Elsevier

Journal of Theoretical Biology

Volume 243, Issue 2, 21 November 2006, Pages 191-204
Journal of Theoretical Biology

Natural variation in HIV infection: Monte Carlo estimates that include CD8 effector cells

https://doi.org/10.1016/j.jtbi.2006.05.032Get rights and content

Abstract

Viral load and CD4 T-cell counts in patients infected with the human immunodeficiency virus (HIV) are commonly used to guide clinical decisions regarding drug therapy or to assess therapeutic outcomes in clinical trials. However, random fluctuations in these markers of infection can obscure clinically significant change. We employ a Monte Carlo simulation to investigate contributing factors in the expected variability in CD4 T-cell count and viral load due solely to the stochastic nature of HIV infection. The simulation includes processes that contribute to the variability in HIV infection including CD4 and CD8 T-cell population dynamics as well as T-cell activation and proliferation. The simulation results may reconcile the wide range of variabilities in viral load observed in clinical studies, by quantifying correlations between viral load measurements taken days or weeks apart. The sensitivity of variability in T-cell count and viral load to changes in the lifetimes of CD4 and CD8 T-cells is investigated, as well as the effects of drug therapy.

Introduction

Changes in CD4 lymphocyte counts and viral load (RNA copies) are widely used as indicators of HIV progression in infected patients. These measures inform decisions regarding the initiation or continuance of antiretroviral therapy and are likewise used to assess therapeutic effects in clinical trials. It is therefore critical to define the expected variance between consecutive measurements of T-cell count and plasma RNA levels, since random fluctuations may obscure clinically significant changes.

Total variability for sequential CD4 T-cell determinations from asymptomatic (clinically latent) HIV infected individuals has been examined clinically (Hughes et al., 1994, Paxton et al., 1997). Hughes et al. (1994) found that the within-patient mean standard deviation (SD) of T-cell count was 0.25logecells/μL, corresponding to a coefficient of variation (CoV) of 25%, when derived from data at 0 and 8 weeks. Paxton et al. (1997) determined that the long-term within-individual variability for T-cell measurements had a CoV of 16%, equivalent to a SD of 0.071log10cells/μL.

Similar investigations measuring instead the variation among consecutive measurements of HIV RNA in plasma have revealed a total variability, expressed as SD, of 0.19–0.4log10copies/mL (Bartlett et al., 1998, Brambilla et al., 1999, Coombs et al., 1998, Deeks et al., 1997, Hughes et al., 1997, Paxton et al., 1997, Raboud et al., 1996, Yamashita et al., 2003). The wide range reported might be attributed to the sizes of the clinical cohorts—between 11 (Deeks et al., 1997) and 387 (Bartlett et al., 1998) individuals—or to variations in antiretroviral drug therapy among individuals participating in each study. Perhaps more importantly, the intervals between consecutive measurements in these studies vary from days to over a year, and thus the reported range may be an effect of differences in sampling intervals. For example, Deeks et al. (1997) reported the highest variability of 0.4log10 RNA copies/mL for measurements one day apart. Bartlett et al. (1998) reported a slightly lower SD of 0.31log10 RNA copies/mL for two consecutive measurement 2–4 weeks apart. Raboud et al. (1996) reported an even lower value still of 0.29log10 RNA copies/mL derived from subjects seven days apart and Hughes et al. (1997) reported the lowest variability of 0.19log10 RNA copies/mL based on paired baseline measurements with a median interval of 6 days apart (2–15 days).

A handful of mathematical models of HIV dynamics have attempted to quantify the inherent variability in HIV infection (Heffernan and Wahl, 2005, Kamina et al., 2001, Tan and Wu, 1998, Tuckwell and Le Corfec, 1998, Wu and Ding, 1999). Tuckwell and Le Corfec (1998) used SDEs to estimate the intrapatient variability in both the time to peak viremia and the viral load at this peak, two important characteristics of primary infection. Tan and Wu (1998) used a Monte Carlo simulation to estimate the probability of clearing an initial inoculum of virus. Kamina et al. (2001) later used the same Monte Carlo approach to estimate the time to peak and peak viremia. None of these investigations consider variability during the clinically latent stage of infection.

In a previous study, using a Monte Carlo model of in-host HIV infection, we investigated the variability in T-cell count and viral load that could be attributed to interactions between CD4 T-cells and free virus during both primary infection and clinical latency (Heffernan and Wahl, 2005). The simulation proceeded at the level of individual infected and uninfected CD4 T-cells and infectious and non-infectious virions. We also examined the effects of antiretrovirals, such as reverse transcriptase and protease inhibitors (RTIs and PIs), and the effects of more realistic lifetime distributions for T-cells. A limitation of this previous study was that many of the underlying processes contributing to variation in CD4 T-cell and virus concentrations in vivo were not included in the model.

In particular, the processes of activation and proliferation of CD4 T-cells directly affect cell and virus population dynamics. Naive CD4 T-cells, brought into the lymph tissues from the thymus, are stimulated by viral antigen to become activated CD4 T-cells. Activated CD4 T-cells, if they are uninfected, proliferate and aid the rest of the immune system in fighting the infection. If they are infected, activated CD4 T-cells bud virus.

Similarly, CD8 T-cells (cytotoxic T lymphocytes or CTL) also affect steady state variability. Activated CD8 T-cells (or effector cells) are produced by the activation of naive CD8 T-cells by viral antigen and CD4 T-cells. Effector cells kill infected cells displaying viral antigens (Janeway and Travers, 2005).

Analytical models based on differential equations (Culshaw and Ruan, 2000, De Boer and Perelson, 1995, Ho et al., 1995, Kirschner, 1996, McLean, 1994, Nelson and Perelson, 2002, Nelson et al., 2004, Nowak and May, 2000, Perelson, 2002, Stafford et al., 2000, Wei et al., 1995, Wodarz et al., 1998, Wodarz et al., 1999) have been used to predict the dynamics of in-host HIV infection as have numerous simulation studies (Chao et al., 2004, da Silva and Hughes, 2002, Kamina et al., 2001, Kousignian et al., 2003; Ruskin et al., 2002, Tan and Wu, 1998, Tuckwell and Le Corfec, 1998, Wu and Ding, 1999). Few of these studies, however, have incorporated the effects of the CD8 T-cell population and fewer still have included the effects of activation and proliferation in the CD4 or CD8 T-cell populations. This is sensible since the inclusion of numerous biological processes into analytical models can prove unwieldy, often requiring a numerical solution. Also, it has only recently become computationally feasible to incorporate realistic levels of biological detail in simulation models (da Silva and Hughes, 2002). Recently, Chao et al. (2004) developed an agent-based simulation to study cytotoxic T-cell responses at the individual level. This study included the effects of activation and proliferation of the CD8 T-cell population, however, it did not incorporate the activation and proliferation of CD4 T-cells. Also, the inherent variability in HIV dynamics was not examined.

In this study we further explore variability during the clinically latent phase of infection (steady state variability) by extending our Monte Carlo simulation to include the CD8 T-cell population, as well as the processes of activation and proliferation of CD4 and CD8 T-cells. Biologically motivated lifetime distributions describing cell death and CD8 T-cell activation are incorporated, and variability during drug therapy is examined using a pharmacokinetic model of drug dynamics. The correlations we observe between measures of variability separated by days or weeks may help explain some apparent discrepancies in the clinical literature.

Section snippets

Model

Our Monte Carlo simulation progresses at the level of individual uninfected CD4 T-cells (naive or activated), infected CD4 T-cells (naive or activated), virions (infectious or non-infectious), and CD8 T-cells (naive or activated). We may also include the effects of antiretroviral drugs, a RTI and a PI. The simulation moves forward in time using “event times” that are assigned to the individual cells and virions from the appropriate probability distributions i.e. death time, infection time or

Natural variability

Fig. 1 shows ten sample runs of the Monte Carlo simulation. This figure demonstrates that the viral dynamics can change dramatically from one run of the simulation to another, due to the inherent stochasticity of the model. To quantify the variability around the equilibrium for each population we determine the variance of T-cell count and viral load over the last ten months of a single run, well after equilibrium is reached. The variance is then divided by the mean uninfected T-cell count to

Discussion

We have developed a detailed Monte Carlo model of in-host HIV dynamics that includes both CD8 T-cells and T-cell activation and proliferation. The model allows us to quantify the contribution of these immune system components to the underlying variability (steady state variability) in CD4 T-cell count and viral load. These results follow from a previous study where we quantified this variability due solely to the CD4 T-cell and virus populations (Heffernan and Wahl, 2005).

The Monte Carlo

Acknowledgements

This work is supported by the Natural Sciences and Engineering Research Council of Canada, the Ontario Ministry of Science, Technology and Industry and by the SHARCNET parallel computing facility. We are indebted to Olga Krakovska for many helpful discussions and to Baolai Ge, Gary Molenkamp and Mark Hahn for their invaluable technical expertise. We also thank two anonymous referees for many insightful suggestions.

References (58)

  • R. Smith et al.

    Distinct effect of protease and reverse transcriptase inhibition in an immunological model of HIV-1 infections with impulsive drug effects

    Bull. Math. Biol.

    (2004)
  • R. Smith et al.

    Drug resistance in an immunological model of HIV-1 infection with impulsive drug effects

    Bull. Math. Biol.

    (2005)
  • M.A. Stafford et al.

    Modeling plasma virus concentration during primary HIV infection

    J. Theor. Biol.

    (2000)
  • W.-Y. Tan et al.

    Stochastic modeling of the dynamics of CD4+ T-cell infection by HIV and some Monte Carlo studies

    Math. Biosci.

    (1998)
  • H.C. Tuckwell et al.

    A stochastic model for early HIV-1 population dynamics

    J. Theor. Biol.

    (1998)
  • D. Wodarz et al.

    Dynamics of macrophage and T cell infection by HIV

    J. Theor. Biol.

    (1999)
  • J.A. Bartlett et al.

    Variability in repeated consecutive measurements of plasma human immunodeficiency virus RNA in persons receiving stable nucleoside reverse transcriptase inhibitor therapy or no treatment

    J. Inf. Dis.

    (1998)
  • Brambilla, D., Reichelderfer, P.S., Bremer, J.W., Shapiro, D.E., Hershow, R.C., Katzenstein, D.A., Hammer, S.M.,...
  • R.M. Coles et al.

    Progression of armed CTL from draining lymph node to spleen shortly after localized infection with herpes simplex virus 1

    J. Immunol.

    (2002)
  • R.W. Coombs et al.

    Association between culturable human immunodeficiency virus type 1 (HIV-1) in semen and HIV-1 RNA levels in semen and blood: evidence for compartmentalization of HIV-1 between semen and blood

    J. Inf. Dis.

    (1998)
  • da Silva, J., Hughes, A., 2002. Monte Carlo simulation of HIV-1 evolutions in response to selection by antibodies....
  • S.G. Deeks et al.

    Variance of plasma human immunodeficiency virus type 1 RNA levels measured by branched DNA within and between days

    J. Inf. Dis.

    (1997)
  • A.V. Gett et al.

    A cellular calculus for signal integration by T cells

    Natl Immunol.

    (2000)
  • A.T. Haase et al.

    Quantitative image analysis of HIV-1 infection in lymphoid tissue

    Science

    (1996)
  • D.D. Ho et al.

    Rapid turnover of plasma virions and CD4 lymphocytes in HIV-1 infection

    Nature

    (1995)
  • M.D. Hughes et al.

    Within-subject variation in CD4 lymphocyte count in asymptotic human immunodeficiency virus infection: implications for patient monitoring

    J. Inf. Dis.

    (1994)
  • M.D. Hughes et al.

    Monitoring plasma HIV-1 RNA levels in addition to CD4 lymphocyte count improves assessment of antiretroviral therapeutic response

    Ann. Int. Med.

    (1997)
  • G. Iezii et al.

    The duration of antigenic stimulation determines the fate of naive and effector cells

    Immunity

    (1998)
  • C. Janeway et al.

    Immunobiology: The Immune System in Health and Disease

    (2005)
  • Cited by (23)

    • Environmental variability in a stochastic HIV infection model

      2023, Communications in Nonlinear Science and Numerical Simulation
    • In-host modeling

      2017, Infectious Disease Modelling
    • Bifurcation analysis of macrophages infection model with delayed immune response

      2016, Communications in Nonlinear Science and Numerical Simulation
    • Bifurcations of a mathematical model for HIV dynamics

      2016, Journal of Mathematical Analysis and Applications
    View all citing articles on Scopus
    View full text