Table of Contents Author Guidelines Submit a Manuscript
Journal of Applied Mathematics
Volume 2018, Article ID 8276317, 11 pages
https://doi.org/10.1155/2018/8276317
Research Article

Modelling In Vivo HIV Dynamics under Combined Antiretroviral Treatment

1Department of Mathematics, Laikipia University, P.O. Box 1100-20300, Nyahururu, Kenya
2Department of Mathematics, Masinde Muliro University of Science and Technology, P.O. Box 190-50100, Kakamega, Kenya

Correspondence should be addressed to B. Mobisa; ek.ca.aipikial@asibomb

Received 19 September 2018; Revised 19 November 2018; Accepted 26 November 2018; Published 10 December 2018

Academic Editor: Qingdu Li

Copyright © 2018 B. Mobisa et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

In this paper a within host mathematical model for Human Immunodeficiency Virus (HIV) transmission incorporating treatment is formulated. The model takes into account the efficacy of combined antiretroviral treatment on viral growth and T cell population in the human blood. The existence of an infection free and positive endemic equilibrium is established. The basic reproduction number is derived using the method of next generation matrix. We perform local and global stability analysis of the equilibria points and show that if , then the infection free equilibrium is globally asymptotically stable and theoretically the virus is cleared and the disease dies out and if , then the endemic equilibrium is globally asymptotically stable implying that the virus persists within the host. Numerical simulations are carried out to investigate the effect of treatment on the within host infection dynamics.

1. Introduction

Human Immunodeficiency Virus (HIV) remains a major threat to human life for the last three and half decades. HIV infection in humans causes Acquired Immunodeficiency Virus (AIDS), a disease that has ravaged human population all over the world. Since its discovery in the early 1980s, there has been tremendous research work on how to contain or eradicate the disease. Mathematical modelling of viral infections has led to greater understanding of virus dynamics and helped in predicting and controlling the spread of viral diseases such as HIV, Hepatitis B Virus (HBV), Hepatitis C Virus (HCV), and Dengue Fever. One of the early models of HIV infection known as the basic model was used by Nowak and May [1] and by Perelson and Nelson (1999) and was successful in numerically reproducing the dynamics of the early stages of HIV and its target CD4+ cells following an infection event. Recent studies have focused on HIV viral and cellular infections incorporating dynamics such as intracellular delays, latent infection and viral mutation, and spatial heterogeneity [25]. For instance, [6] investigated the global stability of within host virus models with cell-to-cell viral transmission and obtained a complete analytic description of equilibria. A four-dimensional system of delayed differential equations, where the production and removal rates of the virus and cells are given by general nonlinear functions, was proposed by [4]. Their model investigated the dynamical behaviour of virus target and cell target incidences incorporating humoral immune response. They established three key equilibrium results, an infection free equilibrium, a chronic free equilibrium with inactive humoral immune response, and chronic infection equilibrium with active humoral immune response. With dynamics governed by two bifurcation parameters basic reproduction numbers and the humoral immunity numbers and using Lyapunov functionals and Lasalle’s invariance principle, the authors proved the global stability of the equilibria.

The inclusion of treatment, at within and between host levels in mathematical modelling, has gained considerable attention in recent years. For instance, at between host levels, epidemiological models with saturated treatment function have been proposed by [7, 8]. Research on within host models that incorporate treatment has been carried out over the years, with early models highlighting the effects of AZT on viral replication [9]. Among the key findings, viral decline is drug dependent. A study of combined drug therapy of HIV infection was conducted by [10]; the mathematical model developed was used to simulate chemotherapy treatment of HIV infection. The simulations were based on clinical data of treatment with combinations of antiviral drugs involving reverse transcriptase inhibitors (RTI) and protease inhibitors (PI) and focused on the timing of treatment. The findings revealed that the success of treatment is based on longer survival times equated to the CD4+ T cells. Global dynamics of delay distributed HIV infection models with differential drug efficacy in cocirculating target cells was investigated [3]. Recent work by [11] sought to mathematically analyze the potential of Prophylaxis treatment in preventing and slowing the spread of HIV/AIDS in the population. In this study early use of Prophylaxis drug was shown to slow the rate of HIV transmission.

Whereas extensive research on HIV transmission dynamics has been carried out, mathematical modelling of HIV with combined treatment still remains an area of active research among mathematicians and biologists.

In this paper, we propose a within host HIV infection model with a logistic incidence rate that explicitly incorporates the two levels of antiretroviral treatment, namely, the reverse transcriptase inhibitors (RTIs) which prevent the reverse transcription of viral RNA into DNA. In this way the RTIs serve to reduce the rate of infection of activated cells. The other category is protease inhibitors (PIs) which prevent HIV-1 protease from clearing the HIV polyprotein into functional units, thereby causing infected cells to produce immature virus particles that are not capable of infecting additional cells; hence PIs decrease the number of newly infectious virus (virions) that are produced [12]. A mathematical analysis of the effects of treatment on the within host infection dynamics is carried out.

2. Model Description and Formulation

A mathematical model of within host HIV infection dynamics is considered. The model is composed of three interacting variables, namely, uninfected CD4+ T cells , actively infected cells , and free virus particles . The uninfected CD4+ T cells are produced at rate and die naturally at the rate . The total number of T cells in the body remains bounded; thus the growth of T cells is governed by the logistic proliferation term which limits T cell growth as the cell population approaches the limit . The uninfected CD4+ T cells become infected by free virus and actively infected cells according to the simple mass infection terms and , respectively. This generates actively infected cells, , which die naturally at the rate . The infected cells produce free viruses at the rate and are cleared from circulation at rate per virus. This viral decline is a function of the efficiency of the combined treatment of Reverse Transcriptase Inhibitor (RTI) and Protease Inhibitor (PI), which are represented by the parameters and , respectively. From the description and definitions made, the infection dynamics are summarized by the following system of ODEs:

3. Analysis of the Model

Since the model describes cell and virus populations dynamics, all the model variables are nonnegative for . In the absence of the virus, the T cell population has a steady state value ; hence the initial conditions for the model (1) are , , and . It can be shown that with positive initial data the solutions of model (1) will remain positive and bounded in the feasible region , .

3.1. Basic Reproduction Number

The basic reproduction number is defined as the average number of secondary infections produced by one infectious virion and one infected cell over the course of their infectious period in uninfected CD4+ T cell population. We compute for model (1) using the next generation matrix method as used in [13, 14]. Model (1) has two infected compartments and . Let be the rate of appearance of new infections in compartment and as the transfer of individuals out of compartment for the two compartments, respectively, and are given in partitioned form as follows:andThe Jacobian of and evaluated at the Infection Free Equilibrium yieldswhere is nonnegative and is nonsingular. The basic reproduction number is thus given by , where is the spectral radius of the matrix . Hence

3.2. Local Stability Analysis of the Infection Free Equilibrium

We investigate the local stability properties of the infection free equilibrium by approximating the nonlinear system of the differential equations (1) with the linear system at the infection free equilibrium .

Theorem 1. The infection free equilibrium is locally asymptotically stable if and only if .

Proof. Evaluating the Jacobian of model (1) at , we obtainand clearlyis one of the eigenvalues of the matrix in (7), which is negative because for a population that is growing in numbers; the rate of production (birth rate) is greater than the death rate, that is . The nature of the remaining roots of (7) can be determined from the reduced matrix:Using Routh-Hurwitz stability criteria, matrix in (9) will have negative real roots if and only if the and ; thusandand using (6), (11) reduces to From (10) and (12), and if and only if . Thus is locally asymptotically stable whenever and unstable otherwise.

This means that if a small number of free virus particles enter the blood stream, each virus will infect on average less than one uninfected cell in its entire period of infectivity whenever . Theoretically this shows that the virus is cleared from the body if .

3.3. Global Stability Analysis of the Infection Free Equilibrium

In this section we study the global stability of the infection free equilibrium of model (1) using the theorem by Castillo-Chavez et al. [13]. We rewrite model (1) in the formwhere denotes the number of uninfected cells and denotes the number of actively infected cells and free virus particles, respectively. The infection free equilibrium (IFE) is now denoted byThe conditions and below must be met in order to guarantee global asymptotic stability:(i)() For is Globally Asymptotically Stable (GAS)(ii)() , for

where is an M-matrix (the off-diagonal elements of are nonnegative) and is the region where the model makes biological sense. If system (13) satisfies conditions and , then the following theorem holds.

Theorem 2. The fixed point is Globally Asymptotically Stable equilibrium of (13) provided that and that assumptions () and () are satisfied.

Proof. Let , , , and whereFrom (16) this implies that . Therefore, is globally asymptotically stable when .

This means that, at any perturbation of the equilibrium point by the introduction of free virus particles, the model solutions will always converge to the IFE, whenever .

3.4. Existence of the Endemic Equilibrium (EE)

Theorem 3. A positive endemic equilibrium exists provided .

Proof. The endemic equilibrium satisfiesFrom (19) we haveSubstituting (20) in (18) we getSubstituting and in (17) we obtainThe endemic equilibrium (EE) is given as Clearly if and only if .

3.5. Local Stability Analysis of the Endemic Equilibrium

Theorem 4. The endemic equilibrium is locally asymptotically stable whenever .

Proof. The Jacobian matrix of model (1) at is as follows:The characteristic equation of (24) is in the formwhere The number of possible negative real roots of (25) depends on the signs of , , and . This can be established by applying Descartes Rule of Signs as used in [11].According to this rule the number of negative real zeros of is either equal to the number of sign changes of or less by an even number, as shown in Table 1.
From Table 1 the maximum number of sign changes in is 2; hence the characteristic polynomial (27) has two negative roots. Thushas negative real roots. Hence for and if cases 1 to 8 are satisfied then the endemic equilibrium is locally asymptotically stable.

Table 1: Roots of characteristic equation (27).

Therefore if and given a small number of free virus particles, each virus, in the entire period of its infectivity, will produce on average more than one infected cell, implying viral persistence.

3.6. Global Stability Analysis of the Endemic Equilibrium

In this section we investigate the global stability of the endemic equilibrium using geometric approach, as developed by Li and Muldowney in [15]. For a brief description of this approach, see [6, 8]. Consider the autonomous dynamical systemwhere , is an open set and is simply connected and , , . Let be an equilibrium point, then is said to be globally stable in if it is locally stable in and all trajectories in converge to . In this method the equilibrium is locally asymptotically stable provided the following conditions hold: (i)(H1) is simply connected(ii)(H2) There exists a compact absorbing set (iii)(H3) Equation (29) has a unique equilibrium in

Let be a matrix-valued function that is on and consider where the matrix is and let the matrix be the second additive compound matrix of the Jacobian matrix ; that is, , is a matrix, and in our case n=3; henceConsider the Lonziskii measure of with respect to a vector norm in where It is proved in [15] that if (H1), (H2), and (H3) hold and conditionis satisfied, then the unique equilibrium is globally asymptotically stable

Lemma 5. Assume that conditions (H1), (H2), and (H3) hold, then is globally asymptotically stable in provided that a function and a Lonziskii measure [16] exist such that condition (32) is satisfied.

Theorem 6. The endemic equilibrium is globally asymptotically stable in if .

Proof. Consider the Jacobian of model (1)and the second compound additive matrix of (33) is given aswhere We define an auxiliary matrix function on assetting everywhere in , and is smooth and nonsingular. and are given aswhere and . Matrix is given asThus the matrix as defined in (4.4) of [15] can be written in block form aswhere , , , and . Let the vector norm in be chosen asThe Lozinskii measure with respect to can be estimated as follows:where and are operator norms associated with the linear mappings and , respectively, where is endowed with the vector norm in both cases. Specifically , , , and . From model (1) we findand recalling expressions for , , , and we obtainSince we find that , then (42) implies that ; thusSince is the limiting value of , then this implies that ; therefore expression (49) reduces toIntegrating (50) we obtainHence the endemic equilibrium (EE) is globally asymptotically stable whenever .

This implies that, regardless of any starting solution, the solution of the model will converge to EE whenever . Immunologically, it means that, at any perturbation of the equilibrium point as a result of the introduction of the free virus particles, the model solutions will converge to the endemic state.

4. Numerical Simulation of the Model

In this section we perform a numerical simulation of model (1) using MATLAB with the parameter values given in Table 1. The primary purpose of the numerical simulation is to analyze the change in state of virus progression with time and also to outline the impact of the variation of treatment efficacy on the transmission dynamics of HIV. This is achieved by varying the parameter values and while keeping the other parameters constant.

4.1. Effect of Variations in RTI and PI Treatment Efficacy on the Asymptotic Behaviour of the Equilibrium Points

The effect of variations in combined treatment efficacy on the stability of equilibria is investigated using the parameter values given in Table 2; this is achieved by choosing four different initial conditions(i)IC1: , , (ii)IC2: , , (iii)IC3: , , (iv)IC4: , ,

Table 2: Parameters values for model (1).

and varying parameters and while keeping other parameters constant. From Theorems 2 and 6, the corresponding stabilities of model (1) are as follows:(1)If , the IFE is globally asymptotically stable.(2)If , the EE is globally asymptotically stable.

(i) and . For this set of parameters we obtain . By Theorem 6 the endemic equilibrium is globally asymptotically stable and the states of the system converge to , for the four initial conditions IC1-IC4, implying that the virus persists in the host. This shows that the low treatment efficacy of 0.1 cannot effectively combat the virus. This is illustrated by Figures 1, 2, 3, and 4.(ii) and . This set of parameters yields and the system state approach for all the four initial conditions IC1-IC4. This means that exists and is globally asymptotically stable according to Theorem 6. Again despite the increase in treatment efficacy the virus persists in the host. Figures 1, 2, 3, and 4 illustrate this scenario.(iii) and . This treatment efficacy yields and by Theorem 2 exists and is GAS for all initial values IC1-IC4, as can be observed from Figures 1, 2, 3, and 4 implying that the high treatment efficacy of over 0.82 helps in stifling viral replication within host.

Figure 1: Effect of treatment efficacy on , and .
Figure 2: (a)-(c) The graph trajectory of ,, and for and with the initial conditions IC2-IC4. and is GAS.
Figure 3: (a)-(c) The graph trajectory of ,, and for and with the initial conditions IC2-IC4. and is GAS.
Figure 4: (a)-(c) The graph trajectory of ,, and for and with the initial conditions IC2-IC4. and is globally asymptotically stable.
4.2. Effect of Variations in RTI and PI Treatment Efficacy on the CD4+ T Cell Population and Viral Load

From Figures 1(a), 1(b), and 1(c), it is observed that variations in the treatment efficacy have significant effect on the number of CD4+ T cells. For instance, with a treatment efficacy of 0.1 at the onset of HIV infection, the CD4+ T cells undergo a sharp decline, while treatment efficacy of 0.5 only slows down CD4+ T cell depletion and takes longer (approximately 40 days) to reduce to zero. However with a drug efficacy of 0.9 the number of CD4+ T cells is kept above 700 . This demonstrates that treatment efficacy is directly proportional to the number of CD4+ T cells.

Figures 1(b) and 1(c) illustrate the effect of treatment efficacy on the number of actively infected cells and free virions, respectively. It is clear that from the time of infection if treatment efficacy is kept at a minimum of 0.1, then both actively infected cells and free virions replicate rapidly attaining a primary peak of 519 infected cells within the first 17 days, while the virions reach a peak of within the same period of time. With an increase in treatment efficacy both infected cells and viral replication rates are significantly reduced; for instance, at 0.9 efficacy level the number of virus and infected cells approach zero within the first 90 days.

5. Conclusion

In this paper, a mathematical model for within host HIV infection with virus-to-cell and cell-to-cell treatment has been formulated and analyzed. Target cell production has been modelled by a logistic incidence rate. The global dynamics have been shown to be dependent on , in which for values of the infection free equilibrium is globally asymptotically stable, and for the endemic equilibrium is globally asymptotically stable. The global asymptotic stability of IFE and EE has been proved using the method by Castillo-Chavez and the geometric approach, respectively. Numerical simulations have shown that combined treatment has significant effect on the dynamics of HIV infection, in which high levels of treatment efficacy for both RTI and PI treatment was observed to stifle viral replication. This observation is in agreement with the findings of [9] which showed that if before treatment then the virus will increase as would be the number of infected cells, but after treatment if , then and would both decline.

Data Availability

Previously reported data were used to support this study and are available at [9]. These prior studies (and datasets) are cited at relevant places within the text as references.

Conflicts of Interest

The authors declare that there are no conflicts of interest regarding the publication of this paper.

References

  1. M. A. Nowak and R. M. May, Virus Dynamics: Mathematics Principles of Immunology and Virology, Oxford University Press, London, UK, 2000. View at MathSciNet
  2. S.-S. Chen, C.-Y. Cheng, and Y. Takeuchi, “Stability analysis in delayed within-host viral dynamics with both viral and cellular infections,” Journal of Mathematical Analysis and Applications, vol. 442, no. 2, pp. 642–672, 2016. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  3. A. M. Elaiw and N. A. Almuallem, “Global dynamics of delay-distributed {HIV} infection models with differential drug efficacy in cocirculating target cells,” Mathematical Methods in the Applied Sciences, vol. 39, no. 1, pp. 4–31, 2016. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  4. A. M. Elaiw and A. A. Raezah, “Stability of general virus dynamics models with both cellular and viral infections and delays,” Mathematical Methods in the Applied Sciences, vol. 40, no. 16, pp. 5863–5880, 2017. View at Publisher · View at Google Scholar · View at MathSciNet
  5. Y. Yang, L. Zou, and S. Ruan, “Global dynamics of a delayed within-host viral infection model with both virus-to-cell and cell-to-cell transmissions,” Mathematical Biosciences, vol. 270, no. part B, pp. 183–191, 2015. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  6. H. Pourbashash, S. S. Pilyugin, P. De Leenheer, and C. McCluskey, “Global analysis of within host virus models with cell-to-cell viral transmission,” Discrete and Continuous Dynamical Systems - Series B, vol. 19, no. 10, pp. 3341–3357, 2014. View at Publisher · View at Google Scholar · View at Scopus
  7. X. Zhang and X.-N. Liu, “Backward bifurcation of an epidemic model with saturated treatment function,” Journal of Mathematical Analysis and Applications, vol. 348, no. 1, pp. 433–443, 2008. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  8. J. Zhang, J. Jia, and X. Song, “Analysis of an SEIR epidemic model with saturated incidence and saturated treatment function,” The Scientific World Journal, vol. 2014, 2014. View at Google Scholar · View at Scopus
  9. A. S. Perelson, D. E. Kirschner, and R. D. Boer, “Dynamics of HIV infection of CD4+ T cells,” Mathematical Biosciences, vol. 114, no. 1, pp. 81–125, 1993. View at Publisher · View at Google Scholar · View at Scopus
  10. D. E. Kirschner and G. F. Webb, “A Mathematical Model of Combined Drug Therapy of HIV Infection,” Journal of Theoretical Medicine, vol. 1, no. 1, pp. 25–34, 1997. View at Publisher · View at Google Scholar
  11. F. K. Tireito, G. O. Lawi, and C. O. Okaka, “Mathematical analysis of HIV/AIDS prophylaxis treatment model,” Applied Mathematical Sciences, vol. 12, no. 18, pp. 893–902, 2018. View at Publisher · View at Google Scholar
  12. S. Pankavich, “The effects of latent infection on the dynamics of HIV,” Differential Equations and Dynamical Systems, vol. 24, no. 3, pp. 281–303, 2016. View at Publisher · View at Google Scholar · View at MathSciNet
  13. C. Castillo-Chavez, Z. Feng, and W. Huang, “On the computation of R0 and its role on global stability,” in Mathematical Approaches for Emerging and Reemerging Infectious Diseases: An Introduction, C. Castillo-Chavez, S. Blower, P. van den Driessche, D. Kirschner, and A.-A. Yakubu, Eds., vol. 125, pp. 229–250, Springer, New York, NY, USA, 2002. View at Publisher · View at Google Scholar · View at MathSciNet
  14. Van Den Driessche and J. Wu, Mathematical Epidemiology, Springer-Varleg, Berlin Heidelberg, Germany, 2008.
  15. M. Y. Li and J. S. Muldowney, “A geometric approach to global-stability problems,” SIAM Journal on Mathematical Analysis, vol. 27, no. 4, pp. 1070–1083, 1996. View at Publisher · View at Google Scholar · View at MathSciNet · View at Scopus
  16. W. A. Coppel, Stability and Asymptotic Behavior of Differential Equations, D. C. Heath, Boston, Mass, USA, 1965. View at MathSciNet