Christian Hove Claussen (1), Anetta Hove Claussen (1)
(1) PKPD Consult
Introduction: Treatment of cancer requires careful consideration of the effect of the drug on the tumor while taking into account the adverse effects on sensitive tissues such as the bone marrow. While the drug dose is the focus of optimization for effect and safety, the cycle time and number of cycles is often set based on past clinical experience. In this work, we present an approach to describe the growth and inhibition of tumors in the mouse xenograft model mathematically with the purpose of utilizing this rich source of preclinical information in drug research projects to predict optimal doses, cycle times, and number of cycles. We hypothesize that the timescales derived from this optimization relates to the one-dimensional nature of tumor spheroids and the rate of vascularization of tumors following a challenge.
Methods: Literature data from several sources (such as [1, 2]) were selected based a) the associated drug being a chemotherapeutic or radiotherapeutic drug, b) tumor size profiles for growth, challenge, and regrowth phases being available, and c) sufficient dosing information being available to construct approximate pharmacokinetic models based on dose and half-lives for the drug. The data was digitized and a semi-mechanistic tumor-growth model driven by the simulated kinetics of the drug was fitted to the data, allowing for growth, inhibition, and subsequent regrowth or death of the tumor, the latter based on a probabilistic component of the model with a size threshold for survivability of the tumor. Tumor inhibition time (until regrowth) and fraction of dead tumors were simulated as a function of dose, cycle time, and number of cycles to allow for optimization.
Results: The analysis indicated that a) tumor growth is a one-dimensional process such that the tumor radius or diameter is a linear function of time after the tumor has grafted, b) tumors regrow at a linear rate similar to the pre-challenge rate following a delay caused by inhibition, and c) cycle time and number of cycles are limiting factors determining this delay as well as whether the tumor is eradicated as long as the cycle dose is sufficiently high. Cycle times in the order of weeks were found to be optimal and it is likely that this timescale is related to physiologically determining factors such as tumor (re-)vascularization rate, removal rate of dead or damaged cancer cells by the immune system, and surface-based growth and vascularization of a tumor with hypoxic dormant centers.
Conclusions: By describing tumor growth as a linear function of time with drug-induced inhibition, a semi-mechanistic model incorporating probabilistic tumor death was able to provide a generic description of tumor growth inhibition in mice. This modeling approach should be used to optimize treatment cycles in oncology based on the operating timescales of the tumor physiology and immune system derived in mice and translated to humans. Thus, the flexibility of the mouse xenograft model can be utilized to explore and optimize cycles for the benefit of clinical studies and patients.
References:
[1] Pedley, R. Barbara, et al. Eradication of colorectal xenografts by combined radioimmunotherapy and combretastatin a-4 3-O-phosphate. Cancer research 61.12 (2001): 4716-4722.
[2] Chou, Ting-Chao, et al. Therapeutic effect against human xenograft tumors in nude mice by the third generation microtubule stabilizing epothilones. Proceedings of the National Academy of Sciences 105.35 (2008): 13157-13162.
Reference: PAGE 30 (2022) Abstr 10232 [www.page-meeting.org/?abstract=10232]
Poster: Drug/Disease Modelling - Oncology