Speeding the Pace of Influenza Vaccine Development

Ingrid Hein

September 02, 2016

In influenza research, the gap between phase 1 clinical trials conducted to demonstrate the safety and immunogenicity of a vaccine and large efficacy trials involving tens of thousands of people is too long, according to researchers.

"This is holding back the development of novel influenza vaccines," Sarah Gilbert, PhD, from the University of Oxford, United Kingdom, said at the Options for the Control of Influenza 2016 Conference in Chicago.

There is another way to go about testing influenza vaccines, she told Medscape Medical News.

"What I'm proposing is a change to the clinical end point in influenza vaccine efficacy trials conducted for initial proof of concept of the protective efficacy," she explained.

Currently, to show efficacy in the field, patients are vaccinated with either the vaccine or placebo. During the following influenza season, participants report cases of influenza-like illness, and the influenza virus is verified with polymerase chain reaction (PCR) testing of nasal swabs.

The studies have to be large to detect a decrease in the vaccine group because the number of positive identifications in the placebo group will be low.

"It requires a minimum of 30,000 people to get statistically significant results, and often requires 70,000," Dr Gilbert said. If you need that many people to study vaccine efficacy, it becomes expensive. And if there aren't many infections that season, the study will be underpowered, she pointed out.

In the approach proposed by Dr Gilbert, researchers would conduct a small cost-effective clinical trial with about 2000 patients after initial safety and immunogenicity testing. They would compare the number of influenza-like illnesses that occurred during the influenza season in the vaccine and placebo groups.

Large trials are a big roadblock for vaccine development.

Only about half the illnesses will be caused by influenza, but if the vaccine is highly effective, the difference in the number of infections between the vaccine group and the placebo group will be large enough to detect. The researchers could also cross-reference their findings with independently collected data from the weeks when influenza is circulating in the study areas, and analyze all the results from just that time period.

The idea for this type of study arose when Dr Gilbert and her team were reviewing data from the large, multiyear Flu Watch Study of people who did and did not receive the seasonal influenza vaccine. The data showed that PRC-confirmed influenza is very rare in people older than 65 years, which makes this end point impractical for early efficacy studies.

The research team used the data from that study to conduct a novel phase 2b trial in which the number of influenza-like illnesses was the primary end point.

Although an influenza challenge study provided preliminary evidence of the vaccine's efficacy in healthy adults 18 to 45 years of age — the cohort least likely to suffer from severe disease or death after infection with the influenza virus — fewer than half the people in the placebo group became infected, making it difficult to detect efficacy.

But by looking at influenza-like illness, the researchers could assess the effect of the vaccine on all age groups in their relatively small initial study.

More Proof-of-Concept Testing

The research team concluded that any vaccine that produces a detectable change in any measure of immunity to influenza — such as T-cell response to nucleoprotein or antihemagglutinin stem antibody response — could be tested in this manner, which would allow proof-of-concept testing of different immune mechanisms in the target age group. It is also a way of measuring efficacy against currently circulating strains.

This method could help researchers decide whether to move forward with the big trial that is required before a vaccine can be licensed.

Even in large trials, sometimes the results are not evident. "Recording PCR-confirmed influenza requires that the nasal swabs are taken correctly, at the right time, are stored and transferred to a lab without damage, and that the lab has a very sensitive assay to detect the influenza virus in the sample," Dr Gilbert explained. "In real life, none of that will be perfect, and the number of detections of the virus by PCR will be lower than the number of infections that are genuinely caused by influenza."

By adding a step early in the process, a lot more vaccines could get tested.

"Large trials are a big roadblock for vaccine development. It's a huge barrier to go from a first clinical study to demonstrating efficacy in a large trial," she said. "I'm definitely not saying this is a final test — we still have to do big trials — but this would cost a whole lot less and would be more likely to happen. It would enable companies to get a first indication of likely efficacy of a novel vaccine before committing funds to a trial using PCR-confirmed flu as the clinical end point."

No Lab Confirmation

There will definitely be a lot more skepticism about efficacy when there is no lab confirmation, said Kathleen Neuzil, MD, from the University of Maryland School of Medicine in College Park.

It's going back in history to the way we used to do influenza trials before PCR.

"It's a little like back to the future," she said. "It's going back in history to the way we used to do influenza trials before PCR. We had nonspecific outcomes," she pointed out, "and a lot of skepticism.

However, this could be very useful for companies that want proof of concept before a decision is made to go forward and invest in a trial. "This could be a parameter they could use," Dr Neuzil added. "But if it's for regulatory or policy use, then there are different standards."

Is the ratio of lab-confirmed flu to influenza-type illness consistent from year to year? Dr Neuzil questioned. And "are there data to assess that?"

Dr Gilbert told Medscape Medical News that there is. "The data were collected over 5 years in Flu Watch, in the same age group, and that was part of the reasoning behind the trial design."

Dr Gilbert is a cofounder of and consultant for Vaccitech. Dr Neuzil has disclosed no relevant financial relationships.

Options for the Control of Influenza 2016 Conference: Abstract 054. Presented August 26, 2016.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: