The World Health Organization held a conference to assess the status of testing and eventual licensing of two candidate Ebola virus vaccines. The agenda and list of participants and the final report are available. I was interested in the following list of key expected milestones:

October 2014:
Mechanisms for evaluating and sharing data in real time must be prepared and agreed upon and the remainder of the phase 1 trials must be started

October–November 2014:
Agreed common protocols (including for phase 2 studies) across different sites must be developed

October–November 2014:
Preparation of sites in affected countries for phase 2 b should start as soon as possible

November–December 2014:
Initial safety data from phase 1 trials will be available

January 2015:
GMP (Good Manufacturing Practices) grade vaccine doses will be available for phase 2 as soon as possible

January–February 2015:
Phase 2 studies to be approved and initiated in affected and non-affected countries (as appropriate)

As soon as possible after data on efficacy become available:
Planning for large-scale vaccination, including systems for vaccine financing, allocation, and use.

I wonder how a phase 2 study will be conducted, the goal of which is to determine if it is effective and further evaluate its safety. Will this be done in west Africa, where protection against Ebola virus infection can be assessed? If so, will there be controls who receive placebo?

If indeed an Ebola virus vaccine is our best hope in limiting the current outbreak, it won’t be distributed for a while, according to the optimistic expectations of WHO – assuming all proceeds on time, and that the results are favorable.


Dr. Francis Collins, the head of the National Institutes of Health, believes that we would have an Ebola virus vaccine if not for the past ten years of flat budgets for life science research:

NIH has been working on Ebola vaccines since 2001. It’s not like we suddenly woke up and thought, ‘Oh my gosh, we should have something ready here.’ Frankly, if we had not gone through our 10-year slide in research support, we probably would have had a vaccine in time for this that would’ve gone through clinical trials and would have been ready. (Source: Huffington Post)

I do understand that Collins needs to be a champion of life sciences research, but to promise that a vaccine would be ready by now is overly optimistic. Vaccines are not easy to design, as the efforts to make an HIV-1 vaccine illustrate. There is no guarantee that even unlimited resources would have produced an approved vaccine. However, more money might have allowed clinical trials of the Ebola virus vaccine candidates currently beginning phase I testing.

I believe that Collins should take the Ebola virus outbreak as an opportunity to emphasize the need for continuous, strong support of basic life sciences research. Michael Eisen, who is particularly annoyed with Collins’ statement, is right about what Collins should have said:

But what really bothers me the most about this is that, rather than trying to exploit the current hysteria about Ebola by offering a quid-pro-quo “Give me more money and I’ll deliver and Ebola vaccine”, Collins should be out there pointing out that the reason we’re even in a position to develop an Ebola vaccine is because of our long-standing investment in basic research, and that the real threat we face is not Ebola, but the fact that, by having slashed the NIH budget and made it increasingly difficult to have a stable career in science, we’re making it less and less likely that we’ll be equipped to handle all of the future challenges to public health that we’re going to be face in the future.

Don’t get me wrong. I get what Collins is trying to do. I just think it’s a huge mistake. Every time I see testimony from NIH officials to Congress, they are engaged in this kind of pandering – talking about how concerned they are about [insert pet disease of person asking question] or that and how, if only they could get more money, we’d be able to take make amazing progress. But guess what? It hasn’t worked. The NIH budget is still being slashed. It’s time for the people who run the biomedical research enterprise in this country to make basic research the center of their pitch for funding. Collins had a huge opportunity to do that here, but he blew it.



Enterovirus D68 by Jason Roberts

An outbreak of respiratory disease caused by enterovirus D68 began in August of this year with clusters of cases in Missouri and Illinois. Since then 691 infections have been confirmed in 46 states in the US.

The number of confirmed infections is likely to increase in the coming weeks, as CDC has developed a more rapid diagnostic test. Previously it was necessary to amplify the viral genome by polymerase chain reaction, followed by nucleotide sequencing to determine the identity of the agent. The new test utilizes real time, reverse transcription PCR which is specific for the EV-D68 strains that have been circulating this summer.

Since its discovery in California in 1962, EV-D68 has been rarely reported in the United States (there were 26 isolations from 1970-2005). Beginning in 2009 it was more frequently linked to respiratory disease outbreaks in North America, Europe, Asia, and Africa. It seems likely that the virus was always circulating, but we never specifically looked for it.

The current EV-D68 outbreak is the largest ever reported in North America. Enterovirus infections are not rare – there are millions every year in the US – but why EV-D68 has been so frequently isolated this year is unknown. One possibility is that the CDC, after the initial outbreak in August 2014, began looking specifically for the virus.

Sequence analysis of the EV-D68 viral genomes indicate that 3 different strains are involved in the US outbreak. These viruses are related to EV-D68 strains that have previously circulated in the US, Europe, and Asia. The sequences are available at GenBank as follows: US-IL-14/18952, US-KY-14/18951, US-MO-14/18950, US-MO-14/18949, US-MO-14/18948, US-MO-14/18947, and US-MO-14/18946.

Most of the illness caused by EV-D68 in the US has been respiratory disease, mainly in children. Five of the 691 confirmed EV-D68 cases were fatal, but whether the virus was responsible is not known.

There have also been some cases of polio-like illness in children in several states associated with EV-D68. In Colorado the virus was isolated from four of 10 children with partial paralysis and limb weakness. Previously there had been one report of an association of EV-D68 with central nervous system disease. In this case viral nucleic acids were detected in cerebrospinal fluid. EV-D68 probably does not replicate in the human intestinal tract because the virus is inactivated by low pH. If the virus does enter the central nervous system, it may do so after first replicating in the respiratory tract, and then entering the bloodstream.

There are no vaccines or antivirals to prevent or treat EV-D68 infection. Most infections will resolve without intervention save for assistance with breathing. As the fall ends in North America, so will infections with this seasonal virus.


Ledipasvir and SofosbuvirThe Food and Drug Administration has approved the use of a single pill containing two different antiviral drugs for the treatment for hepatitis C. It is the first combination pill approved for the disease, and also the first treatment that does not contain interferon or ribavirin.

The new hepatitis C drug, called Harvoni, is a mixture of the antiviral drugs ledipasvir and sofosbuvir. Ledipasvir (pictured) is an inhibitor of the hepatitis C virus protein NS5A, which has multiple roles in the viral replication cycle that include RNA synthesis and virus particle assembly. The mechanism of NS5A inhibition by ledipasvir is not known. Sofosbuvir is a previously licensed inhibitor that targets the viral RNA-dependent RNA polymerase. It is an analog of the nucleoside uridine, one of the four building blocks of RNA. Sofosbuvir is utilized by the viral RNA polymerase, leading to inhibition of viral RNA synthesis.

The use of single antiviral drugs (monotherapy) to treat RNA virus infections is always problematic because resistance usually arises rapidly. Dual-therapy pills like Harvoni are better, but the best are triple-therapy pills. Triple therapy formulations such as Atripla have been used successfully to treat infections with HIV-1, and presumably there will be mixtures of three antiviral drugs for treating hepatitis C.

Let’s use HIV-1 to illustrate the value of treating infections with multiple antiviral drugs. The HIV-1 viral genome, like that of HCV, is slightly less than 10,000 bases long. Assume that one mutation in the viral genome is needed for drug resistance. If the RNA polymerase mutation rate is 1 out of every 10,000 bases synthesized, then each base in the viral genome is substituted in a collection of 10,000 viruses. An HIV-1 infected person can make as many as 10,000,000,000 virus particles each day, so 1010/104 = one million viruses will be produced each day with resistance to one drug.

If we use two antiviral drugs, developing resistance to both occurs in every 104 x 104 = 108 viruses. In this case 1010/108 = 100 viruses will be produced each day with resistance to two drugs.

If we use three antiviral drugs, developing resistance occurs in every 104 x 104 x 104= 1012 viruses, which is more than what is produced each day.

This is why triple antiviral therapy has been so successful for the treatment of AIDS.

And yes, I’m sure someone has tested Sofosbuvir for inhibition of Ebola virus replication.


I cannot pass up the opportunity to point out this wonderful quote by Ginia Ballafante in her NY Times piece, Fear of Vaccines Goes Viral. The article starts by noting an article on plummeting vaccination rates in Los Angeles:

The piece had the virtue of offering New Yorkers yet another opportunity to feel smugly superior to their counterparts in L.A., because of course here on the East Coast we like our science to come from scientists, not from former Playboy models and people who feel entitled to pontificate about public health because they drink kefir.

As a scientist who works in New York, I can’t help but think that this is not entirely true. This idea is supported by a quote in the article from a New York City pediatrician, who says that 10 percent of parents in his practice express opposition to vaccination. If they oppose vaccination, they can’t be getting their information from scientists.

Here is the second best quote from the article, which comes from another New York City pediatrician after a discussion of current Ebola virus and enterovirus D68 outbreaks:

My feeling is that it will take something like that on a very large scale to get upper-middle-class people to realize that this is serious stuff…most of the deaths in the world are from contagious diseases. Not ISIS.


On episode #306 of the science show This Week in Virology, the Grand Masters of the TWiV discuss Ebola virus transmission, air travel from West Africa, Ebola virus infectivity on surfaces, the Dallas Ebola virus patient, and Ebola virus in dogs.

You can find TWiV #306 at

brincidofovirThe Liberian man who was diagnosed with Ebola virus infection after traveling to Dallas, Texas, was treated with an antiviral drug called brincidofovir. This drug had originally been developed to treat infections with DNA-containing viruses. Why was it used to treat an Ebola virus infection?

Brincidofovir (illustrated) is a modified version of an antiviral drug called cidofovir, which inhibits replication of a variety of DNA viruses including poxviruses and herpesviruses. When cidofovir enters a cell, two phosphates are added to the compound by a cellular enzyme, producing cidofovir diphosphate. Cidofovir is used by viral DNA polymerases because it looks very much like a normal building block of DNA, cytidine. For reasons that are not known, incorporation of phosphorylated cidofovir causes inefficient viral DNA synthesis. As a result, viral replication is inhibited.

Cidofovir was modified by the addition of a lipid chain to produce brincidofovir. This compound (pictured) is more potent, can be given orally, and does not have kidney toxicity, a problem with cidofovir. When brincidofovir enters a cell, the lipid is removed, giving rise to cidofovir. Brincidofovir inhibits poxviruses, herpesviruses, and adenoviruses, and has been tested in phase 2 and 3 clinical trials. The antiviral drug is being stockpiled by the US for use in the event of a bioterrorism attack with smallpox virus.

Ebola virus is an RNA virus, so why was brincidofovir used to treat the Dallas patient? According to the drug’s manufacturer, Chimerix,  with the onset of the Ebola virus outbreak in early 2014, the company provided brincidofovir, and other compounds, to the CDC and NIH to determine if they could inhibit virus replication. Apparently brincidofovir was found to be a potent inhibitor of Ebola virus replication in cell culture. Based on this finding, and the fact that the compound had been tested for safety in humans, the US FDA authorized its emergency use in the Dallas patient.

Unfortunately the Dallas patient passed away on 8 October. Even if he had survived, we would not have known if the compound had any effect. Furthermore, the drug is not without side effects and these might not be tolerated in Ebola virus-infected patients. It seems likely that the drug will also be used if other individuals in the US are infected.

Looking at the compound, one could not predict that it would inhibit Ebola virus, which has an RNA genome. RNA polymerases use different substrates than DNA polymerases – NTPs versus dNTPs. NTPs have two hydroxyls on the ribose sugar, while dNTPs have just one (pictured). The ribose is not present in cidofovir, although several hydroxyls are available for chain extension. I suspect that the company was simply taking a chance on whether any of its antiviral compounds in development, which had gone through clinical trials, would be effective. This procedure is standard in emergency situations, and might financially benefit the company.

Update: The NBC news cameraman is being treated with brincidofovir in Nebraska.


Poliovirus by Jason Roberts

Poliovirus by Jason Roberts

Vaccine-associated poliomyelitis caused by the oral poliovirus vaccine is rare, but its occurrence in a healthy, immunocompetent 6-month old child was highly unusual because the child had been previously immunized with two doses of the injected, inactivated poliovirus vaccine (IPV).

The three poliovirus vaccine strains developed by Albert Sabin (OPV, oral poliovirus vaccine) contain mutations which prevent them from causing paralytic disease. When the vaccine is ingested, the viruses replicate in the intestine, and immunity to infection develops. While replicating in the intestinal tract, the vaccine viruses undergo mutation, and OPV recipients excrete neurovirulent polioviruses. These so-called vaccine-derived polioviruses (VDPV) can cause poliomyelitis in the recipient of the vaccine or in a contact. During the years that the Sabin poliovirus vaccines were used in the US, cases of poliomyelitis caused by VDPV occurred at a rate of about 1 per 1.4 million vaccine doses, or 7-8 per year. Once the disease was eradicated from the US in 1979, the only cases of polio were caused by the Sabin vaccine.

To prevent vaccine-associated poliomyelitis, in 1997 the US switched to an immunization schedule consisting of two doses of IPV followed by one dose of OPV. The US then switched to using IPV exclusively in 2000. The child in this case essentially had a polio immunization course similar to that utilized in the US from 1997-2000: two doses of IPV, one dose of OPV. Why did the child develop poliomyelitis?

One clue comes from the fact that after the switch to an IPV-OPV schedule in 1997, there were still three cases of VAPP in 1998 and three in 1999. Another hint comes from a study of immune responses in children given multiple doses of IPV. Most of the children receiving two doses of IPV produced antibodies against types 1 and 2 poliovirus (92 and 94%), but only 74% of children produced antibodies against type 3 poliovirus.

The final piece of information needed to solve this puzzle is that the child in this case had vaccine-associated poliovirus caused by the type 3 strain, which was isolated from his feces.

Therefore, the child in this case most likely did not produce sufficient antibodies to type 3 poliovirus after receiving the two doses of IPV. As a consequence, when he was given OPV, he developed type 3 vaccine-associated poliomyelitis.

This case of VAPP could have been prevented: the child was born in Canada, and as customary in that country since 1995, he received two doses of IPV. At 5 months of age the child and his family visited China, where his parents decided to continue his immunizations according to the local schedule. China still uses OPV, so that is what the child received.


Incubation periodThe time before the symptoms of a viral infection appear is called the incubation period. During this time, viral genomes are replicating and the host is responding, producing cytokines such as interferon that can have global effects, leading to the classical symptoms of an acute infection (e.g., fever, malaise, aches, pains, and nausea). These symptoms are called the prodrome, to distinguish them from those characteristic of infection (e.g. paralysis for poliovirus, hemorrhagic fever for Ebolaviruses, rash for measles virus).

Whether or not an infected person is contagious (i.e. is shedding virus) during the incubation period depends on the virus. For example, Ebola virus infected patients do not pass the virus on to others during the incubation period. This fact explains why Tom Frieden said there was ‘zero chance’ that the passenger from Liberia who was diagnosed with Ebola virus infection in Dallas would have infected others while on an airplane. He had no symptoms of infection because he was still in the incubation period of the disease.

In contrast to Ebolaviruses, poliovirus and norovirus are shed during the incubation period – in the feces, where they can infect others.

Remarkably, viral incubation periods can vary from 1 or 2 days to years (Table; click to magnify). Short incubation times usually indicate that actions at the primary site of infection produce the characteristic symptoms of the disease. Longer incubation times indicate that the host response, or the tissue damage required to reveal the symptoms of infection, take place away from the primary site of infection.

The table was taken from the third edition of Principles of Virology. Missing from the table (which will be corrected in the next edition) is the incubation period of Ebola virus, which is 2 to 21 days. I would also argue that the incubation period of HIV is not 1-10 years, but 2-4 weeks, the time until the prodromal symptoms occur. The characteristic symptom of HIV-1 infection, immunosuppression, occurs much later.



Enterovirus D68 by Jason Roberts

In February 2014 I wrote about children in California who developed a poliomyelitis-like paralysis, also called acute flaccid paralysis or AFP. However, the cause of this paralysis was not known. The CDC has released its study of these cases and concludes “The etiology of AFP with anterior myelitis in the cases described in this report remains undetermined.”

A total of 23 cases of AFP* in California were reported to CDC during the period June 2012 through June 2014. These cases were from diverse geographic regions of the state. Specimens from 19 of the patients were available and tested for poliovirus, aroboviruses, herpesviruses, parechoviruses, adenoviruses, rabies virus, influenza virus, metapneumovirus, respiratory syncytial virus, parainfluenza viruses, Mycoplasma pneumoniae, Rickettsia, and amoebas. Rhinovirus was detected in one patient, and enterovirus D68 in two patients; all others were negative for potential etiologic agents.

All 23 patients with AFP also had anterior myelitis, inflammation of the grey matter of the spinal cord, which is characteristic of poliomyelitis. While the rate of AFP in California betweeen 1992-1998 was 1.4 cases per 100,000 children per year,  anterior myelitis was not described in any of 245 cases reviewed by CDC. However, poliovirus was ruled out as a cause in the 19 individuals who could be tested.

The cause of AFP is often difficult to determine because there infectious and non-infectious etiologies. Only 2 of the 19 clinical specimens met CDC guidelines for poliovirus detection (two stool specimens collected ≥24 hours apart and <14 days after symptom onset) and the others were likely taken too late to detect the presence of virus. The finding of enterovirus D68 in two of the samples is difficult to interpret, as the virus was detected in respiratory specimens and could have been a coincidental infection.

This investigation began with a request from a San Francisco area physician to the California State Department of Public Health to determine whether poliovirus was present in a 29 year old male with AFP and anterior myelitis. Subsequently this department posted alerts for AFP with anterior myelitis to  local health departments, and it is from the cases submitted that the 23 were drawn. Therefore the number of cases of AFP with anterior myelitis might be a consequence of this surveillance.

We are left with the unsatisfying conclusion that these 23 cases of AFP with anterior myelitis were either caused by an undetected infectious agent, or by something else.

*Defined by CDC as “at least one limb consistent with anterior myelitis, as indicated by neuroimaging of the spine or electrodiagnostic studies (e.g., nerve conduction studies and electromyography), and with no known alternative etiology”.