Roche just heralded a great set of placebo comparisons for their new flu drug — too bad it flopped against generic Tamiflu

Genentech just rolled out some of the hard data around its Phase III CAPSTONE-2 trial for their new flu medicine baloxavir marboxil, and it doesn’t look good.

Once you dig through the bonny remarks and keep going on to the statistically significant placebo comparisons, the pharma giant throws in a major snafu. Their flu drug failed to prove significantly better than Tamiflu in a head-to-head comparison on the time it took to reduce symptoms of the flu. And that’s the money shot.

The median time to improvement of symptoms for baloxavir was 73.2 hours. The Tamiflu comparator arm: 81 hours for an awful p value of 0.8347. Roche optimistically called this outcome “numerically superior,” but a payer would say the relatively modest difference could have been due entirely to chance.

That’s not the sort of thing payers like to pay for, especially as cheap copies of Tamiflu abound, erasing a franchise Roche had hoped to keep. During epidemics, Tamiflu had achieved multibillion dollar sales figures. Now it looks to remain an occasional bestseller, at a fraction of the cost.

The fly in the flu data helps explain why peak sales forecasts for this drug have tended to vary dramatically. 

There is a wide range of data to illustrate how this drug does better than a placebo, but baloxavir won’t be competing against a sugar pill. In the meantime, Roche has a Christmas eve PDUFA date with the FDA and strong odds they’ll get an OK for a record-breaking year at the agency.

Roche got rights to this drug from Shionogi, which sells it in Japan.

On the bright side, Roche did have some good news for a subpopulation, with their drug beating Tamiflu in the influenza subtype B category, with a median time to improvement of 74.6 hours compared to Tamiflu’s 101.6 hours (p<0.05). Their drug also reduced “the time that the virus continued to be released from the body (viral shedding; median time of 48.0 hours for baloxavir marboxil versus 96.0 hours for both placebo and oseltamivir; p<0.0001).”

How about adverse reactions? There again, there was no significant difference, with baloxavir at 25.1% and Tamiflu at 28%. That too is weak, with the pharma giant falling back to its argument of numerical superiority.

Roche’s Genentech had repeatedly boasted that baloxavir was the biggest new thing to come along in the flu-fighting arena in 20 years. And they’re still saying it.

“This study adds to the growing body of evidence supporting baloxavir marboxil as a potential first-in-class antiviral flu treatment, and we plan to discuss these data with health authorities around the world,” Sandra Horning, Genentech’s chief medical officer and head of Global Product Development.

The best place to read Endpoints News? In your inbox.

Comprehensive daily news report for those who discover, develop, and market drugs. Join 44,900+ biopharma pros who read Endpoints News by email every day.

Free Subscription

Sr. Manager, Regulatory Affairs, CMC
CytomX Therapeutics San Francisco, CA
Marketing Associate - Demand Generation
Catalytic Data Science Charleston, SC
Associate Principal, Life Sciences Partnerships
Flatiron Health New York City or San Francisco

Visit Endpoints Careers ->