Paper # 996 
Effect of Expanded ART Strategies on the MSM HIV Epidemic in San Francisco
Edwin Charlebois*1, T Porco1,2, M Das-Douglas1,3, and D Havlir1
1Univ of California, San Francisco, US; 2FI Proctor Fndn, San Francisco, CA, US; and 3San Francisco Dept of Publ Hlth, CA, US
Background: Expanded use of ART may reduce HIV
incidence and eventually HIV prevalence at the population level, assuming
significant behavioral HIV risk disinhibition does not occur. In San Francisco, where HIV prevalence among MSM is 23% and the majority is aware of their HIV
serostatus, we sought to estimate the effects of expanded ART treatment
strategies on the San Francisco MSM HIV epidemic.
Methods: We used a mathematical model to quantify
the effect of 3 expanded ART treatment strategies on the MSM HIV epidemic in
San Francisco: 1) ART treatment for all adults currently in care at CD4<500,
2) ART for all adults in care regardless of CD4 count, 3) ART for all adults in
care plus community-wide annual HIV testing, comparing these strategies to the
current practice of ART for CD4 <350. Modeling was done in R using ordinary
differential equations structured as a compartment model with uninfected
persons and 4 stages of HIV disease (CD4:<200, 200 to 350, 350 to 500, 500+)
and 4 subcategories (undiagnosed HIV, diagnosed HIV untreated, treated partially
suppressed, and treated fully suppressed) and high and low behavioral risk populations.
Model input parameters, population sizes, and progression rates were taken from
observed epidemiologic data collected in San Francisco and published data.
Significant assumptions included 80% on ART would achieve viral load
suppression, 100% reduction in HIV transmission with viral load suppression, 80%
reduced infectivity in partially treated individuals, and 38% of MSM in the
high behavioral risk group.
Results: All 3 treatment strategies resulted in
reduction of new HIV infections compared to current practice. The percent
reduction in incident infections seen with the 3 strategies was 49%, 71%, and
91% over the first 10 years for strategies 1, 2, and 3, respectively. The
number of new infections averted by the 3 strategies is estimated at 3559, 5153,
and 6560 at 10 years and 6146, 10512, and 13729 at 20 years. HIV prevalence is
predicted to decline for all three strategies as seen in the figure below:

Conclusion: In a city with a mature HIV epidemic
where the majority of individuals have known HIV status, the successful
application of more widespread ART would result in large numbers of HIV
incident cases averted and decline in prevalence when combined with annual HIV
testing to detect new cases.
|