November 17, 2023
8 min read
- Out of 13 panelists, 12 voted against the approval of gefapixant for adults with unexplained chronic cough.
- Panelists expressed concern over the small reduction in cough frequency between gefapixant and placebo.
In a 12-to-1 vote, the FDA’s Pulmonary-Allergy Drug Advisory Committee voted against approving gefapixant 45 mg tablets to treat adults with refractory or unexplained chronic cough.
The FDA is not required to follow the recommendations of its advisory panels, but it usually does. At the moment, there are still no FDA-approved therapies for this patient population.
In January 2022, the FDA issued a complete response letter to Merck declining to approve a new drug application for gefapixant, an antagonist of the P2X3 receptor, in this patient population and requested additional validation of the wearable cough counting system (VitaloJAK, Vitalograph) used to assess the drug’s efficacy.
The FDA also expressed concerns about the clinical benefit of the drug, as it was unclear based on study results whether the primary endpoints conferred clinically meaningful improvements, and on patient-reported outcome data.
Following a validation study of VitaloJAK, Merck recounted the cough data from their two pivotal randomized, double-blind, placebo-controlled clinical trials, P030 and P027, and resubmitted their application.
In P030, 409 patients received twice-daily gefapixant 45 mg and 419 patients received twice-daily placebo for 24 weeks. In P027, 217 patients received twice-daily gefapixant 45 mg and 222 patients received twice-daily placebo for 12 weeks. Both trials also included a 15 mg gefapixant arm but, due to a lack of efficacy in that group, the focus of the application is on the 45 mg dose.
Based on data from the VitaloJAK cough counting system, researchers compared cough frequency (number of cough events in 24 hours/total recording duration) between the gefapixant and placebo groups, according to the FDA briefing document.
Both trials found only a small decrease in cough frequency with gefapixant vs. placebo from baseline, with a significant relative reduction of 14.6% (95% CI, –26% to –1.5%) at week 24 in P030 and nonsignificant 17% (95% CI, –31.5% to 0.6%) relative reduction at week 12 in P027.
The FDA asked panelists to consider the large placebo effect observed in both trials when considering these data, but also whether such small effects conferred clinically meaningful improvements.
Due to the “complicated presentation of the data” that used natural log-transformed cough frequency, the FDA also conducted a post-hoc analysis of the absolute cough frequency. Results showed that prior to receiving gefapixant, patients from both trials had mean and median cough frequencies between 20 to 30 coughs per hour, and the FDA briefing document notes that gefapixant only lowered the number of coughs by one to two per hour in the two trials relative to placebo. Again, the FDA asked the panelists whether this represents a clinically meaningful change for patients suffering with chronic cough.
When evaluating how patients’ perception of their cough improvement — measured using the Patient Global Impression of Change scale — relates to their cough frequency, researchers found a low correlation between the two factors in P030 and P027.
Further, the placebo effect was large when comparing scores from the Leicester Cough Questionnaire (LCQ), Cough Severity Diary and cough severity VAS between patients taking gefapixant vs. placebo.
Out of these patient-reported outcomes (PROs), researchers only found a significant improvement in the total LCQ score in the P030 trial, with more patients in the gefapixant arm achieving at least a 1.3-point increase from baseline (n = 262 of 342 vs. n = 245 of 355; OR= 1.4; 95% CI, 1-2).
In terms of safety, more than half (65%) of those receiving the 45 mg dose of gefapixant reported disturbance or loss of taste, with 14% discontinuing the drug for this reason.
Prior to voting, panelists discussed the evidence of the effectiveness of gefapixant in this patient population.
One of the main concerns was the small reduction in cough frequency seen with the drug. John M. Kelso, MD, clinical professor in the division of allergy, asthma and immunology at Scripps Clinic, said the reduction by one to two coughs per hour “seems not meaningful or not relevant.”
Sally A. Hunsberger, PhD, biostatistician at the NIH, added that after listening to patients with chronic cough talk about how they frequently cough in clusters, she is unsure if cough frequency is the best endpoint to assess gefapixant.
“My concern is that we don’t really have the right endpoint to establish whether this is a beneficial drug or not,” Hunsberger said.
Several panelists also expressed that the PRO results did not appear to be clinically meaningful. Edwin H. Kim, MD, MS, associate professor of medicine and division chief of the pediatric allergy and immunology division at University of North Carolina School of Medicine, stated his concern over the lack of correlation between the PROs and cough frequency.
“I would like to think that the way that the drug works, decreasing the frequency of cough should correlate with those [PROs], and so to not see that correlation is worrisome to me that the medication at least the way it’s supposed to be working is not effective in actually improving those PROs, so any improvement seen may be coming from some other factors other than the medication itself,” Kim said.
When discussing the common adverse event of taste disturbance, Jennifer Schwartzott, MS, patient representative, said this was not a major concern in her eyes.
“If the tastes disturbance is only minor, then to me, the reduction in cough even if it’s a small one might be worth it,” she said.
“[Given] the fact that there’s no major safety issues, a patient is going to be more inclined to go with something that may not be perfect, but at least is something in the short term,” Schwartzott added.
On the other hand, Mark S. Courey, MD, professor in the department of otolaryngology-head and neck surgery at Icahn School of Medicine, expressed his worries over the high prevalence of this adverse event among those taking gefapixant.
“The fact that the taste disturbance was so present in two-thirds of the patients when you have a minimal response, and much of that is judged on PRO, it’s very concerning to me,” he said.
The 28% dropout rate observed in this study was another topic of interest among the panelists, specifically when discussing the drug’s benefit-risk ratio.
“Assuming many [patients] are applying to this trial looking for help, and then having a 28% dropout rate, suggests there’s something may be off with this benefit-risk ratio if up to almost a third of these patients don’t stay on,” Kim said.
During her summary of this discussion, Paula Carvalho, MD, FCCP, chairperson of the meeting and professor of medicine in the division of pulmonary, critical care and sleep medicine at University of Washington, noted that the panelists’ concerns over the assessed endpoints suggest that they will possibly need to be reworked.
“We’re [in] a little bit of uncharted territory because we don’t have prior experience with the interpretation of these kinds of results,” she said. “We don’t have a good precedent for endpoints, and we are hearing loud and clear that endpoints do need to be rethought and reconsidered.”
When asked if the evidence presented for gefapixant shows a clinically meaningful benefit in the studied patient population, one panelist voted yes, and 12 panelists voted no.
Schwartzott was the only panelist who voted yes, and when discussing the rationale behind her vote she admitted that she was on the fence but wanted to present an option to patients suffering with this condition.
“With this drug, any reduction of cough symptoms for many patients would be worthwhile to them as long as the risk is low, which I felt that it was,” she said. “I wanted to give the patients a chance, to give them something that could potentially work at least a little bit until the perfect drug comes along, which hopefully won’t be that far from now.”
Notably, she did mention the need for further study on the drug and more definition on the protocols used.
Although the rest of the panelists voted no, each one recognized that chronic cough is a severe illness that has a great impact on a patient’s life.
“My heart goes out to this patient population who remain hopeful for a therapy that would make a difference, but I am just concerned … we don’t want to be providing just hope, we want to be providing predictably effective pharmacologics that are likely to make meaningful differences,” Leonard B. Bacharier, MD, Janie Robinson and John Moore Lee Chair in Pediatrics at Vanderbilt University Medical Center, said.
For Courey, the small difference between placebo and gefapixant had a major impact on his vote.
“I don’t think the level of evidence supports that the drug makes a significant difference,” he said. “It’s unfortunate. I am concerned that if the drug is readily available, it could lead to a delay in diagnosis of … other illnesses because cough, while it can be very debilitating, is a symptom, not a disease in of itself. I think this would delay the evaluation of the patients for other diseases and could be potentially harmful that way.”
Kim also expressed the same rationale for voting no and wondered if the study design could be altered in order to distance the placebo effect seen.
“Is there a way to separate out the placebo effect from the treatment itself? Whether that might be in a clinical trial design, I’m not sure if some sort of crossover design or something like that might be able to tease out placebo vs. an actual medication effect.”
Bacharier added, “I think we’re really at a loss for what an outcome really would compel us that an agent in this condition made our patients meaningfully and predictably better.”
When explaining her rationale, Emma H. D’Agostino, PhD, consultant at the Cystic Fibrosis Foundation, said that study endpoints should capture what these patients are dealing with, which, based on the public hearing, seems to be cough clusters.
“Moving forward … [researchers should be] rethinking the endpoints to capture what’s most meaningful to patients, so rethinking that cough frequency,” she said. “Maybe instead of looking at overall frequency over 24 hours, [we should be] looking at cough clusters … to really capture the most meaningful manifestations of cough.”
Nicole Hamblett, PhD, professor in the department of pediatrics at University of Washington School of Medicine, also mentioned the need for more robust endpoints in trials studying treatments for chronic cough.
“When we think ahead in terms of what data do we need, I think as long as we have a study drug that is at risk for potentially unblinding, then we need designs and we need endpoints that are robust to that,” Hamblett said. “Maybe it’s taste matching. If that’s not feasible, then we really do need to invest in more objective endpoints.”
With this potentially being the first FDA-approved drug for adults with chronic cough, Scott E. Evans, MD, FCCP, ATSF, professor and chair ad interim in the department of pulmonary medicine at The University of Texas MD Anderson Cancer Center, voted no because he wants to make sure he isn’t settling on treatment for these patients.
“I understand the need, [and] I’m sympathetic to the folks that presented today, but I do want to be careful and resist my own urge to think that something is better than nothing because I think we are establishing precedents here, and if we adopt the wrong markers and outcomes, I think we actually may limit our ability to identify the best drug,” Evans said.
At the end of voting discussion, Carvalho reiterated that although the majority voted against gefapixant, they recognize the need for approved therapies for those suffering with chronic cough.
“We all agree that this needs to be something that we continue … to pursue because we all know that these patients are highly uncomfortable and that their quality of life could be improved,” she said.