Jan 16, 2009 (CIDRAP News) – Picking the influenza strains to put in the flu vaccine each year is always a gamble, given the unpredictable prevalence of different strains, but choosing the influenza B strain has become particularly vexing in recent years.
Since the 2001-02 flu season, two distinct lineages of influenza B—Victoria and Yamagata—have been circulating in the United States, and experts have found it impossible to predict which one would dominate in any winter.
Because a vaccine against one lineage offers little protection against the other, the government's flu vaccine advisors have been discussing for several years the possibility of putting both lineages in the seasonal vaccine. That would turn the standard three-strain, or trivalent, vaccine into a quadrivalent vaccine. Trivalent vaccine contains two subtypes of influenza A—H3N2 and H1N1—and one of influenza B.
"It looks like we're stuck with two lineages for the time being," says Dr. Anthony Fiore, a medical epidemiologist in the Influenza Division at the Centers for Disease Control and Prevention (CDC). "So we can either ignore the one lineage or try to make another strain to protect against that lineage. There's just not much cross-protection."
Putting both B families into the vaccine would offer protection against both B strains, which is regarded as particularly important for children. But the idea raises questions about the effects on vaccine manufacture and supply, cost, and side effects. Some vaccine producers are studying the possibility, but they will need a formal government recommendation before they would proceed.
Vaccine often misses
The problem with B viruses is nicely illustrated by events of the past couple of years. For the 2007-08 flu season in the United States, the government's experts picked a Victoria strain for the vaccine. But by the end of the season, 98% of the 350 B viruses that were antigenically characterized by the CDC belonged to the Yamagata family.
With Yamagata viruses dominant last winter, the government's advisors, meeting last February, recommended a Yamagata virus for the B component of this season's vaccine. But the early returns this season suggest that Victoria strains may be back on top: 35 of 52 B viruses analyzed by the CDC through Jan 10 belonged to the Victoria clan. (But 30 of those 35 came from only two states.)
Looking further back, CDC records show the B strain in the vaccine missed the dominant B lineage in circulation in 4 of the 7 seasons from 2001-02 through 2007-08.
The spotty record has led to what Melinda Wharton, MD, MPH, of the US Food and Drug Administration's (FDA's) Vaccines and Biological Products Advisory Committee (VRBPAC), called "the annual agony of trying to flip the coin" to pick a B strain for the vaccine, according to a transcript of the panel's Feb 21, 2008, meeting.
The FDA committee is expected to consider the issue again when it meets in February to select flu strains for the 2009-10 US season. Fiore said the VRBPAC has asked the CDC's flu experts to make a presentation on the question at the Feb 18 session.
How much harm?
How much harm results when the vaccine misses the predominant B strain is not very clear, but experts say children are likely to get little protection against B viruses when this happens. The CDC now recommends that all children from age 6 months on up receive annual flu vaccination.
In general, B viruses cause less severe disease than influenza A, and they seem to be associated with smaller clusters or outbreaks, according to William Schaffner, MD, president-elect of the National Foundation for Infectious Diseases and chair of the Department of Preventive Medicine at the Vanderbilt University School of Medicine in Nashville.
Fiore agreed, then added, "But there are definitely deaths and severe outcomes caused by influenza B. It's generally thought of as being more a disease that circulates in children, but adults do get it and have hospitalizations and deaths from it. . . . For any given person you can have a severe illness with B and a mild illness with A. On a population level, A is more severe."
Schaffner said he himself had a nasty case of what he believed to be influenza B not covered by the vaccine last winter, when it was circulating in Nashville. "I was laid low for about 48 hours last year. I'm a poster child for the importance of doing this," he said, referring to putting both lineages in the vaccine.
Robert B. Couch, MD, a flu expert at Baylor College of Medicine in Houston, analyzed the influenza B issue and reported his findings to the VRBPAC at its February 2007 meeting.
Calling B viruses "the major cause" of an epidemic about every 2 to 4 years, he said, according to the transcript, "Infections occur in all age-groups. Illness is most prominent among older children and young adults. Illness in infants and young children appears to be more common among the low-income groups. Infections are prominent in the elderly in some epidemics, with excess mortality, but not in all epidemics. Overall impact is less than H3N2, but greater than H1N1."
Couch also examined research findings on how much protection the vaccine for one B lineage offers against the other lineage. He reported that protection is reduced in adults and more so in children. "By and large, very young children get very negligible benefit from the opposite lineage of the influenza B that's in the vaccine," he told the committee.
However, some experts suggest that more data are needed on the effects of B mismatches. "Evidence of the actual effect of mismatches on the flu B vaccine performance is still being gathered," John Treanor, MD, a flu vaccine expert at the University of Rochester in New York, told CIDRAP News. "Last year, it did look like the mismatch resulted in very little efficacy against B, but to make a major decision it may be necessary to accumulate more information from additional years of experience."
Why issue is heating up
The VRBPAC has been talking about the problem of the two B lineages for years. What seems to be heating up the issue now is the belief that both viruses are likely to keep on co-circulating indefinitely, plus an increase in the supply of flu vaccine.
All US-licensed flu vaccines are grown in chicken eggs, supplies of which are finite. But production has increased substantially in recent years with more producers entering the market.
Fiore said the surpluses of flu vaccine in recent years—millions of doses have gone unused in the last few seasons—make the possibility of a quadrivalent vaccine look more practical.
Donna Cary, a spokeswoman for vaccine manufacturer Sanofi Pasteur, agreed. "From a supply perspective, the increase in influenza vaccine capacity in recent years has made this option viable," she told CIDRAP News.
When the CDC's flu experts appear at the Feb 18 VRBPAC session, they expect to give an assessment of the potential impact of including both B lineages in the vaccine, according to Fiore.
"We have to weigh in the idea that there might be less vaccine available, against the fact that the extra strain might be preventing some extra amount of infection or hospitalizations or death," he said.
The CDC team looked at data from the past 10 years and tried to estimate the effects of having a quadrivalent vaccine in those years, Fiore explained.
"We haven't reached any conclusions about it yet," he said. "There are differences by year. In some years when we were in relatively tight supply, reducing the supply was a problem. Having a better vaccine didn't make up for the loss of supply.
"Now that we have an oversupply, the calculation may be changing," he added. "But as far as actual numbers of hospitalizations prevented, it's hard to come up with numbers."
Disadvantages and uncertainties
Adding a second B strain would not be without disadvantages, experts say.
"It would be more complex for the manufacturers, and it would probably cost more," said Fiore. And given that a producer can grow only so much vaccine, switching from three strains to four will mean fewer doses in the end.
Cary, of Sanofi Pasteur, acknowledged that going to a quadrivalent vaccine would reduce overall production. "But we do not anticipate any supply issue in transitioning to a quadrivalent influenza vaccine," she said. "This is due to the global increase in influenza vaccine supply, as illustrated by the opening of our new influenza manufacturing facility in the US in 2009," she added, referring to Sanofi's Swiftwater, Pa., plant.
Fiore said he doesn't expect the manufacturers would face major technical challenges in making vaccine with the two B lineages: "The manufacturers know how to make both of these lineages and have done it in recent years, so it's not a big manufacturing challenge."
Cary said Sanofi Pasteur is evaluating a possible transition to a vaccine with two B strains, including both technical and regulatory hurdles, but she declined to specify what they might be.
A spokesperson for MedImmune, maker of FluMist, the nasal spray vaccine, said the company is studying the possibility of making a vaccine with two B strains, but likewise declined to offer further details.
Other questions the prospect raises include the chance of increased side effects and whether clinical trials of a quadrivalent vaccine would be necessary.
The trivalent vaccine contains 15 micrograms (mcg) of each strain, or 45 mcg total. Assuming that the same dosage would be used for a second B strain, the total amount of antigen in a dose of a four-strain vaccine would be 60 mcg.
Fiore noted that the licensed prepandemic H5N1 vaccine in the US stockpile has been tested in 90-mcg doses. "So it's not that much of an increase in volume that we'd expect to see a lot more reactogenicity," he said. "But the FDA might require some studies about that."
There has been discussion of using just 7.5 mcg of each B strain, leaving the total B content in the vaccine at 15 mcg. But some experts doubt that the reduced amount would generate an adequate immune response.
Schaffner said another question is whether adding a fourth strain to the vaccine will leave the immune response to the other three unimpaired. "That's what we anticipate, but it'll have to be formally tested," he said.
Still another question is what the World Health Organization (WHO) will do. Experts convened by the WHO make recommendations for the flu vaccine strains for the northern and southern hemispheres each season, and the VRBPAC normally follows the WHO's lead.
The WHO's position is that much more information is needed, according to Gregory Hartl, the agency's infectious-disease spokesman in Geneva. "There's been a lot of discussion, but all the possibilities require safety and immune-response data which are not yet available, and after that there's regulatory approval which would be needed. So any change is not going to happen tomorrow," he said.
Couch said the FDA needs more input from vaccine makers to help sort out the issue. "Thus far, VRBPAC has only heard informal and 'soft' comments by industry," he told CIDRAP News.
Other ways to cover both strains
Offering a quadrivalent vaccine to everyone is not the only way to provide some protection against both B lineages. Several other options have been proposed, including alternating lineages from year to year, making a quadrivalent vaccine intended only for children and the elderly, and offering a supplemental B vaccine to cover the strain not in the trivalent vaccine. Couch described these in his VRBPAC presentation in 2007.
The idea of alternating lineages would be that people who get vaccinated each year might have some residual cross-protection in case the dominant strain in a given year is not in the vaccine, said Fiore. "But there's really no way to predict what the impact of that would be," he said.
Feb 14, 2008, CIDRAP News story "WHO advises total makeover for 2008-09 flu vaccine"
CDC's weekly flu surveillance report
CDC summary of 2007-08 season
Transcript of Feb 21, 2008, VRBPAC meeting:
Transcript of Feb 28, 2007, VRBPAC meeting with discussion of influenza B strains and vaccine
VRBPAC meeting documents site