In This Issue
Engineering and Vaccine Production for an Influenza Pandemic
September 1, 2006 Volume 36 Issue 3

Engineering and Vaccine Production for an Influenza Pandemic (editorial)

Friday, September 1, 2006

Author: Arthur H. Heuer

Editor’s Note

Every year, influenza A and B cause epidemics that lead to an average of 36,000 deaths and 114,000 hospitalizations in the United States alone. Periodically, new influenza strains evolve with the capacity to cause pandemics (epidemics that spread around the globe). The recent spread of avian influenza (H5N1) among birds and more than 200 cases of transmission to humans, indicating the potential evolution of a pandemic influenza virus, have stimulated a great deal of discussion of measures to combat it.

Pharmaceutical companies have been very successful at producing safe, effective vaccines to protect against seasonal influenza epidemics, but current methods may not be sufficient to produce large amounts of vaccine rapidly enough to combat a pandemic. Despite ongoing efforts to predict the nature of the next pandemic strain, correct identification cannot be assured in advance; thus, advance stockpiling of vaccine may not be helpful. In addition, the efficacy of available antiviral drugs, such as amantadine (Symmetrel™) and oseltamavir (Tamiflu™), is limited by resistant strains and a short window of opportunity for effective therapy.

Thus, the options for treating pandemic influenza are limited, and the main defense is still prophylactic vaccination. However, once a pandemic strain has been identified, it is estimated it will take at least five to six months using current production methods to produce enough vaccine to protect a substantial proportion of the population. Depending on many variables, including the seasonal timing of onset, a pandemic could spread and resolve within six months, leading to the grim prospect that vaccine would be available mainly for survivors.

Current vaccine production involves numerous time-consuming steps: reassortant virus must be generated and then grown in embryonated eggs (trivalent vaccines, the usual vaccine for seasonal [epidemic] influenza, require separate growth campaigns for the three different viruses); the virus must be purified and inactivated; the virus must then be split and the antigenic components enriched via biochemical procedures; the vaccine must be formulated to combine the three antigen preparations (for a trivalent vaccine) plus adjuvant. The last three steps—filling, packaging, and distribution—involve many engineering considerations that may not have been given enough attention. In addition, numerous quality-control tests must be done along the way (many of them time consuming and expensive) to ensure the safety and efficacy of the vaccine.

In a pandemic, resources and production capacity will be focused on producing a monovalent vaccine targeted for the pandemic virus, rather than a trivalent vaccine. Current worldwide production capacity for trivalent influenza vaccine is 300 million doses per year; a theoretical equivalent of monovalent vaccine would be 900 million doses, enough to vaccinate only 10 to 15 percent of the world’s population with a single dose. Decreasing the dose per person might be one way to stretch vaccine supplies, but this would have only a marginal impact on the time to delivery, and the effectiveness of reduced doses would have to be carefully tested.

Currently, vaccine producers have little excess production capacity because, under current conditions, industry would incur significant costs in building standby production capacity and maintaining standby readiness.

With these distressing facts in mind, a recent meeting, “Vaccine Production: Potential Engineering Approaches to a Pandemic,” sponsored by the National Academy of Engineering (NAE) and the Institute of Medicine (IOM) and hosted by Case Western Reserve University (CWRU), was held to explore issues related to vaccine production and alternative manufacturing processes. The meeting provided a venue for a “community” of academic and industrial engineers, government regulators, and other scientists and physicians to bring an engineering perspective to bear on critical issues in vaccine production. The meeting was attended by 275 participants representing 6 countries, 60 companies, and 44 universities. There were also about 1,000 web-cast viewers.

The papers included in this issue of The Bridge cover the major issues discussed at the meeting. The first article by Roy Anderson, a noted epidemiologist from Imperial College, London, is based on his keynote lecture, “Scientific Methods Underpinning Policy Formulation for an Influenza Pandemic,” which helped set the stage for the meeting. The next article, “Pharmaceutical Preparedness for a Pandemic,” is by Pat Scannon, chief biotechnology officer of XOMA, a small pharmaceutical company. He explains why it takes so long to introduce vaccines and other pharmaceuticals, identifies “choke points” inherent in current methods of pharmaceutical development, and makes an eloquent appeal to the engineering profession for help in meeting the challenges of an influenza pandemic. James Matthews of Sanofi Pasteur, a major vaccine manufacturer, describes “Egg-Based Production of Influenza Vaccine: 30 Years of Commercial Experience.”

The fourth article, “Cell-Culture-Based Vaccine Production: Technological Options,” by Rino Rappuoli of Novartis (formerly Chiron Vaccine), argues that cell-culture-based vaccine production is likely to obtain FDA approval in the future and to replace egg-based methods. Alan Shaw of VaxInnate, a small biotech company, then reviews “Alternative Methods of Vaccine Production,” all of which are still in the R&D phase (mostly R). At this point, it is difficult to predict which of these alternative technologies, if any, will make it to market.

The final article, “Adapting Industry Practices for Rapid, Large-Scale Manufacture of Pharmaceutical Proteins,” is by David Estell of Genencor, a medium-sized biotech company. He describes the biotechnology of industrial protein production (e.g., for commercial-grade enzymes), which produces materials that sell for ~$1/gm active protein; this is compared to vaccine production costs of ~$1,000/gm active protein. Estell argues that there are no obvious technological barriers to adapting industrial protein-engineering technologies to the production of vaccines and other pharmaceuticals. However, a major unanswered question is whether regulatory and other constraints will permit these commercial technologies to be used in vaccine production.

Based on the discussions at the meeting summarized in this issue, the real possibility of an influenza pandemic clearly raises enormous engineering challenges. I urge NAE and IOM to use their resources to promote more discussions and critical analyses to meet these challenges.

Note: I gratefully acknowledge the help of Clifford Harding and John Angus (both of Case Western) in preparing this editorial. Archived presentations from the workshop are available at The website also identifies 21 academic, government, and industry co-sponsors.

About the Author:Arthur H. Heuer is University Professor and Kyocera Professor of Ceramics, Case Western Reserve University, and a member of NAE.