A missed opportunity

3 Oct

[Note: I originally wrote this article in Dutch for a newspaper for biologists (201409EbolaBionieuws) and translated it to publish here as well.]

"Ebola virus virion" by CDC/Cynthia Goldsmith - Public Health Image Library, #10816

“Ebola virus virion” by CDC/Cynthia Goldsmith – Public Health Image Library, #10816

Randomized placebo-controlled clinical trials are often large and very complex, but they don’t have to be, according to professor Joanna Masel. In an article for Scientia Salon, Masel argues that the few doses of ZMapp that were available to treat ebola patients should have been used for a small clinical trial. Even a very small trial is better than no trial at all. That this didn’t happen is a missed opportunity.

Ebola is a horrible disease and, as you can’t have missed, there is an epidemic going on in West Africa. About two-and-a-half thousand people already died of ebola in the last couple of months. There is no drug to treat ebola patients. At least, there is no drug of which we know that it works. There are a few experimental drugs, of which ZMapp is the best known. ZMapp consists of a mixture of three monoclonal antibodies. The press wrote a lot about whether ZMapp should be used and who should get the first available doses.

Because Ebola is associated with a very high mortality (>50%), experts agreed quickly that ZMapp should be made available to patients, even though it hasn’t gone through all the usual tests. Another question was who should get the first doses. If Africans would be the first to be treated, then the risk would be that it would be seen as using Africans as guinea pigs for a new drug. However, if non-Africans would be treated first, it could look as if whites were given priority. This is not easy decision.

ZMapp in a small trial?

Another discussion received less attention. ZMapp has never been tested in a clinical trial. But if we are going to give ZMapp to people, shouldn’t we use the opportunity to do a randomized trial right away? In such a study we can find out whether ZMapp is saving lives and we can potentially save lives at the same time. Professor Joanna Masel from the University of Arizona suggested the following: if you only have six doses , and many more sick people , then obviously most of the sick will not get the medicine. That’s terrible. But, she says, let us try to make the best of this bad situation. Let’s not pick six, but twelve people who qualify for the medications. And then lets give a placebo to half of the twelve and the real medicine to the other half. In this way we do a very small randomized, placebo-controlled clinical trial. And if we are lucky, the results of that trial will tell us whether the drug works.

But wait, we can hear the critics say, isn’t a study of twelve people far too small? In her article, Masel uses standard calculations (see also here) to determine whether a trial with twelve patients would make sense. It turns out that if the drug works extremely well, so that everyone gets the drug also survives, then in a study with twelve people we have an 80% chance that we find a significant difference between the placebo and ZMapp. And 80% is pretty decent for a clinical trial.

By now, the available doses have all been used. It is said that ZMapp has saved lives (link), but no one knows for sure. The patients who received ZMapp were probably in better condition and were treated in better hospitals than the average ebola patient. We don’t know what the real effect of ZMapp was. That’s a missed an opportunity.


There are simple tools that calculate the power of clinical trials. The screenshot is from a website http://www.sealedenvelope.com .


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: