223   Viral Burden in Genital Secretions Determines Male-to- Female Sexual Transmission of HIV-1: A Probabilistic Empiric Model.

H. Chakraborty*, P. K. Sen, R. W. Helms, P. L. Vernazza, S. A. Fiscus, J. J. Eron, B. K. Patterson, R. W. Coombs, J. N. Krieger, and M. S. Cohen.
Emory Univ., Atlanta, GA and Univ. of North Carolina at Chapel Hill.

Background:Most published epidemiological and mathematical models developed to estimate the likelihood of HIV-1 transmission during a single episode of sexual intercourse assumed constant infectivity within and between couples. We developed a probabilistic model to estimate the male-to-female penile-vaginal per-sexual- act HIV-1 transmission probability as a function of changes in seminal viral load and receptor cell numbers.

Methods:We used conditional and unconditional probability theory to develop the model. We also used a bootstrap re-sampling method to account for repeated observations and successive approximation method to estimate model parameters. This model assumes preferential transmission of HIV-1 isolates and considers the NSI/SI phenotype of HIV-1 in semen as well as the actual number of CD4 and CCR5 receptors in endocervical tissue.

Results:We evaluated data sets from three different centers (n = 88, 165, and 100) for seminal plasma RNA concentrations and one data set (n = 28) for receptor cell counts. The model demonstrates a sharp increase in transmission probability as seminal viral load and/or receptor cells count increases. At low concentrations of HIV-1 observed in semen in men (e.g. <5,000 copies of HIV-1/ml of seminal plasma), HIV-1 transmission is very unlikely to occur (probability of <1/10,000/episode of intercourse). Conversely, in subjects with greater concentrations of HIV-1 in semen (e.g. 50,000—1,000,000 copies of HIV-1/ml of seminal plasma) the probability of HIV-1 transmission increases to 0.007 and 0.03, respectively.

Conclusions:The results also suggest a per-contact rate of sexual transmission of HIV-1 that better explains the magnitude of the epidemic than older epidemiological models. Our model can be used to examine the biological basis for accelerated spread of HIV-1 in some developing countries and the effects of different prevention strategies that influence viral burden, viral phenotype, and expression of receptor cells.