GASTROENTEROLOGY
`BASICANDCLINICAL
`
`REVIEWSIN
`
`REVIEWS IN BASIC AND CLINICAL
`GASTROENTEROLOGY
`
`GASTROENTEROLOGY 2008;135:1877–1891
`
`Challenges to the Therapeutic Pipeline for Irritable Bowel Syndrome: End
`Points and Regulatory Hurdles
`
`MICHAEL CAMILLERI* and LIN CHANG‡
`
`*Clinical Enteric Neuroscience Translational and Epidemiological Research (CENTER), Mayo Clinic, Rochester, Minnesota, and ‡Center for Neurobiology of Stress,
`David Geffen School of Medicine at UCLA, VAGLAHS, Los Angeles, California
`
`Recent advances in our understanding of basic neu-
`roenteric mechanisms and the role of effectors and
`transmitters in the brain-gut axis have provided op-
`portunities to develop new therapeutic agents for ir-
`ritable bowel syndrome (IBS). Furthermore, human
`pharmacodynamic studies utilizing transit, colonic,
`or rectal sensitivity and brain imaging have been use-
`ful in determining therapeutic efficacy (particularly
`for drugs that act on motor function). This review
`provides an overview of medications that have not yet
`been approved for treatment of patients with IBS yet
`have shown promise in phase IIB trials. These include
`drugs that act on the serotonin receptor and transporter
`system: antidepressants, norepinephrine reuptake in-
`hibitors, opioids, cholecystokinin antagonists, neuro-
`kinin-antagonists, chloride channel activators, guan-
`ylate cyclase C agonists, atypical benzodiazepines,
`probiotics, and antibiotics. The changing landscape
`in the regulatory approval process has impacted the
`development of IBS drugs. Guidance documents from
`regulatory agencies in Europe and the United States
`have focused on patients’ reported outcomes and as-
`sociated quality of life. After a decade of experience
`with different end points that have generated some
`data on psychometric validation and unprecedented
`information about responsiveness of the binary or
`global end points to drug therapy, it is necessary to
`pursue further validation studies before or during
`pivotal phase IIB or III trials. The hope of providing
`relief to patients should galvanize all parties to
`achieve these goals.
`
`Irritable bowel syndrome (IBS) involves a broad range
`
`of physiologic and psychologic alterations that affect
`brain-gut dysregulation, gut function, visceral percep-
`tion, and mucosal integrity and function. In the absence
`of a reliable biologic marker of IBS, it has been challeng-
`ing to develop well-targeted, effective IBS drugs and op-
`timal end points for clinical trials. Despite advances in
`
`our understanding of basic neuroenteric mechanisms
`and the role of effectors and transmitters in the brain-gut
`axis, the pipeline of drugs for IBS and lower functional
`gastrointestinal (GI) disorders (Table 1), and relevant
`pharmacodynamics end points to predict proof of effi-
`cacy, the changing landscape in the regulatory approval
`process, particularly the expectations of IBS trial end
`points, have impacted the development of IBS drugs.
`This review addresses 3 main topics: the pipeline for IBS
`and lower functional GI disorders, approaches to the
`development of medications for IBS, and IBS trial end
`points and insights into regulatory affairs.
`
`What Therapeutic Agents Are in the
`Pipeline for IBS?
`There are a number of novel agents with different
`mechanisms of action that are in various stages of devel-
`opment. Several of the drugs in development that are in
`ongoing or planned clinical trials for IBS are presented in
`Table 1. The rationale, putative action, pharmacodynam-
`ics, and results in clinical trials1–90 are summarized in
`Table 2.
`
`Appraisal of Drugs That Affect GI
`Motility, Sensation, Secretion, or
`Central Actions
`Although there is a greater understanding of the
`basic neuroenteric mechanisms and the role of effectors
`and transmitters within the brain-gut axis, which provide
`opportunities for development of new therapeutic agents
`in IBS, there are still significant conceptual and practical
`barriers. IBS is a complex multifactorial disorder with
`distinct but often interrelated pathophysiologies. These
`
`Abbreviations used in this paper: 5-HT, serotonin; IBS, irritable bowel
`syndrome; IBS-C, irritable bowel syndrome with predominant consti-
`pation; IBS-D, irritable bowel syndrome with predominant diarrhea.
`© 2008 by the AGA Institute
`0016-5085/08/$34.00
`doi:10.1053/j.gastro.2008.09.005
`
`MYLAN - EXHIBIT 1031
`
`

`

`1878 CAMILLERI AND CHANG
`
`GASTROENTEROLOGY Vol. 135, No. 6
`
`Table 1. Drugs in Development for IBS in Open or Planned and Classified by Phase I to III Clinical Trials
`
`Drug/agent in development
`
`SSR241586
`SAR102779
`Octreotide
`Guanilib (SP304)
`RR210
`BMS 562086
`GW876008
`LX1031
`Dextofisopam
`Citalopram
`AST 120 (kremezin)
`Traditional Chinese medicine
`AGN 203818
`VSL#3
`Flora-Q
`Lactobacillus farciminis
`Tianeptine
`DDP733
`DDP225
`Mesalamine
`Ibaconda (olsalazine/colchicine)
`Gastrafate IB (sucralfate)
`Arverapamil
`
`Linaclotide
`Rifaximin
`Saccharomyces boulardii
`
`Mechanism
`
`IBS patients
`
`Phase
`
`NK2/NK3 antagonist
`NK2/NK3 antagonist
`Somatostatin analog
`Guanylate cyclase-C agonist
`5-HT3 partial agonist
`CRF-1 antagonist
`CRF-1 antagonist
`Tryptophan hydroxylase inhibitor
`2,3-Benzodiazepine agonist
`Selective serotonin reuptake inhibitor
`Adsorbs bile acids and bacterial toxins
`Herbal medicine
`Alpha 2B agonist
`Probiotic combination
`Probiotic
`Probiotic
`Enhances serotonin reuptake
`Partial 5-HT3 agonist
`Serotonin and noradrenaline reuptake inhibitor
`5-Aminosalicylate
`5-aminosalicylate/intestinal secretion
`Sucrose sulfate-aluminum salt: cytoprotection
`Enantiomer of verapamil; calcium channel blocker;
`5-HT2b and melatonin (MT1) binding
`Guanylate cyclase-C agonist
`Antibiotic
`Probiotic
`
`IBS
`IBS
`Women only
`IBS-C
`IBS
`IBS
`Women only
`IBS
`IBS
`IBS
`Non-IBS-C
`IBS
`Pain predominant IBS
`IBS-D
`IBS-D
`IBS-D
`IBS
`IBS
`IBS-D
`PI-IBS
`IBS-C
`All subtypes
`IBS-D
`
`IBS-C
`IBS-D
`IBS-D
`
`I
`I
`I
`I
`I
`I
`I/II
`I/II*
`II
`II
`II
`II
`II
`II
`II
`II
`II
`II
`II
`II
`II
`II/III
`III
`
`III
`III
`III
`
`CRF, corticotropin releasing factor; NK, neurokinin.
`
`pathophysiologic processes and associated symptom phe-
`notypes can change within an individual over time. Fur-
`thermore, several putative mechanisms may control the
`pathophysiologic processes that might underlie the gen-
`eration of symptoms. Although a significant number of
`IBS patients report meal-related symptoms, the interac-
`tion of food and intraluminal content with secretory,
`motor, and sensory mechanisms in IBS is poorly under-
`stood. The approach to development of medications for
`IBS has been based on specificity of targets, analogous to
`that of, for example, hypertension. The difference is that,
`whereas hypertension is dominated by the biology of
`vascular tone, IBS does not have a dominant mechanistic
`pathway to symptom generation. Moreover, there are
`single targets that appear to regulate multiple functions,
`including gut motor function and sensitivity in animal
`models, such as specific serotonergic (5-HT) receptor
`subtypes. However, despite the apparent relevance of
`such targets, efficacy and safety are not always clearly
`demonstrable in the IBS patient population. As a result
`of the approach based on targeting specific receptors in a
`disease that does not have a malfunction of a single
`receptor or transmitter deficiency, approaches that target
`one receptor or pathophysiologic mechanism cannot be
`expected to affect the broad spectrum of patients. Thus,
`approaches directed at changing motor function could
`alleviate bowel dysfunction, or induction of intestinal
`secretion might alleviate constipation, but the compo-
`
`nent of bloating and pain might be left unattended.
`Antiinflammatory approaches that have been investi-
`gated in small trials, even in those that include many
`patients with postinfectious IBS, have been disappoint-
`ing. Therefore, multitargeted approaches are often used
`in clinical management, particularly in patients with
`moderate to severe IBS. Centrally acting agents appear to
`be promising because they might correct disturbances in
`the brain-gut axis. However, their efficacy has generally
`been limited in clinical trials, and many patients prefer to
`avoid taking “mind-altering” medications for symptoms
`that disturb their quality of life but are not life-threat-
`ening. The risk-benefit ratio of any new medication for
`IBS is clearly a determining factor in the approval and
`marketing of such compounds. It is understandable,
`therefore, that approaches with probiotics and antibiot-
`ics have reached a level of acceptance in practice that
`exceeds the available evidence of efficacy in support of
`their use.
`Some of the challenges are, in our current state of
`knowledge, not easily resolved. Given the high prevalence
`and disease burden associated with IBS, there needs to be
`continued and vigorous basic and translational research
`in the field, rigorous pharmacologic assessment of can-
`didate drugs, or other therapies using validated biomar-
`kers and relevant clinical end points. It is also important
`to recognize that multiple, complex, and dynamic patho-
`physiologic processes that underlie IBS might be best
`
`BASIC AND CLINICAL
`GASTROENTEROLOGY
`
`REVIEWS IN
`
`

`

`GASTROENTEROLOGY
`BASICANDCLINICAL
`
`REVIEWSIN
`
`December 2008
`
`CHALLENGES TO THE IBS THERAPEUTIC PIPELINE 1879
`
`Table 2. Summary of Rationale, Mechanisms, and Efficacy of Medications in Pipeline for IBS
`
`Rationale or
`putative action
`
`Pharmacodynamic
`(intestine or colon)
`
`Clinical efficacy: phase IIB or III
`primary end points
`
`Safety issues/comments
`
`Reference
`No.
`
`Drug class
`
`Examples
`
`5-HT3-agonist
`
`DDP-733
`
`5-HT4-agonists
`
`Prucalopride
`
`ATI-7505
`
`TD-5108
`
`NARI and 5-HT3-
`antagonist
`
`DDP-225
`
`Antidepressants
`
`Stimulate intrinsic
`cholinergic neurons to
`enhance motility
`
`Stimulate intrinsic
`cholinergic neurons to
`enhance motility
`
`Stimulate intrinsic
`cholinergic neurons to
`enhance motility
`Stimulate intrinsic
`cholinergic neurons to
`enhance motility
`
`May increase synaptic
`levels of norepinephrine
`to reduce visceral pain;
`inhibit intrinsic
`cholinergic neurons
`May reduce visceral
`sensation and relieve
`depression associated
`with IBS
`
`4-mg dose delayed fasting
`migrating motor complexes,
`accelerated small intestinal
`transit, and induced softer
`stools or diarrhea in 15% of
`subjects
`Increases SB, colon motility and
`transit in healthy controls and
`patients with chronic
`constipation
`Increases colon transit in healthy
`controls
`
`Dose-related increase in SB and
`colon transit in healthy
`controls
`
`Uninterpretable
`
`IIB, dose-ranging study in 91 IBS-C
`patients: 1.4-mg dose
`associated with significantly
`greater proportion of responders
`(subject global assessment of
`relief)
`IIB and III in CC (thousands of
`patients): BM frequency and
`satisfaction with bowel function
`both improved
`None reported
`
`IIB, dose-ranging study in 401 CC
`patients increased BM
`frequency and proportion with
`adequate relief
`IIB, dose-ranging study in 87 IBS
`patients increased proportion
`with adequate relief
`
`SSRIs, fluoxetine and paroxetine,
`and TCA, amitryptiline, do not
`reduce visceral sensitivity, in
`contrast to the SNRI,
`venlafaxine; SSRI accelerates
`and TCA slows SB transit
`
`Small studies with SSRI or TCA
`equivocal; large study had no
`significant benefit of
`desipramine over placebo in ITT
`analysis, but did in per-protocol
`analysis (completed treatment)
`
`No known vascular effects
`
`1, 2
`
`Greater selectivity for
`5-HT4 than 5-HT1B or
`hERG channel
`
`Greater selectivity for
`5-HT4; not metabolized
`by CTP 3A4
`Greater selectivity for
`5-HT4
`
`No constipation reports
`suggest low expectation
`for 5-HT3 antagonist
`activity
`
`Side effects common with
`TCA. Post hoc analysis
`for desipramine showed
`benefit in those with
`moderate symptoms,
`abuse, no depression,
`and IBS-D
`
`Possibly more events of
`worsening abdominal
`pain; headaches were
`more frequent with
`placebo
`
`Nausea that is usually
`mild; FDA approved
`
`3–10
`
`11
`
`12–14
`
`15
`
`16–28
`
`33–36
`
`37, 38
`
`45–50
`
`51–53
`
`51, 52,
`54, 57
`
`58, 59
`
`62–67
`
`73–75
`
`76–84
`
`85–90
`
`␬-opioid agonist
`
`Asimadoline
`
`␬-opioid receptors in
`visceral perception
`
`2,3-Benzodiazepine
`modulator
`
`Dextofisopam
`
`Potential to reduce
`stimulation-induced
`colonic motility and
`visceral sensitivity
`
`Reduce sensation in response to
`colon distentions in the
`nonnoxious range; relax colon
`tone in healthy controls;
`increase sensory thresholds in
`patients with IBS
`
`None reported
`
`CCK1 antagonist
`
`Dexloxiglumide
`
`Competitive antagonist of
`the CCK1-receptor
`
`Slower ascending colon emptying
`with no significant effect on
`overall colonic transit
`
`Reduce the emotional response
`of IBS patients to rectosigmoid
`distention
`Reduce contraction frequency
`and amplitude on MMC in SB
`in healthy males
`No effect on rectal compliance,
`sensory thresholds, or
`intensity ratings in healthy
`controls
`Accelerates SB and colonic
`transit in healthy controls
`
`Accelerated ascending colonic
`transit and altered bowel
`function in 36 women with
`IBS-C
`Slow colonic transit in IBS-D
`
`NK antagonists
`
`NK1 antagonist,
`ezlozipant
`
`NK1-receptors’ role in
`nociception
`
`NK2-antagonist,
`nepadutant
`
`NK3-antagonist,
`talnetant
`
`NK2-receptors’ influence
`on smooth muscle
`contractility
`NK3-receptors’ role in
`nociception
`
`Increases intestinal water
`and electrolyte
`secretion
`Increases intestinal water
`and electrolyte
`secretion
`
`Potential mechanisms:
`immune, barrier,
`fermentation
`
`Cl-C2 channel
`activator
`
`Lubiprostone
`
`Guanylate cyclase-
`C agonist
`
`Linaclotide
`
`Probiotics
`
`Antibiotics
`
`Several, eg,
`Bifidobacteria,
`Lactobacillus,
`Saccharomyces
`species, or
`combinations
`Neomycin,
`metronidazole,
`rifaximin
`
`On-demand dosing not effective in
`reducing severity of abdominal
`pain in 100 IBS patients; IIB,
`dose-ranging study, 596 IBS
`patients: at least average
`moderate pain benefit in IBS-D
`and IBS-A
`IIB study in 140 IBS patients:
`increased number of months of
`adequate overall relief of IBS
`symptoms; efficacy lower over
`time
`Two initial IIB or III trials: not
`efficacious in IBS-C; a
`randomized withdrawal design
`trial showed longer time to loss
`of therapeutic response, longer
`for dexloxiglumide
`None
`
`None
`
`Two IIB trials in 1350 IBS
`patients: no benefit
`
`Two phase III in several hundred
`CC and IBS-C patients:
`efficacious
`IIA and IIB studies in CC or IBS:
`increased BM frequency
`
`Several IIB studies: efficacy in
`overall IBS and single
`symptoms, eg, flatulence, pain
`
`Changes in gut microflora
`may be present in IBS
`
`No consistent reduction in breath
`hydrogen excretion after
`lactulose load in those with
`symptom relief
`
`IIB trials of various sizes: efficacy
`for global symptoms in some,
`gas and bloating in others
`
`BM, bowel movements; CC, chronic constipation; Cl-C2, chloride channel type 2; IBS, irritable bowel syndrome; ITT, intention to treat; MMC, migrating motor complexes; NARI, norepinephrine
`reuptake inhibitor; NK, neurokinin; SB, small bowel; SSRI, selective serotonin reuptake inhibitor; SNRI, serotonin and norepinephrine reuptake inhibitor; TCA, tricyclic antidepressant.
`
`

`

`1880 CAMILLERI AND CHANG
`
`GASTROENTEROLOGY Vol. 135, No. 6
`
`addressed with a multicomponent approach. This has to
`be coupled with awareness of safety signals in drug de-
`velopment programs in IBS. For almost all of the drug
`classes described here (Table 2), rigorous phase III trials
`are still awaited.
`
`Approaches to Proof of Concept for
`Novel IBS Drugs
`There are at least 3 different approaches to deter-
`mining the efficacy of new treatments of IBS. The tradi-
`tional path is based on identifying the molecular targets in
`animal models that are thought to mediate the human
`phenotype, such as visceral hyperalgesia and rapid gut
`transit.91 If a candidate drug has been shown to be
`effective in preclinical studies, and it is safe in phase I
`trials in humans, it is moved into trials in healthy human
`subjects and subsequently in patients with IBS during
`different phases of clinical trials.
`Other approaches occur at later stages of drug devel-
`opment. For example, a drug in development or one that
`is already approved for another condition that has an associated
`effect on GI function or symptoms can be tested in patients
`with IBS. For example, if a drug has been found to be
`effective in treating patients with constipation, it could
`be further investigated as a treatment for patients with
`IBS with constipation. It is also possible to assess the
`efficacy of a drug that is used to treat a condition that
`commonly coexists with IBS and/or is thought to have
`shared pathophysiology, such as fibromyalgia, anxiety, or
`depression. Examples of these agents include selective
`serotonin reuptake inhibitors and serotonin-norepineph-
`rine reuptake inhibitors.
`An alternative approach to drug discovery and devel-
`opment is pharmacologic brain imaging in animal mod-
`els and humans.91 Brain responses can reflect global IBS
`symptoms; this approach to drug development is the
`subject of ongoing study.
`
`The Drug Development Path
`Before regulatory approval, candidate drugs move
`through a long and complex development path that
`includes toxicology,
`toxicokinetics, pharmacokinetics,
`and in vivo efficacy testing in animals as well as 3 phases
`of clinical trials. Phase I trials are dose-ranging studies
`designed to measure the safety, tolerability, pharmacoki-
`netics, and pharmacodynamics of a test drug. If the drug
`is found to be safe and tolerable in phase I trials, phase II
`studies are then conducted in relatively larger numbers of
`subjects. Phase IIA trials are designed to assess the dosing
`in patients and serve as proof-of-concept studies. In
`phase IIB trials, the efficacy of the drug is determined at
`specific prescribed doses. Definitive evaluation of efficacy
`is determined in phase III studies, which are multicenter,
`randomized, controlled trials in large numbers of pa-
`tients. To date, few studies have collected biologic sam-
`ples (or other measurements) for ancillary evaluations of
`
`efficacy in IBS. This is a vestige of the belief that IBS is a
`disorder of function with no valid biologic marker. How-
`ever, evidence with physiologic (eg, transit), biochemical
`(eg, serum or other markers of immune activation), and
`even pharmacogenetic modulation suggests that there is
`a need to reassess the optimal drug development path.
`
`Current Considerations in Study Design
`IBS has no “gold standard” of treatment, so can-
`didate drugs are usually compared with placebo. The
`study protocol specifies all end points that will be mea-
`sured, including each domain score that is targeted to
`support a specific claim.92 Drug approval by regulatory
`agencies is based on achieving the primary end point in
`phase III trials.
`The Rome III guideline on design of trials for func-
`tional GI disorders recommended the use of validated
`instruments as primary outcome assessment tools in IBS
`clinical trials.93 Secondary end points in clinical trials
`support or explain the results of the primary outcome
`analysis (particularly if a global end point or composite
`score is used). Improvements in secondary end points can
`help to characterize the response to a global end point
`because they represent the multiple manifestations of the
`global or multidomain measure.94 Primary end points
`have been described as what is of interest to patients,
`whereas secondary variables are generally of interest to
`clinical researchers.95
`In the Patient-Reported Outcome (PRO) Guidance
`Document, released in 2006, the US Food and Drug
`Administration (FDA) mandated that outcome measures
`for clinical studies be validated.92 The process for devel-
`oping a new PRO instrument or modifying an existing
`instrument is shown in Figure 1. The starting point in
`developing a valid and meaningful outcome measure is
`to establish a conceptual framework (ie, a path diagram)
`for IBS. This framework can be developed using patient-
`reported information to characterize the full disease ex-
`perience, factors related to severity, impact on daily ac-
`tivities, and treatment response. In addition, published
`studies in well-characterized IBS patients that have ad-
`dressed disease mechanisms or treatment response can
`help establish a multidimensional conceptual frame-
`work. It is recommended that this framework guide the
`development and measurement of valid, reliable, and
`reproducible patient reported end points and objective
`biomarkers. This is followed by creation or modification
`of the instrument including the generation of items;
`choice of the data collection method; choice of the recall
`period; choice of response options (eg, visual analog
`scale, Likert scale, numeric rating scale, checklist of bi-
`nary end points); assessment that patients understand
`the instrument; development of format, instructions, and
`training of those collecting the instrument data; identi-
`fication of scoring of items and domains; and assessment
`of respondent and administrator burden.
`
`BASIC AND CLINICAL
`GASTROENTEROLOGY
`
`REVIEWS IN
`
`

`

`GASTROENTEROLOGY
`BASICANDCLINICAL
`
`REVIEWSIN
`
`December 2008
`
`CHALLENGES TO THE IBS THERAPEUTIC PIPELINE 1881
`
`Figure 1. The process recommended by the PRO guidance document for developing of new or for modifying existing instruments for clinical trials
`(reproduced from US Department of Health and Human Services FDA Center for Drug Evaluation and Research; US Department of Health and
`Human Services FDA Center for Biologics Evaluation and Research; US Department of Health and Human Services FDA Center for Devices and
`Radiological Health92).
`
`Biomarkers Used in IBS Treatment Studies
`and Their Validity
`A number of physiologic outcome measures to
`assess treatment responses have been studied in IBS.
`These include measures of visceral perception (eg, rectal
`or colonic pain thresholds and perceptual ratings) and
`intestinal transit (eg, orocecal and colonic transit times).
`A recent review of the literature determined that the
`correlations between biomarkers obtained in preclinical
`and clinical models and respective symptoms are rela-
`tively small, and the ability to predict drug effectiveness
`for specific as well as for global IBS symptoms is limit-
`ed.91 On the other hand, colonic transit measurements
`correctly predict the effects of agents on bowel function
`and are generally associated with global, binary end
`points such as adequate or satisfactory relief of IBS pain
`and discomfort in patients with constipation- or diar-
`rhea-predominant IBS.96,97
`GI and colonic transit. Transit is a clinically rele-
`vant and noninvasive measure of GI function, primarily
`
`related to motility and secretion. Colonic transit time has
`been shown to correlate with stool form, as measured by
`the Bristol Stool Form Scale.98 It can be measured by
`different techniques including radiopaque markers99,100
`and breath hydrogen tests.101
`However, the most robust and consistent results for de-
`tailed GI transit measurements have been reported with
`scintigraphy, which allows for regional transit assessments.
`Colonic transit is accelerated in IBS-diarrhea (IBS-D) pre-
`dominant patients, compared with healthy individuals, and
`those with IBS-constipation (IBS-C) predominant range
`from normal to slow transit times.102 Approximately 35% of
`patients with IBS have abnormal overall colonic transit,
`including 48% of those with IBS-D.
`The effect of medications that affect GI transit time
`has been studied in patients with IBS. These include
`bulking agents,103,104 cimetropium bromide,105 imipra-
`mine,101 alosetron,106,107 tegaserod,108 renzapride,109,110
`the guanylate cylase-C agonist linaclotide,73 and the pro-
`biotic combination VSL#3.80 Correlations between effects
`
`

`

`1882 CAMILLERI AND CHANG
`
`GASTROENTEROLOGY Vol. 135, No. 6
`
`of medications on regional or total colonic transit and
`stool number or stool form have been shown for a num-
`ber of drugs, including 5-HT4 agonists such as prucalo-
`pride6 and tegaserod,108 the combined 5HT4 agonist-
`5HT3 antagonist renzapride,109 the guanylate cyclase-C
`agonist linaclotide,73 and the chloride channel activator
`lubiprostone.63
`Importantly, the results from phase IB or IIA studies
`using scintigraphic colonic transit have correctly pre-
`dicted efficacy of treatments in phase IIB or phase III
`studies in IBS with the following medications: alosetron,
`tegaserod, prucalopride, renzapride, linaclotide, and lu-
`biprostone. For example, the efficacy of lubiprostone63 in
`improving colonic transit and stool form in pharmaco-
`dynamic studies in healthy volunteers accurately pre-
`dicted clinical efficacy in patients with IBS-C.66 At
`present, the only discrepancy between results of scinti-
`graphic transit111 and phase IIB clinical trial data per-
`tains to the drug solabegron (preliminary data, in press);
`results of larger phase II trials are eagerly awaited. None-
`theless, GI transit assessment can be a useful biomarker
`for predicting the effects of a test on bowel habits and
`overall response in patients with IBS.
`Visceral hypersensitivity. Twenty percent to 94%
`of IBS patients are more sensitive than healthy individ-
`uals to balloon distention of the colon or rectum via
`barostat, a computerized distention device used to mea-
`sure visceral compliance, tone, and perception.91 Variabil-
`ity in the prevalence of visceral hypersensitivity across IBS
`patients (reviewed in Camilleri et al102) suggests that level
`of rectal sensitivity is not a biologic marker of IBS, as was
`previously claimed.112 A number of factors affect visceral
`perception: gender,113 predominance of bowel dysfunc-
`tion,114,115 psychologic factors,116,117 and specifics of the
`experimental study protocol, such as inclusion of data on
`stress118,119 or food ingestion.120,121 Whereas levels of
`pain and bloating have high correlation values with mea-
`sures of visceral hypersensitivity,122 perceptual thresholds
`and ratings have only moderate correlation values with
`IBS symptom severity or response to treatment.91 It is
`possible that, with more effective visceral analgesics, test-
`ing of visceral sensitivity would have utility similar to
`that of transit measurement in predicting efficacy of
`drugs affecting intestinal transit.
`Barostat testing has been used in patients with IBS to
`measure the changes in visceral sensitivity to opiates,
`serotonergic agents, and antidepressants. Small crossover
`studies support the validity of the barostat test as a valid
`human experimental model for visceral pain, especially
`with the ␮-opioid agonist fentanyl and, to a lesser extent,
`the ␬-opioid agonist fedotozine.29,32 However, visceral
`barostat testing might not predict the effect of a candi-
`date drug on IBS symptoms.17,91 The barostat studies
`that correctly identified clinical efficacy of alosetron uti-
`lized brain imaging.123
`
`Brain imaging. Neuroimaging techniques such as
`positron emission tomography or functional magnetic
`resonance imaging show that brain activation patterns
`elicited in anticipation of a noxious distention and/or
`actual intestinal distention are different in IBS patients
`than controls.112,124 –126 The mechanisms proposed to ex-
`plain an altered brain response in patients with IBS are
`unclear but may include increased signaling from the gut
`or amplification of a normal signal during transmission
`from the gut or in the brain.127 Brain activation patterns
`can be influenced by gender,128,129 bowel function,130
`psychologic symptoms,131 and adaptation to experimen-
`tal conditions over time.132
`There are limited data evaluating brain activation re-
`sponses to efficacious IBS treatments. In a H215O positron
`
`emission tomography study performed in nonconstipated
`IBS patients, 3 weeks of treatment with the 5-HT3 antago-
`nist alosetron improved IBS symptoms and was associated
`with reduced regional cerebral blood flow in limbic and
`paralimbic regions and central autonomic circuits (which
`contain 5-HT3 receptors), particularly during anticipation
`of distention.123 In the alosetron-treated patients, lower
`activity in bilateral orbitofrontal cortex and medial tempo-
`ral gyrus during pretreatment scans suggested changes in
`regional cerebral blood flow and psychologic symptoms
`predicted treatment response.123,133
`In a functional magnetic resonance imaging study134
`conducted in 19 women with painful IBS, amitriptyline was
`associated with decreased activation in the rostral anterior
`cingulate cortex and the left posterior parietal cortex during
`the combined rectal distention and psychologic stress but
`not during the relaxation condition. There was no reported
`association of brain responses with symptom ratings in this
`small study. Lackner et al135 measured brain responses us-
`ing H2
`15O positron emission tomography in 6 women with
`IBS before and after treatment with cognitive behavioral
`therapy and found a reduction in regional cerebral blood
`flow in the parahippocampal gyrus and the ventral portion
`of the anterior cingulate cortex. These reductions correlated
`with improvement in GI symptoms and psychologic func-
`tioning. Thus, IBS treatments exert beneficial effects via
`brain regions concerned with pain modulation and auto-
`nomic responses; neuroimaging is a promising modality for
`proof-of-concept studies.
`An average of 50% of patients with IBS experience
`abnormal visceral hypersensitivity or abnormal transit,
`and approximately 20% have evidence of both pathophys-
`iologies.102 Further research in central processing and its
`modulation with therapy is necessary to understand its
`utility in drug development.
`
`IBS Trial End Points and Regulatory
`Insights
`Almost a decade ago, the path to successful devel-
`opment of therapeutics for IBS seemed fairly clear: ap-
`propriate trial design, clear pharmacologic efficacy,
`
`BASIC AND CLINICAL
`GASTROENTEROLOGY
`
`REVIEWS IN
`
`

`

`GASTROENTEROLOGY
`BASICANDCLINICAL
`
`REVIEWSIN
`
`December 2008
`
`CHALLENGES TO THE IBS THERAPEUTIC PIPELINE 1883
`
`safety, and use of binary end points that assessed the
`global effects of medications on a global improvement
`scale.54 Academics, clinical trial designers, biostatisti-
`cians, and the FDA had agreed on this path; there was a
`significant pipeline of medications with potential for
`treating patients with IBS and other bowel dysfunction.
`With the withdrawal of cisapride, alosetron, and tega-
`serod and lack of approval of cilansetron for alleged
`safety concerns, the risk-benefit ratio of new drugs for
`IBS was reexamined, and the psychometric validation of
`end points used in previous clinical trials was questioned.
`The need for psychometric validation of patient response
`outcomes has been increasingly emphasized by regula-
`tory agencies to be consistent with standards set for other
`disorders characterized predominantly by patient re-
`sponses rather than objective biomarkers. It is assumed
`that application of principles and validation of end
`points used in chronic pain, fibromyalgia, or cancer pain
`will prove effective in IBS. The clinical trial experience
`and responsiveness observed in almost 20,000 IBS pa-
`tients with at least 5 different medications (alosetron,
`cilansetron,
`tegaserod,
`lubiprostone, dextofisopam)
`tested with binary or global improvement end points
`have been devalued on the grounds that some aspects of
`psychometric validation are not available.
`
`What End Points Were Used in Clinical
`Trials in the Past Decade?
`Table 3 summarizes end points used in clinical
`trials involving at least 100 patients with IBS. There is
`compelling evidence in favor of the binary or global end
`points used in the past 10 years and involving 20,000
`people studied with 5 different drugs. First, these trials
`showed that the end points are responsive: a metaanalysis
`of 6000 patients in clinical trials of 5-HT3 antagonists
`with several different outcomes/end points showed that
`global end points can document responsiveness to this
`class of medication.97 Second, participants in trials were
`able to relate the level of relief to the level of their
`symptoms, determined at baseline, and provided consis-
`tent responses. In the trial of alosetron, 80% of patients
`reported consistent
`responses in subsequent weeks.
`Third, the performance measures (validity and reliability)
`were independently assessed and found to be acceptable;
`binary or global response end points159 and global im-
`provement scale160 and responses were typically driven by
`the patients’ most bothersome symptoms.159
`A review of the validation of the diverse end points
`(Table 4) on the basis of 12 prespecified criteria con-
`cluded that the weight of evidence is in favor of adequate
`relief,161 confirmed by an independent group.162 The
`7-point Likert global improvement scale has also been
`validated for clinical, satisfaction, and productivity sig-
`nificance,160 and responsiveness of the global improve-
`ment scale in several trials with alosetron and cilansetron
`has been reviewed.97 Fourth, drugs that are effective,
`
`based on the binary or global response end points, were
`also found to improve general or disease-specific quality
`of life.163
`The Rome III guidance on clinical trials93 has endorsed
`use of a global measure that integrates the symptom data
`i