Scheduling collateral sensitivity‐based cycling therapies toward eradication of drug‐resistant infections

Abstract

Drug resistant pathogens are a global public health threat and their control has become a challenging task. A new health paradigm has been proposed in recent years through clinical research, this is the sequential use of drugs where resistance to one drug induces sensitivity to another drug, a concept called collateral sensitivity and its converse is known as cross resistance. However, the order and time of cycling between drugs need to be tailored to the pathogen population presented in the host. Here, by abstracting mutation networks of collateral sensitivity based on switched systems, we explore the control theoretical aspects and implications of collateral sensitivity on the impact of eradication of drug-resistant pathogens. Our numerical simulations illustrate the potential implications of our approach to mitigate drug resistance or even eradicate pathogenic populations.

Invariant set theory for predicting failure of antibiotic cycling

“…A network with n sates ( x 1 , x 2 , …, x n ) can be linked to a switched system, as demonstrated in a prior study [64] and by this we can investigate the impact of sequential drug exposures on population dynamics. The effect of therapy is modeled by drug-induced death rate framework, i.e., the growth rate for variant x i under exposure of drug σ is given by with a carry capacity of ; death and mutation from x i to x j rates are given by and , respectively.…”

“…In this study, we introduce a combinatorial mutation network that models qualitative collateral sensitivity data for the assessment of sequential drug regimens. We employed a switched system framework [42], extensively utilized in biomedical control problems [24, 62, 2] and recently adapted for sequential drug regimens with mutation dynamics [64], with the goal of targeting phenotype states associated with specific collateral effects to improve treatment success in chronic infections. The model suggests that collateral sensitivity profiling can forecast the emergence and proliferation of MDR strains.…”