Neurokinin-1 Receptor Inhibition and Cough.

There is a need for better treatment for cough. Unexplained or chronic refractory cough (CRC) is the focus of several recent and ongoing large drug trials, with particular interest in antagonists of ATP-stimulated P2X receptors (1). Although such drugs appear very promising, there are nonresponders, and side effects may be unacceptable to some (2). Hence, a requirement for alternative approaches, not only for CRC but also for cough associated with chronic and incurable diseases of the lung, such as lung cancer, is needed. In this issue of the Journal, Smith and colleagues (pp. 737–745) present their findings on aprepitant for cough in non–radically treatable lung cancer, which, considering the current need, are very welcome (3).

Aprepitant is an antagonist of NK1 (neurokinin 1), a G protein–coupled receptor triggered by the ligand Substance P (SP). NK1 receptors are present in the central and peripheral nervous system as well in other tissues, with apparently varied physiological functions (4). Of note is the possible involvement of SP in sensory disorders, including overactive bladder and chronic itch (5). An important role of NK1 in cough has also been postulated. Within the nucleus tractus solitarius in the brain stem, there have been repeated observations in animal models of the activity of SP and NK1 inhibition on the cough reflex (6). Vagal afferent C fibers in the airway appear to produce SP, and selective NK1 antagonism specifically blocks C fiber–dependent coughing in guinea pigs (7). In humans with respiratory disease, inhaled SP can induce cough (8).

Previous trials of NK1 antagonists in humans with airway disease have failed to impact cough. The selective antagonist CP-99,994 did not demonstrate an effect in 14 subjects with mild asthma on hypertonic saline–induced numbers of coughs or bronchoconstriction (9). However, this may relate to minimal pathologic cough in the study subjects or to the selection of an inappropriate trial endpoint; experimentally induced cough correlates poorly with daily cough frequency (10). The NK1/2 inhibitor DNK333 apparently failed to demonstrate efficacy on self-reported cough over 2 weeks in 28 individuals with chronic obstructive pulmonary disease, although these findings are not fully reported (11).

Aprepitant is a member of the class that probably has more central nervous system penetration than other NK1 inhibitors. The drug is licensed in the United Kingdom as an antiemetic in patients receiving systemic anticancer therapy, likely acting centrally. Patients with lung cancer are therefore a logical group in whom to explore potential antitussive effects. Smith and colleagues’ study participants were a real-world sample of patients from a UK lung oncology clinic who were bothered by cough (3). The group was mixed, with either small cell or stage 3–4 non–small-cell lung cancer of several histological subtypes. Tumors were either peripherally or centrally located (the latter presumably being more likely to produce airway-related symptoms). Current smokers were included, as were patients receiving drugs known to impact cough, including opiates, gabapentin, and ACE inhibitors (albeit at an established and fixed regular dose). Coexisting lung diseases or other relevant pathologic conditions such as acid reflux were not excluded. The particular underlying triggers to cough were therefore likely varied in this heterogenous group.

In a randomized double-blinded crossover study design, 20 patients received either aprepitant for 3 days (at the standard antiemetic dosing of 125 mg and then 80 mg once daily) followed by placebo for another 3 days, or vice versa, with a 3-day washout period in between. Change in cough frequency during waking hours was the primary endpoint, measured with a portable acoustic cough monitor for 24 hours at baseline and on Day 3 of each treatment. Patient-reported symptom scores were the main secondary outcome.

A significant reduction of 22% in mean awake cough frequency was observed with the drug compared with placebo. There were also significant and clinically meaningful average improvements in all three subjectively reported measures of cough. No significant adverse events were associated with the study drug.

These results are very encouraging and have formed part of a renewed interest in NK1 inhibitors in cough. However, this is clearly a small pilot study in a mixed group. As well as showing the interindividual differences between subjects at baseline, Figure 3 in the paper by Smith and colleagues demonstrates the large heterogeneity in outcomes (3). Awake cough frequency in perhaps half of participants was either very similar or higher after receiving aprepitant than after receiving placebo (Figure 3A in Reference 3). Some of this similarity could potentially relate to inadequate washout between treatments in the participants who were randomized to the study drug first; cough frequency measurement was not repeated between the end of treatment 1 and the start of treatment 2. Only larger studies of longer duration will cast light on the interactions between potential antitussive effects of aprepitant and individual triggers and determinants of coughing. Furthermore, doses higher than those used in this trial may lead to more than the modest overall reductions in cough counts reported here.

Evidence for the efficacy of NK1 inhibitors for cough in patients with conditions other than lung cancer is so far very limited, although there are recent data. In an open-label study of 13 patients with CRC, orvepitant at 30 mg once daily for 4 weeks was associated with a 26% reduction in cough frequency (12). With three different doses, preliminary reports from the phase 2b randomized controlled trial VOLCANO-2 (A Double-Blind, Randomized, Placebo-controlled Study of the Efficacy and Safety of Three Doses of Orvepitant in Subjects with Chronic Refractory Cough) are of a failure to demonstrate overall efficacy of orvepitant on cough frequency in CRC. However, in a prespecified analysis, there was a trend for reductions in cough counts in those with higher cough frequencies. Furthermore, across all subjects, there were significant improvements in patient-related outcomes with the higher dose (30 mg once daily) and, again, no significant adverse events compared with placebo (13). Meanwhile, a randomized controlled trial of serlopitant at 5 mg once daily has failed to demonstrate any effect on cough frequency or symptoms in CRC (14). Whether higher doses of both drugs would have achieved different outcomes is not clear; full NK1 receptor blockade may be required for clinical efficacy (15).

In terms of potential mechanisms of action of aprepitant in cough, these are explored in the second part of Smith and colleagues’ study (3). SP applied to tissue samples of the vagus nerve from both guinea pigs and humans induced depolarization, a response that was specifically inhibited by exposure of the samples to aprepitant. This raises the possibility of important peripheral antitussive sites of action of NK1 inhibitors in the airway as well as, or instead of, at the level of the brainstem. There could therefore be an analogy with the suggested significant role of SP and NK1 in itch signaling, in this case all along the pathway from sensory afferents (including C fibers), to dorsal horn cells, ascending sensory spinal neurons and higher centers (5). The current study has demonstrated that there is still a lot to learn about NK1 antagonists, which may still yet have potential as much-needed novel treatments for cough.