Issue Date: September 28, 2009
Flu Vaccine Race Against The Clock
The influenza pandemic caused by the novel H1N1 virus that emerged in April isn’t the one that was supposed to occur. For several years now, health agencies and governments around the world have been preparing to fight the pathogenic avian H5N1 virus. That virus circulates largely in Asia, and since 2003, it has killed 262 people out of 440 confirmed cases, according to the World Health Organization.
What the world faces instead, at least so far, is a more contagious, but much less deadly, virus to which almost no one is immune. These factors will result in transmission among one-third of the world’s population, WHO predicts. Antiviral drugs can go far in treating the disease and limiting its spread (see page 15), but immunization with a vaccine offers the best means to prevent H1N1 infections.
Racing against the clock, vaccine developers have succeeded in making an effective vaccine against the novel H1N1 virus. Although worldwide capacity for the vaccine is conservatively pegged at 1 billion to 2 billion doses per year, according to WHO, only a fraction will be ready for distribution before the flu pandemic peaks.
Flu vaccines are good at what they do, despite having to be created anew each year. This is necessary because flu viruses are a mutating, reassorted mix of subtypes and genetic components from pigs, birds, and humans. For each flu season, manufacturers create vaccines around what health officials believe will be the most widely circulating seasonal virus strains.
Flu viruses have the same basic form—most notably, they have a surface spiked with hemagglutinin and neuraminidase proteins. The pairing of 16 hemagglutinin and nine neuraminidase subtypes gives viruses their labels—for example, H1N1 or H5N1. The 2009 H1N1 virus is a unique variant that hasn’t previously circulated but has become the predominant form worldwide.
The outer proteins are the antigens against which the body raises an immune response. Vaccines typically contain inactivated or live, but weakened, viruses, in whole or in part, to present these antigens to the immune system. Even among flu viruses, a vaccine’s protective effect, not surprisingly, is the strongest against the viral strain to which it is matched.
The upshot is that seasonal flu vaccines, which were in production when the pandemic was declared in June, won’t protect against the 2009 H1N1 virus, pointed out Nancy Cox, director of the Influenza Division at the Centers for Disease Control & Prevention, during the meeting of a Food & Drug Administration advisory group earlier this summer. Any stockpiled H5N1 vaccines likely won’t work either.
Pandemic-preparedness planners and vaccine developers had to quickly shift gears to create new vaccines. Despite a lot of nail-biting along the way, recent clinical results indicate success. The next hurdle is manufacturing and distributing the vaccines in time to have an effect during the impending flu season.
On Sept. 15, just a few days after the National Institutes of Health and companies announced these initial results, FDA approved 2009 H1N1 vaccines made by Novartis, Sanofi Pasteur, MedImmune, and CSL. Data had shown that the vaccines induced a robust immune response in most healthy adults eight to 10 days after a single dose. GlaxoSmithKline is awaiting approval after reporting similar results. The European Medicines Agency (EMEA) is reviewing data for authorization as well.
Following up on the good news, the federal government is preparing for a voluntary national vaccination campaign. “With unprecedented speed, we have completed key steps in the vaccine development process: We have characterized the virus, identified a candidate strain, expedited manufacturing, and performed clinical trials,” said Department of Health & Human Services (HHS) Secretary Kathleen Sebelius in testimony on Sept. 15 before the House of Representatives’ Energy & Commerce Committee.
U.S. authorities were working with firms that already supply millions of doses of seasonal flu vaccines every year. FDA decided to treat 2009 H1N1 vaccines just as it would a seasonal “strain change,” which is to say it will allow new antigens to be put into vaccine formulations made using the same process as for previous years’ vaccines. Companies still must conduct limited clinical studies to get immunogenicity data on the correct antigen dose and dosing regimen for protection.
In Europe, the procedure is slightly different. As part of pandemic planning there in 2004, manufacturers developed vaccines and production capabilities using available influenza strains with pandemic potential, such as H5N1. After clinical evaluation and approval of these “mock-up vaccines,” fast-track licensing of similar vaccines using a real pandemic strain is possible with the submission of limited clinical data.
With the regulatory pathways more or less laid out, companies set about trying to make 2009 H1N1 vaccines. In late May, world health authorities provided a modified seed-virus strain to manufacturers, who then had to acclimate it to grow optimally in fertilized eggs. After being injected into the eggs, the virus is grown and then extracted for purification and inactivation. Moving from the virus strain to the first commercial lots is expected to take four to six months.
Although egg-based production has been used for decades, initial results for the 2009 H1N1 virus were disappointing to say the least. People assumed that the yield would be equivalent to that for seasonal vaccines, says Eric Althoff, Novartis’ director of global media relations. “And we are actually seeing around 50% of that,” he adds.
Other manufacturers were facing the same productivity problem, reporting yields 30 to 50% below normal to an FDA vaccine advisory committee in late July. At the time, it was unclear how large the doses would have to be, how many would be required, and thus what quantity of antigen had to be produced. “The total volume of vaccines can still be made, it will just take longer to make,” Althoff points out.
Novartis estimates that it can make about 150 million doses of pandemic flu vaccine per year. The company has discussed the supply of vaccines with about 35 countries, including the U.S., France, Switzerland, Spain, and the Netherlands. Likewise, as of early August, GSK had contracts with 10 governments totaling 291 million doses, most of which are slated for the U.S.
In May, FDA licensed Sanofi Pasteur’s new flu vaccine production plant in Swiftwater, Pa. The plant triples the company’s annual U.S. capacity to 150 million doses. The firm can produce another 120 million doses in France and will supply the U.S., Brazil, France, and other countries.
Although capacity is adequate for making seasonal vaccines, it falls short for pandemic needs. “The big issue of capacity is not the bulk manufacturing but filling individual doses,” explains Vijay B. Samant, chief executive officer of Vical, a San Diego-based biotech firm developing DNA vaccines. Finishing facilities are expensive to maintain, he says, “so nobody has a lot of idle capacity.”
CSL got FDA approval in August for its new vaccine-filling and -finishing facility in Kankakee, Ill. The Australian firm also has finishing plants at home and in Germany. MedImmune is trying to win approval for a second high-speed filling line that it says is required to meet its commitment to the U.S. government.
For MedImmune, which has a vaccine delivered as an intranasal spray, the problem has been the number of sprayers. Contrary to what others were reporting, “our yields have been very good,” Raburn Mallory, the firm’s senior director for clinical development, said at the July meeting. “Our capacity to produce bulk doses exceeds our capacity to fill them,” he added. Able to make about 200 million doses and fill only about 40 million, the company is working with regulators to put its vaccine into an alternative delivery device.
To facilitate filling and finishing the vaccines, vaccine producers have also turned to third parties, explained Robin Robinson, director of the HHS Biomedical Advanced Research & Development Authority, at a National Vaccine Advisory Committee meeting two weeks ago. To have vaccine ready as soon as possible, health authorities decided to start filling vaccines with 15 μg of antigen, as is used in seasonal vaccines, when bulk material became available but before trial results were out. In retrospect, he said, the decision was a good one based on the subsequent clinical results that this was enough.
Some vaccine suppliers are exploring alternative production methods. HHS awarded Novartis a $486 million contract in January to build a mammalian-cell-culture plant in North Carolina. With a capacity of 150 million doses, the plant is scheduled to open in 2012. Novartis already has a cell-culture plant in Germany and is licensed to use the technology in Europe. Cell-based methods are expected to shave weeks off production times because they avoid having to adapt the virus to grow in eggs.
Baxter Healthcare uses a Vero cell-culture method to grow the virus for vaccines that it makes in a facility in the Czech Republic. Although it is among the major vaccine producers to receive an EMEA mock-up license, the company is not licensed in the U.S. because it doesn’t have approval for a seasonal flu vaccine, a Baxter spokesman explains. He says that the company has purchase agreements for about 80 million doses from five countries, including the U.K., Ireland, and New Zealand.
Most producers had made enough bulk vaccine to start clinical trials by early August. In mid-August, however, health authorities announced that only about one-third of the doses expected, or about 50 million, would be ready in October (C&EN, Aug. 24, page 8). The plan is to first vaccinate the most at-risk groups: health care workers, pregnant women, and children and young adults with preexisting medical conditions. In the U.S., this amounts to about 160 million people.
Meanwhile, Chinese vaccine producer Sinovac Biotech was moving faster. After receiving the CDC seed virus, it readied a vaccine and started clinical trials in July. On Aug. 17, Sinovac announced that its vaccine showed good immunogenicity after one dose. About two weeks later, it won Chinese State FDA approval and got an initial government order to supply 3.3 million doses by Sept. 15. Chinese regulators have also approved vaccines from Hualan Biological Engineering and Shanghai Institute of Biological Sciences.
Vaccine experts were expecting that it would take two doses to get an immune response—one to prime and another to boost the immune system—because people lacked any previous exposure to the 2009 H1N1 virus. Authorities were only truly enthusiastic later when other trials gave corroborating results.
Although one dose appears to work for more than 90% of adults, early clinical-trial results suggest that children under age 10 will likely need two shots because their immune systems are more naïve to flu or flu vaccines, according to CDC officials. As occurs with seasonal flu vaccines, the response in elderly adults was less robust, at only about 60%, said National Institute of Allergy & Infectious Diseases Director Anthony S. Fauci when he reported the results in a Sept. 11 briefing.
Fortunately, the results were good even when the vaccines didn’t contain any adjuvant—an additive that enhances the immune response with a small amount of antigen. Oil-in-water emulsions are widely used as adjuvants in Europe and are being tested there in flu vaccines. Although HHS is interested in adjuvants in principle and has ordered supplies as a cautionary measure, FDA has never licensed an adjuvanted flu vaccine and would have to issue an Emergency Use Authorization to use one in the current pandemic.
At the July advisory committee meeting, Bruce Innes, who leads GSK’s global influenza development team, advocated for the use of adjuvanted flu vaccines. Not only does the adjuvant make antigen quantities go further, requiring as little as 3.75 μg per dose, he explained, but data from H5N1 studies also show that the response is protective even if the virus mutates slightly and its antigens change. Although the 2009 H1N1 virus hasn’t mutated over the past six months, researchers are concerned that it might.
Broadly protective vaccines, along with the speed of their development, are being addressed by new vaccine technologies. An inherent weakness in making vaccines in eggs or other cell cultures is the series of manipulations needed to adapt the virus and handle it during processing, Vical’s Samant explains. Vical considers its process safer because, he says, “we don’t handle the pathogens; all we need is the gene sequence.”
To make its DNA vaccines, Vical engineers plasmids to contain viral genes and then produces large quantities of them through simple fermentation methods. Within two months of receiving the 2009 H1N1 sequence, Vical was reporting results of immunogenicity testing in animals. It has an R&D agreement with the U.S. Naval Medical Research Center to develop its flu vaccine and move it into human testing.
After the plasmids are injected, the body produces viral antigens that elicit an immune response. Because the sequence used is exactly that of the target virus, and because the antigen hasn’t changed during adaptation or manufacturing, the immune responses should be strong and cross-protective against different strains, Samant says. Plasmids can also be stored for five or more years.
Similarly, Inovio Biomedical is working with NIH’s Vaccine Research Center to develop its DNA vaccines. Instead of using one viral gene sequence, the company constructs “consensus” genes from several strains. Calling it a “universal vaccine,” Inovio believes the resulting antigen mix will provide broad cross-strain protection against existing viruses and against new, unmatched viruses that may emerge, CEO J. Joseph Kim says.
Meanwhile, Protein Sciences and Novavax genetically engineer insect viruses to produce flu vaccine components in insect cells. Novavax’ viruslike particles (VLPs) contain three influenza proteins: hemagglutinin, neuraminidase, and an internal matrix protein. These VLPs are similar in size and structure to the virus, which is good for eliciting an immune response, but they are not infectious. Novavax made its first batch of H1N1 VLPs three weeks after receiving the gene sequence and manufactured bulk clinical-grade material for filling doses in about 11 weeks. Since June, it has been working with NIH to evaluate its 2009 H1N1 vaccine candidate.
Protein Sciences takes a similar route, but its vaccines contain purified recombinant hemagglutinin, which by itself takes on an immunogenic particle-like structure. The company is waiting for FDA approval of its seasonal flu vaccine, after which it hopes FDA might approve a 2009 H1N1 strain change so it can become a pandemic vaccine supplier, Chief Operating Officer Manon M. J. Cox says. Protein Sciences made its first batch in six weeks and can produce up to 200,000 15-μg doses in about a week at its 500-L facility.
Cox believes that scale-up should be relatively straightforward in standard facilities. In June, HHS awarded Protein Sciences a $35 million contract for the advanced development of its production method. The contract could be extended up to five years and a total of $147 million. If the company’s technology is approved for flu vaccine production, the contract requires it to set up manufacturing capabilities that can provide finished vaccines within 12 weeks and at least 50 million doses within six months of the start of a pandemic.
It’s unlikely that new methods will find use in the current pandemic, especially in light of the recent success for conventional vaccines. The robust response of clinical-trial participants to 2009 H1N1 vaccines is good news for vaccine availability and has eased worries about adequate supplies. Delivery will be easier because people need only one shot. And as Sebelius pointed out at the Sept. 11 briefing, many people should be immunized and protected well before Thanksgiving, a previous target date that would have been well into winter flu season.
Still, the 2009 H1N1 virus continues to spread rapidly, and for some, the vaccine will come too late. As it stands now, health authorities and vaccine makers are targeting mid-October, if not a bit earlier, for vaccinations to begin. HHS will be managing a rolling inventory from five producers through a central distribution channel.
Earlier in the summer, the U.S. alone contracted for more than $1 billion worth of bulk antigen, or about 195 million doses. At two doses per person, this would have covered only about 60% of the priority groups. Now, at one dose per person, it is enough for 60% of the entire U.S. population.
Just over a week ago, WHO applauded moves by nine countries—Australia, Brazil, France, Italy, Norway, New Zealand, Switzerland, the U.K., and the U.S.—to donate vaccines. Previously, Sanofi and GSK had earmarked a portion of their production for developing countries or donations to WHO. Given that current demand outstrips supply, the donations and pledges, WHO said, “will help increase supplies of pandemic vaccines to populations that otherwise would not have access.”
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society