A study billed as the first large, randomized clinical trial of a quadrivalent (four-strain) influenza vaccine showed that GlaxoSmithKline's FluLaval Quadrivalent was about 55% efficacious in protecting children ages 3 to 8, similar to some previous findings for trivalent vaccines, according to a report published this week in the New England Journal of Medicine.
The trial also suggested that the vaccine was somewhat better at protecting children against "moderate to severe" flu, with an efficacy of more than 70% for such cases. But because of flu circulation patterns during the trial, it couldn't answer a key question about quadrivalent vaccines: whether they achieve the intended goal of improving protection against influenza B, compared with trivalent vaccines.
Unlike trivalent vaccines, quadrivalent vaccines contain two influenza B strains, with the aim of improving protection by targeting both of the common B lineages. In recent years it has been difficult to predict which lineage, Victoria or Yamagata, would predominate in any given season. But in the eight countries where the trial was conducted in 2011, very few B/Yamagata viruses were circulating, leaving it unclear whether the vaccine could provide broader protection.
Nonetheless, experts are hailing the new study as a worthy addition to the very short list of randomized controlled trials assessing the efficacy of inactivated flu vaccines in children.
The trial was sponsored by GSK, which won US Food and Drug Administration (FDA) approval of FluLaval Quadrivalent in August. The company also makes a second 4-strain flu vaccine, Fluarix Quadrivalent.
In a Dec 11 statement, GSK described the study as "the first large-scale clinical trial conducted specifically to review the safety and effectiveness of vaccinating children with a 4-strain flu vaccine." The company said the study was one of the pivotal studies leading to FDA approval of the vaccine.
5,000 children in eight countries
The trial was conducted at 15 centers in mostly tropical and subtropical countries: Bangladesh, the Dominican Republic, Honduras, Lebanon, Panama, the Philippines, Thailand, and Turkey. The authors recruited 5,220 children from ages 3 to 8 and assigned them to receive FluLaval Quadrivalent (QIV) or a hepatitis A vaccine as a control.
The QIV targeted influenza A/H1N1 (2009 H1N1) and A/H3N2 as well as the two B strains. The QIV group consisted of 2,584 children and the control group of an equal number. Members of the treatment group received one or two doses, depending on their vaccine priming status.
Recruitment of the children began in December 2010, and the authors conducted active and passive surveillance for flu-like illness for at least 6 months, through the end of October 2011. The trial's primary end point was influenza A or B confirmed by real-time polymerase chain reaction (rt-PCR). As secondary end points, the authors used PCR-confirmed moderate-to-severe flu and PCR-positive, culture confirmed flu. Moderate-to-severe illness was defined as a body temperature above 39 degrees C, acute otitis media, lower respiratory tract illness, or serious extrapulmonary complications.
The authors reported separate sets of results for the overall QIV and control groups and for the per-protocol cohorts, referring to children who were successfully contacted at least once after vaccination and who adhered to the protocol. The per-protocol cohorts consisted of 2,379 children in the QIV group and 2,398 in the control group.
A total of 563 flu-like illnesses occurred in the QIV group and 657 in the controls, the report says. Nearly all (96%) of these patients were tested for flu.
In the per-protocol results, the authors confirmed 58 flu cases in the QIV group (2.44%) and 128 cases in the control group (5.34%). That translated into a vaccine efficacy of 55.4% (95% confidence interval [CI], 39.1% to 67.3%). For the groups overall, vaccine efficacy was 59.3% (95% CI, 45.2% to 69.7%).
The researchers counted 14 moderate-to-severe flu cases (0.59%) in the per-protocol QIV group, versus 52 cases (2.17%) in the control group, which signaled a vaccine efficacy of 73.1% (97.5% CI, 47.1% to 86.3%) against such cases. For the overall group, efficacy against moderate-to-severe flu was estimated at 74.2%.
The authors estimated that this protection against moderate-to-severe illness was associated with 69% fewer medical visits, 75% fewer hospitalizations, and 77% fewer school absences among the vaccinees.
The team also found that the quadrivalent vaccine generated good immune (antibody) responses against all four flu strains.
Few B/Yamagata cases
As for influenza B strains, 23 B/Victoria infections occurred in the QIV group and 43 in the control group. Only two B/Yamagata cases were detected, both in the control group. This low number precluded a meaningful estimate of vaccine efficacy against that strain, but the immune response to the Yamagata strain was as high as the response to the other strains, the authors said.
"The true value of the QIV will be seen in years when the two B lineages are cocirculating or if there is a shift from one lineage to another between the time the vaccine is developed and the beginning of the influenza season," they commented.
As for safety, the report says there were no notable differences between the QIV group and the control group, except that pain at the injection site was more common in the QIV group (47.7% versus 34.8%).
The authors note that there is limited evidence from randomized trials to support the use of inactivated flu vaccines in healthy children. They say their results are in line with those of three other randomized trials of trivalent inactivated vaccines: 43% efficacy in children 6 months to 6 years old, 51% in children 18 months to 6 years, and 56% in children 3 to 9 years (H1N1 only).
"This randomized study provides additional evidence of the efficacy of QIV against influenza" as confirmed by PCR, the researchers say. They stress the vaccine's efficacy against moderate and severe cases, saying the greatest value of vaccination is in preventing such cases.
Study draws praise, with qualifiers
Lisa A. Jackson, MD, MPH, senior investigator with the Group Health Research Institute in Seattle, called the study "a welcome addition to the scant literature on efficacy of inactivated influenza vaccine in children. A well designed and well done study. They achieved a remarkably high rate of testing of children with ILI supporting the validity of the findings."
But she also mentioned a few caveats. One is that the majority of severe (hospitalized) flu cases in children are in those under age 3, who were not included in this study. "So we cannot say that QIV is 60% effective in 'children.' We know every little about vaccine efficacy in the under 3-year-old age group," she said in e-mailed comments.
"Also notable is that nearly all of the children who should have received 2 doses of vaccine did receive 2 doses," Jackson said. "This is in sharp contrast to the real world, where less than half do. So, again, this trial may not reflect real world vaccine effectiveness."
Further, she said the moderate-to-severe illness category "primarily reflected higher body temperatures, not truly severe things like pneumonia or hospitalization. The estimate of efficacy against this endpoint was not significantly different than the estimate against all influenza, a point not emphasized by the authors."
The study also drew praise from Michael T. Osterholm, PhD, MPH, director of the University of Minnesota's Center for Infectious Disease Research and Policy, which publishes CIDRAP News. He was the senior author of a major report on flu vaccines, released last year, and of a meta-analysis of flu vaccine studies published in the Lancet Infectious Diseases in 2011.
"It's a very well-done study, and it's very much welcome," he said, adding that only two previous studies met strict criteria for assessing the efficacy of inactivated flu vaccines in children.
Osterholm said the finding of 59% vaccine efficacy in the whole trial cohort points up, again, the need for better flu vaccines. "The fact that we're seeing only 59% reminds us that 41% were not protected," he said.
In addition, he questioned the assumption that quadrivalent vaccines are likely to improve protection against influenza B. He noted that studies by the Centers for Disease Control and Prevention (CDC) during the 2012-13 flu season showed no significant difference in protection against B/Victoria and B/Yamagata strains, even though the vaccine included only B/Victoria. Effectiveness against Yamagata was estimated at 64%, versus 56% against Victoria.
Neither the study authors nor the author of an accompanying editorial mentioned those CDC findings, Osterholm said.
He added that it is unclear whether the accepted model of how virus subtype affects vaccine effectiveness applies to influenza B the same way as it does to A. With type A, the hemagglutinin type (H1, H2, or H3) is very important for vaccine effectiveness, but it's not certain that the same is true for B lineages, he said.
"All the predictions about B [and vaccine effectiveness] were theoretical and based on serology, with no real data," he said.
Jain VK, Rivera L, Zaman K, et al. Vaccine for prevention of mild and moderate-to-severe influenza in children. N Engl J Med 2013; online publication Dec 11 [Abstract]
Baden LR. For an influenza vaccine, are two Bs better than one? (Editorial). N Engl J Med 2013 online publication Dec 11 [Full text]
Dec 11 GSK press release
Oct 15, 2012, CIDRAP News story about major report on flu vaccines