The current flu vaccine shortage in the United States is symptomatic of the slow and cost-prohibitive process of research and development for vaccines. Government involvement has made that enterprise far less profitable, and one consequence is fewer manufacturers of vaccines. Legislators should adopt policies that encourage companies to make vaccines more quickly and less expensively, such as applying federal liability preemption to vaccine products and reimbursing people for getting vaccinated rather than having government purchase vaccines outright.
The World Health Organization is aiming to fix our nation's problems with the production of flu vaccines. In early November, the international health agency convened an unprecedented summit meeting of flu-vaccine makers and government health ministers. The predictable outcome was a flotilla of proposals for government-run solutions to the growing threat of a flu pandemic, among other issues.
The World Health Organization is late to this game. So is the U.S. government, which had ample warning that the market for flu vaccines had nearly collapsed under a withering combination of regulated prices, runaway liability suits, and an overabundance of caution by the Food and Drug Administration, according to a 2004 report by the General Accountability Office. Everyone involved in the current debate is searching for solutions in their rearview mirror.
It is time to look forward, especially at the opportunities offered by improved technology for developing and manufacturing vaccines. And while this year's shortage of flu vaccine is dominating public policy discussions, the same systemic problems that created this crisis plague the entire market for vaccines against infectious diseases. In that regard, flu vaccine is a good window into why the research and development (R&D) of new, improved vaccines has been painfully slow compared with other sectors of drug and medical technology markets.
In fact, observers do not have to search hard to find the sources of this predicament. The first few pages of a 1985 Institute of Medicine (IOM) report on "Vaccine Supply and Innovation" lament "technical problems, high R&D costs, the expense and logistics of clinical testing and surveillance of reactions, the risk of litigation over untoward events associated with vaccine use (whether causally related or not), and limited sales" as among the pressures weighing on this market. What the IOM said in 1985 is still a good description of the causes of the flu vaccine shortage that dominated headlines during the closing days of this year's presidential race. That these same conclusions formed the core of a report issued about twenty years ago should provide no comfort that these problems are anywhere near being resolved.
Instead, the problems have been getting worse for a long time. From 1967 to 1984, the number of U.S. vaccine manufacturers fell from 37 to 15, while the number of licensed vaccines declined from 380 to 88. Twenty years later, we are down to three large manufacturers in the United States and only a few dozen vaccine products.
Even if these troubles with the U.S. vaccine market are not new, the public's frustration over this situation is, and that has motivated policymakers to seek quick fixes to calm an irritated electorate. Policymakers should tread carefully, because the vaccine industry is in its current state precisely because of bad policymaking, and the current ideas being circulated for how to fix it do not hold out much hope of righting the industry's fundamental woes.
What Has Gone Wrong?
As everyone now knows, the current public anger was triggered by news that nearly half of the nation's expected supply of flu vaccine will not be available this winter because the British government suspended the manufacturing license at the factory in Liverpool that makes it. As a result, Chiron Corporation, the American biotech firm that owns the factory, announced that it would not be able to supply any of the almost 50 million doses it had planned to sell in the United States.
The British action appeared hasty and born mostly from process violations that the British inspectors found inside the plant. Without evidence of widespread contamination, U.S. regulators--were they in charge of the inspections--would likely have found ways to filter some of the Chiron flu vaccine and salvage some of the serum (assuming Chiron would have been willing to take on the added liability risks). Getting past the finger-pointing, the real culprit can be traced much farther back. Failed policies have created a fragile market for most new vaccines, where science has stagnated and left the United States too reliant on a few large suppliers and too dependent on antiquated and needlessly expensive production processes.
Pharmaceutical giant Merck & Co. abandoned the flu vaccine business almost twenty years ago. Wyeth pulled out last year after throwing away about one-third of the flu vaccine it made for the 2002-2003 season because it did not sell. That leaves just three established companies in the business--the French firm Aventis Pasteur SA, Chiron Corp. (the company whose entire output for the United States had to be destroyed because of contamination), and MedImmune, Inc.
Why didn't a lot of the big companies get back into the vaccine business? Most of the vaccine industry long ago ceased to behave as a competitive industry focused on creating better products and being able to earn profits through that innovation. In many cases, vaccine makers are constrained to producing largely the same vaccines according to government specs, using manufacturing processes that have been held in place for years by regulations, and selling their finished product mostly for a single government price. These impediments are especially evident when it comes to flu vaccine.
The end result was the near-dissolution of a once pioneering industry. And if the ingredients that led us here sound familiar, that is because the same ideas are embedded in many of today's plans for how to "solve" the problem of high drug prices. The vaccine industry's fate is worth taking stock of because the rest of our drug industry is heading in the same direction. Bad history has an awful way of being repeated, as do bad government policies. If we are not mindful, we will visit on the rest of the drug industry the same failed policies that saddled the vaccine makers and killed off a once vibrant industry.
Aggressive Liability Discourages Investment. Vaccines are given to millions of healthy people, and as biologics (which are grown as living organisms rather than being synthesized chemicals), they are subject to substantial uncertainties in the manufacturing process and require careful handling. These distinctions make them easy targets for tort litigation on behalf of anyone who suffers--or even thinks they suffer--any sort of illness after vaccination.
After litigation nearly drove childhood vaccines from the marketplace in the 1980s, they were salvaged by congressional action that greatly restricted the scope for liability by establishing a science-based, no-fault compensation system. This Vaccine Injury Compensation Program (VICP) was enacted in the late 1980s through the collaboration of various groups, including legislators, the American Academy of Pediatrics, and parent-advocacy lobbying groups, to make sure that the children who might be injured as a result of a routinely recommended vaccine would be quickly, easily, and appropriately compensated. Prior to the enactment of the legislation, litigation led to shortages as manufacturers left the marketplace.
Recent developments, however, threaten to expand liability beyond the boundaries of the no-fault compensation plan, thus undermining the stability of the childhood vaccine market itself. The plaintiff bar has been exploring ways around the system, including litigation after the time limit for non-litigation expires, merging of many individual claims for under a thousand dollars into class action suits, and recent thimerosal litigation. The National Vaccine Advisory Committee working group has responded with recommendations to maintain and even strengthen the VICP. These measures would expand the number of vaccines covered and recognize that the entire mix in the vial should be considered the vaccine-that is to say, the vaccine consists of not only the active component, but also preservatives and additives that make immunization effective.
Even so, the United States is down to a single manufacturer for some of the country's most important childhood vaccines, compared with more than a dozen before the liability crisis. Once companies got out of this business, they lost specialized skills and workers, and reacquiring these unique capabilities became a high barrier to re-entry. Especially at today's vaccine prices, it has become prohibitively expensive to invest the resources needed to get back into the business. One consequence is that only three firms were manufacturing this year's flu vaccine for the United States when the shortage hit.
Heavy Regulation Thwarts Innovation. Another problem plaguing the market for new vaccines is the way they are made. Across the board, the United States has a very expensive process for regulating the manufacture of complex biologics. This regime is at its pinnacle when it comes to flu vaccine because companies must effectively gain approval for a new product every year. But while techniques and equipment for manufacturing other biologics have advanced dramatically, allowing companies to improve quality and reduce costs, the processes for manufacturing vaccines--especially the flu vaccine--have often been held in place by excessive regulation of development and manufacturing.
Chicken eggs are at the heart of a protracted process of producing flu vaccine, a practice that has changed little since the first flu shot was introduced in the 1940s. It takes between six and eight months to collect millions of these delicate incubators from special chicken farms, and then use them to grow the viruses that make up the next season's vaccine. That means the United States needs to guess, nearly a year in advance, what the prominent strains of flu will be and how many lots it should purchase.
This technique is hardly ideal, and new technologies--reverse genetics and mammalian cell culture--promise to improve on the cumbersome and costly process. With cell culture, a flu virus is made to infect a cell line able to replicate itself over and over in large fermenting vats filled with special growth nutrients. This production process bypasses the egg as a factory altogether. The advantage is it can be scaled up more quickly to meet rising demand, and it does not rely on chickens, so it can be done in a shorter time frame, perhaps shaving off a few months or more from the total development time. It should also be less expensive, encouraging more vaccine makers to enter the market for flu vaccine.
In reverse genetics, the traditional process for isolating the gene that causes the infection and then creating the killed vaccine particles that compose the vaccine is turned right on its head. Regarding flu vaccine, this technique is being applied to the development of experimental vaccines against SARS. In the conventional egg method, two flu strains with the preferred features for a new vaccine are injected into an egg, and their genes re-assort naturally. Researchers sift through the 250-plus possible combinations of viruses to find one that displays the desired proteins on its surface--the right hemagglutinin (HA) and neuraminidase (NA) antigens--and is still able to grow inside an egg. Checking each re-assorted virus to identify one with the desired gene sequences takes a lot of time, however. From the moment a new strain is identified, it can take two to three months before the vaccine production process in eggs can even begin. Reverse genetics cuts down on that time dramatically. Researchers clone individual gene sequences of the virus, hand-selecting the sequences for the two viral surface proteins to combine with master strain sequences. The only thing that comes out is the particular combination you are looking for. There is no need to go through a lot of selection of individual viruses. The technique also gets around another hurdle. Some strains of flu, like the Spanish flu, SARS, or of course, the avian flu (which typically kills chickens) are so pathogenic that they kill chicken eggs, making it impossible to grow it to the volumes and concentrations needed for commercial-scale vaccine development.
Widespread adoption of these technologies is within our grasp. Polio vaccine is already made using mammalian cell culture, as are some pediatric vaccines. And reverse genetics was used last year to recreate a SARS virus. New cell-culture methods, which rely on genetic engineering, are already being used to make smallpox vaccine, but they still have not been adequately demonstrated and validated for flu vaccine production.
With smarter regulation aimed at enabling the science to surmount the evident and theoretical concerns with these new processes, flu vaccines developed using the processes could be on the market in a relatively short time frame--a few years in some best-case scenarios. Baxter International, a medical products company, recently announced that it is finishing the preparations for a plant in the Czech Republic, where it plans to produce a flu vaccine using the same cell-culture technology it used to produce a smallpox vaccine stockpile. The technology is based on vero cells, a cell line that is originally derived from African green monkeys. The company expects to launch the new flu vaccine in Europe in late 2005 or early 2006 and to begin seeking approval in the United States at about the same time.
A flu vaccine made from Chiron's cell-culture method is entering phase III studies of effectiveness in Europe. (Phase III studies are the large-scale, definitive studies of effectiveness and safety that companies use in seeking marketing approval.) Chiron announced that it has filed an application with the FDA to begin studies on its cell-culture vaccine in the United States.
One might ask why the Europeans are, in some cases, ready to approve vaccine technology that is not close to FDA approval. The answer is that regulators inside the Food and Drug Administration view these techniques nervously and maintain tight, and sometimes inefficient and inhibitory, regulations on the manufacturing of new biologics like vaccines. Among other things is the theoretical fear that genes could jump from the cell cultures into the vaccine. Instead of standing askance, our regulators should work proactively to develop the science that will give them comfort to certify this process.
There is a precedent for the FDA itself developing the science to prove that a product is safe in the face of theoretical concerns. When monoclonal antibodies were first being developed there were concerns that retroviruses found in these drugs could pose a human health hazard. Development of these drugs was stymied. The FDA took the lead in developing methods to test for the presence of these viruses and to demonstrate when these products could be certified as safe. As a result of this scientific work, the entire industry was able to move forward, and countless useful products have been brought to the market.
Of course, manufacturing tools and processes can come from the private sector, too. But incentives can be weak when the necessary work includes basic research whose benefits will be spread among many entities, including competing manufacturers. The FDA's work on validating the safety of monoclonal antibodies is an example of the most useful application of government spending on science--developing tools and processes that facilitate and accelerate the fruition of private development efforts, rather than spending government money on erecting competing development projects that crowd out private R&D investment.
Right now, the capital required to get into the business is a real barrier to new suppliers. It costs about $300 million to build a chicken egg factory, and it takes almost five years to get a facility built and inspected because of all of the specialized equipment that is needed and the frequent inspections required at every stage of construction. And unlike the economics of drugs, with vaccines grown in chicken eggs, the variable costs--those that need to be paid every year on each vaccine that is produced--are high.
The end result is that without the ability to innovate more easily through development of cheaper and better production methods, the only way vaccine makers can improve profitability is by squeezing manufacturing costs, which is almost impossible to do in the current regulatory environment.
High Costs and Low Prices Limit Profit Potential. Building factories to FDA specifications is only part of what makes the vaccine business expensive. Vaccines also pose unique inventory and logistical challenges, long production cycles, higher storage and handling costs, shorter shelf lives, and seasonal peaks in demand.
Vaccine manufacture also requires highly specialized equipment and production materials, and a highly specialized and uniquely skilled workforce to operate this equipment. This specialization competes with the same skill set needed to produce other biologics, which are high-margin businesses. Many vaccine operators have bachelor's degrees in science (although this is not the practice at all firms). Pharmaceutical manufacturing does not require highly educated operators. Also, support engineers for vaccines are more likely to have Ph.D.s while pharmaceutical engineers tend to have master's degrees. The vaccine production labor force, therefore, must be well paid. In the past, companies have worked to spread out the high fixed costs involved in building these sophisticated plants and hiring specially trained workers by producing many different kinds of vaccines that could share production space and workforce.
As various segments of the vaccine market have become unprofitable through government-influenced prices and litigation threats, vaccine makers have been forced to exit individual lines of vaccine development. Making a profit has become very difficult. While there are between one and two dozen small, start-up companies working on developing new vaccines against infectious diseases, most of these companies are pursuing niche products like vaccines against sexually transmitted diseases or products with broad regulatory and government support such as AIDS medications. Even the vaccine makers going after more traditional targets are having a difficult time attracting the kind of venture capital needed to sustain research and development. (See table 1.)
Investors have shunned the early-stage vaccine market . . .
Early Stage Vaccine Companies, Selected Venture Capital Investments, 1994-2004
|Company||Total Investments (in millions)|
|Protein Sciences||$ 2.0|
|Vaxin||$ 11.0 *|
|Targeted Genetics||$ 16.0|
|Therion Biologics||$ 90.0|
*Excluding government grants
SOURCE: BioCentury, American Enterprise Institute
. . . even while money has flowed into other
sectors of biotech
Selected Venture Capital Investments
by Business Category, 1994-2004
|Business Category||Total Investments (in millions)|
|Gene/Cell therapy||$ 1,145.8|
*Excluding VC vaccine funds (totaled at $250.6 billion)
SOURCES: BioCentury; American Enterprise Institute
To these ends, the single most significant event to discourage new investors may well have been the Vaccines for Children program (VFC) signed into law by President Clinton. The VFC is a vaccine entitlement program for eligible children aged eighteen and younger. Originally crafted in response to the 1989-1991 measles epidemic in the United States, the program was finally passed as part of the leading edge of Clinton's failed health care program. In addition to the program's stated goals (such as improving child vaccination rates), the VFC program sought to implement a framework for government intervention in the segment of the drug market that would provide the least political opposition and the most plausible case for the government's role.
Under the VFC, the government would contract directly with the manufacturers themselves and, in the largest public-private partnership involving the Centers for Disease Control (CDC), handle distribution to that population of children who often went unvaccinated. The VFC went into effect on October 1, 1994, and, as of 2002, had served approximately 41 percent of the childhood population with the following vaccines: diphtheria, haemophilus influenza type b (Hib), hepatitis A and B, measles, mumps, pertussis (whooping cough), pneumococcal disease, polio, rubella, tetanus, and varicella (chickenpox).
But government involvement in procuring pediatric vaccines meant that this line of vaccine development soon became even less profitable and therefore less attractive for vaccine makers. It made it harder for companies to assemble a portfolio of profitable vaccines over which to amortize the high cost of vaccine production.
The role of government in driving down vaccine prices, and therefore profit margins that fuel continued R&D and attract competing manufacturers, is apparent throughout the vaccine industry. The standard flu vaccine has been made into a commodity product similar to generic drugs with minimal seller margins despite the clear difficulties in manufacturing and distribution. Taking advantage of the fact that injectable flu vaccines are not patentable, the federal government has become a large purchaser at low prices, and manufacturers are expected to sell at the same very low prices to clinics, hospitals, physicians, and pharmacies. The government uses its monopsony buying clout with no frills to extract the lowest possible price on the vaccines it procures--even when the target is an innovative vaccine under patent. Recently, a manufacturer's proposed price of $58 a dose for its pneumococcal vaccine was rejected by the CDC, which demanded a discount of more than $10 a dose.
All these market flaws have come into play in the flu vaccine saga. Buyers avoided binding contracts for supplies, and manufacturers had no incentive to produce extra supplies to hedge against disruptions. Hastily mounted trials have reinforced earlier studies that demonstrated a strong likelihood that supplies could easily be more than doubled by employing different methods to inject smaller, but still effective doses. It is hard to imagine, however, that manufacturers feel free to charge the higher prices necessary to make such late adjustments and to promote what they had achieved.
A Four-Step Plan for Reform
The vaccine market is not inherently unworkable. Where patents are available and markets work normally, innovative and profitable vaccine development is possible. This has been demonstrated by the successful introduction of Wyeth's Prevnar, which prevents pneumococcal (viral pneumonia) infection, and by Wall Street's favorable reaction, presumably in anticipation of future profits, to successful trials for Merck's new vaccine for HPV, a common infection that causes certain sexually transmitted diseases and is a leading cause of cervical cancer.
Spending more government money to purchase vaccine stockpiles in bulk is not the answer. Neither is ratcheting up retail vaccine prices by a few bucks, nor giving more money to government research arms so that they can try to develop their own vaccines.
Some critics of private vaccine development have recently cited the case of the inhaled flu vaccine FluMist as proof that the public sector can develop its own vaccines, given sufficient resources. This view misunderstands the economics of vaccine development. FluMist was indeed based on a University of Michigan discovery dating back to the Johnson administration. The private biotechnology company Aviron licensed the vaccine technology in 1995 from the National Institute of Allergy and Infectious Diseases, where it had already undergone discovery research and preliminary clinical testing. Aviron subsequently spent seven years and endured numerous setbacks developing the nasal-spray flu vaccine, including clinical trials involving over 20,000 subjects and the building of a manufacturing and quality infrastructure that ultimately required hundreds of people and hundreds of millions of dollars. That investment--almost $1 billion by the time the product was launched in 2003--came entirely from the private sector. More importantly, it was fraught with risk, not just in the science (Will it work? And will it be safe despite using a different form of the flu virus?) but in market reaction. MedImmune, the firm that purchased rights to FluMist and brought it to the market in 2003, lost a great deal of money that first year and has yet to break even, as it has had to pursue additional clinical trials to obtain broader FDA approval.
The answer to the problems of the flu vaccine market rests not with more government efforts to produce vaccines but with policies aimed at freeing companies to innovate so they can produce better vaccines that will command premium prices, and so they can adopt new ways of making the vaccines more quickly and less expensively. And then, if the government is still itching to spend some money on solving the problem, it can start by refining the basic science that will give it comfort enough to sign off on the improved manufacturing processes that have been percolating in the industry for years.
We propose the following four-step plan for creating an environment for new vaccine development:
- The government should spend money on research into solving regulatory challenges, not manufacturing vaccines inside the NIH. Many of the problems that plague the development of new vaccine technology flow from regulatory uncertainty about demonstrating the safety and effectiveness of new products, as well as a general overabundance of caution. Many of these regulatory concerns stem from technical challenges in proving that vaccines are sterile and safe, and that production methods are predictable and reproducible. Better technology for demonstrating these things could allow for faster, more certain development pathways. The FDA needs to adopt such technology to keep pace with the growing sophistication of the vaccine products that manufacturers would like to present to the regulatory agency. To the extent that European regulatory agencies have appropriate technology in place, the FDA should adopt it.
- In order to limit frivolous and non-science-based liability claims, federal liability preemption should apply to all vaccine products approved by the FDA. The government's solution for childhood vaccines saved that market, but that solution is being undercut by creative trial lawyers and only applies to a small slice of the vaccine market. Liability protection needs to be extended to any vaccine that targets an infectious disease capable of being transmitted person-to-person. The creation of these products is a public good that ends up producing benefits for all Americans, whether they receive the products or merely benefit from the protection of living in an environment where other people have received them. As a public good, it deserves protection from frivolous torts and state juries that are not proficient in the complex science at stake.
- The federal government should reimburse people for vaccinating themselves rather than purchasing vaccines outright. Monopsony buying by federal agencies ends up setting artificially low prices that get adopted by other purchasers and drive down the profit margins on new vaccines. Rather than cutting costs by setting prices, government should use targeted subsidies to make vaccines part of the safety net for low- income consumers, with the goal of advancing the "herd immunity" that makes vaccines a social good.
- Public health officials should shed their bias for interchangeable vaccines and little product differentiation, and regulators should allow companies to promote new vaccines based on scientifically proven advantages of new vaccines. Public health agencies and regulators have had a bias against differentiation in this market. Instead, they need to encourage it. Without the ability to produce better proprietary vaccines with qualities that can appeal to consumers, such as longer protection windows, easier administration, greater efficacy, or improved safety, vaccine manufacturers are unable to earn premium pricing that contributes to the profit margins that attract new capital and greater R&D. Having more unique products would also encourage direct-to-consumer and professional advertising and would drive more people to their doctors seeking vaccines.
The vaccine market is not inherently unworkable. Giving serious manufacturers incentives to assume the high fixed costs needed to get into this business will require policy solutions aimed at creating a truly competitive market that embraces innovation and rewards it.
Right now, new entrants into the vaccine market are all small, under-funded companies. Not one of the big pharmaceutical companies that exited the vaccine market is getting back into this line of work. And while these entrepreneurial ventures are experimenting with some very promising technology, the market remains undersized relative to other fields of medicine. This is another area of medicine that is not improved by government's direct intervention. There is little hope that the National Institutes of Health can become the nation's vaccine manufacturers, as their own track record is littered with failures.
Our best hope then is to create an environment where private capital once again is willing to place big bets on producing fundamentally better vaccines and then promoting them to the consumers who stand to benefit. This will require entrepreneurs with daring--of which there are plenty in America. It will also require policymakers with a penchant for embracing these private-sector solutions and the people who produce them, over the more linear route of more money for the NIH.
1. Janet Heinrich, "Health Care-Public Health Issues," GAO Testimony before the U.S. Senate Special Committee on Aging, 108th Congress, Infectious Disease Preparedness: Federal Challenges in Responding to Influenza Outbreaks (September 28, 2004).
2. Institute of Medicine, Division of Health Promotion and Disease Prevention, Vaccine Supply and Innovation (Washington, D.C.: National Academy Press, 1985).
3. John R. La Montagne and Anthony S. Fauci, "Intradermal Influenza Vaccination-Can Less Be More?" New England Journal of Medicine, early release, November 25, 2004 (available at www.njem.org).
John E. Calfee is a resident scholar at AEI. Scott Gottlieb, M.D., is a resident fellow at AEI.