Veru yesterday heralded a win in a small trial of its oral microtubule inhibitor VERU-111 in hospitalised Covid-19 patients, and consequently enjoyed a 28% share price bump. But a close read of the data reveals that the findings are not quite as rosy as the company claims.
The central issue is that, to claim a hit by the thinnest of statistical margins, Veru had to exclude one crucial patient from its primary analysis. The group confirmed to Evaluate Vantage that this patient had suffered respiratory failure, a component of the study’s primary endpoint. There still might be a signal in the data, but on current showing it does not stand up to a rigorous statistical analysis.
Veru also reported an intriguing mortality benefit with VERU-111. But this was not a primary endpoint, and conversely to make this claim the company did use the full intent-to-treat population, which in this case slightly flattered the numbers.
VERU-111 warrants further investigation in Covid-19, but whether the company can repeat the effect in an upcoming pivotal trial is unclear.
Veru said yesterday that its 39-patient phase II trial of VERU-111, in hospitalised Covid-19 patients at high risk for acute respiratory distress syndrome, had met its primary endpoint, the proportion of patients alive without respiratory failure.
According to the company, 94% of the VERU-111 group met this measure versus 70% of the placebo group; Veru said the difference only just met statistical significance, with p=0.05.
However, it revealed that this analysis was on a modified intent-to-treat population that excluded one of 19 patients given VERU-111. A spokesperson told Evaluate Vantage that this patient had only received one dose of VERU-111, but, crucially, had suffered respiratory failure and then recovered.
Thus, if this subject is included, the proportion of active cohort patients alive without respiratory failure becomes 17 of 19, or 89%. Given that the stated analysis was barely on the threshold of statistical significance, it seems clear that a difference of 89% versus 70% would have failed to clear the bar.
If the study was technically a bust, the admittedly interesting findings on mortality, a secondary endpoint, can only be considered exploratory. Veru yesterday pointed to an 82% relative reduction in deaths with VERU-111 versus placebo.
In this case the company did consider all 19 active cohort subjects.
|Summary of Veru's study of VERU-111|
|Intent-to-treat (ITT) population||19||20|
|Developed respiratory failure and died||1||6|
|Developed respiratory failure and recovered||1||0|
|Proportion alive without respiratory failure per ITT||89%||70%|
|P value for primary endpoint per ITT||Not given*|
|Relative reduction in death or respiratory failure per ITT||Not given**|
|Relative reduction in patient mortality per ITT||82%|
|*p=0.05 if using mITT population of 18 in active cohort;
**81% relative reduction if using mITT population of 18 in active cohort.
Drugs that prevent deaths in hospitalised Covid-19 patients are desperately needed; the only product that has shown a clear benefit here is the steroid dexamethasone.
VERU-111 is thought to combat Covid-19 by disrupting the microtubules that make up the cytoskeleton of cells, thus perturbing the “highways” through which the coronavirus is transported.
Veru now plans a 200-patient pivotal trial, with a similar design to the phase II study and the same primary endpoint. This is set to start in April and finish by the fourth quarter, so it will not be too long until it becomes apparent whether VERU-111 could indeed become part of the anti-Covid-19 armamentarium.
If it fails, Veru can always fall back on its original plan of developing VERU-111 for prostate cancer.