Effect of drug resistance on an HIV epidemic in heterogeneous populations

The emergence of drug resistance (DR) may impair the control of the human immunodeficiency virus (HIV) epidemic. We analyze the transmission and drug-resistance dynamics of HIV using a stochastic model in combination with Monte Carlo simulations, which considers population heterogeneity, set in terms of the set-point viral load (SPVL) distribution. In our modeling framework, once a SPVL is sampled, it determines the transmission rate and survival time for each new infected individual. On the other hand, DR emergence randomly occurs with a fixed probability in an individual receiving antiretroviral therapy (ART). In addition, we implicitly assume that the SPVL serves as a proxy for the replication pressure and resistance risk, as the DR fitness-cost proportionally reduces the SPVL in our model (an assumption motivated by modeling convenience instead of an actual biological mechanism present in the DR dynamics). Finally, we simplify the dynamics by assuming no ART regimen switching in cases of treatment failure (which we assumed to be only due to DR). Our results show that for a high treatment coverage, DR causes a higher stationary infection prevalence in populations with moderate-to-high mean viral loads. Moreover, DR could produce more infections for greater ART coverage, especially when treatment is administered early after contagion; however, an earlier initiation of ART allows for the possibility of epidemic extinction. Additionally, we evaluate the effect on the epidemic dynamics of other parameters, such as the probability of DR emergence and the relative fitness of the DR strain. Overall, our analysis contributes to the understanding of drug resistance evolution in situations where ART is inadequately managed.