Minerva claims 'positive' PhII data on depression drug while skeptics push stock down
It might have come in a bit of an unorthodox shape, but Minerva Neurosciences says it has a Phase II win in its hands.
When measured against the bar of a 2-sided type I error level of 0.1, the biotech said, one of three tested doses of its depression drug seltorexant passes muster for a statistically significant improvement over placebo. In their words:
The least squares mean (LS mean) difference from placebo of the change in [Montgomery-Asberg Depression Rating Scale] total score at the end of week 6 was 3.1 for the 20 mg dose of seltorexant, and the 2-sided p-value was 0.083, which is below the pre-specified 2-sided type I error level of 0.1.
The 20mg dose is the only arm that Minerva was prepared to report on. Investigators halted enrollment in the 40 mg group earlier, while the 10 mg arm had too few patients to be informative.
That sounds like cherry picking for some investors, sending shares $NERV down 7% to $6.42.
Jefferies analysts, on the other hand, endorsed the rosy view: “We think confusion on efficacy data caused the initial stock reaction, however, the data suggests a path forward w/ PIII starting early ’20.”
In the trial, the oral pill was given to adult patients with major depressive disorder alongside standard antidepressant therapies, including serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs), to which they have not responded adequately.
Minerva wanted to boost the response rates (only 30% to 40% by its count) by tinkering with the orexin system, which is associated with feeding, homeostasis, arousal, modulation of sleep-wake cycles and motivation. Janssen signed on as a partner to seltorexant, an ORX2 inhibitor, on two indications in 2014, but returned the rights for insomnia three years later.
The amended deal left Minerva solely in charge of developing seltorexant as an insomnia treatment. It took the opportunity to highlight the connection between the two conditions by singling out one of 19 secondary analyses: a subgroup with clinically significant insomnia. Those patients, who struggle to sleep, saw an LS mean difference versus placebo of 4.9 on the MADRS score with a 2-sided p-value of 0.050.
According to the Jefferies note, around 105 patients between the drug and placebo arms belonged in that group.
“While the 3.1 point difference in the overall population does not appear to differentiate from other antidepressants approved for adjunctive use (2.8 point difference from pbo observed on avg), the 4.9 point difference in the pts with significant insomnia appears clinically meaningful,” the analysts wrote.
A separate Phase IIb trial for insomnia has completed enrollment, CEO Remy Luthringer added with topline results expected later this quarter.