Speaker
Description
Anti-cancer drug treatment can induce tumor cells to adopt reversible drug-tolerant phenotypic states, setting the stage for the evolution of resistance and eventual treatment failure. This challenges conventional maximum tolerated dose (MTD) regimens by creating a trade-off between short-term tumor reduction and long-term tumor control. Here, we show how mathematical model-informed treatment scheduling can be used to balance this trade-off. Across a family of simple models of drug-induced tolerance, we characterize when and why intermittent treatment is preferable to continuous treatment and low doses are preferable to high doses. On the one hand, our results reveal distinct classes of optimal strategies depending on how the drug affects cell proliferation and phenotypic state transitions, indicating a need for patient-specific approaches. On the other hand, we identify a handful of robust schedules that outperform MTD treatment across a broad range of possible dynamics, suggesting that outcomes can be improved even when the tolerance-inducing effects of the drug are only partially understood.
