On listening to the podcast ‘From the Director’s Desk, a discussion between NIH Director Jayanta Bhattacharya and NIAID Director Jeffery Taubenberger’

Dr. Taubenberger is a distinguished flu researcher best known for resurrecting the 1918 Spanish flu virus. At 15:08 he says that …the 1957 pandemic, the so called H2N2 [virus]…The problem with that is since 1968 nobody on earth has seen a H2 virus. That virus was completely human adapted. It is not a select agent. We work with it under the same conditions as we would the 1918 virus, but that’s not an obligation and it probably sits in clinical diagnostic and basic lab virology freezers all over the world, and that really scares me actually because I think we’re concerned about how a new flu pandemic would occur but here’s one that was actually a pandemic. Now, the vast majority of the world’s population born after the 1968 pandemic does not have immunity to H2, and so that’s something I worry about.

Ouch, he’s both scared and worried, so let’s kick off with a little background. Pandemic flu viruses strike anywhere between 11 to 41 years after the last one, which means that we’re already in the next window: 2026 – 2009 (last pandemic) = 15 years. Flu viruses are characterized by two proteins on their surfaces called the hemagglutinin and the neuraminidase, aka H and N. H comes in one of 16 types numbered H1, H2 etc. while there are 7 for N aka N1 to N7. Pandemics result from a new flu virus with a different type of H protein. N is less important. By contrast, seasonal flu viruses are characterized by mutations within the same type of H protein.

The 1918 Spanish flu virus went by the epithet H1N1, that of 1957 Asian flu was a H2N2 virus while the 1968 Hong Kong flu virus was a H3N2 combination. The 2009 pandemic was a repeat performance by a H1N1 virus, albeit different to Spanish flu. For those interested in the genetics of these pandemic viruses, see “The 2009 Pandemic Influenza Virus: Where Did It Come from, Where Is It Now, and Where Is It Going?”

The Asian flu pandemic killed from 1-4 million people worldwide. It settled down as a seasonal flu virus until the Hong Kong flu virus showed up in 1968. Since then, nada. It went extinct naturally.

In the meantime, the world’s population has more than doubled, up from 3.5 Bn in 1968 to 8.2 Bn today. This means that most of the world’s population (around 84%) has never seen the H2N2 virus and so they are immunologically naïve. Those born before 1968, like yours truly, will have antibodies to the H2N2 virus although their immune systems are less robust due to age. In short, just about everyone would be vulnerable were it to come back.

Next, It is not a select agent. Select agents have the potential to pose a severe threat to public health and safety. They are so dangerous or worrisome that work with them must be highly regulated. In the case of the smallpox viruses work is limited to just two labs worldwide. The Department of Health and Human Services (DHSS) has oversight of select agents for humans in the US. For another take see Wikipedia.

The same goes for animal and plant viruses with the US Department of Agriculture playing top dog.

Select agents, aka viruses, bacteria and parasites come in for further classification, with the most dangerous being called Tier 1 agents. Not surprisingly this is where you’ll find the extinct smallpox virus along with Ebola virus and Nipah virus even though the latter has caused ‘only’ 435 deaths out of 754 infections to date worldwide. It’s the high death rate that’s scary.

The resurrected infectious Spanish flu virus is on the select agent list as is the rinderpest virus of cattle that was eradicated by vaccination the last case being documented in 2001 (Rinderpest). Like the smallpox viruses, rinderpest virus is a Tier 1 agent.

So you might have thought an extinct human pandemic flu virus would be on the select agent list like smallpox and rinderpest viruses, perhaps even a Tier 1 agent. Yet it is not, which makes no sense at all - that’s something I worry about.

As already mentioned in several essays, in the virocene viruses can come back by way of the freezer (A viral reawakening in China). And this is where we come back to Dr Taubenberger being scared - the H2N2 virus probably sits in clinical diagnostic and basic lab virology freezers all over the world, and that really scares me actually.

It just so happens that before March 2005 Samples of the influenza A(H2N2) virus were sent to 3,747 labs, the vast majority of them in the United States, the World Health Organization (WHO) said in a statement. They were inadvertently sent out by the College of American Pathologists. By April 20 2005, 98.8% of the samples had been destroyed.

If Dr Taubenberger is scared it is because of a lab leak or a lab accident leading to an infection of a researcher. We have seen that Lab Acquired Infections (LAIs) occur and are generally underreported, especially outside of the US which does a much better job. Not surprisingly, LAIs occur with select agents.

It so happens, again, that While a few H2N2 laboratory-acquired infections have been documented in the past, the likelihood of laboratory-acquired influenza infection is considered low when proper biosafety precautions are followed. The risk for the general population is also considered low. Without knowing when these LAIs occurred it is difficult to appreciate these lines from the World Health Organization. If they occurred between 1957 and 1968 then the risk was low for the virus was all around. If after, the whole world was at risk for a moment. Some very close shaves. Fortunately, it didn’t go bigtime.

It is always possible to eke out a little more data about the virus by continuing research on it. However, H2N2 is human adapted meaning highly efficient at respiratory transmission. A single lab acquired infection could put 8.2 Bn people in danger. Indeed, if it did a repeat performance today of what id did back in 1957 it would claim between 2-9 million lives. Catastrophic risk by any metric.

Yet wait a moment. This is a conversation between arguably the two most powerful people in infectious disease research in the US if only by virtue of the size of the budgets and influence at their command. The world has just gone through COVID and instead of saying something like ‘this scary situation should be cleaned up, indeed we could clean it up’, the conversation moved on.

The Federal list of Select Agents comes under review every two years. Human pathogens are managed by the CDC with input from the NIH which makes sense. So they could make a difference.

For some reason there seems to be a blind spot when it comes to the H2N2 virus.

• A 2022 paper used a multi-criteria decision analysis to try and reduce expert bias in the assessment of Select Agents, which was certainly worth trying. It seems that H2N2 wasn’t a strong candidate for the Select Agent list because it wasn’t a severe killer like, say, the Ebola virus. A virus that kills ‘only 0.5%’ of those infected it is often considered by many not to be a dangerous agent. Yet the key point is how infectious the virus is.

If 30% of 8.2 Bn people were infected by a virus that ‘only’ killed 0.5% that would result in 8.2 Bn x 0.3 x 0.005 = 12.3 million deaths. Ebola virus has killed ‘only’ 15,266 people between 1976-2020. Concentrating on virulence and putting aside transmissibility is a common fallacy (An evolutionary void).

• A May 2025 RAND Organization report entitled Identifying and Closing Gaps in the Federal Select Agent Program didn’t even mention H2N2 flu. That Dr. Taubenberger was worried and scared in August 2025 means there is at least one gap in the Federal Select Agent Program.

Conclusions

• It’s hard to imagine any lab acquired infection that carries more danger than human H2N2 flu because we know what the virus is capable of. It’s an agent of mass destruction. The risks massively outweigh any tiny benefits of continuing research on an extinct virus.

• Gents, use all the leverage you have to get human H2N2 flu virus on the Federal Select Agent List as a Tier 1 agent. And fast. Please. Other countries should do likewise.

• As it has been extinct for 56 years, embark on destroying stocks of the H2N2 virus and any tissues harboring the virus. This should have started years ago following smallpox and rinderpest jurisprudence. There is no excuse for keeping H2N2 flu virus stocks. The H2N2 genome is known so the virus could be recovered by reverse genetics if ever it was needed.

Aside 1

The conversation from 23:00 is also worth noting. They are talking about influenza in the context of pandemic preparedness. Dr. Taubenberger let’s drop we’re not at a point yet, we do not have the fund of knowledge to predict what even next year’s seasonal flu will be like, much less when the next pandemic will occur. So, in my view, the only thing we can do is react to a pandemic when it happens, sadly, but that’s the truth.

Exactly. We must react as quickly as possible to an exploding pandemic. There may be no vaccine or if there is one, stocks will be in the process of being expanded. Drugs might exist but we cannot bank on them. It depends on how transmissible and lethal the pathogen is. A million people might die before measures start kicking in. It’s horrible but this is the nature of severe pandemics. This is why it is so important for politicians and public health people to appreciate the dangers up front.

Don’t listen to those who pretend to predict the next pandemic or to have pandemics sewn up. Just as you’d not listen to the predictions of a soothsayer.

Aside 2

Experiments with influenza viruses containing genes or segments from 1918-1919 H1N1 (1918 H1N1), human H2N2 (1957-1968)… shall be conducted at BL3 enhanced containment (see Appendix G-II-C-5, Biosafety Level 3 Enhanced for Research Involving Risk Group 3 Influenza Viruses). The USG document is dated April 2024. Hence it is not clear what the remark We work with it under the same conditions as we would the 1918 virus, but that’s not an obligation. It could just be slip of the tongue as happens in data rich, fast moving conversations.