The latest clinical trial flop from Novavax (NASDAQ:NVAX) wasn't nearly as bad as the failure it announced in 2016, but it was still a long way from what most vaccine developers would consider a success. There's still a glimmer of hope, though, because the respiratory syncytial virus (RSV) isn't like most viruses.
Despite decades of effort, we still don't have an effective vaccine that prevents the virus from spreading. Were Novavax's results good enough to convince the FDA its vaccine deserves a stamp of approval? Here's what you need to know.
How bad was it?
In 2016, older adult patients who received ResVax were hit by RSV-related lower respiratory tract infections (LRTI) slightly more often than those given a placebo. That's about as bad as it gets for a vaccine efficacy trial, but the company pushed on with a phase 3 study for expectant mothers. Investors will be glad to know babies belonging to mothers who were vaccinated were less susceptible to RSV LRTI than those who weren't.
Unfortunately, the protective benefit observed wasn't strong enough to convince regulators of anything good, and a much larger trial will almost certainly be necessary to move ResVax forward. Novavax reported a 39% efficacy rate for ResVax, which might seem pretty good to the uninitiated. Unfortunately, 4,636 subjects aren't nearly enough to be sure a vaccine efficacy rate as low as 39% will provide a benefit in the real world.
What the FDA wants to see
The primary end point of the Prepare study was the prevention of RSV infections severe enough to cause abnormal breathing or low levels of circulating oxygen. Investigators are 97.5% sure that ResVax's efficacy rate is somewhere between -1% and 64%, which is an extremely wide confidence interval that regulators probably won't accept.
Potential new vaccines that don't have an available standard for comparison need the low end of their confidence intervals to reach a level that's still likely to provide a clinically meaningful benefit. Novavax didn't state what minimum the FDA might be looking for, but we can be sure that it's a positive number.
Novavax also reported results that included additional data ascertained from hospitalization records, which makes it look as if a lot of patients who went to the hospital with something that might have been a severe RSV infection weren't included. Including extra data, ResVax had a 41% efficacy rate, and investigators are 95% confident that vaccine efficacy in the real world would fall between 16% and 58%. That isn't a great confidence interval, but you could probably argue a vaccine that's 16% effective at prevention provides a clinically meaningful benefit.
How much is this going to cost?
Unfortunately, the FDA's rarely willing to bend the rules in a way that could result in millions of expecting mothers receiving useless vaccinations. That means Novavax will need to run a new study with ResVax that includes a definition of "medically significant" RSV infections that will give statisticians more to work with.
If regulators don't loosen up on the definitions of medically significant RSV infections, Novavax will need to run a trial with more patients than it can possibly afford to enroll. Calculating just how many patients you need to produce acceptable confidence intervals is tricky, but a vaccine that's 50% effective at stopping an infection that 3% of the placebo group catches needs to enroll at least 10,450 subjects. Severe RSV infections were reported in less than 2.4% of unvaccinated infants during the Prepare study, which means a 10,450-subject study probably wouldn't be nearly large enough.
Where does that leave Novavax?
There's a good chance that ResVax provides some preventive benefit, but Novavax will never be able to run a study large enough to prove it. After burning through $135.4 million during the first nine months of 2018, Novavax finished September with just $145.6 million in cash and investments.
Investors hoping a big pharma with deep pockets will look at the Prepare results and decide to take a big risk partnering with Novavax will probably be disappointed. During the Prepare study, around 998 test subjects were enrolled in the U.S. and they didn't do well. In fact, severe RSV infections were 32.7% more common among vaccinated infants in the U.S. Without a stronger 57% efficacy rate among 2,179 patients enrolled in South Africa, overall results from the Prepare study would have been an unequivocal failure.
The odds of Novavax salvaging something from this failure are worse than before, but that probably won't stop the company from trying to raise more cash by touting numbers that look impressive but don't really mean anything to regulators. While there's nothing necessarily wrong with highlighting interesting secondary objectives that have been met, they aren't going to convince the FDA to approve ResVax.