The discovery and development of gefapixant

Abstract

Gefapixant is the approved generic name for a compound also known as MK-7264, and prior to that AF-219 and RO-4926219. It is the first-in-class clinically developed antagonist for the P2X3 subtype of trimeric ionotropic purinergic receptors, a family of ATP-gated excitatory ion channels, showing nanomolar potency for the human P2X3 homotrimeric channel and essentially no activity at related channels devoid of P2X3 subunits. As the first P2X3 antagonist to have progressed into clinical studies it has now progressed to the point of successful completion of Phase 3 investigations for the treatment of cough, and the NDA application is under review with US FDA for treatment of refractory chronic cough or unexplained chronic cough. The molecule was discovered in the laboratories of Roche Pharmaceuticals in Palo Alto, California, but clinical development then continued with the formation of Afferent Pharmaceuticals for the purpose of identifying the optimal therapeutic indication for this novel mechanism and establishing a clinical plan for development in the optimal patient populations selected. Geoff Burnstock was a close collaborator and advisor to the P2X3 program for close to two decades of discovery and development. Progression of gefapixant through later stage clinical studies has been conducted by the research laboratories of Merck & Co., Inc., Kenilworth, NJ, USA (MRL; following acquisition of Afferent in 2016), who may commercialize the product once authorization has been granted by regulatory authorities.

Introduction

The program to identify novel P2X3 antagonists at Roche Palo Alto was first initiated in 1996 as part of an initiative to identify novel drug discovery targets for treatment of a variety of poorly addressed medical conditions, particularly those related to the peripheral nervous system, and at that time recently identified P2X family members were medicinally ‘unprecedented’. The neurobiology discovery unit at Roche was then focused on novel therapeutic approaches for management of chronic pain conditions as well as disorders associated with peripheral visceral organ dysfunction and sensitization. Specific patient targets included chronic inflammatory pain, joint pain, neuropathic pain as well as a range of genitourinary conditions such as overactive bladder, prostatism, urinary incontinence, and pelvic pain syndromes, though the anticipation of broader potential for impact was noteworthy.

In the preceding 3–4 years there had been an explosion in the gene sequencing, cloning and molecular characterization among many families of receptors for intercellular signaling that included those responding to purine and pyrimidine nucleotides. In the case of ionotropic ATP channels, between 1994 and 1996, seven distinct subunits for this new family of trimeric receptors (P2X₁ – P2X₇, encoded by the human genes P2RX1-P2RX7), were cloned, sequenced, localized and characterized and, in some cases, aligned with preexisting pharmacologically-defined response profiles (Brake et al., 1994; Valera et al., 1994; Bo et al., 1995; Chen et al., 1995; Lewis et al., 1995; Collo et al., 1996; Surprenant et al., 1996). These cloning efforts began with breakthroughs from groups led by David Julius at UCSF (P2X2), Alan North and Anne-Marie Surprenant at Glaxo Institute of Molecular Biology, Geneva, as well as the laboratories of Geoff Burnstock and John Wood at University College London. This ion channel family was seen by many to offer a potential bounty for ‘drug-hunters’ to deliver much needed therapeutic options for many poorly met diseases, syndromes, and symptoms. The promise of targeting novel drug substrates is always highly appealing, and ligand gated channels have been successfully targeted before. Nevertheless, in the absence of selective chemical probes, challenges remained in determining whether individual trimeric channels from this family were specifically involved in symptom presentation or disease pathology, whether their distributions revealed special relevance in particular patients populations and ascertaining if their chemical targeting may introduce potential for safety or tolerability limitations. Our initial perspective was that this target class offered many clear discovery opportunities due to the emerging distribution profiles and characteristics: pain, sensory and autonomic physiology, and visceral dysfunction were clearly in play, with evidence revealing the common association of ATP with pathobiological processes and tissue/organ responses; thus, our attentiveness to this new field was fully piqued. The expression analysis across mammalian tissues of P2X1 subunits, particularly abundant in smooth muscles, P2X3 found in sensory ganglia, P2X2 found associated with P2X3 in autonomic and sensory ganglia (plus a much broader non-sensory distribution), and P2X7 being associated with immune cells and inflammatory cytokine secretion cascades (Burnstock et al., 1972; Valera et al., 1995; Vulchanova et al., 1997; Ferrari et al., 1997; Bradbury et al., 1998), left our team, and many others across the industry, excited by the possibilities of targeting for therapeutic benefit these newly discovered proteins. Nevertheless, a primary driver that provided so much context for these opportunities was the inexorable and fascinating work of Geoff Burnstock: decades in the making, origins in the deep hollow viscera (colon, duodenum, bladder, ureter, vasculature), with studies and reviews galore (Burnstock, 1972; Burnstock, 1978; Burnstock and Kennedy, 1985; Ralevic and Burnstock, 1998).