04-27-2020, 04:50 PM
I think there is a role for challenge trials, and though there may be some ethical issues, you can probably achieve equipoise in a low-risk population. That being said, while they are informative, I don't think they are a substitute for broad clinical trials, since challenges won't fully replicate the real-world way that COVID is transmitted. Some issues that a challenge trial can't tell you:
* What is the performance of the vaccine over time (in a challenge trial you typically perform the challenge at peak immunity)?
* Does the vaccine work with repeated exposures?
* Does the vaccine work with different types of exposures (large particles deposited in the oropharynx, small particles deposited in the lower respiratory tract?
* What is the vaccine safety in high risk groups (you probably can't ethically test a challenge model in patients with comorbidities or higher risk of death from COVID-19)
* How does the vaccine risk compare to the benefit, particularly in low COVID-19 risk populations?
To give you an idea, COVID-19 doesn't yet have a well-characterized IFR, but it's probably somewhere between 0.1% and 1%. For low risk groups (those under 40), the IFR is probably lower, such that those patients die due to COVID at a rate of between 1 in 1000 and 1 in 10,000. In order to prove the vaccine is safe enough in those low risk groups, you would have to show that harm due to the vaccine was lower than the disease it is preventing. In order to do that, as a rough guide, you need to test 3 times as many patients as the proportion of the outcome (the so-called rule of three). So, if the number of deaths in low-risk populations is 1 in 5,000, you would need to test the vaccine in at least 15,000 people to establish its safety.
That's not to say that there isn't a role for something like this. You might, for example, use a challenge trial to ungate vaccine deployment in very high risk settings such as nursing homes or for healthcare workers. In parallel with that, you would start a large conventional trial to establish traditional efficacy, as well as an open-label safety study (everyone gets vaccine) to make sure rates of adverse events in lower risk populations before you deploy it broadly.
BC
* What is the performance of the vaccine over time (in a challenge trial you typically perform the challenge at peak immunity)?
* Does the vaccine work with repeated exposures?
* Does the vaccine work with different types of exposures (large particles deposited in the oropharynx, small particles deposited in the lower respiratory tract?
* What is the vaccine safety in high risk groups (you probably can't ethically test a challenge model in patients with comorbidities or higher risk of death from COVID-19)
* How does the vaccine risk compare to the benefit, particularly in low COVID-19 risk populations?
To give you an idea, COVID-19 doesn't yet have a well-characterized IFR, but it's probably somewhere between 0.1% and 1%. For low risk groups (those under 40), the IFR is probably lower, such that those patients die due to COVID at a rate of between 1 in 1000 and 1 in 10,000. In order to prove the vaccine is safe enough in those low risk groups, you would have to show that harm due to the vaccine was lower than the disease it is preventing. In order to do that, as a rough guide, you need to test 3 times as many patients as the proportion of the outcome (the so-called rule of three). So, if the number of deaths in low-risk populations is 1 in 5,000, you would need to test the vaccine in at least 15,000 people to establish its safety.
That's not to say that there isn't a role for something like this. You might, for example, use a challenge trial to ungate vaccine deployment in very high risk settings such as nursing homes or for healthcare workers. In parallel with that, you would start a large conventional trial to establish traditional efficacy, as well as an open-label safety study (everyone gets vaccine) to make sure rates of adverse events in lower risk populations before you deploy it broadly.
BC