Introduction
Over the last three millennia, civilizations have attempted to cure sickness with products derived from animal, plant and microbial origins or from mineral sources [1]. About 80% of the populations of some Asian and African countries are estimated to rely on traditional medicines as sources for their primary healthcare. Annual revenues for herbal medicines in Western Europe reached EUR 3.5 billion in 2003–2004 [2]. Although the underlying use of many of these so-called natural products clearly lacks scientific logic, several of them nevertheless became founding platforms for modern drug development throughout the twentieth century [1,3].
Three classes of analgesic medications are derived from plant ingredients: opiate receptor agonists from poppy seeds, cyclooxygenase inhibitors from willow bark, and capsaicinoids from chili peppers. Capsaicinoid-containing preparations are popular natural medicines for pain syndromes and although the chili peppers from which they are derived (Capsicum spp.) were originally found in the Americas, they are now widespread [4,5]. Capsaicin of various purities and grades has been widely available in pharmacies as low-to-moderate concentration creams and gels [5,6,7]. Increased local perfusion and the resulting warming or capsaicin-induced pain leading to counterirritation were initially thought to account for the analgesic effects of capsaicin [5].
Cloning of the capsaicin receptor, i.e. the transient receptor potential channel of the vanilloid receptor family subtype 1 (TRPV1), provided the molecular basis for understanding the various actions of capsaicin [8]. TRPV1 was discovered by an expression cloning strategy utilizing capsaicin-induced increases in intracellular calcium concentration as a marker. The calcium signal not only indicates opening of a nonselective cation channel and depolarization, but also initiation of receptor desensitization and downregulation, and degeneration of epidermal nerve fibers (ENFs), which is referred to as defunctionalization [5,9,10]. As a result of neurite degeneration, ENFs become less sensitive to a variety of stimuli (including capsaicin itself), leading to reduced pain responses [5,10]. A challenge in capsaicinoid drug development was harnessing the desensitizing action of capsaicin without unnecessarily exposing patients to TRPV1 activation-induced pain.
An early therapeutic approach using a low-dose capsaicin cream formulation to defunctionalize nerve endings in human skin required repeated daily applications over weeks [11]. Although pain arising from such topical capsaicin applications is mild, mucous membranes are highly sensitive to the effects. Thus, each topical exposure to capsaicin can result in irritation of the eyes and respiratory tract [12]. A systematic review of clinical trials using low-dose capsaicin creams described moderate-to-poor efficacy in the treatment of chronic neuropathic pain with a number needed to treat of 5.7 [7]. The suboptimal ratio of defunctionalization to irritation with the low-dose capsaicin creams and the lengthy application procedure limited their widespread use [5,13].
Development of structural analogs of capsaicin that retained its antinociceptive attributes without inducing irritation failed in the late 1980s [5]. Although the capsaicin derivative olvanil was 10 times more potent than capsaicin in TRPV1 activation assays and devoid of pungency, it did not reduce peripheral fiber responsiveness in vitro, possibly because of an alternative mechanism of receptor activation and desensitization [5,14,15].
Initiatives to enhance the clinical effectiveness of capsaicin explored use of high doses (10% weight by volume [w/v] and weight by weight [w/w] formulations), which fully desensitize TRPV1-expressing neurons in human skin within a few days or even upon a single application. These creams were useful to determine the actions of TRPV1 expressing neurons in experimental studies of human volunteers [16,17]. Whereas pain during application is moderate and tolerable in healthy subjects, it may be too strong for patients with sensitization or upregulation of TRPV1. Therapeutic use of high-dose creams was halted because of severe application site-associated pain and the risk of capsaicin contamination to the patient and healthcare professional [16].
Recently, a new dermal patch has been developed containing 8% w/w pure trans-capsaicin in the adhesive layer. This formulation, which enables administration of high-dose capsaicin with a single treatment, is licensed for use in peripheral neuropathic pain in nondiabetic adults in the European Union and for postherpetic neuralgia (PHN) in the USA [18,19]. Pretreating the affected area of skin with a local anesthetic is advised, although doubts have been raised regarding the benefits of such pretreatment [20,21].
The aim of the current review is to discuss aspects of capsaicin pharmacology and advances in this challenging area of analgesic drug development. After appraising the neurophysiological effects and mechanism of analgesic action established in preclinical studies, we show that the early concepts regarding capsaicin use have been translated into clinical application through evidence from randomized controlled trials.
Capsaicin Treatment for Hyperalgesia
TRPV1 Receptor
Peripheral nociceptive neurons are pseudo-unipolar neurons situated in sensory ganglia that have a single axon which branches within the ganglion into one peripheral and one central axon. TRPV1 functions as a polymodal receptor on the peripheral terminals of primary afferent neurons [22,23]. On the central terminals, TRPV1 may be involved in modulating signal transmission at the first synapse between the dorsal root ganglion or trigeminal ganglion neurons and the dorsal horn or caudal spinal trigeminal nucleus neurons [24,25,26,27].
TRPV1 is a ligand-gated, nonselective cation channel and is one of 30 known members of the TRP ion channel family [8,28]. TRPV1 is a key receptor involved in the transmission and modulation of pain signals [28], and is an important transducer of noxious stimuli (e.g. heat, low pH) and certain chemicals (including capsaicin). TRPV1 is sensitized by inflammatory mediators responsible for inflammatory pain arising from tissue injury [23,29]. TRPV1 receptor expression increases in some clinical pain conditions, possibly resulting from increased retrograde transport of nerve growth factor to the cell body [30,31].
Preclinical Studies Targeting the TRPV1 Receptor
Several molecular biology and pharmacological techniques have been employed to elucidate the role of the TRPV1 receptor in somatosensory pathways. Rodents lacking the TRPV1 channel are not only insensitive to vanilloid-evoked acute pain, but also exhibit impairment in their ability to detect thermal stimuli and to develop thermal hyperalgesia (table 1) [32,33,34,55]. However, responsiveness to noxious heat stimuli is not completely lost in TRPV1 knockout mice. Whereas TRPV1 knockout animals lack a component of sensory transduction in an otherwise intact neuronal circuit, intrathecal injection of capsaicin breaks the connection that all capsaicin-sensitive fibers have with the spinal cord [37]. It has been known for some time that intrathecal administration of capsaicin in rats causes long-lasting loss of heat sensitivity [35,36,56] and can induce selective degeneration of a distinct population of primary sensory neurons involved in the mediation of chemogenic pain [57]. More recent findings in the mouse indicate that pharmacological ablation of the central branches of TRPV1 nociceptors with capsaicin results in a more complete loss of acute sensitivity to heat pain than that observed in knockout constructs (table 1) [37]. However, it should be noted that 80–90% of neonatal rats desensitized to capsaicin in this way developed wounds, scabs and areas of alopecia [58].
Interestingly, intrathecal administration of resiniferatoxin, a more potent agonist of TRPV1 than capsaicin, to rats and monkeys not only rapidly inhibits nociceptive synaptic transmission but also provides long-lasting analgesia in behavioral models (due to destruction of TRPV1-expressing central sensory nerve terminals) [59,60]. These data support development of novel analgesics targeting TRPV1-expressing neurons either centrally or peripherally.
The striking difference between the limited effects of TRPV1 gene knockout and the pronounced effects of ablation of TRPV1-positive neurons on nociceptive signal processing is highly relevant for analgesic drug development [61]. While TRPV1 antagonists are anticipated – at best – to mimic the effects of TRPV1 knockout, TRPV1-positive neuron ablation can be mimicked by topical application of TRPV1 agonists to target tissues (skin or mucous membranes).
Studies in Human Volunteers
It is interesting to compare the behavioral effects of topical or intradermal capsaicin in humans with the pronounced changes observed in mice after intrathecal injection of capsaicin, as these procedures involve exposure to capsaicin at the two terminals of the same neuron. Topical exposure to capsaicin in humans leads initially to nociceptor firing and a period of enhanced sensitivity to painful heat stimuli. A refractory period follows during which individuals are relatively resistant to capsaicin and heat but not pinprick stimuli [17,38,62,63], consistent with the existence of a TRPV1-negative population of nociceptors that appear to be specialized for mechanical pain. Defunctionalization following continuous capsaicin exposure is accomplished via a TRPV1 receptor-mediated massive influx of calcium ions into the neuron [64] and the ensuing activation of calcium-sensitive proteases (among other mechanisms), which results in reversible neurite degeneration that can be visualized as reduction in ENFs (fig. 1) [10,42,43].
Fig. 1
Intraepidermal nerve fibers retract following high-dose topical capsaicin treatment. Reprinted from Kennedy et al. [43] with permission from Elsevier.
The rate and magnitude of reversible neurite degeneration was demonstrated in skin biopsies of healthy volunteers who received either repeated topical applications of low-dose capsaicin cream (0.075% w/w) over 7-day and 3-week treatment periods [10,40] or a single application of a high-concentration capsaicin patch (8% w/w) for 60 min [43]. Overall, the reduction in ENF density was slower and less extensive after repeated application of the low-concentration capsaicin creams than after a single application of the high-concentration capsaicin patch (table 1). Sensory testing revealed a pronounced reduction in thermal pain sensitivity but not pinprick sensitivity within the first few days after high-dose capsaicin cream [17,65], consistent with animal data. One week after application of the high-dose patch, however, thermal pain sensitivity was normal [43], suggesting relatively rapid recovery of nociceptive function in healthy subjects. A study using laser-evoked potentials and skin biopsies after low-dose capsaicin cream treatment also found a more rapid recovery of thermal pain sensation than ENF density [41].
The laser-evoked potential data indicate that heat-sensitive A-fiber nociceptors are also defunctionalized by capsaicin, whereas capsaicin was not associated with alterations in light touch, deep pain or mechanical pain detection thresholds [39]. Sharp mechanical (pinprick) pain is primarily mediated via capsaicin-insensitive A-fiber nociceptors, which would explain why topical capsaicin had only modest effects on this stimulus [17]. In turn, these data indicate that painful heat is the most adequate stimulus to monitor defunctionalization by capsaicin.
Therapeutic Use
Low-Concentration Topical Capsaicin
Despite promising pharmacodynamic data describing hypoalgesia accompanied by ENF degeneration in human volunteers [10], low-concentration capsaicin creams (≤0.1%) have demonstrated poor-to-modest efficacy in neuropathic pain syndromes over the short term (table 1) [44,45,46,47,48,49]. Explanations for the disappointing results include the low concentration of pure capsaicin used and poor patient adherence [9,66,67,68].
High-Concentration Topical Capsaicin
High-concentration capsaicin was first used as a 10% w/v solution in 10 patients with intractable pain due to bilateral peripheral neuropathy but this approach was terminated due to severe pain on application (despite regional analgesia) and risk of aerosol contamination [16]. More recently, a dermal patch has been developed containing a high concentration of synthetic pure trans-capsaicin (8% w/w) in the adhesive layer (QUTENZA™; NGX-4010). Advances in stereoselective medicinal chemistry have facilitated synthesis of pure trans-capsaicin (99%), which is identical to the naturally occurring molecule [9,67]. The synthetic process delivers higher concentrations of active trans-capsaicin than can be extracted from capsicums [67]. The high-concentration patch demonstrated efficacy in patients with PHN and in those with HIV-associated distal sensory polyneuropathy (HIV-DSP) (table 1) [51,52,53,54]. Lack of unnecessary exposure to capsaicin during the application and removal procedure represents a practical tolerability advantage of the patch over capsaicin creams. In addition, only one treatment is required every 3 months (or longer depending on pain control status).
In contrast to the clinical trial data for low-concentration capsaicin creams, the clinical efficacy data for the 8% capsaicin patch are entirely consistent with skin biopsy data. In double-blind randomized trials that used a low-dose capsaicin patch (0.04% w/w) as an active control to maintain blinding, a single application of the 8% capsaicin patch produced rapid and prolonged pain relief in patients with moderate-to-severe PHN or HIV-DSP, irrespective of whether they were receiving concomitant medications for neuropathic pain (table 1) [51,52,53].
A 60-min application of the 8% capsaicin patch – shown to be effective in patients with PHN [50] – exerted its maximum therapeutic effect as early as 2 days after application [51]. Significantly more patients treated with the 8% capsaicin patch reported a ≥30% pain decrease from baseline than control patients (42 vs. 32%; p = 0.04) [51]. Furthermore, the proportion of patients reporting an improvement on the Patient Global Impression of Change scale was significantly higher among those treated with the 8% capsaicin patch compared with those treated with the control at both week 8 (53 vs. 42%; p = 0.03) and week 12 (55 vs. 43%; p = 0.04) [51]. Similarly, a dose-finding study in patients with HIV-DSP revealed that a single 30- or 120-min (but not 60-min) application of the 8% capsaicin patch provided meaningful pain relief within 2 weeks, which persisted throughout the 12-week observation period [53]. The proportion of patients who reported a ≥30% pain decrease from baseline was >2-fold greater in the group treated with the 8% capsaicin patch than the control group (42 vs. 18%; p = 0.01) [53]. The importance of these positive trial data cannot be understated as there are very few, if any, other treatments with proven effectiveness in HIV-DSP [69]. Two open-label studies, which followed patients for a total of 48 weeks after their first treatment, have shown that re-application of the 8% capsaicin patch results in reproducible tolerability and safety, and suggest that a comparable degree of pain relief can be achieved with each retreatment (table 1) [52,54].
The clinical efficacy of the 8% capsaicin patch for treatment of PHN and HIV-DSP has been confirmed in two independent meta-analyses [70,71] and the European Federation of Neurological Societies have included the 8% capsaicin patch in recent guidelines on the pharmacological treatment of neuropathic pain [69]. Further studies are now needed to identify those patients who get the most pain relief and to identify other peripheral neuropathic pain syndromes that may respond to treatment with the 8% capsaicin patch. Patients with heat hyperalgesia, who make up about 18% of patients with neuropathic pain [72], are likely candidates.
Capsaicin Safety
Intradermal capsaicin administration leads to irreversible loss of TRPV1 neurons due to systemic distribution in newborn rodents [73,74], and to transient TRPV1 denervation at the injection site in human adults [75]. Repeated, long-term application of topical capsaicin to the rat hind paw did not damage the sciatic nerve or result in neuron loss in the dorsal root ganglion [76].
High-level dietary intake of capsaicin is not associated with any safety concerns in humans. The estimated highest capsaicin dietary exposure (in India, Mexico and Thailand) is 25–200 mg/day [77] and the total capsaicin content of one 8% capsaicin patch is within that range (179 mg per patch, of which only a fraction is absorbed). Systemic absorption of capsaicin is very low following a 60- or 90-min application of the 8% capsaicin patch, with a maximum plasma concentration of 17.8 ng/ml observed in any patient [78]. Systemic exposure to capsaicin following 8% capsaicin patch application is also transient as capsaicin levels decline rapidly with a mean population elimination half-life of 1.6 h. Furthermore, detectable levels of capsaicin metabolites were not observed in any patients. Therefore, treatment with the 8% capsaicin patch is unlikely to result in systemic side effects or to alter the systemic metabolism of concomitant medications. In clinical trials, topical application of capsaicin was generally well tolerated and associated with transient application-site reactions such as pain and erythema (table 1).
It is important to evaluate the safety of topical capsaicin in patients with diabetes as they likely have ongoing neurodegenerative processes that might be exacerbated by capsaicin. Case-control data indicate that ENF regeneration after low-dose capsaicin is slower among patients with diabetes even when these patients had no evidence of neuropathy [79]. Moreover, caution should be exercised when applying topical capsaicin to skin at risk of ulceration [80]. Two small studies of patients with painful diabetic neuropathy reported the encouraging finding that topical capsaicin cream had no adverse effects on sensory function or neurovascular control [81,82], but more safety data, using both morphological (ENF) and functional assessments (heat pain, laser-evoked potentials), are needed before this treatment can be recommended for patients with diabetes (currently a contraindication).
Discussion
A better understanding of the molecular and cellular effects of capsaicin has helped rationalize the popularity of this molecule as a means of assuaging pain. Theories on the mechanism of action of capsaicin have shifted from increased perfusion and counterirritation to defunctionalization. Defunctionalization of the entire population of TRPV1-expressing neurons is more efficient in reducing pain behavior in animals than knockout of TRPV1. Likewise in humans, defunctionalization by high-dose patches has been found effective in meta-analyses of clinical trials, whereas TRPV1-antagonists have yet to be approved (they are also burdened by adverse effects such as fever, hypertension and gastric ulceration [29]). Both low-dose and high-dose topical capsaicin are listed as effective in the most recent meta-analysis on neuropathic pain therapy [71]. Of all the effective therapies, the high-dose patch is unique in that it can deliver up to 12 weeks of pain relief from a single application, whereas all other therapies were evaluated for continuous treatment over the same period of time. At the clinic, activation of TRPV1 by topical application of the 8% capsaicin patch translated into meaningful pain relief from PHN and HIV-DSP.
In conclusion, the capsaicin story is a valuable case study describing how pharmacognosy can provide valuable and necessary inputs into rational drug design, and how medicinal chemistry and neurobiology can improve traditional medicines. For the capsaicin patch, the difficult step of translating efficacy in defunctionalizing nociceptive neurons in animal models to pain relief in human patients was greatly facilitated by pre-existing clinical experience with topical capsaicin preparations that had been in use before the understanding of its molecular basis of action.
Acknowledgements
Editorial assistance and evidence tables were provided by Malcolm Darkes at Adelphi Communications, supported by Astellas Europe Ltd.
Disclosure Statement
M.H. has lectured or served on advisory boards for the following companies in the past 24 months: Abbott, Astellas, Astra Zeneca, Eli Lilly, Janssen Cilag, Medtronic, MSD, Mundipharma, Pfizer and Sanofi-Pasteur, and has participated in international congresses as a guest of Astellas, Medtronic and Pfizer. RDT has lectured or served on advisory boards for or received grants from the following companies in the past 24 months: Astellas, Boehringer Ingelheim, Galderma, Grünenthal, Kade, Eli Lilly, Merz, Nycomed, Pfizer.
Get Permission
PubMed ID
