Introduction

It has been well established that capsaicin, the algesic component of chili peppers, produces tissue injury leading to neuronal activation and sensitization [1–4]. Psychophysical studies have demonstrated that mechanical and heat hyperalgesia and a cutaneous flare response follow intradermal injection of capsaicin into human skin and are correlated with pain magnitude and duration [1,4]. Neurophysiological studies in monkeys identified heat-induced sensitization of A and C fiber mechano-heat-sensitive (AMH, CMH) fibers and capsaicin-induced sensitization of spinal dorsal horn neurons conducting along the spinothalamic tract [3,5,6]. Interestingly, heat-evoked pain in human subjects and heat-evoked neural activity in monkey CMH fibers overlapped, with an activation threshold of approximately 45 °C and a stimulus-response function that increased monotonically with increasing stimulus intensity [6]. Later, heat activation of the heterologously expressed cloned capsaicin receptor was defined to be in the noxious range with an activation threshold of 42 °C, thereby providing a molecular mechanism by which heat and capsaicin exert physiological effects [7,8]. Following this seminal discovery, pharmaceutical and academic laboratories have spent more than 15 years researching this protein as a therapeutic target for analgesia.

The cloned capsaicin receptor, initially called the vanilloid receptor subtype 1 because of the unique gating by vanilloids such as capsaicin, was recognized as the first member of the transient receptor potential vanilloid family and was designated TRPV1. Structurally, TRPV1 is a tetrameric six-transmembrane ion channel protein with nonselective permeability to cations. Consistent with early neurophysiological studies demonstrating that capsaicin activated AMH and CMH fibers that travel along the spinothalamic tract, TRPV1 resides primarily on peptidergic small- and medium-diameter neurons in the peripheral nervous system [9–11]. Initially classified as the capsaicin receptor, TRPV1 functions as a molecular integrator of a variety of stimuli, including endogenous lipids, noxious heat, and acid. In addition to capsaicin, TRPV1 can be activated by plant products, including resiniferatoxin (RTX), piperine, gingerol, zingerone, as well as camphor and eugenol, ethanol, and spider and jellyfish venom ([12,13], and reviewed in Refs. [14,15]). Channel activation results in electrical and chemical activity in neurons. Channel opening results in signal transduction of nociceptive stimuli in neurons. Pronociceptive mediators including substance P, glutamate, and calcitonin gene-related peptide (CGRP) are released in response to TRPV1 activation, which in turn contributes to the development of neurogenic inflammation. The resulting proinflammatory milieu can sensitize TRPV1, leading to enhanced channel opening and thereby contributing to neuronal sensitization. Initially thought to exert effects primarily in the peripheral nervous system, TRPV1 expression and function in the central nervous system (CNS) is now widely reported, although the full extent of its physiological function in the CNS has not been elucidated [16–19].

TRPV1 functions as a molecular integrator of multiple physical and chemical stimuli, consistent with its localization on polymodal nociceptors in the peripheral nervous system. First-generation TRPV1 antagonists were designed to block all modes of TRPV1 activation. Clinical investigations of these compounds unveiled thermosensory deficits as a key limitation to further development. A breakthrough in the field was realized when TRPV1 antagonists could be designed to selectively block different modes of activation. The field quickly shifted toward the discovery of selective TRPV1 antagonists that could differentially block heat and capsaicin, but not acid activation of the channel to pharmacologically separate analgesic and thermoregulatory effects. Such modality-specific pharmacology has since been the focus for development of a novel analgesic-targeting TRPV1.

Preclinical Overview of TRPV1 Antagonists

In an effort to achieve analgesia through modulation of TRPV1 without the burning sensation associated with agonists such as capsaicin and RTX, the Sandoz group (now Novartis) reported on discovery of antagonists of TRPV1 and their assessment as pain relievers [20,21]. Synthesis of small molecules with structural resemblance to capsaicin led to the development of capsazepine, the first TRPV1 antagonist pharmacological tool compound widely utilized in early TRPV1 research (Figure 8.1).

The beginning of the 21st century witnessed a burst of TRPV1 drug discovery activities in major pharmaceutical and smaller biotechnology companies. The potential role of TRPV1 antagonists as analgesic agents was suggested based in part on attenuation of pain-like behaviors in TRPV1 knockout (KO) mice [22]. However, the first generation of TRPV1 antagonists was limited by chemotype-independent hyperthermia in preclinical species and humans ([23], and reviewed in Refs. [24,25]). These findings, along with hypothermia-inducing properties of TRPV1 agonists and absence of hyperthermia in KO mice treated with TRPV1 antagonists, unambiguously established a thermoregulatory role for the TRPV1 channel [26–28]. Therefore, pharmacological separation of analgesic and hyperthermic effects became the key challenge in developing TRPV1 antagonists as viable agents for pain management.

One of the approaches to decouple undesired thermoregulatory effects from desired analgesic effects consisted of preventing TRPV1 antagonists from penetrating the brain where the hypothalamus is known to be involved in thermoregulation [29]. However, potent peripherally restricted TRPV1 antagonists still caused hyperthermia in rats, suggesting that peripheral restriction was not sufficient to attenuate the thermoregulatory effects and that the site of action for the hyperthermic effect was predominantly outside the CNS [30]. Direct administration of a selective but undifferentiated TRPV1 antagonist into the medial preoptic area of the hypothalamus did not affect core body temperature [18]. However, TRPV1-antagonist-induced increase in core body temperature was blocked by systemic RTX-mediated desensitization of TRPV1 [27]. Together, these studies suggest that visceral TRPV1 receptors are responsible for regulation of core body temperature.

Selective pharmacological blockade of some but not all modes of TRPV1 activation emerged as a more promising direction toward discovery of TRPV1 antagonists that could provide pain relief without affecting body temperature. AMG-8562 (Figure 8.2) from Amgen was one of the early and most characterized modality-specific TRPV1 antagonists [31].

AMG-8562 blocked capsaicin activation of rat TRPV1 with an IC₅₀ of 1.75 nM, did not affect heat activation, and potentiated pH 5 activation in a ⁴⁵Ca^2 + uptake assay using cells expressing recombinant rat TRPV1 [31]. Oral administration of AMG-8562 in rats either did not induce hyperthermia or had small hypothermic effect, but still showed efficacy in several preclinical pain models, albeit at rather high plasma concentrations. It should be noted that the pharmacological profile of AMG-8562 at human TRPV1 differs from the rat profile in that it partially blocked pH 5 activation in humans, whereas it potentiated acid activation at rat TRPV1. Significance of the blockade of proton activation for the thermoregulatory effects was confirmed in a systematic study of hyperthermic responses of rats, mice, and guinea pigs to TRPV1 antagonists displaying different pharmacological profiles [32]. It was concluded that TRPV1 antagonists that do not block low pH activation would exhibit a hyperthermia-free profile, even if they are potent blockers of heat activation.

AS-1928370 (Figure 8.2) from Astellas Pharma displays differential pharmacology in blocking the activation of TRPV1 [33]. This compound inhibited capsaicin-induced Ca^2 + flux in rat TRPV1 with an IC₅₀ value of 880 nM and capsaicin-induced currents in electrophysiological assay with an IC₅₀ value of 32 nM, but showed very small inhibitory activity against pH 6 activation (< 20% block at 10 μM). This profile was responsible for lack of effect on rectal body temperature in rats up to 10 mg/kg oral dose, although 30 mg/kg dose induced significant hypothermia. At an oral dose of 1 mg/kg, AS-1928370 fully attenuated pain-like behaviors evoked by intradermal capsaicin in a model of secondary hyperalgesia and exerted full efficacy in the rat spinal nerve ligation (SNL) model of mechanical allodynia (1-3 mg/kg dose range). These data provided evidence that analgesic effects in the SNL model of neuropathic pain were mediated by TRPV1. Such a conclusion was supported by high brain drug concentration, which sufficiently covered the IC₅₀ value obtained from electrophysiological recordings (355 versus 32 nM). This was the first demonstration that a TRPV1 antagonist displaying no inhibitory effect on proton-induced activation can exhibit high efficacy in neuropathic pain model. Potent effects in rats were recapitulated in mice, where AS-1928370 significantly suppressed both capsaicin-induced acute nocifensive and withdrawal responses in the hot plate test at oral doses of 10-30 mg/kg [34]. Significant efficacy was obtained in the SNL model at lower oral doses of 0.3-1.0 mg/kg. Despite favorable preclinical pharmacological and safety profiles, there was no information reported on whether AS-1928370 or any of its analogs entered clinical trials. It should be noted that AS-1928370 did inhibit proton activation of human TRPV1 with an IC₅₀ value of 1.5 μM, while having no effect at rat TRPV1 (IC₅₀ > 20 μM) [33]. Given the correlation established between inhibitory effects on proton-evoked activation of TRPV1 in vitro and changes in body temperature in vivo, there should be considerable caution in predicting a hyperthermia-free profile for AS-198370 in humans.

Grünenthal described several chemotypes of TRPV1 antagonists, some representatives of which displayed differential pharmacology against capsaicin, low pH, and heat activation of TRPV1. Compound 41 (Figure 8.2), a diaryl acetamide, is a very potent TRPV1 blocker against capsaicin activation (K_i = 0.1 nM), but much weaker against pH 6 activation (IC₅₀ = 87 nM) [35]. Several compounds from their propanamide series also exhibited large potency gaps inhibiting in vitro capsaicin and low pH. For example, compounds 12 (Figure 8.2) [36] and 15 (Figure 8.2) [37] were potent inhibitors of capsaicin activation with IC₅₀ values of 8 and 2 nM, but weak inhibitors of low pH activation with 0% and 15% effects correspondingly at 5 μM. Additional TRPV1 antagonists with differentiated pharmacology were discovered within carboxamide and urea containing series [38–40]. Grünenthal did not describe effects on core body temperature; therefore, it is unknown if modality specific in vitro pharmacology for some TRPV1 antagonists was predictive of lack of hyperthermia in preclinical species.

In a series of patent applications AbbVie claimed TRPV1 antagonists of different chemical classes, several representatives of which did not produce thermosensation deficits in rats. Lack of thermosensory effects was demonstrated by increased average response latency for tail withdrawal from a 55 °C water bath. For example, Compound 10 (Figure 8.2) blocked capsaicin activation of TRPV1 with an IC₅₀ value of 54 nM, but blocked only 28% of pH 5 activation at 37.5 μM [41]. As a result of differentiated pharmacology, Compound 10 had no effect on latency to tail withdrawal in the rat tail immersion assay, suggesting no in vivo effect on thermosensation. Similarly, no thermosensation deficits were observed for Compound 6 (Figure 8.2) in the same tail immersion assay [42]. Compound 6, representing a different chemical class of TRPV1 antagonists than Compound 10, was characterized as a full blocker of capsaicin activation with an IC₅₀ = 4 nM and partial blocker of acid activation with only 38% inhibition at 37.5 μM. Compound 6 also attenuated pain-like behaviors with 88% effect in the capsaicin-induced flinching model at an oral dose of 100 μmol/kg and with 59% effect in capsaicin-induced secondary mechanical hypersensitivity model at an oral dose of 10 μmol/kg.

A more detailed account was reported from AbbVie on an additional series of differentiated TRPV1 antagonists [43]. A-1165442 (Figure 8.2) is a potent blocker of capsaicin activation of rat and human recombinant TRPV1 in FLIPR assays with IC₅₀ values of 35 and 17 nM, respectively, as well as in whole cell patch clamp electrophysiology studies in rat dissociated dorsal root ganglia neurons (IC₅₀ value 2.7 nM). In contrast, A-1165224 exhibited only a partial blockade of acid-evoked response at both rat and human TRPV1 measured in FLIPR (14% and 61% at 11 μM) and electrophysiology assays (66% block at 10 μM). Interestingly, biochemical analysis indicated that A-1165442 was less efficacious in blocking acid-evoked CGRP release compared with capsaicin-evoked CGRP release (22% vs. 100% block at 10 μM). A-1165442 and related analogues had a similar differentiated pharmacological profile highlighted by acid-sparring inhibition of TRPV1 and lack of or diminished body temperature elevations of < 0.5 °C in telemetrized rats. The relationship between acid blocking of TRPV1 in vitro and hyperthermic effects of TRPV1 antagonists in vivo is consistent with results of earlier studies with other classes of TRPV1 antagonists [31–33] and was demonstrated with a structural analogue, A-1106625 (Figure 8.2), which is a potent TRPV1 blocker at all modes of activation [43]. Although acid partial blocker A-1165442 did not change rat core body temperature at plasma concentrations ~ 8.5-fold higher than its capsaicin blocking IC₅₀, A-1106625 induced 1 °C temperature increase at lower multiples of plasma concentration. Both compounds exhibited comparable efficacy in a rat osteoarthritis (OA) pain model (ED₅₀ ~ 30-35 μmol/kg). However, unlike A-1106625, acid partial blocker A-1165442 was ineffective in attenuating allodynia in a mouse bone cancer pain model. The latter result can be explained by the importance of osteoclast-induced acidosis for bone cancer pain generation [44,45].

Clinical Overview of TRPV1 Antagonists

The first selective TRPV1 antagonist that was tested in humans was SB-705489 from GlaxoSmithKline. The compound effectively blocked TRPV1 activation in vitro by capsaicin, heat, and low pH and reduced inflammatory pain in rodents [46,47]. In target engagement studies during Phase 1 clinical trials, SB-705498 (Figure 8.3) at a single 400-mg oral dose significantly reduced the area of capsaicin-evoked flare versus placebo [48]. The same dose also reduced ultraviolet-B (UVB)-evoked flare area and heat hyperalgesia compared with placebo, albeit to a lesser degree. Reduction of flare area correlated with plasma exposure levels of SB-705495, suggesting effects were drug related and via a TRPV1-mediated mechanism. However, SB-705498 did not reduce the intensity of both capsaicin- and UVB-evoked flare as well as capsaicin-evoked thermal hyperalgesia. It is possible that the combination of capsaicin and heat resulted in more pronounced activation of TRPV1, blockade of which would require higher concentration of SB-705498. At the time point of pharmacodynamics assessments (~ 6 h postdose), total plasma concentration of SB-705498 was 0.5 ± 0.22 mg/mL, comparable to those predicted to be efficacious based on preclinical models. The maximum tolerated dose of 400 mg resulted in t_max of 2 h (0.75-4 h) and half-life of 54 h (35-93 h) with SB-705498 remaining quantifiable for the 168 h postdose sampling period. Effects of SB-705498 on heat and taste thresholds were also investigated. Heat threshold was elevated into the noxious range, in line with the role of TRPV1 as thermosensor. Taste experiment results were not conclusive, but the threshold may have not changed after administering a series of diluted capsaicin solutions to volunteers [49]. Following Phase I, SB-705498 was investigated in clinical studies of acute migraine, dental pain, and rectal pain, but the results of these studies were not published.