Explaining the Determinants of First Phase HIV Decay Dynamics through the Effects of Stage-dependent Drug Action

A recent investigation of the effect of different antiretroviral drug classes on first phase dynamics of HIV RNA plasma virus levels has indicated that drugs acting at stages closer to viral production, such as the integrase inhibitor raltegravir, can produce a steeper first phase decay slope that may not be due to drug efficacy. Moreover it was found that for most drug classes the first phase transitions from a faster (phase IA) to a slightly slower decay region (phase IB) before the start of the usual second phase. Neither of these effects has been explained to date. We use a mathematical model that incorporates the different stages of the HIV viral life cycle in CD4+ T cells: viral entry, reverse transcription, integration, and viral production, to investigate the intracellular HIV mechanisms responsible for these complex plasma virus decay dynamics. We find differences in the phase IA slope across drug classes arise from a higher death rate of cells when they enter the productively infected stage post-integration, with a half-life of approximately 8 hours in this stage, whereas cells in earlier stages of the infection cycle have half-lives similar to uninfected cells. This implies any immune clearance is predominantly limited to the productive infection stage. We also show that the slowing of phase IA to phase IB at day 2 to 4 of monotherapy, depending on drug class, is a result of new rounds of infection. The level at which this slowing occurs is a better indicator of drug efficacy than the slope of the initial decay. Author Summary: The infection of a cell by HIV proceeds through a series of stages and each stage can now be inhibited by an available antiretroviral drug class. It is known that different drug classes can result in different decay curves of plasma viral levels that are not well explained by current mathematical models of HIV dynamics. Here we develop a mathematical model that incorporates these stages of infection and show how it successfully reproduces plasma decay curves for the five classes of currently available antiretroviral drugs. Our modeling indicates that the efficacy of antiretroviral drugs is not solely described by the rate of decay of plasma viral levels as currently thought. Drugs such as the integrase inhibitor raltegravir will result in a faster initial decline of plasma viral levels compared to a drug that acts further from viral integration and production such as the CCR5 inhibitor maraviroc, even though they may have the same efficacy. Moreover, we find that infected cells only die at rates above the background level when they are in the productive phase, indicating that immune clearance is mostly absent from the early stages of HIV cellular infection. This is of particular concern given that most infected cells are in these early stages of infection.