Deriving Optimal Treatment Timing for Adaptive Therapy: Matching the Model to the Tumor Dynamics

Adaptive therapy (AT) protocols have been introduced to combat drug resistance in cancer, and are characterized by breaks from maximum tolerated dose treatment (the current standard of care in most clinical settings). These breaks are scheduled to maintain tolerably high levels of tumor burden, employing competitive suppression of treatment-resistant sub-populations by treatment-sensitive sub-populations. AT has been integrated into several ongoing or planned clinical trials, including treatment of metastatic castrate-resistant prostate cancer, ovarian cancer, and BRAF-mutant melanoma, with initial clinical results suggesting that it can offer significant extensions in the time to progression over the standard of care. Prior AT protocols apply drug treatment when the tumor is within a specific size window, typically determined by the initial tumor size. However, this approach may be sub-optimal as it does not account for variation in tumor dynamics between patients, resulting in significant heterogeneity in patient outcomes. Mathematical modeling and analysis have been proposed to optimize adaptive protocols, but they do not account for clinical restrictions, most notably the discrete time intervals between the clinical appointments where a patient's tumor burden is measured and their treatment schedule is re-evaluated. We present a general framework for deriving optimal treatment protocols that account for these discrete time intervals, and derive optimal schedules for several models to avoid model-specific personalization. We identify a trade-off between the frequency of patient monitoring and the time to progression attainable, and propose an AT protocol that determines drug dosing based on a patient-specific threshold for tumor size. Finally, we identify a subset of patients with qualitatively different dynamics that instead require a novel AT protocol based on a threshold that changes over the course of treatment.