|Home | About | Journals | Submit | Contact Us | Français|
Irritable bowel syndrome (IBS) is a functional gastrointestinal disorder affecting up to 3-15% of the general population in western countries. It is characterized by unexplained abdominal pain, discomfort, and bloating in association with altered bowel habits. The pathophysiology of IBS is multifactorial involving disturbances of the brain-gut-axis. The pathophysiology provides the rationale for pharmacotherapy: abnormal gastrointestinal motor functions, visceral hypersensitivity, psychosocial factors, autonomic dysfunction, and mucosal immune activation. Understanding the mechanisms, and their mediators or modulators including neurotransmitters and receptors have led to several therapeutic approaches including agents acting on the serotonin receptor or serotonin transporter system, antidepressants, novel selective anticholinergics, α-adrenergic agonists, opioid agents, cholecystokinin-antagonists, neurokinin-antagonists, somatostatin receptor agonists, corticotropin releasing factor antagonists, chloride-channel activators, guanylate-cyclase-c agonists, melatonin, atypical benzodiazepines, antibiotics, immune modulators and probiotics. The mechanisms and current evidence regarding efficacy of these agents are reviewed.
Irritable bowel syndrome (IBS) is one of the most common functional gastrointestinal disorders (1,2) characterized by abdominal pain and discomfort in association with altered bowel habits; symptoms are not explained by structural abnormalities using current standard diagnostic tests (3). The pathophysiology of IBS is still not well understood, but is most likely multifactorial. Several factors such as motor and sensory dysfunction, neuroimmune mechanisms, psychological factors, and changes in the intraluminal milieu appear to play a role [Figure 1 (3,4)]. The increased release of serotonin in the circulation especially in the D-IBS group (5-7) and increased serine proteases (derived from mast cells) in stool of patients with IBS (8) provide evidence for the potential role of neurotransmitters or chemical mediators such as proteases in the disorder.
Changes in mucosal serotonin levels may affect sensory and motor functions possibly contributing to IBS symptoms (9). There is increasing evidence of mucosal dysfunction, immune activation or inflammation (10-12) and, possibly, alterations in colonic bacterial flora. Increased number of activated mast-cells in the proximity of colonic nerves in the lamina propria is associated with abdominal pain severity (13). Mucosal immune activation is associated with systemic evidence for a pro-inflammatory state (decreased IL10/IL12 ratios) in IBS patients (14) and changes in local defense mechanisms in the sigmoid and colonic mucosa in IBS (15). These provide novel promising targets for future therapies. Traditional IBS therapies (Figure 1) are mainly directed at the relief of individual symptoms, e.g. antidiarrheals for diarrhea, laxatives for constipation or smooth muscle relaxants for pain. They are often of limited efficacy in addressing the overall symptom complex.
This article reviews recent and novel approaches in IBS therapy (Table I, Figure 2) that have preclinical efficacy, Phase IIa studies of pharmacodynamics or proof of concept, and Phase IIb or phase III studies that show evidence of efficacy.
Serotonin (5-hydroxytryptamine) is an important neurotransmitter of the enteric nervous system [ENS (16,17)], the peristaltic reflex (chiefly 5-HT4 receptor), the brain-gut axis and vagal and visceral afferent cell bodies or pathways, chiefly 5-HT3 receptors (16-18). 5-HT plays a pivotal role in the modulation of multiple gut functions such as motility, sensation, blood flow and secretion (19). In humans, the highest level of specific binding has been found in the amygdalae, which are integral to the emotional responses to visceral stimulation.
Activity at 5-HT receptors is regulated by the 5-HT reuptake transporter protein [SERT (20)]. It is controversial whether there is reduced expression of SERT (mRNA or immunohistochemistry) in IBS (21,22).
5-HT3 antagonists and 5-HT4 agonists have demonstrated efficacy and effectiveness in the treatment of multiple or global symptoms of IBS. The prototype medications were alosetron and tegaserod respectively.
5-HT3-receptor antagonists retard colonic transit, reduce secretion, and increase colonic compliance in response to distension (23) and have central effects that contribute to their beneficial effects on sensation in IBS (24).
Alosetron and subsequently cilansetron were efficacious in several large, multicenter, randomized controlled clinical trials, reviewed in an updated meta-analysis (25). 5-HT3 antagonists more effective than the comparators in achieving global improvement in IBS symptoms (pooled relative risk, 1.60; 95% confidence interval [CI], 1.49-1.72; heterogeneity index, I2 = 0%) and relief of abdominal pain and discomfort (pooled relative risk, 1.30; 95% CI, 1.22-1.39; I2 = 22%). Benefit was apparent for both agents, in both genders. These agents were more likely to cause constipation (pooled relative risk, 4.28; 95% CI, 3.28-5.60, I2= 65%); 0.2% using 5-HT3 antagonists had possible ischemic colitis versus none in control groups. Cilansetron was never approved; alosetron is associated with clear benefits on quality of life (26), but warnings about potential lack of safety has resulted in only a minority of patients being prescribed the medication.
The 5-HT3 agonist, MKC-733, also named DDP733 and pumosetrag in the literature, stimulated fasting antroduodenal migrating motor complex activity and accelerated small intestinal transit (27). A preliminary report of a phase 2a study in IBS-C showed a significant overall benefit in overall clinical response rate (54%) in patients receiving 1.4 mg t.i.d. versus to a 15% for placebo (28).
5HT4-receptor agonists generally accelerate small bowel and colonic transit.
Tegaserod is a selective partial agonist of the 5-HT4-receptor. Several trials in IBS and chronic constipation were summarized in a Cochrane meta-analysis (29). In IBS-C, tegaserod did not significantly improve the patients’ individual symptoms of abdominal pain and discomfort although bowel habit showed a statistically significant improvement with tegaserod 4 mg and there was a non-significant trend in this outcome in favor of tegaserod, 12 mg. In patients with chronic constipation, the RR of being a responder in terms of complete spontaneous bowel movements per week with tegaserod 12 mg was 1.54 (95% CI 1.35 to 1.75). The drug was essentially withdrawn because of possible, rare cardiovascular adverse effects.
ATI-7505 is a novel, potent agonist of the 5-HT4 receptor and is chemically related to the 5-HT4-receptor agonist, cisapride. However, it has been chemically designed to eliminate cardiac liabilities (e.g., QT prolongation, tachycardia) and CYP450-dependent metabolism at therapeutically relevant concentrations. Pharmacodynamic data in healthy human volunteers (30) indicate prokinetic effects with acceleration of gastric emptying (20 mg tid) and colonic transit (10 mg tid; overall and ascending colon emptying), as well as inducing looser stool consistency. Thus, it appears to be promising for C-IBS. Extensive human safety data are not available. To date, there are no reports of adverse cardiac side effects.
Prucalopride is a 5-HT4-receptor agonist that has been extensively studied, including pharmacodynamics and large clinical trials in patients with severe chronic constipation (31-34). The medication has greater selectivity (~150-fold) for 5-HT4-receptors than for hERG channel, and it is anticipated that this will be sufficient to prove safe from a cardiac arrhythmia perspective. Certainly, administration for approximately 1 year in more than 80 people with cardiovascular co-morbidity proved safe (35).
TD-5108 (or velusetrag) is a potent agonist at human 5-HT4 receptors with high intrinsic activity that displays preferential binding to the 5-HT4 receptor compared with other 5-HT receptor subtypes or hERG channel (36). In vivo studies have shown that TD-5108 increases colonic transit in guinea pigs, relaxes esophageal body musculature in rats, and increases contractility in multiple regions of the GI tract including the gastric antrum, duodenum, jejunum and proximal colon in conscious dogs with implanted strain gauges (37). In healthy volunteers, single dose, 30- and 50-mg, TD-5108 accelerated colonic and small bowel transit; with multiple doses, TD-5108 30-mg also accelerated colonic transit (38). A 400 patient, randomized, controlled trial demonstrated efficacy on bowel frequency and consistency in chronic constipation (39).
Many patients with IBS receive psychoactive agents for their co-morbid psychiatric illnesses, including anxiety, mood, and somatoform disorders, and for the potential effects on visceral sensation. Some SSRIs accelerate small bowel transit; others have effects on colorectal sensation, compliance or tone [e.g. citalopram, venlafaxine (40-42)].
There may also be racial or ethnic differences, as efficacy appears greatest in trials conducted in Iran (43,44). In intention-to-treat analysis, complete responses were significantly more common in the amitriptyline group. Efficacy of antidepressants in treatment of IBS has been appraised in a recent meta-analysis (45). At first evaluation, one might conclude that this class of drugs is extremely effective. The calculated relative risk of IBS symptoms persisting or remaining unimproved after treatment with antidepressant therapy versus placebo was 0.66 (95% CI 0.57 to 0.78), with the NNT with antidepressant therapy to prevent IBS symptoms persisting of 4 (95% CI 3 to 6). However, it is worth noting that very few (typically small single center) trials individually show significance, and the meta-analysis involves diverse medications and doses, study designs, endpoints, questionable generalizability, and small studies. The analysis included 13 studies comparing antidepressants (total n=432) to placebo (n=357) for IBS; 7 studies were in secondary, and 6 in tertiary care with none in primary or community cohorts. Moreover, there was marginal statistically significant heterogeneity detected between studies (I2 = 26.4%, P = 0.17) and funnel plot asymmetry (P = 0.02), suggestive of publication bias.
In the meta-analysis, only three papers are associated with clear efficacy and they involve small samples of patients: two in Iranian patients (43,44) and one in German patients (46). Intriguingly, the most efficacious treatment attributed to a tricyclic agent was reported with a trial of amitryptiline, 10 mg, that did not significantly affect abdominal pain, bloating or diarrhea, relative to placebo. Similarly, the trial of the SSRI, fluoxetine, (43) in 44 patients in Iran shows very large differences in patients with significant abdominal discomfort and bloating, but the proportion of placebo responders over 12 weeks was uncharacteristically low (<15% average), and it is unclear whether these results are generalizable.
Gastrointestinal effects of acetylcholine are mediated by nicotinic receptors in the myenteric plexus, and by muscarinic receptor subtypes M1-3, in the myenteric plexus, as well as the neuromuscular junction (47,48). Acetylcholine is the main excitatory neurotransmitter in the gastrointestinal (GI) tract. Nonselective anticholinergics or specific antimuscarinic agents reduce bowel motility and associated pain. Relaxant drugs act directly on the smooth muscle e.g. by blocking smooth muscle Ca 2+ channels. Several compounds (mebeverine, otilinium, pinaverium, and cimetropium), used in Europe, have never been approved in the U.S. Several meta-analyses suggest global improvement with this class of agents (49,50), with a mean odds ratio of 2.13 [95% CI: 1.77-2.58]. The low quality (Jadad Scores) of the individual clinical trials compromises the overall interpretation of efficacy of this drug class.
Newer anticholinergics specifically targeting the muscarinic type-3 receptor (M3) on smooth muscle decrease non-specific anticholinergic side effects (dry mouth or increased heart rate) and have promising gastrointestinal effects in animal models (51). Among M3-selective antagonists, darifenacin (used to treat overactive bladder) is associated with constipation (52), and zamifenacin (53) significantly reduced colonic motility without other anticholinergic effects in 36 IBS patients.
The adrenergic nervous system provides extrinsic tonic inhibitory control of non-sphincteric gut motility. The α2-adrenergic agents such as clonidine affect human colonic and rectal motor and sensory functions (54-56), increasing colonic compliance, reducing fasting tone without altering the colonic response to a meal or colonic transit, and significantly reducing sensation within a dose range of 0.1 to 0.3 mg bid. In a study of 44 patients with D-IBS (57), 67% of the patients treated with clonidine 0.1.mg twice daily compared to 46% in the placebo group achieved satisfactory relief, the primary endpoint. Clonidine also significantly improved bowel functions without altering gastrointestinal transit. Drowsiness, dizziness and dry mouth were observed at doses >0.1 mg bid. This suggests benefits of α2-adrenergic agents in IBS.
Gabapentin, a 3-alkylated analogue of γ-amino butyric acid (GABA), has been shown to reduce elements of central sensitization in human experimental hyperalgesia. Forty patients with D-IBS were randomized for 5-day treatment with gabapentin, 300 mg/day and then 600 mg/day, which reduced rectal sensory thresholds through attenuating rectal sensitivity to distension and enhancing rectal compliance (58).
Preclinical studies suggest that the α2δ ligand, pregabalin, reduces both visceral allodynia and hyperalgesia, but is inactive on basal sensitivity. The perception of rectal distension was tested in 26 hypersensitive IBS patients randomized to increasing doses of oral pregabalin for 3 weeks or placebo control. Pregabalin significantly increased the sensory thresholds from baseline for first sensation desire to defecate and pain compared with placebo; it also increased rectal compliance, an effect unrelated to the changes in sensitivity (59). Studies of GABA-ergic agents and α2δ ligands on clinical outcomes in IBS are not available.
Three major opioid receptors, μ, δ, and κ, are distributed in the peripheral and central nervous systems. μ-agonists modulate visceral nociception and to slow down gastrointestinal transit resulting in constipation or central side effects (60-62). κ-opioid receptors are involved in visceral perception (63).
The κ-opioid agonist, asimadoline (64), was shown to reduce sensation in response to distensions in the non-noxious range applied to the colon and to relax colonic and gastric tone during fasting. There were no significant effects on gastrointestinal or colonic transit or postprandial tone response to meal ingestion. Delvaux et al. (65) showed that this drug increased sensory thresholds in patients with IBS. On demand treatment of IBS pain with asimadoline was not significantly better than placebo (66); however, a recent trial using daily administration of asimadoline demonstrated efficacy in IBS-D and IBS-M patients (67).
Alvimopan, a novel, peripherally-restricted μ is effective in the treatment of opioid-induced constipation and postoperative ileus (68-70). It reverses the peripheral effects of narcotics without influencing the pain relief desired by concomitant opioid administration. A pharmacodynamic study in healthy volunteers (71) confirmed that alvimopan normalized the colonic transit delay induced by co-administered codeine. Interestingly, alvimopan alone accelerated colonic transit, suggesting that the μ-opiate mechanisms participate in the physiological control of colonic transit (71).
Oral naltrexone does not accelerate colonic transit (72), but oral methylnaltrexone accelerated gut transit in opiate-naïve controls (73) and it reversed the prolongation of orocecal transit induced by long-term methadone (74). Trials to test the effects of peripheral μ-opioid antagonists such as alvimopan and methylnaltrexone in patients with constipation, C-IBS, and opiate-induced constipation are warranted.
Corticotropin releasing factor (CRF) is a key mediator of stress response in the brain gut axis (75). Two CRF receptor subtypes have been cloned: CRF1 and CRF2. In animals, stress activation of the CRF receptors alters gastrointestinal motility (76), effects that can be blocked with selective CRF1-receptor antagonists [antalarmin (77)] and the nonselective CRF-receptor antagonist α-helical CRF 9-41 [αhCRF (78)]. The increased colonic motility response to rectal electrical stimulation in IBS patients was significantly suppressed after αhCRF infusion (79). However, the CRF1 antagonist, pexacerafont, was not effective in normalizing colonic transit or bowel function in patients with diarrhea-predominant IBS (80). The role of CRF modulation requires further study.
Dextofisopam (81) is the R-enantiomer of the nonsedating homophthalazine anxiolytic, tofisopam. It binds to 2,3-benzodiazepine receptors located in subcortical and hypothalamic regions. In animal models, dextofisopam reduced stimulation-induced colonic motility and visceral sensitivity (81). A 12-week, placebo-controlled, phase II study in patients with diarrhea-predominant or alternating-IBS showed more months with adequate relief in the dextofisopam compared to the placebo group (82). The effects were most prominent within the first 4 weeks of treatment, and also included an improvement of stool consistency and a reduction of stool frequency. The drug was well tolerated, with only 3% reporting constipation; however, 12% of the patients experienced a worsening of abdominal pain (versus 4% in the placebo group).
Melatonin is a pineal gland neurohormone involved in the regulation of the sleep-wake cycle; it is also synthesized in the gastrointestinal tract, and may participate in the regulation of gastrointestinal motility and sensitivity (83), possibly by blocking nicotinic channels (84) or Ca2+ activated K+ channels (85). Two small studies of the effects of melatonin in IBS patients (86,87) reported improvement of abdominal pain and IBS symptom score or a reduction of rectal pain sensitivity. This appears to be a peripheral effect, since there was no alteration of sleep or anxiety or depression. The role of melatonin modulation in IBS requires further study.
Cholecystokinin [CCK (88)], a neuropeptide released by endocrine cells within the duodenal and jejunal mucosa in response to a variety of nutrients, has been proposed as a mediator of IBS symptoms. Colonic responses to CCK in vivo and in vitro were increased in IBS patients (89). The effects of CCK are mediated by two distinct receptors, CCK1 and CCK2 (90,91), which are located predominantly in the periphery and the central nervous system respectively. The CCK1 (also called CCKA) receptors are present in gastrointestinal tract smooth muscles and vagal afferents. Blockade of CCK1 receptors in the gastrointestinal tract was proposed as an approach to stimulate gut motility and to change colonic transit time in patients with C-IBS.
Loxiglumide is a highly specific, competitive antagonist of the CCK1 receptor (92). Earlier pharmacodynamic (93) and clinical (94) studies suggested that loxiglumide might improve constipation. A twelve-week, phase II trial of 200 mg/day dexloxiglumide (95) in female IBS-C patients showed improvement in abdominal pain and discomfort compared to placebo, but two large phase III trials in IBS-C showed no efficacy for dexloxiglumide (96). The latter result was consistent with a pharmacodynamics study in C-IBS (97) which showed no significant effect on overall colonic transit or satisfactory relief of IBS. However, in a phase III randomized withdrawal study, dexloxiglumide demonstrated sustained relief of symptoms in female (not male) patients (98). Further studies with this mediation are awaited.
The somatostatin analog, octreotide, activates predominantly somatostatin type 2 receptors, and reduces gastrointestinal secretion, retards gastrointestinal transit and has antinociceptive properties. Octreotide retards orocecal and small bowel transit time in patients with D-IBS (99,100) and normalizes visceral perception and discomfort thresholds in IBS patients without changing rectal compliance (101,102). There are no clinical trials with octreotide in IBS.
NK1-receptors play a role in nociception (103), whereas NK2 receptors have a greater influence on smooth muscle contractility (103-105) and of NK2-antagonists on gut motility (105). In a small pharmacodynamic study of IBS patients, the NK1-receptor antagonist, ezlozipant (106), reduced the emotional response to rectosigmoid distension but did not significantly decrease rectal sensitivity. In healthy controls, the NK2-receptor antagonist, nepadutant (107), reduced contraction frequency and amplitude on migrating motor complexes in the small intestine and effectively antagonized the motility-stimulating effects of infused NKA. The NK3 antagonist, talnetant, was tested in pharmacodynamics (108) and clinical trials (109), and proved ineffective.
Intestinal Cl- secretion is critical for intestinal fluid and electrolyte transport. In the gastrointestinal tract, chloride channels type 2, ClC2, has been found in gastric parietal cells, small intestinal and colonic epithelia. Lubiprostone is a prostone that acts as a selective ClC2 activator and increases intestinal water secretion (110). In a pharmacodynamic study, lubiprostone was shown to accelerate small bowel and colonic transit time in healthy volunteers (111). In several clinical trials, lubiprostone, 24 μg b.i.d., had positive effects on stool consistency, frequency and straining, and was safe and effective for treating constipation (112-114). Interestingly, lubiprostone accelerated colonic transit without accelerating ascending colon emptying, suggesting it affects distal colonic function. Lubisprostone also reduced abdominal pain and improved bowel dysfunction in IBS-C (115,116). Side effects of lubiprostone include diarrhea and nausea that is usually mild and transient. In January 2006, the FDA approved lubiprostone.
Linaclotide is a novel agonist of the human guanylate cyclase-c (GC-C), a transmembrane protein located in the gut epithelium. Activation of the GC-C induces secretion of fluid, sodium and bicarbonate in the intestinal lumen (117-120). In animal and early human safety studies, linaclotide has been observed to increase stool frequency, decrease stool consistency and decrease visceral pain (121). A phase IIA pilot, randomized, controlled study shows efficacy of this drug in treatment of chronic constipation (122). Therefore, the drug seems to be promising for the treatment of C-IBS and chronic constipation.
Crofelemer reduces excess chloride ion secretion via the CFTR channel. Crofelemer does not affect gut motility and is not absorbed systemically to any significant level. It has been used in the past to treat diarrhea associated with intestinal secretion. In a dose-ranging study of crofelemer versus placebo b.i.d. for 12 weeks in 242 IBS-D patients (123), crofelemer did not produce improvement in the primary endpoint related to bowel function, urgency or adequate relief. On the other hand, female IBS-D patients exposed to the highest dose of 500 mg b.i.d. crofelemer had a higher proportion of pain and discomfort free days. These post-hoc observations suggest further studies are indicated to assess the visceral analgesic potential of crofelemer.
Probiotics may have beneficial effects on altered colonic inflammation or barrier function that may be found in some IBS patients. The precise mechanism of action of probiotics is unclear and change of intraluminal milieu and modification of fermentation processes and gas production, inactivation of bile acids with decreased effect of endogenous bile acids on colonic fluid secretion and motility, and alteration of gastrointestinal motility may also contribute to the symptom improvement (124). A recent meta-analysis (125) identified 19 randomized, controlled trials (18 papers) in 1650 IBS patients. Trial quality was generally good. Among 10 randomized, controlled trials involving 918 patients providing outcomes as a dichotomous variable, probiotics were significantly better than placebo with an NNT = 4 (95% CI = 3 to 12.5); however, there was significant heterogeneity (I2 = 68%, p=0.001) and possible funnel plot asymmetry. Among 15 trials assessing 1351 patients that reported an improvement in IBS score as a continuous outcome, the standard mean difference was -0.34 (95% CI -0.60 to -0.07); however, there was statistically significant heterogeneity (I2 = 79%, p=0.001). These data suggest overall benefit for probiotics, but the precision of the NNT is uncertain. The safety profile of probiotics suggests they may be considered for the treatment of bloating and flatulence.
The role of bacterial overgrowth of the small intestine (SIBO) in IBS is controversial. A placebo-controlled study (126) showed that antibiotic treatment with neomycin normalized the lactulose hydrogen breath test (LBT), and this was associated with significant reduction of IBS symptoms over the short term (~1 week). More convincing evidence of the efficacy of antibiotics is provided by the randomized, controlled trials (127) and a preliminary report from Lembo et al. (128). Although there are many unanswered questions, regarding long-term effects, and frequency of retreatment, some patients may benefit from rifaximin treatment.
Sodium Cromoglycate is a mast cell stabilizer; there is experimental evidence of increased mast cell infiltration and greater proximity of mast cells to intramural nerve endings. Proteases which may arise from mast cells have been associated with activation of visceral algesia in experimental studies. Sodium cromoglycate alone or in association with dietary exclusion of suspected allergens has been tested in trials conducted in IBS-D (129) and childhood IBS (130).
Given the efficacy of budesonide in microscopic or collagenous colitis (131-133) and the increasing evidence of immune activation in IBS, we anticipate more formal trials of these agents, especially in post-infectious IBS. In a recent meta-analysis, Chande et al. (133) report that, among 3 trials with budesonide for collagenous colitis, the odds ratio for inducing a response was 12.32 (95% CI 5.53-27.46), maintaining response was OR 8.82, and NNT of 2. Similar efficacy was observed for trials that included both collagenous and lymphocytic colitis (NNT of 3). Mesalazine and bismuth compounds were less effective for microscopic colitis. However, a small, pilot clinical trial in IBS suggests that mesalazine reduced total number of immune cells and mast cells and pain severity score, and increased global relief relative to placebo treatment (134). These interesting and provocative data suggest that 5-ASA compounds deserve further study.
The field of IBS therapeutics is vibrant, with greater understanding of neuroenteric mechanisms, effectors and transmitters in the brain-gut axis. These should provide opportunities for development of new therapeutic agents for IBS treatment. For almost all of the drug classes described here, rigorous phase III trials and assessment of drug safety are eagerly awaited.
The study was supported in part by grant RO1-DK54681 (MC) from National Institutes of Health. The excellent secretarial support of Mrs. Cindy Stanislav is gratefully acknowledged.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.