Abstract A017: Computational and mouse models of adaptive therapy with multiple drugs in breast cancer

Abstract

Abstract Adaptive therapy is an evolutionary approach to cancer therapy designed to maintain control of therapeutically resistant cells and thereby extend both time to progression and patient survival. Most of the previous work in adaptive therapy has used a single cytotoxic drug, and yet there are still many variations of adaptive therapy protocols. Although multidrug therapy is standard for most cancers, there is a combinatorial explosion of possible protocols with an increase in the number of drugs used. We set out to explore which protocols are likely to work best, under which conditions, using computational models. We explored 2 drug therapies but allowed the drugs to be cytotoxic or cytostatic. We tested the effects of varying the fitness cost of resistance, the amount of turnover in the tumors, and the ability of cells to replace their neighbors under crowded conditions. Adaptive therapy tends to work best when there is more competition between cells in the cancers. We also explored different dosing schedules with both fixed doses and with varying the doses as a function of tumor burden dynamics. We also selected ER+ breast cancer (MCF7) cells to be resistant to both fulvestrant (an anti-estrogen therapy) and palbociclib (a CDK4/6 targeted therapy). These endocrine-resistant ER+ breast cancer cells represent the target population for a clinical trial in endocrine-resistant ER+ metastatic breast cancer that we are seeking to open at the Mayo Clinic. We tested a few of the adaptive therapy protocols with those cells in immune-compromised mice, using either gemcitabine, capecitabine or both drugs. In general, our computational models suggest that the less drug that a protocol uses, the longer we are able to keep control over the cancers. This was corroborated in our mouse experiments. However, if too little drug is used it can’t even control the sensitive cells in the tumors, leading to a sort of Goldilocks effect for the dosing. Capping the maximum dose at a relatively low level, and modulating the dose below that level, led to long-term control for many of the protocols. Many of the tumors in the mice remained dormant for many weeks after we stopped adaptive therapy, for unknown reasons. Our results support adaptive therapy as a promising approach for both extending time to progression and improving quality of life by reducing toxicity. We are using these results to design our clinical trial. Citation Format: Carlo C. Maley, Sareh Seyedi, Daniel Saha, Alexander Anderson. Computational and mouse models of adaptive therapy with multiple drugs in breast cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Cancer Evolution and Data Science: The Next Frontier; 2023 Dec 3-6; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_2):Abstract nr A017.