# American Institute of Mathematical Sciences

## Modeling within-host viral dynamics: The role of CTL immune responses in the evolution of drug resistance

 1 Department of Mathematics, Nanjing University of Science and Technology, Nanjing 210094, China 2 Department of Mathematics, Nanjing University of Science and Technology, Nanjing 210094, China, Department of Mathematics, University of Florida, Gainesville, FL 32611, USA

* Corresponding author: Zhipeng Qiu

Received  February 2020 Revised  June 2020 Published  August 2020

Fund Project: T. Guo was supported by the CSC (201806840119) and the NSFC (11971232). Z. Qiu was supported by the NSFC (11671206). L. Rong was supported by the NSF Grant DMS-1758290

To study the emergence and evolution of drug resistance during treatment of HIV infection, we study a mathematical model with two strains, one drug-sensitive and the other drug-resistant, by incorporating cytotoxic T lymphocyte (CTL) immune response. The reproductive numbers for each strain with and without the CTL immune response are obtained and shown to determine the stability of the steady states. By sensitivity analysis, we evaluate how the changes of parameters influence the reproductive numbers. The model shows that CTL immune response can suppress the development of drug resistance. There is a dynamic relationship between antiretroviral drug administration, the prevalence of drug resistance, the total level of viral production, and the strength of immune responses. We further investigate the scenario under which the drug-resistant strain can outcompete the wild-type strain. If drug efficacy is at an intermediate level, the drug-resistant virus is likely to arise. The slower the immune response wanes, the slower the drug-resistant strain grows. The results suggest that immunotherapy that aims to enhance immune responses, combined with antiretroviral drug treatment, may result in a functional control of HIV infection.

Citation: Qi Deng, Zhipeng Qiu, Ting Guo, Libin Rong. Modeling within-host viral dynamics: The role of CTL immune responses in the evolution of drug resistance. Discrete & Continuous Dynamical Systems - B, doi: 10.3934/dcdsb.2020245
##### References:

show all references

##### References:
The schematic diagram of HIV transmission with wild-type and resistant virus
The dynamics of infection before converging to steady states. $(a)-(d)$ represent uninfected cells, wild-type virus, resistant virus, and CTL cells, respectively. The blue solid line is the case in which CTL cells are present. The parameters are from Table 1 and $b = 0.1$. We have $R_{s} = 3.13>R_{r} = 1.74>1$, $R_{cs} = 68.1>1$. Solution trajectories of the system converge to the positive equilibrium $E^{\star} = (415257, 14994.3, 586870, 1.01219, 26.4110, 44.9845)$. The red dotted line refers to the case in which CTL cells are absent. The parameter $c$ is 0. This yields $R_{s} = 3.13>R_{r} = 1.74>1$, $R_{cs} = 0<1$. Solution trajectories of the system converge to the CTL immune-free equilibrium $E^{f} = (318355, 23056.2, 902393, 1.55640, 40.6104, 0)$
Sensitivity analysis for four reproduction numbers. (a) The PRCC of $R_{s}$ for seven parameters, (b) the PRCC of $R_{r}$ for six parameters, (c) the PRCC of $R_{cs}$ for nine parameters, and (d) the PRCC of $R_{cr}$ for eight parameters
Predicted dynamics of system (4) under treatment. $(a)-(d)$ represent uninfected cells, wild-type virus, resistant virus, and CTL cells, respectively. The blue solid line represents the case in which CTL cells are present. The parameters given in Table 1 are used and $b=0.1$. This yields $R_{s}^{\prime}=1.57>R_{r}^{\prime}=1.54>1$ and $R_{cs}^{\prime}=36.30>1$. Thus, solutions trajectories of the system converge to the interior equilibrium $E^{\star\prime}=(733912, 7673.78, 243257, 13.8008, 346.384, 23.0628)$. The red dotted line refers to the case in which CTL cells are absent. Parameters from Table 1 are used and $b=0.1, c=0$. In this case, $R_{s}^{\prime}=1.57>R_{r}^{\prime}=1.54>1$ and $R_{cs}^{\prime}=0<1$. Solutions trajectories of the system converge to the CTL immune-free equilibrium $E^{f\prime}=(635966, 12174.6, 476875, 21.8671, 571.018, 0)$
Invasion of drug-resistant virus. (a) Effects of the infection rate $\beta_{s}$ and overall drug efficacy $\eta_{s}$ on $R_{s}^{\prime}$ and $R_{r}^{\prime}$. (b) The projection of (a) on the $\beta_{s}-\eta_{s}$ plane. (c) Part of (b) when $\eta_{s}$ is from 0.4 to 0.6. The vertical line in magenta represents $\eta_{s} = 0.51$. In the simulation, we assume that $\eta_{r} = 0.23\eta_{s}, \beta_{r} = 0.83\beta_{s}$. The other parameters are given in Table 1
Predicted dynamics of system (4) when drug resistance invades during treatment. Figure $(a)-(b)$ shows the wild-type virus and the resistant virus, respectively. $\eta_{s} = 0.6876, \eta_{r} = 0.1635$, $R_{s}^{\prime} = 0.9778<1<R_{r}^{\prime} = 1.455, R_{cr} = 312.7>1$. The solutions of the system converge to the infected equilibrium with drug-resistant strain and CTL immune response $E^{rc\prime} = (928780, 0, 0, 1756.85, 45817.3, 52.7080)$. Figure $(c)-(d)$ is similar to $(a)-(b)$ except $\eta_{s} = 0.4330, \eta_{r} = 0.0957$. In this case, $R_{r}^{\prime} = 1.573<R_{s}^{\prime} = 1.775, R_{cs}^{\prime} = 436.6>1$. Both strains of virus persist and the solutions of the system converge to the interior equilibrium $E^{\star\prime} = (872574, 2744.46, 86963.0, 0.722593, 18.1288, 82.2710)$. The parameters are from Table 1
Dynamics of system (4) with different values of $c$ (the proliferation rate of CTLs). Green dashed-dot line is for $c = 0$, black solid line is for $c = 0.00003$, blue dashed line is for $c = 0.0002$, and red dot line is for $c = 0.001$. We chose $\eta_{s} = 0.5328, \eta_{r} = 0.1221$ and the other parameters can be found in Table 1
Drug resistance dynamics under different drug efficacies. (a) The immune clearance rate is $b = 0.1$ and (b) $b = 0.01$. Green curve is for $\eta_{s} = 0.52, \eta_{r} = 0.12$, red curve is for $\eta_{s} = 0.69, \eta_{r} = 0.16$, and blue curve is for $\eta_{s} = 0.96, \eta_{r} = 0.29$. $c$ is fixed at $0.001$ and other parameters are given in Table 1
(a) The difference between non-infectious and infectious wild-type virus concentrations. (b) The difference between non-infectious and infectious drug-resistant virus concentrations. Shortly after initiation of potent antiretroviral therapy, the difference between non-infectious and infectious viral levels (logarithm with base 10) approaches a constant. The parameters can be found in Table 1
Parameter descriptions and sources for their values
 Parameter Value Ranges Description Reference $\lambda$ $10^{4}ml^{-1}day^{-1}$ $100\sim200000$ Production rate of uninfected cells [42] $d$ $0.01day^{-1}$ $0.0001\sim 0.8$ Death rate of uninfected cells [34] $\beta_{s}$ $2.4\times10^{-8}ml\cdot day^{-1}$ $10^{-6}\sim2.4\times10^{-6}$ Infection rate of uninfected cells by wild-type virus [34] $\beta_{r}$ $2.0\times10^{-8}ml\cdot day^{-1}$ $10^{-6}\sim2.0\times10^{-6}$ Infection rate of uninfected cells by drug-resistant virus [42] $\delta$ $0.3day^{-1}$ $0.1\sim0.9$ Death rate of infected cells [42] $k$ $0.002mm^{3}day^{-1}$ - Clearance rate of infected cells by CTL killing [48, 1] $p_{s}$ $900day^{-1}$ $90\sim1000$ Generation rate of wild-type virus [8, 17] $p_{r}$ $600day^{-1}$ $60\sim1000$ Generation rate of drug-resistant virus [8, 17] $d_{1}$ $23day^{-1}$ $10\sim50$ Clearance rate of wild-type and resistant virus [39] $c$ $0.0003ml^{-1}day^{-1}$ $0\sim0.001$ generation rate of CTL [55, 61] $b$ $0.01day^{-1}$ $0.01\sim1$ death rate of CTL [48, 61] $\mu$ $3\times10^{-5}$ $10^{-6}\sim10^{-4}$ Single mutation rate [24] $\varepsilon_{RTI}$ Varied $0\sim1$ Efficacy of RTI see text $\varepsilon_{PI}$ Varied $0\sim1$ Efficacy of PI see text $\sigma_{RTI}$ Varied $0\sim1$ Resistance ratio of RTI see text $\sigma_{PI}$ Varied $0\sim1$ Resistance ratio of PI see text
 Parameter Value Ranges Description Reference $\lambda$ $10^{4}ml^{-1}day^{-1}$ $100\sim200000$ Production rate of uninfected cells [42] $d$ $0.01day^{-1}$ $0.0001\sim 0.8$ Death rate of uninfected cells [34] $\beta_{s}$ $2.4\times10^{-8}ml\cdot day^{-1}$ $10^{-6}\sim2.4\times10^{-6}$ Infection rate of uninfected cells by wild-type virus [34] $\beta_{r}$ $2.0\times10^{-8}ml\cdot day^{-1}$ $10^{-6}\sim2.0\times10^{-6}$ Infection rate of uninfected cells by drug-resistant virus [42] $\delta$ $0.3day^{-1}$ $0.1\sim0.9$ Death rate of infected cells [42] $k$ $0.002mm^{3}day^{-1}$ - Clearance rate of infected cells by CTL killing [48, 1] $p_{s}$ $900day^{-1}$ $90\sim1000$ Generation rate of wild-type virus [8, 17] $p_{r}$ $600day^{-1}$ $60\sim1000$ Generation rate of drug-resistant virus [8, 17] $d_{1}$ $23day^{-1}$ $10\sim50$ Clearance rate of wild-type and resistant virus [39] $c$ $0.0003ml^{-1}day^{-1}$ $0\sim0.001$ generation rate of CTL [55, 61] $b$ $0.01day^{-1}$ $0.01\sim1$ death rate of CTL [48, 61] $\mu$ $3\times10^{-5}$ $10^{-6}\sim10^{-4}$ Single mutation rate [24] $\varepsilon_{RTI}$ Varied $0\sim1$ Efficacy of RTI see text $\varepsilon_{PI}$ Varied $0\sim1$ Efficacy of PI see text $\sigma_{RTI}$ Varied $0\sim1$ Resistance ratio of RTI see text $\sigma_{PI}$ Varied $0\sim1$ Resistance ratio of PI see text
Summary of the stability results of system (1)
 Conditions System (1) $R_{0<1}$ $E^{0}$ $is$ $L.A.S$ $R_{0}>1$ $R_{s}1}$ $E^{rc}$ $is$ $L.A.S$ $R_{s}>R_{r}$ $R_{cs<1}$ $E^{f}$ $is$ $L.A.S$ $R_{cs>1}$ $E^{\star}$ $is$ $L.A.S$ $(\hbox{by simulation})$
 Conditions System (1) $R_{0<1}$ $E^{0}$ $is$ $L.A.S$ $R_{0}>1$ $R_{s}1}$ $E^{rc}$ $is$ $L.A.S$ $R_{s}>R_{r}$ $R_{cs<1}$ $E^{f}$ $is$ $L.A.S$ $R_{cs>1}$ $E^{\star}$ $is$ $L.A.S$ $(\hbox{by simulation})$
 [1] Seung-Yeal Ha, Shi Jin. Local sensitivity analysis for the Cucker-Smale model with random inputs. Kinetic & Related Models, 2018, 11 (4) : 859-889. doi: 10.3934/krm.2018034 [2] Rui Hu, Yuan Yuan. Stability, bifurcation analysis in a neural network model with delay and diffusion. Conference Publications, 2009, 2009 (Special) : 367-376. doi: 10.3934/proc.2009.2009.367 [3] Baba Issa Camara, Houda Mokrani, Evans K. Afenya. Mathematical modeling of glioma therapy using oncolytic viruses. Mathematical Biosciences & Engineering, 2013, 10 (3) : 565-578. doi: 10.3934/mbe.2013.10.565 [4] Jan Prüss, Laurent Pujo-Menjouet, G.F. Webb, Rico Zacher. Analysis of a model for the dynamics of prions. Discrete & Continuous Dynamical Systems - B, 2006, 6 (1) : 225-235. doi: 10.3934/dcdsb.2006.6.225 [5] Sohana Jahan. Discriminant analysis of regularized multidimensional scaling. Numerical Algebra, Control & Optimization, 2021, 11 (2) : 255-267. doi: 10.3934/naco.2020024 [6] Rafael Luís, Sandra Mendonça. A note on global stability in the periodic logistic map. Discrete & Continuous Dynamical Systems - B, 2020, 25 (11) : 4211-4220. doi: 10.3934/dcdsb.2020094 [7] Lakmi Niwanthi Wadippuli, Ivan Gudoshnikov, Oleg Makarenkov. Global asymptotic stability of nonconvex sweeping processes. Discrete & Continuous Dynamical Systems - B, 2020, 25 (3) : 1129-1139. doi: 10.3934/dcdsb.2019212 [8] Michael Grinfeld, Amy Novick-Cohen. Some remarks on stability for a phase field model with memory. Discrete & Continuous Dynamical Systems - A, 2006, 15 (4) : 1089-1117. doi: 10.3934/dcds.2006.15.1089 [9] Guangying Lv, Jinlong Wei, Guang-an Zou. Noise and stability in reaction-diffusion equations. Mathematical Control & Related Fields, 2021  doi: 10.3934/mcrf.2021005 [10] Qiang Guo, Dong Liang. An adaptive wavelet method and its analysis for parabolic equations. Numerical Algebra, Control & Optimization, 2013, 3 (2) : 327-345. doi: 10.3934/naco.2013.3.327 [11] Vieri Benci, Sunra Mosconi, Marco Squassina. Preface: Applications of mathematical analysis to problems in theoretical physics. Discrete & Continuous Dynamical Systems - S, 2021, 14 (5) : i-i. doi: 10.3934/dcdss.2020446 [12] Gloria Paoli, Gianpaolo Piscitelli, Rossanno Sannipoli. A stability result for the Steklov Laplacian Eigenvalue Problem with a spherical obstacle. Communications on Pure & Applied Analysis, 2021, 20 (1) : 145-158. doi: 10.3934/cpaa.2020261 [13] Scipio Cuccagna, Masaya Maeda. A survey on asymptotic stability of ground states of nonlinear Schrödinger equations II. Discrete & Continuous Dynamical Systems - S, 2021, 14 (5) : 1693-1716. doi: 10.3934/dcdss.2020450 [14] Pengfei Wang, Mengyi Zhang, Huan Su. Input-to-state stability of infinite-dimensional stochastic nonlinear systems. Discrete & Continuous Dynamical Systems - B, 2021  doi: 10.3934/dcdsb.2021066 [15] Liangliang Ma. Stability of hydrostatic equilibrium to the 2D fractional Boussinesq equations. Discrete & Continuous Dynamical Systems - B, 2021  doi: 10.3934/dcdsb.2021068 [16] Akio Matsumot, Ferenc Szidarovszky. Stability switching and its directions in cournot duopoly game with three delays. Discrete & Continuous Dynamical Systems - B, 2021  doi: 10.3934/dcdsb.2021069 [17] Martial Agueh, Reinhard Illner, Ashlin Richardson. Analysis and simulations of a refined flocking and swarming model of Cucker-Smale type. Kinetic & Related Models, 2011, 4 (1) : 1-16. doi: 10.3934/krm.2011.4.1 [18] Israa Mohammed Khudher, Yahya Ismail Ibrahim, Suhaib Abduljabbar Altamir. Individual biometrics pattern based artificial image analysis techniques. Numerical Algebra, Control & Optimization, 2021  doi: 10.3934/naco.2020056 [19] Dan Wei, Shangjiang Guo. Qualitative analysis of a Lotka-Volterra competition-diffusion-advection system. Discrete & Continuous Dynamical Systems - B, 2021, 26 (5) : 2599-2623. doi: 10.3934/dcdsb.2020197 [20] Hailing Xuan, Xiaoliang Cheng. Numerical analysis and simulation of an adhesive contact problem with damage and long memory. Discrete & Continuous Dynamical Systems - B, 2021, 26 (5) : 2781-2804. doi: 10.3934/dcdsb.2020205

2019 Impact Factor: 1.27

## Tools

Article outline

Figures and Tables