α2δ ligand: a new, smart pill for visceral pain in patients with hypersensitive irritable bowel syndrome?

The heteromultimeric voltage‐sensitive calcium (Ca²⁺) channel is involved in the function of excitable tissues including sensory nerves. It comprises a primary α1 subunit and auxiliary α2δ, β and γ subunits, and the associated α2δ ligand‐binding site (see fig 1). The α1 subunit forms the ion‐conducting pore and serves as voltage sensor. The α2δ subunit consists of the highly N‐glycosylated extracellular α2 protein attached to the membrane‐spanning δ protein that binds divalent metal cations and interacts directly with the α1 subunit. The β subunit is an intracellular protein that mediates modulation of Ca²⁺ channel function by intracellular kinases. The γ subunit is a glycoprotein with four transmembrane segments.¹

Figure 1 Three dimensional reconstruction of the L‐type calcium channel showing α₂δ ligand‐binding site. This figure was published in Trends in Pharmacological Sciences, volume 28; Dooley, Taylor, Donevan et al; Ca²⁺ channel α₂δ ligands: novel modulators of neurotransmission, 75–82; copyright Elsevier, 2007.⁸

Pregabalin is a second‐generation α2δ ligand that is approved in Europe for the treatment of neuropathic pain and epilepsy, and in the US for the management of neuropathic pain associated with diabetic peripheral neuropathy, post‐herpetic neuralgia and as an adjunctive treatment for adults with partial onset seizures. It is 2–10 times more potent and has more predictable pharmacological effects than the prototype α2δ ligand, gabapentin. Although it is structurally related to γ‐aminobutyric acid (GABA), a major inhibitory neurotransmitter in the central nervous system, it is functionally unrelated and inactive at GABA_A, GABA_B or benzodiazapine receptors and is not converted metabolically into GABA or a GABA agonist. Furthermore, clinically effective concentrations of pregabalin have been shown to have no effect on GABA uptake or degradation.²,³

Pregabalin binds potently to the α2δ auxiliary protein associated with voltage‐gated calcium channels,⁴ reducing depolarisation‐induced calcium influx at the nerve terminals and, consequently, the release of several excitatory neurotransmitters including glutamate, norepinephrine, substance P and calcitonin gene‐related peptide (CGRP).

Pregabalin has been shown to be effective in several animal models of inflammatory and neuropathic (somatic) pain. It blocks both thermal and mechanical hyperalgesia induced by inflammatory, surgical and nerve injuries and inhibits both the static and dynamic components of mechanical allodynia induced by streptozocin.⁵ However, there are data suggesting the efficacy of α2δ ligands depends on the underlying nature or the mechanism of the pain.

First, neuropathic pain has both sympathetically maintained pain and sympathetic independent pain. Pregabalin exerts its antiallodynic effect mainly by acting at the spinal cord, and intrathecally administered pregabalin has more potent antiallodynic effects in sympathetically maintained pain. The α2δ subunit might be less involved in cold allodynia associated with sympathetic independent pain.⁶

Second, “sensitised” conditions are prerequisites to observe the effects of α2δ ligands on neurotransmission—for example, pregabalin reduces capsaicin‐evoked substance P and CGRP release from rat spinal cord slices.⁷ Such findings suggest the existence of a novel mechanism whereby α2δ ligands have minimal effects on physiological transmitter release but significantly inhibit sensitised or abnormal release.⁸

The same principle applies in animal models of visceral pain. Pregabalin has been shown to dose‐dependently reduce trinitrobenzene sulfonic acid‐induced colonic allodynia and to suppress lipopolysaccharide‐induced hyperalgesia, measured as a reduction in the amount of abdominal contractility in response to rectal distension in lipopolysaccharide‐treated animals, but to be inactive on basal sensitivity.⁹,¹⁰ However, it is important to note that the analgesic effect of pregabalin was only noted at 30 mg/kg intra‐peritoneally, not at 10 mg/kg intra‐peritoneally, and that the oral dose response suggested greater efficacy with 3 and 10 mg/kg than with 30 mg/kg, which was not effective.¹⁰ Similarly, in the study of Diop et al,⁹ increased sensory thresholds occurred at 60–200 mg/kg subcutaneously, but not with 30 mg/kg.⁹ It is also worth noting that the effects of the drug on colonic compliance were not evaluated in the animal studies that proposed a significant effect on sensation.

Because pregabalin appears to have a broad spectrum of antihyperalgesia activity in diverse animal models, Houghton et al(see p 1218 September issue of Gut) recruited 41 patients with Rome II‐defined irritable bowel syndrome (IBS; 28 female, 13 male); 26 with evidence of rectal hypersensitivity (rectal pain sensation threshold <28 mm Hg) were studied.¹¹ Participants received 21 days' treatment with an ascending dose of pregabalin (up to 200 mg three times daily) or placebo. Barostat studies were compared at baseline and after 3 weeks' treatment. The data suggest that change in sensory thresholds post‐treatment from baseline for first sensation, desire to defecate and pain were all significantly greater with pregabalin versus placebo. It is unclear how doses (30–200 mg/kg parenterally) found to be effective in rats treated with lipopolysaccharide⁹ or trinitrobenzene sulfonic acid colitis¹² relate to the 200 mg three times daily oral dose in humans used in the pharmacodynamic study reported by Houghton et al.¹¹

Interpretation of these human studies is potentially confounded. First, there may be effects of the drug on rectal compliance, which is significantly different (steeper, that is, more compliant) with pregabalin than placebo (see fig 5 of Houghton et al, p 1222). The authors observed similar changes in pain threshold even among those patients with steep pressure–volume relationships. One can also be somewhat reassured by the fact that thresholds were based on pressure distensions, rather than volume‐based distensions which were used in the prior animal studies of rectal sensation.⁹,¹⁰ A weakness of the study is the incomplete assessment of compliance with a limited range of intraballoon pressures (see fig 4 of Houghton et al, p 1222), as the protocol required stopping distensions at the threshold of moderate pain. A second confounder in interpretation of the results is the clear disparity in nervous system adverse effects (especially dizziness in 10, somnolence in 5 and incoordination in 3) with pregabalin relative to placebo. These adverse effects may have unblinded participants in the study, because such adverse effects should have been included in the consent form as potential drug effects. Again, it is somewhat reassuring that, for the 5 participants who developed both dizziness and somnolence, the authors found no difference in pain threshold change from baseline compared with those without these adverse effects.

Given these pitfalls in interpretation, the authors appropriately conclude that the precise mode of action of pregabalin on rectal sensitivity in humans is still to be determined and that further studies are required to explore clinical potential of this class of medications for visceral pain syndromes. This study enriched the patient population for those with demonstrated hypersensitivity. In the view of basic science observations discussed above, it would also be interesting to know whether patients with high sympathetic arousal or those with greater tissue inflammation in association with IBS would have a greater response to pregabalin. Experience with such studies also suggests that statistical analysis comparing sensory endpoints should include covariates of interest such as gender, baseline values, baseline and post‐treatment rectal compliance and, ideally, a validated, on‐line measurement of “central” function (anxiety, fear of pain, somnolence level).

How do these data of the effects of pregabalin on sensation in humans compare with those of other approved or experimental medications in IBS? Comparisons in this discipline are difficult because the methods used, endpoints (thresholds vs sensory ratings vs RIII reflex (the latter being a test of spinal processing of visceral nociception)), definition of thresholds, groups studied (health vs patients with IBS) and concomitant perturbation (eg, lipid infusion or sigmoid painful conditioning distension) differ drastically among the studies. Thus, no direct comparisons are possible among studies. The effects of pregabalin are reported as changes relative to baseline and subtracting effects of placebo. The average differences in sensation thresholds for first sensation (2 mm Hg), desire to defecate (6 mm Hg) and pain (5.4 mm Hg) are consistent with the effects (typically differences in pain threshold of ∼5 mm Hg relative to placebo) reported for alosetron,¹²,¹³ asimadoline and fedotozine¹⁴,¹⁵ and octreotide,¹⁶,¹⁷,¹⁸ and are greater than the effects reported for the NK₃ antagonist, talnetant.¹⁹ However, the effects on discomfort thresholds are lower than those observed with high‐dose (average 18 mm Hg) and comparable to those of low‐dose fentanyl (average 6 mm Hg) on discomfort thresholds.²⁰ High‐dose fentanyl and 0.3 mg dose clonidine,²¹ which induce somnolence and other central effects such as euphoria, also have significant effects on sensory ratings²⁰,²¹ in response to distension. The magnitude of these effects is not comparable with that induced by pregabalin. However, by comparison with the central effects observed by Houghton et al with pregabalin, they emphasise the potential that central antinociception may also contribute to the effects of these agents.

The authors are to be congratulated on a thorough and rigorous study. Their paper illustrates the complexity of such pharmacodynamic studies and the difficulty of extending observations in animal models to proof of principle studies in humans. The data on clinical efficacy presented on 14 patients randomised to pregabalin and 12 to placebo for an average of 3 weeks' treatment are essentially of “curiosity” value. Yet, Houghton et al have provided important insights, such as on dose and patient selection, which enhance the design of phase IIB clinical trials.