Expanding the use of the most potent anti-retroviral drug therapies would help
limit new HIV infections and AIDS deaths among gay men in San Francisco over
the next ten years even if drug resistance and dangerous sexual behavior soar,
according to a new mathematical model developed by scientists at UC San
Increased risky behavior always will cause infection and death rates to rise
but, contrary to what some in the field have predicted, the new model projects
a consistently better future if advanced anti-retroviral therapy is extended to
more HIV-infected gay men.
“It’s a complex interaction, but a simple outcome, really,” says Sally Blower,
PhD, associate professor of medicine, microbiology and immunology at UC San
Francisco and lead author of a report on the study. “The more you treat, the
better you will do, even if risk behavior and drug resistance go way up.”
In the model’s worst-case projection—with risk behavior doubling and drug
resistance severely diluting the therapy’s effectiveness—infection rates and
deaths would still be lower after ten years than if treatment were not
The report, by a team of UCSF virologists and mathematical biologists, is
published in the January 28 issue of the journal Science.
About half of the HIV-infected men in San Francisco are taking a combination of
anti-retroviral drugs, in a treatment regime known as Anti Retroviral Therapy
(ART) or Highly Active Anti Retroviral Therapy (HAART), the authors note.
Since treated men are likely to retain some degree of infectivity, some AIDS
experts have warned that expanding ART treatment could lead to an increased
infection rate. In addition, drug-resistant HIV strains have already emerged
and risky sexual behavior has begun to increase in San Francisco. These trends
have led some to conclude that expanding use of the advanced ART or HAART
therapy - particularly to those less prone to practice safe sex or to follow
the difficult drug regimen carefully - would lead to increases in drug
resistance, higher HIV infection rates and soaring AIDS deaths.
The new model shows otherwise. “Our model found that over ten years, expanding
ART treatment in San Francisco will reduce HIV incidence rates and AIDS death
rates below what they would be without expanded treatment,” Blower stresses.
Although the model finds a beneficial effect of expanded drug treatment, it
starkly reveals that increased risk behavior always leads to more infections
and more AIDS deaths.
“Our predictions clearly show that expanding use of viral suppressing drugs to
more gay men will always reduce HIV incidence and save lives compared to not
expanding treatment, no matter what the rate of drug resistance or risk
behavior is,” Blower says. “But even small changes in risk behavior will reduce
the benefits in the short run. Doubling of risky sexual behavior will negate
benefits in the short run and limit them in the long term.”
“Our results indicate it is imperative that bath houses not be re-opened and
that people practice protected sex, particularly those who are on treatment,”
Co-author on the paper is Robert Grant, MD, assistant professor of medicine at
UCSF and an AIDS specialist at the Gladstone Institute of Virology and
Immunology. Hayley Gershengorn, a former research assistant in Blower’s lab and
now a Harvard medical student, is also a co-author.
Blower and colleagues developed the model to help AIDS experts assess drug
treatment strategies given the high degree of uncertainty in future rates of
drug resistance and risky sexual behavior as well as the effectiveness of
treatment on both rates of infectivity and long-term survival. It is the first
such model to project long-term drug effectiveness while incorporating so many
types of uncertainty.
“We used the model to predict the unpredictable,” Blower says.
The model includes the potential effects of ART on the transmission dynamics of
both drug-sensitive and drug-resistant HIV strains. It examines both a
“best-case condition” (10 percent annual rate of emergence of resistance; no
increase in risky behavior) and a “worst-case condition” (up to 60 percent of
cases become drug resistant per year; risky sexual behavior doubles).
Reflecting the range of possible outcomes provided by the model, the Science
paper is titled “A Tale of Two Futures: HIV and Anti-Retroviral Therapy in San
The model identifies which factors affect how well anti-retroviral therapy
works, and uses these factors to suggest how to ensure that the outcome is “the
best of times,” rather than “the worst of times,” Blower points out. The key
factor is the rate of risky sexual behavior.
Blower, Gershengorn and Grant have applied what is currently known about HIV
biology and infectivity to knowledge of viral transmission and survival
generally to develop the new mathematical model.
“We must move the study of epidemics beyond description to make it a predictive
science,” says Blower. “This is the only certain way to improve strategies to
fight infectious diseases.”
Grant cautions that although the researchers have used all available
information, at this point there are only three years of data on the use of ART
and the possibility of viruses becoming resistant to it.
In addition, he says, “the model indicates that relative virulence and
transmissibility of HIV are critical determinants in predicting the epidemic
behavior of a drug-resistant virus. More information on these two factors will
be crucial in the coming years.”
The research is funded by the National Institute of Allergies and Infectious
Diseases, part of the National Institutes of Health.