Winter 2018​

​In 1918, no one even knew for sure that influenza was a viral disease, but the field of public health laboratory practice was still in its infancy. One hundred years later, public health is in a much better place, but critical preparedness gaps still persist.

Protection against influenza, circa 1918. Soldiers wearing gazue masks walk down a street in Seattle, WA.

Protection against influenza, circa 1918. Soldiers wearing gauze masks walk down a street in Seattle, WA. Photo: Science Photo Library

by Nancy Maddox, MPH, writer

Coffin Supply is Exhausted; Undertakers Must Wait Turn for Caskets to Bury “Flu” Victims
Headline, The Courier-Journal, Louisville, October 23, 1918

“I thought I was going to die. I remember lying here in the front room and watching hearse after hearse pass by my window outside.”

So Barbara Burkett-Halapin, PhD, remembers her grandmother telling her as she recounted her illness with the influenza virus that ravaged the world in 1918-19.

Burkett-Halapin’s grandmother, whose experience is captured in the US Centers for Disease Control and Prevention’s (CDC’s) Pandemic Influenza Storybook, was lucky. She survived the virus at age 20 in small-town Shenandoah, VA. But an estimated 675,000 Americans—and 50 million to 100 million people worldwide—succumbed to the first pandemic influenza of the 20th Century, 100 years ago this year.

The toll of that pandemic—described as the “greatest medical holocaust in history”—was so great that it eclipsed the loss of 18 million lives in World War I and actually decreased the population of the US. Because the “grippe,” as influenza was also known, disproportionately killed those in the prime of life, it shortened US life expectancy by 12 years.
After the virus slowed in 1919, this particular lineage disappeared into the pig population—a known virus reservoir, considered a “universal mixing bowl” where avian, swine and human flu viruses reshuffle their genomes. Then, nearly a century after its deadly debut, a descendent of the 1918 virus re-emerged to prompt the first pandemic of the 21st Century—the 2009 influenza A(H1N1) pandemic. But this time the virus was substantially changed. And so was the world.

100 Years of Detective Work

In 1918, no one even knew for sure that influenza was a viral disease; the potentially grave respiratory illness was widely believed to be bacterial, perhaps caused by a species of Bacillus. But then, the field of public health laboratory practice was still in its infancy.

In 1930, scientists isolated influenza virus for the very first time, from a specimen taken from a pig. This classical swine flu strain was later dubbed influenza A(H1N1), with the “H” and the “N” referring to hemagglutinin and neuraminidase, two glycoproteins that sit on the surface of the virus and play a major role in its transmission and pathogenicity. (Hemagglutinin, additionally, is the major antigenic target that triggers the host’s immune response.)

In 1933, the first human influenza virus was isolated. With laboratory case confirmation now possible, the public health community made rapid gains. A second influenza genus, influenza B, was identified in 1940. Not long thereafter, the first vaccines appeared.

Today, the world enjoys a robust, laboratory-based Global Influenza Surveillance and Response System (GISRS), coordinated by the World Health Organization (WHO), which was itself established in 1948.

Yet, the threat of another 1918-type pandemic persists.

“I wouldn’t call it likely, but I wouldn’t be at all surprised if it happened,” said Pete Shult, PhD, who heads the Communicable Disease Division of the Wisconsin State Laboratory of Hygiene. “We have it in the back of our mind that it could happen at any time. That’s a prudent way to look at it.”

Said Daniel Jernigan, MD, MPH, director of CDC’s Influenza Division, “Influenza is as much of a foe now as it was [in 1918]. We can treat it better, we can prevent it better, we can detect it a lot better. But the potential for a severe pandemic now is just as great as it was then.”

The most worrisome emerging influenza today is the A(H7N9) virus, first detected in China in 2013. H7N9 is the only virus rated as having a “moderate-high” risk for emergence and impact based on the criteria considered in CDC’s Influenza Risk Assessment Tool, including (1) viral properties (e.g., antiviral susceptibility), (2) population attributes (e.g., existing immunity) and (3) viral ecology (e.g., infection rates in animals).

To fully appreciate the risk, it helps to recall the impact of past pandemics.

The 2009 A(H1N1) virus infected 24% of the world population and had a case fatality rate about 0.02%—a relatively mild outcome compared with its 1918 ancestor, which infected a third of the global population and killed over 2.5% of those stricken. In contrast, influenza A(H7N9) has a mortality rate approaching 40%.

Although human H7N9 cases have been mostly associated with exposure to infected poultry, the number and geographic distribution of cases expanded significantly between the first four Chinese outbreaks—involving 798 human infections during March 2013 to September 30, 2016—and the fifth outbreak—involving 759 human infections during October 1, 2016 to August 7, 2017. Should the virus acquire greater facility for human-to-human transmission, GISRS will face its biggest test yet.

“Still not where we should be”

On a scale of 1 to 10, from unprepared to super prepared, Jernigan rates the current state of US influenza preparedness a 5. Shult rates it a “5 or 6.”

But even a 5 is a vast improvement. WHO began its global influenza program in 1947. Today, the organization’s GISRS includes six major collaborating centers—including CDC and St. Jude Children’s Research Hospital in Tennessee (the only center focused exclusively on influenza ecology in animals) in the US—four regulatory laboratories responsible for verifying the potency and safety of vaccines and antivirals, and the 140 or so national influenza centers that collect virus specimens and submit them to their designated collaborating center for advanced antigenic and genetic analysis.

In the United States, after a 2003 SARS scare and a 2003-2004 H5N1 “bird flu” scare, the federal government pumped new resources into infectious disease preparedness. President George H.W. Bush released the National Strategy for Pandemic Influenza in 2005 and signed the Pandemic & All-hazards Preparedness Act the next year. Also in 2005, the US Department of Health and Human Services drafted its own pandemic influenza plan, which was updated last year.

Thanks to this focus and funding, said Jernigan, “a lot of stuff actually improved—surveillance, stockpiling of antiviral drugs, the beginning of making ‘pre-pandemic’ vaccines against viruses like H5N1, improvements in incident response management.” Among other things, CDC funded APHL to begin an ambitious project helping public health laboratories institute real-time, electronic influenza reporting to CDC. And, in 2008, CDC awarded a contract to establish the Influenza Reagent Resource (now the International Reagent Resource), through which authorized users worldwide can access influenza test reagents at no cost via a web storefront.

Many of these activities were still ramping up when—with all eyes looking to the East for threats like SARS and H5N1—the 2009 A(H1N1) virus made a surprise entry into the US via Mexico. It didn’t matter; the laboratory-based surveillance system worked. The Naval Health Research Center in San Diego identified the first two cases of the virus, and the San Diego County Public Health Laboratory identified the next three.

But while the response to that pandemic is considered a success, it clearly taxed the public health system. Shult said, “We had four to six years planning before the 2009 pandemic and we had PCR [in about 18 state labs]. But for a two-to-three-month period, we were running to our very max to deal with H1N1 in the laboratory and in public health, generally. That’s concerning, because by most people’s measures, that was a pretty wimpy pandemic.”

Since 2009, a lot has happened. Molecular, reverse-transcription polymerase chain reaction (PCR) testing for a panel of influenza viruses—with automated, real-time electronic reporting from laboratory information management systems to CDC—is now in place in all state and some local public health laboratories, plus a few US Department of Defense labs.
In 2010, APHL and CDC established three National Influenza Reference Centers (NIRCs): the Wisconsin State Laboratory of Hygiene, New York’s Wadsworth Center and the California Public Health Laboratory. These centers fill three critical roles:

  • Centralization of costly and complex viral culture, as well as drug susceptibility testing and other specialized work on cultured isolates. “It made sense to centralize viral culture in reference labs and get the public health laboratories focused on state-of-the-art PCR methods, which are hands down the most sensitive and specific method for flu diagnostics and subtyping [and don’t require culture],” explained Shult.
  • Since 2015, centralization of next-generation sequencing on the original clinical specimens, with raw data feeds coming off laboratory instruments straight to a processing pipeline on the APHL Informatics Messaging Services platform for cloud-based analysis. Viral genomes are then made available to data submitters, as well as CDC bioinformatics staff in Atlanta. “That has transformed the amount of information we’re able to get on influenza viruses throughout the [flu] season,” said Jernigan.
  • Provision of viral isolates to CDC so the agency can perform further antigenic characterization and grow the candidate viruses for the next season’s flu vaccine.

Thanks to the APHL-CDC “Right Size” project, launched in 2010, CDC can calculate just how many specimens must be collected and forwarded to the NIRCs each week of flu season to assure a representative selection of viruses for surveillance purposes—not too few and not unnecessarily many. The Wisconsin State Laboratory of Hygiene, for example, receives and processes 800 to 1,000 specimens/year from 17 states (AL, AR, IA, IL, IN, LA, KS, MI, MN, MI, MS, ND, NE, OK, SD, TX, WI) for national surveillance, plus additional Wisconsin specimens to support in-state influenza activities.

Nationwide, said Jernigan, roughly 70-80,000 specimens/year are tested for virologic surveillance. Of those, around 30-40% are positive for influenza. The three NIRCs receive a subset of those positive specimens and collectively submit 3,000 to 4,000 viral isolates to CDC each year.

Outside public health—in clinics, hospitals, long-term care facilities and other point-of-care sites—gains have also been made. Perhaps the most salient is the rise of the rapid influenza diagnostic test (RIDT). Although handheld, rapid test devices were around during the 2009 pandemic, they required human intervention to interpret results and were notoriously unreliable.
Current devices, said Shult, are machine read and “a big step up” from first-generation RIDTs. Additionally, the newest devices can report test results directly to the cloud, giving public health authorities access to real-time data. In Wisconsin, where Shult and colleagues are studying the devices, community RIDT sites send patient specimens to the State Laboratory of Hygiene for confirmatory PCR testing. So far, the scientists have found sensitivity in younger individuals in excess of 80% during early infection.

Yet, even with all these improvements—PCR, enhanced surveillance,  next-generation sequencing and  reliable  RIDTs—Jernigan said, “We’re at a better place than we were, but we’re still not where we should be.”

In the United States and worldwide, critical preparedness gaps persist:

  • Inadequate surveillance in birds and swine.
  • A limited selection, and limited availability, of antiviral drugs and other medical countermeasures.
  • Lack of low-cost, high-performing antivirals.
  • Inability to make vaccine quickly or to assure its effectiveness.
  • Suboptimal specimen-sharing among countries.

Many nations face additional challenges. For example, based on attainment of International Health Regulation goals, just a third of countries worldwide are ready for a pandemic response.
And although global influenza surveillance and information-sharing are improving—especially in China and Southeast Asia where virulent avian flu strains have arisen—inadequate laboratory capacity has left influenza “data deserts” across large swaths of the globe.

Seeking A True Game Changer

The holy grail of influenza prevention and control is a universal flu vaccine that would provide long-term immunity against all influenza viruses. And that is still a “good way off,” said Jackie Katz, PhD, head of the CDC-based World Health Organization (WHO) Collaborating Center for Surveillance, Epidemiology and Control of Influenza.

Instead, current influenza vaccines are strain-specific, targeting the viruses projected to predominate six to nine months or so after WHO issues its vaccine recommendation for the Northern Hemisphere in February and for the Southern Hemisphere in September. Those recommendations are based on a painstaking review of “over a thousand pages” of antigenic, genetic and other data from WHO collaborating centers and partners, including mathematical modelers. Yet it is always a gamble.

Katz said, “Current vaccines are based on an understanding of one type of immunity—a neutralizing antibody directed against the head of the hemagglutinin.” And that spiky protein head, she said, “is a moving target,” subject to continual change (“antigenic drift”) as a result of mutations that occur while the virus resides in host organisms and while it is grown in chicken eggs during the months-long vaccine production process.

“I think in most years we’ve had a good antigenic match,” said Katz. “But we’ve been wrong five times in the last 25 years for the H3N2 subtype, the most variable human influenza A virus, with the most recent mismatch in the 2014-15 season.”

To up the odds of success, WHO authorities have begun issuing recommendations for the traditional trivalent (three-strain) vaccine, plus a quadrivalent (four-strain) vaccine.

Thus, today 75% of US flu vaccines are quadrivalent, including two strains of influenza B (one from the Yamagata lineage and one from the Victoria lineage) and two strains of influenza A (typically an H3N2 virus and an H1N1 virus).

In addition, some years ago, the US Biomedical Advanced Research and Development Authority (BARDA) jumpstarted an effort to grow influenza viruses in cell culture to avoid the genetic and antigenic changes that may be introduced into egg-based vaccines. BARDA also invested in bringing a recombinant flu vaccine to the market, another non-egg-based technology.

But to achieve a true game-changer—a universal vaccine—scientists must think outside the box.

“We need a better understanding of the different immune mechanisms and how they can protect against the virus,” said Katz. “The strategy is to focus vaccine response on something that doesn’t change as much [as hemagglutinin].”

So far, she said, “We’re still just scratching the surface.”

In some ways, the 1918 pandemic was unique. Nearly half the flu-related deaths were in adults 20–40 years of age, and more than 99% were among those under age 65. The virus attacked in three waves within a 12-month period. And WWI conditions—overcrowding, mass movements of civilians and military personnel, and reluctance to shut down wartime production facilities—worked in the virus’s favor.

“Today we have vaccines, antivirals, ICUs,” said Katz, to mitigate the impact of any future influenza pandemic—
at least in the developed world. But Katz also recognizes the ease with which a modern traveler could “carry the virus to the other side of the globe.”

Assessing the current state of preparedness, Shult said, “We’ve learned a lot and that’s going to give us an advantage. But so much of the response will be just-in-time, depending on how [a pandemic influenza strain] emerges and begins to spread. Even a moderate pandemic is gonna be a challenge from a surveillance and diagnostic perspective, not considering [drug] stockpiles and vaccine development.”

He said, “Nothing presents the threat in term of scope and possible bad outcomes that influenza does. ...A good comparison is SARS, which emerged quickly, but with most transmission occurring with close personal contact; it didn’t really spread out into the community-at-large. Influenza certainly will.”