The researchers stress, however, that any successful HIV vaccine program should include a behavior-modification program in order to prevent increases in risky behavior after vaccination.
"Most vaccines we use for other diseases have very high efficacy, but this study suggests that even imperfect HIV vaccines can still have a great deal of benefit," said Dr. Douglas Owens, a senior research associate at the Veterans Affairs Palo Alto Health Care System and an associate professor of medicine and of health research and policy at Stanford University School of Medicine.
Owens is a co-author of two research papers on the topic, published in the June 18 issue of the journal AIDS and in the May/June issue of the international journal Interfaces.
Both of the new studies modeled a range of theoretical scenarios and vaccines. The study population consisted of approximately 50,000 gay men in San Francisco. Assumptions regarding this population were based on data from actual surveys and other studies.
"What impressed us was that the vaccines turned out to be cost-effective over such a wide range of conditions — whether the vaccine was therapeutic or preventive, whether the vaccine was moderately or highly effective, or whether the epidemic was in early or late stages," said Donna Edwards, who conducted the studies for her doctoral dissertation in the Department of Engineering–Economic Systems at Stanford.
Edwards is a senior member of the technical staff in the Systems Research Department at Sandia National Laboratories (Livermore, Calif.). Ross Shachter, a Stanford associate professor of engineering–economic systems and operations research, collaborated on the projects.
Both studies looked at two vaccine types: preventive and therapeutic.
Traditional, preventive vaccines aim to prevent new infections by inducing immunity in uninfected individuals. The U.S. Food and Drug Administration this month cleared VaxGen Inc. (South San Francisco) to begin testing a preventive vaccine on 5,000 volunteers in Chicago, Denver, Los Angeles, Philadelphia and St. Louis. Some scientists and AIDS activists have asserted that the VaxGen vaccine is likely to have low efficacy, inducing protection in only apercentage of recipients. This claim is disputed by VaxGen.
Therapeutic vaccines are given to individuals who are already infected, in an attempt to rouse the immune system to fight the virus. No therapeutic vaccines for HIV are commercially available.
Preventive value
In the Interfaces study, the benefit from an imperfect vaccine was particularly noticeable in the case of a preventive vaccine.
Any preventive vaccine that successfully protected 25 percent or more of the recipients would cost less than $50,000 in health-care expenses for each healthy year gained, according to the model. The vaccine appeared most beneficial — resulting in a gain in healthy years as well as a decrease in health-care costs — when efficacy was 75 percent or more and protection lasted at least 10 years, or when efficacy was at least 50 percent with lifetime protection.
When the researchers adjusted their input numbers to model an early-stage epidemic (with fewer people infected, but an increased level of high-risk behavior), they saw mixed effects for the therapeutic vaccine. But for the preventive vaccine, there were greater benefits for a given expenditure and the vaccine more often resulted in savings both in healthy years and in health-care expenses.
This makes sense, said Owens, because the uninfected target population for the preventive vaccine is much larger in an early-stage epidemic. Although much of the AIDS epidemic in the US has the characteristics of a late-stage epidemic, in some US population groups, such as young gay males, it may more closely resembles an early-stage epidemic.
Therapeutic vs. preventive
In the study reported in AIDS, the researchers compared the different impacts of therapeutic and preventive vaccines. For the initial analyses, they presumed that their preventive vaccine would protect 75 percent of recipients and provide 10 years of protection, and that the therapeutic vaccine would give 100 percent of recipients five extra healthy years of life.
There is always the possibility that a therapeutic vaccine will make things worse overall, Owens said, because infected individuals who live longer have more time to transmit HIV. The researchers found that this effect resulted in 210 additional HIV cases in their 20-year study period, compared with the 3,877 fewer HIV cases for the preventive vaccine. The preventive vaccine also resulted in 1,354 fewer lives lost and 15,908 healthy years of life gained.
When one considers only the number of new cases, the therapeutic vaccine looks less promising than the preventive one. But, said Edwards, the therapeutic vaccine is designed to improve health, not necessarily to decrease transmission. Sure enough, the therapeutic vaccine resulted in the loss of 679 fewer lives and the gain of 8854 healthy years of life.
"If you use that kind of measure, it turns out that therapeutic and preventive vaccines could save similar numbers of years of life," Edwards said.
The picture for the therapeutic vaccine improves even more if the vaccine reduces the infectivity of the recipients.
"If the therapeutic vaccine inhibits disease progression by suppressing viral replication, the person may not be as infectious," Owens said. "In terms of causing more cases or preventing cases, it’s this trade-off between how much longer you live and how much less infectious you are."
There are two ways of slowing disease transmission: infected people can be made less infectious (with a therapeutic vaccine), or uninfected people can be made resistant to infection (with a preventive vaccine). In a hypothetical extreme case, a therapeutic vaccine could make vaccinated, infected individuals uninfectious, Owens said. This would achieve the same result as a perfect preventive vaccine. But, he said, "it is unknown to what extent a therapeutic vaccine would reduce infectivity, if at all."
In the model, decreasing the infectivity of therapeutic vaccine recipients by only 8 percent was enough to wipe out the increase in the number of new HIV cases, so that the caseload was equal to that seen with no vaccine. And a decrease in infectivity could put the therapeutic vaccine on par with a preventive vaccine, the model showed. The numbers of healthy years gained were equal for a therapeutic vaccine providing five extra years of healthy life and a 50 percent decrease in infectivity compared with a preventive vaccine providing 10 years of protection to 62 percent of its recipients.
Importance of behavior
The new findings underscore the need for any therapeutic vaccine to be paired with behavioral programs.
Owens says there is a possible danger that recipients of any vaccine will feel a sense of security and will therefore engage in more risk-taking behaviors, such as decreased condom usage. To consider the effects of such a hypothetical response, the Interfaces study presumed a 25 percent decrease in condom usage after either a preventive or therapeutic vaccination program. (No such change was incorporated in the AIDS study.)
In the 20 years of the therapeutic vaccine program, this kept the number of healthy years saved to a modest 2,410. When projected to 150 years (without continuation of the vaccination program), the end result was a loss of 5,290 healthy years.
But if education efforts associated with a therapeutic vaccination program resulted in a 25 percent increase in condom usage, any therapeutic vaccine program would save both healthy years of life and health-care costs, the researchers found.
Even with the 25 percent decrease in condom usage, some therapeutic vaccines appeared valuable. The initial analysis described in Interfaces presumed that the therapeutic vaccine provided five extra healthy years and induced no change in infectivity. By varying these parameters, the researchers found that the therapeutic vaccine could save both healthy years and money if it added at least 10 healthy years per patient, or decreased infectivity by at least 50 percent, or added at least five years and decreased infectivity by at least 25 percent.
All the conclusions from the modeling studies depend on numerous variables — such as numbers of sexual partners and rates of disease transmission — based on studies of the San Francisco population. The researchers stressed that the results must therefore be treated with caution.
"You have to be careful extrapolating this to other populations," Edwards said. "The general ideas about vaccine effectiveness will probably hold up in other populations, but the absolute numbers will not."
Funding for the studies was provided by the Department of Veterans Affairs, through a VA Health Services Research and Development Career Development Award; by the Department of Energy, through the doctoral studies program at Sandia National Laboratories; and by the National Institute on Drug Abuse, through a grant to the Societal Institute of Mathematical Sciences.
Interfaces is an international journal published by the Institute for Operations Research and the Management Sciences (INFORMS), a membership organization based in Providence, Rhode Island, for scientists who apply information technology to achieve informed decision-making.
The Institute for Operations Research and the Management Sciences (INFORMS) is an international scientific society with 12,000 members, including Nobel Prize laureates, dedicated to applying scientific methods to help improve decision-making, management, and operations. Members of INFORMS work primarily in business, government, and academia. They are represented in fields as diverse as airlines, health care, law enforcement, the military, the stock market, and telecommunications.