Influenza vaccines are the best weapons we've got against a disease that each year kills as many as a half-million people, including 3,000 to 49,000 Americans. Yet this season's worse-than-usual flu in the United States underscores the limitations of the existing vaccines.

Shortcomings include the inability to rapidly expand vaccine supply in the event of an especially bad flu and the need to vaccinate people with a new formulation almost every year as the virus mutates. These are problems enough when it comes to coping with the regular seasonal flu. They spell potential disaster in the case of pandemic flu, which occurs sporadically (most recently in 2009) when an animal strain of the virus jumps to humans.

Better vaccines are on the drawing board. Developing them will require a joint effort by government and the pharmaceutical industry.

Most flu vaccines are made using a 1940s-era process in which researchers forecast which strains will prevail in the coming season, and they are then grown in chicken eggs for six months. Manufacturers produce a predetermined quantity based on how much they expect to sell. In recent years, mild outbreaks have meant modest sales, but this year's harsher flu stimulated demand in the U.S., producing spot shortages in many areas.

Because the production process is so slow, vaccines against the 2009-2010 swine flu became plentiful only after the outbreak had subsided. Just a fraction of the doses made it to the developing world, and that was months later. Researchers estimate that swine flu killed as many as 575,400 people globally in just its first year. Vaccines also arrived too late to have much impact on pandemics in 1957 and 1968, which claimed 1.5 million and 750,000 lives respectively.

Innovations have already diversified manufacturing methods somewhat. A system developed by the Swiss company Novartis uses dog kidney cells to grow viral strains and thus mitigates the risk that an avian flu will devastate chicken flocks and restrict the egg supply. Another process, approved in the U.S. last week, may speed production. Using insect cells to grow the vaccine's active ingredient, Protein Sciences Corp. hopes to make doses in three months.

A remarkably fast and cheap option would be to propagate vaccine ingredients in E. coli, which reproduces itself every 30 minutes. The biotech company VaxInnate Corp., which has developed an experimental flu vaccine on this basis, estimates that from a single cubic meter of material as many as 400 million doses could be made for pennies a piece, compared with about $9 for egg-based varieties. This would help ensure that poor as well as rich countries could afford a rapid response to flu.

The greatest breakthrough would be a universal flu vaccine that would protect against all viral strains, eliminating the need for annual and pandemic inoculations. Researchers are experimenting with parts of the virus that don't mutate in the hope of creating vaccines offering lifelong or at least years- long protection.

The U.S. government pays for much of this research. Its scientists are working on a universal vaccine; it helped Novartis build a plant in North Carolina that uses cell-culture technology; and it is partially funding VaxInnate's Phase 2 trial.

Yet the government has limited means and little product- development experience. Making a new vaccine typically takes a decade and can cost $1 billion. A project of that size is better suited to large pharmaceutical companies. Most, however, have been loath to seriously invest in new vaccines, which offer low returns.

Given this market reality, the U.S. government should design incentives to get the industry more deeply involved, and it should encourage other countries with manufacturing capability to follow suit. The National Vaccine Advisory Committee should begin by asking industry leaders what it would take. Among the possibilities they should consider: tax credits for research and development costs, fast-track procedures for product approval, extensions for patents and periods of market exclusivity, and financial prizes for scientific breakthroughs.

By engaging big pharma in creating future flu vaccines, governments can ensure that a market failure doesn't lead to a public health catastrophe.