efficacy cannot be overemphasized. Although some recent trials of motility-modifying agents in GERD patients who had an incomplete response to PPI treatment failed to meet their primary endpoints this may have been due to lack of therapeutic efficacy, or to the inclusion into the trials of patients with the wrong symptom profile(s), or some combination of the two. Eosinophilic esophagitis is a disorder of increasing importance. Long recognized by pediatricians, it is a relative novelty for adult gastroenterologists. The condition remains poorly understood and underappreciated by nonspecialists. It is often confused with GERD (with which it has overlap symptoms), and often goes undiagnosed and untreated for years. There has been debate and disagreement about whether eosinophilic esophagitis and GERD are distinct disorders or whether they form part of a broad spectrum of esophageal conditions. That concept may be better established in pediatric practice than it currently is in adult practice. Furthermore, there has been additional confusion concerning the emerging concept of PPI-responsive esophageal eosinophilia. Patients with PPI-responsive esophageal eosinophilia are currently indistinguishable from those with typical eosinophilic esophagitis on clinical, endoscopic, or histologic criteria. PPIs have traditionally been considered solely as acid-suppressing drugs. However, this may be an oversimplification; discussion at the conference included the possibility of some hitherto unrecognized anti-inflammatory properties of the PPIs. However, why these should be conveniently manifested in the esophagus is unclear. This seems to be a “watch this space” issue. In clinical practice, patients are generally diagnosed with eosinophilic esophagitis when they have symptoms consistent with the diagnosis along with typical endoscopic findings and histologic confirmation of an abnormally high eosinophilic infiltrate on esophageal biopsies (currently defined as ≥15 eosinophils per high-power field). Distinction between PPI-responsive esophageal eosinophilia and “true” eosinophilic esophagitis is generally made on the response—or lack thereof—to a course of high-dose PPI treatment. However, it is unclear how best to follow those patients in clinical practice. Those who do not respond to PPIs—and who are currently considered to have eosinophilic esophagitis—can be managed with swallowed steroids or elimination diets or—when there is esophageal narrowing on endoscopy—by dilation procedures. Steroids, however, are not the final answer; ≤50% of patients do not have an adequate response and the optimal duration of treatment in those who do respond is unclear. Anti-tumor necrosis factor agents—used so successfully in inflammatory bowel disease, psoriasis, and inflammatory arthritides—have been unsuccessful in eosinophilic esophagitis. Future therapeutic targets might include interleukin-5 and interleukin-13. Clinical trials of current and potential therapeutic agents in eosinophilic esophagitis require clear inclusion and exclusion criteria and firmly established therapeutic endpoints. Whereas symptom improvement—such as a decrease in dysphagia scores—is clearly of paramount importance to patients, this may not be an adequate endpoint for clinical trials. Some combination of symptom evaluation and demonstration of a reduction in esophageal eosinophilic infiltrate is likely to be required. Histologic improvement should reflect reduced risk of future complications of eosinophilic esophagitis such as stricturing and rigidity of the esophagus. Although such measures are perhaps not of immediate relevance or interest to patients, they are to researchers and regulatory agencies. Several members of the FDA pointed out the agency’s interest in improving how the patient feels after treatment for a disease, functions after treatment of a disease, or survives a disease. Patients with eosinophilic esophagitis—both children and adults—face enormous challenges and the condition can impose a major financial burden on patients and their families. However, there is clearly hope for patients with this disorder, given the existence of a dedicated group of researchers and active collaboration with the pharmaceutical industry in the development of newer agents with different therapeutic targets and mechanisms of action. Gastroparesis is responsible for considerable morbidity and impairment of patients’ quality of life. Cardinal symptoms of gastroparesis include upper abdominal pain, nausea, and vomiting but no distinctive pathophysiologic abnormality currently explains these adequately. Different measures of gastric emptying have been used in both clinical practice and in clinical trials. However, there is a poor correlation between patients’ symptoms and gastric emptying. Gastroparesis is increasing in prevalence. This is partly explained by the increase in type 2 diabetes associated with obesity. However, there is also what might be called an iatrogenic component. Patients on long-term opioids may experience a variety of GI complaints aside from chronic constipation. Adequate management of opioid-induced constipation is an important part of the management of patients with opioid bowel dysfunction, which may include symptomatic gastroparesis. Symptoms of gastroparesis also overlap with those of functional dyspepsia. Many patients who have participated in clinical trials in gastroparesis would also have fulfilled diagnostic criteria for functional dyspepsia; a good example is that of the NORIG trial that compared nortriptyline with placebo.4Parkman H.P. van Natta M.L. Abell T.L. et al.Effect of nortriptyline on symptoms of idiopathic gastroparesis: the NORIG randomized clinical trial.JAMA. 2013; 310: 2640-2649Crossref PubMed Scopus (115) Google Scholar Currently, management of gastroparesis can be challenging. Patients with gastroparesis secondary to diabetes are encouraged to optimize their diabetic control and, perhaps, to take a “small particle diet” comprising foodstuffs that are easily mashed with a fork. Strict glycemic control in patients with diabetes can be associated with improvement in gastroparesis symptoms and a reduced requirement for hospitalization. Currently, the only medicine that is FDA-approved for gastroparesis is metoclopramide, which is far from optimal treatment. Furthermore, metoclopramide is associated with risks of neurologic disorders, including acute dystonia and chronic dyskinesia. Antiemetics may improve nausea and vomiting in patients with gastroparesis, but do not address other symptoms of the condition. Perhaps future clinical trials will focus on drug combinations (eg, an antiemetic along with an agent designed to facilitate gastric emptying) or on medicines with >1 mechanism of action. Newer agents that are in phase III of development include aprepitant (a NK1-receptor antagonist) and relamorelin (a central ghrelin receptor agonist); some preliminary data on these agents were discussed at the conference. Gastric emptying studies have traditionally been an inclusion factor for gastroparesis trials. Given the lack of consistency with how these studies are performed, and the lack of correlation between symptom response and improvement in gastric emptying measured by these studies, there was a vigorous discussion on whether they should be used as an entry criterion. However, there was broad agreement that they should be used as a secondary objective to enhance our understanding of how symptoms correlate with physiologic changes in gastric function. Patients with functional dyspepsia have pain in the upper abdomen that can be associated with a variety of other symptoms including nausea, upper abdominal burning, fullness after meals, and bloating. Various diagnostic criteria have been used both in clinical practice and in clinical trials. The most recent are the fourth iteration of the Rome criteria. Rome IV defines functional dyspepsia by the presence of one or more of bothersome postprandial fullness, bothersome early satiation, bothersome epigastric pain, and bothersome epigastric burning in the absence of structural disease that might explain these symptoms.5Stanghellini V. Chan F.K.L. Hasler W.L. et al.Gastroduodenal disorders.Gastroenterology. 2016; 150: 1380-1392Abstract Full Text Full Text PDF PubMed Scopus (749) Google Scholar Furthermore, symptoms should have been present for the last 3 months with a symptom onset of ≥6 months ago. The use of the term “bothersome” is meant to indicate that the symptoms are “severe enough to impact on usual activities” and can be scored semiquantitatively as ≥2 on a scale extending from 0 to 5. Although it will be some time before the Rome IV criteria are generally applied, they might form the basis for enrollment into future therapeutic trials in functional dyspepsia. Functional dyspepsia is an extremely prevalent disorder. However, this has not been reflected in the ability to recruit suitable patients into clinical trials.6Irvine E.J. Tack J. Crowell M.D. et al.Design of treatment trials for functional gastrointestinal disorders.Gastroenterology. 2016; 150: 1469-1480Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar Some trials have been unduly restrictive in an attempt to have a patient population of “pure” functional dyspepsia, in which case patients with additional complaints such as heartburn would have been excluded. Although this allows a trial to focus on the improvement of a specific symptoms (or set of symptoms), it might not be representative of the “real-world” setting where patients may have a variety of symptoms that—individually or collectively—can wax and wane in severity and frequency over time. Overlap diagnoses of GERD, irritable bowel syndrome, gastroparesis, and other functional GI and non-GI conditions predominate in the real-world setting of clinical practice, but such patients may have been excluded deliberately from the more rarefied environment of the ivory tower-based randomized, controlled trial.6Irvine E.J. Tack J. Crowell M.D. et al.Design of treatment trials for functional gastrointestinal disorders.Gastroenterology. 2016; 150: 1469-1480Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar One recent placebo-controlled trial of 2 antidepressant agents—namely, amitriptyline and escitalopram—that was funded by the National Institutes of Health, had to be terminated prematurely because of difficulties with patient enrollment.7Talley N.J. Locke G.R. Saito Y.A. et al.Effect of amitriptyline and escitalopram on functional dyspepsia: a multi-center, randomized, controlled study.Gastroenterology. 2015; 149: 340-349Abstract Full Text Full Text PDF PubMed Scopus (216) Google Scholar Interestingly, the single greatest impediment to recruitment was the large numbers of functional dyspepsia patients who were taking—or had recently been taking—an antidepressant agent. Some functional dyspepsia patients have infection with Helicobacter pylori—as do some asymptomatic individuals. H pylori infection is always worth treating because it is associated with possible serious long-term consequences. However, it is now clear that treatment of infected functional dyspepsia patients will only result in sustained symptom improvement in a small proportion. There was some disagreement among experts as to whether H pylori-infected patients with functional dyspepsia should be considered eligible for future therapeutic trials in functional dyspepsia. The best approach to this is likely to be influenced by such factors as the geographical setting of the trial and the corresponding prevalence of H pylori infection, which is typically higher in developing than industrialized nations. The importance of PRO measures in evaluating response to investigational agents has already been highlighted. In functional dyspepsia, there is genuine progress in this regard. Attendees at the conference were updated about the development of a Functional Dyspepsia Symptom Diary. This represents an important advance and is the result of work done through the Patient-Reported Outcome Consortium, which was launched in 2009 in collaboration with the FDA and the pharmaceutical industry. The working group on the Functional Dyspepsia Symptom Diary included experts who participated in the Drug Development Conference. To reiterate, the CDT’s mission statement is “To support the development of therapies and diagnostic tests that will enhance human health and improve the lives of patients with digestive disorders.” This—and, hopefully, future—Drug Development Conferences are one means whereby we can try to achieve this goal. A special feature of the conference was its multidisciplinary attendance, bringing together researchers in the field with FDA personnel, pharmaceutical industry investigators, and patient representatives. We hope that the stimulating discussion will help to encourage progress in these disorders which are responsible for so much morbidity and health care costs. We are grateful to all the invited speakers, moderators, and discussants, and to all who attended the conference in other capacities and contributed to the discussion. We expect the individual white papers from this conference will have a real and lasting impact on these often difficult-to-manage upper GI disorders. We hope these papers will stimulate thought, discussion, and interaction, all of which should serve to advance the mission of the CDT. The authors recognize the 2016 AGA Drug Development Conference faculty, discussants, and moderators: Richard Aranda, Michael Camilleri, Robyn Carson, Evan Dellon, Fred Fiedorek, Glenn Furuta, Martin Golding, Donna Griebel, M. Scott Harris, William Hasler, Ikuo Hirano, Jessica Lee, Douglas Levine, Paul Moayyedi, John Pandolfino, Henry Parkman, P. Jay Pasricha, Nicholas Shaheen, Stuart Spechler, Jan Tack, Juli Tomaino, Michael Vaezi, Marcelo Vela, Priya Venkataraman-Rao, and Timothy Wang.