Topline data from a mid-stage trial of Bellus's chronic cough candidate BLU-5937 not having gone the way it wanted, the group is taking refuge in a subgroup analysis. Though a time-honoured technique for trying to salvage some upside from a failed trial, the data dredge has not mollified investors, and Bellus is nursing a 77% loss so far today.
The P2X3 receptor inhibitor’s failure was unexpected. Merck’s P2X3 inhibitor had already succeeded in phase III, and Bellus’s project was regarded by many as a safe bet. But binary outcome plays are always risky, and with no other projects in development the failure leaves Bellus few places to turn.
The Relief trial of Bellus’s asset enrolled 68 patients with chronic refractory cough, but was closed after 52 subjects completed treatment; Bellus blamed the early termination on the Covid-19 pandemic. None of the four doses of BLU-5937 tested showed a significant reduction in awake cough frequency versus placebo, and no dose response was seen.
Chronic cough is a hard condition to treat. In phase II only the highest dose of Merck & Co’s gefapixant was significantly reduced awake cough frequency. Still, this would be enough to put the Merck asset ahead even without its phase III hit, the magnitude of which has not yet been disclosed (Merck’s cough success highlights pipeline hopes, March 27, 2020).
Still, some had hoped that Bellus's candidate coulf overtake gefapixant, which has unpleasant side effects, altering or reducing patients’ ability to taste, and BLU-5937 appears to have a better profile. Effects on taste were seen in around 9% of the patients versus nearly half in Merck’s phase II trial.
BLU-5937’s better side-effect profile is believed to be related to its high selectivity for the for P2X3 receptor, and indeed the doses tested in Relief achieved estimated receptor occupancy comparable to or greater than gefapixant or Shionogi’s P2X3 inhibitor S-600918. However, it was pointed out on the call that cross-trial comparisons suggest that the tighter a molecule binds to the P2X3 receptor the worse its efficacy is.
Bellus’s damage-limitation strategy has been to play up what it claims is a success in one prespecified subgroup – subjects coughing at least 32.4 times per hour, which was the median figure in the total population at baseline.
In this 31-strong cohort cough frequency was reduced by around 30% versus placebo for all four ’5937 doses tested. Bellus assigned p values to these reductions, all of which were below 0.05, but in a post-hoc subgroup analysis of a failed trial these have no statistical relevance.
|Topline data from Relief trial of Bellus's BLU-5937|
|All patients (n=62)|
|Dose||Placebo-adjusted reduction in awake cough frequency||Stats|
|Patients with awake cough frequency at or above baseline median (n=31)|
|Dose||Placebo-adjusted reduction in awake cough frequency||Nominal stats|
|Source: company press release.|
Bellus now hopes to conduct a phase IIb trial in an “enriched” population of high-frequency coughers in the fourth quarter.
It also intends to push ’5937 forward in itching, with a phase II trial in chronic pruritus associated with mild to moderate atopic dermatitis planned for the second half of 2020. These two trials will cost around $20m and $10m respectively, management said on a conference call. The company had just under $80m of cash at the end of last year.
A phase II trial of Shionogi’s P2X3 inhibitor S-600918 could report a year hence, and the differing fortunes of gefapixant and BLU-5937 make the result hard to predict. It is clear that Merck remains at the top of the pile in chronic cough for now.