Effect of drug efficacy and the eclipse phase of the viral life cycle on estimates of HIV viral dynamic parameters

Fits of mathematic models to the decline in HIV-1 RNA after antiretroviral therapies have yielded estimates for the life span of productively infected cells of 1 to 2 days. In a previous report, we described the mathematic pr operties of an extended model that accounts for imperfect viral suppression and the eclipse phase of the viral life cycle (the intracellular delay bet ween initial infection and release of progeny virions). In this article, we fit this extended model to detailed data on the decline of plasma HIV-1 RN A after treatment with the protease inhibitor ritonavir. Because the therap y in this study was most likely not completely suppressive, we allowed the drug efficacy parameter to vary from 70% to 100%. Estimates for the clearan ce rate of free virus, c, increased with the addition of the intracellular delay (as reported previously) but were not appreciably affected by changes in the drug efficacy parameter. By contrast, the estimated death rate of v irus-producing cells, delta, increased from an average of 0.49 day(-1) to 0 .90 day(-1) (an increase of 84%) because the drug efficacy parameter was re duced from 100% to 70%. Neglecting the intracellular delay, the comparable increase in delta was only about 55%. The inferred increases in delta doubl ed when the model was extended to account for possible increases in target cell densities after treatment initiation. This work suggests that estimate s for delta may be greater than previously reported and that the half-life of a cell in vivo that is producing virus, on average, may be 1 day.