Are Data from Ebola Studies Still Being Misinterpreted?

IN 2014, CDC Director Tom Freiden testified to the US House Select Committee that there will not be an outbreak of Ebola in the US, as long as there are no mutations in this virus, and that “there are none”. Like any evolutionary biologist would, I found this testimony to be odd. But I found it especially odd, given that at the same time that I was listening to the testimony, I was analyzing the 396 mutations that had been found to date, many of which altered amino acids – meaning they could, in fact, change some aspect of how the Ebolavirus functions. The current pathogen is, in fact, 97.3% similar to the 1976 virus – that’s the genetic similarity between humans and orangutans.

Prior to the testimony, I had also read, from an important study published in the New England Journal of Medicine, of a mutation that had gone from low frequency to high frequency in the early phases of the outbreak. This behavior of a rare genetic variant in a growing population is consistent with natural selection driving the variant to fixation. It seems highly unlikely that a variant would drift to fixation by chance in this setting: the virus exists at population sizes of billions per person, and it undergoes repeated bottlenecks during transmission from person to person. Thus, rare variants are very likely to go extinct by chance, and are unlikely to be seen to increase steadily toward fixation.

I was so disturbed by the CDC’s testimony that I was motivated to write a book on Ebola. In the book, I also explore another conversation with a CDC scientist in which I, and scientists from around the country, were told that the current virus was “99.999% similar” to the 1976 virus, after I asked specifically about the 396 mutations published in the NEJM study.

Today, NIAID Director Anthony Fauci was quoted in USA Today as saying that it was “good news” that a new study conducted by the NIH showed that monkeys infected with the 1976 Ebolviruses got sick two days earlier than those infected with the 2014 virus.  This means that the virus was more virulent in 1976. Those involved in the study also interpreted their result as showing that the 2014 virus was not more virulent (deadly). I want to point out that these data also mean the virus has evolved.

In Ebola: An Evolving Story (the final draft of which was sent to the publisher in January, 2015), I included a chapter entitled “Ways in Which Ebola Guinea May Differ from Past Outbreaks: Evolution of Viral Phenotypes”.  The various viral phenotypes I present and explore in that chapter include:

  • Going Airborne
  • Shifting Other Routes of Transmission
  • Replicating More Slowly
  • Asymptomatic Transmission

Most readers will have already heard about various discussions on the likelihood of airborne transmission.  I actually lament (as many others have) the attention that airborne transmission has received, not because it is intrinsically unlikely to be a primary route of transmission in this epidemic, but for other reasons, including (a) “Airborne” is the incorrect term. We should consider all forms of aerosol-mediated transmission, including non-intimate droplet transmission; (b) consideration of this unlikely route has completely distracted from other, potentially more important changes in viral phenotype that might be useful to consider when studying the evolution of pathogenecity of viruses and other infectious agents.

I described how the current virus’ reported symptoms do not match those that historically killed humans via end-stage hemorrhaging, and how the current symptoms are consistent with a shift in the first rapid progression to the tissues of the gastrointestinal tract. I point out that a virus could become more effective at spreading in a population by replicating more slowly, especially if this phenotype is consistent with even small amounts of asymptomatic transmission, of which there is a hint of evidence in overdispersion in the distribution of days to first symptoms after last possible exposure (also presented in the book).  I also describe the types of censoring studies needed to rule out asymptomatic transmission.

Successful viruses are expected to become less virulent during transmission to new hosts, because this way they survive better in the new host – and stick around longer, as infected individuals may shed the virus for longer periods of time and thus are able to infect larger numbers of new host individuals. The data from the monkey study is therefore consistent with viral evolution being an important driving factor in the spread of Ebolavirus disease in the current epidemic, which is the opposite of Dr. Fauci’s conclusion.

Knowing the epidemiological, population ecological and population dynamics involved, I cannot see this as a “good thing”.

Care is needed in the interpretation of results from studies of a virus as dynamic as Ebola. One must think both at the molecular and ecological levels. Delayed replication may have a likely effect on the virulence, yes, but lower lethality to individuals means that the virus may, overall, result in higher mortality due to longer time and larger numbers of opportunities for transmission.

Purchase Ebola: An Evolving Story on Amazon.com. Paperback versions available from the Publisher’s website and from your local independent bookseller.

A portion of the proceeds of the sale of this book are contracted to be donated to Doctors Without Borders (MSF).

NewCover1_5 inches high 300 DPI

One thought on “Are Data from Ebola Studies Still Being Misinterpreted?

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