arXiv subjects

Kathleen P. Wilkie

Publications and source records attributed to Kathleen P. Wilkie.

4 recordsLinked to original sources

arXiv2025

How Mathematical Forms of Chemotherapy and Radiotherapy Bias Model-Optimized Predictions: Implications for Model Selection

The move towards personalized treatment and digital twins for cancer therapy requires a complete understanding of the mathematical models upon which these optimized simulation-based strategies are formulated. This study investigates the influence of mathematical model selection on the optimization of chemotherapy and radiotherapy protocols. By examining three chemotherapy models (log-kill, Norton-Simon, and Emax), and three radiotherapy models (linear-quadratic, proliferation saturation index, and continuous death-rate), we identify similarities and significant differences in the optimized protocols. We demonstrate how the assumptions built into the model formulations heavily influence optimal treatment dosing and sequencing, potentially leading to contradictory results. Further, we demonstrate how different model forms influence predictions in the adaptive therapy setting. As treatment decisions increasingly rely on simulation-based strategies, unexamined model assumptions can introduce bias, leading to model-dependent recommendations that may not be generalizable. This study highlights the importance of basing model selection on a full analysis of bias, sensitivity, practical parameter identifiability and/or inferred parameter posteriors, as a part of the uncertainty quantification process, rather than solely relying on information criterion. Understanding how model choice impacts predictions guiding personalized treatment planning with sufficient uncertainty quantification analysis, will lead to more robust and generalizable predictions.

q-bio.QM↗

arXiv2023

A closer look at parameter identifiability, model selection and handling of censored data with Bayesian Inference in mathematical models of tumour growth

Mathematical models (MMs) are a powerful tool to help us understand and predict the dynamics of tumour growth under various conditions. In this work, we use 5 MMs with an increasing number of parameters to explore how certain (often overlooked) decisions in estimating parameters from data of experimental tumour growth affect the outcome of the analysis. In particular, we propose a framework for including tumour volume measurements that fall outside the upper and lower limits of detection, which are normally discarded. We demonstrate how excluding censored data results in an overestimation of the initial tumour volume and the MM-predicted tumour volumes prior to the first measurements, and an underestimation of the carrying capacity and the MM-predicted tumour volumes beyond the latest measurable time points. We show in which way the choice of prior for the MM parameters can impact the posterior distributions, and illustrate that reporting the highest-likelihood parameters and their 95% credible interval can lead to confusing or misleading interpretations. We hope this work will encourage others to carefully consider choices made in parameter estimation and to adopt the approaches we put forward herein.

q-bio.QM↗

arXiv2022

Stochastic failure of cell infection post viral entry: Implications for infection outcomes and antiviral therapy

A virus infection can be initiated with very few or even a single infectious virion, and as such can become extinct, i.e. stochastically fail to take hold or spread significantly. There are many ways that a fully competent infectious virion, having successfully entered a cell, can fail to cause a productive infection, i.e. one that yields infectious virus progeny. Though many discrete, stochastic mathematical models (DSMs) have been developed and used to estimate a virus infection's extinction probability, these typically neglect infection failure post viral entry. The DSM presented herein introduces parameter $γ\in(0,1]$ which corresponds to the probability that a virion's entry into a cell will result in a productive cell infection. We derive an expression for the likelihood of infection extinction in this new DSM, and find that prophylactic therapy with an antiviral acting to reduce $γ$ is best at increasing an infection's extinction probability, compared to antivirals acting on the rates of virus production or virus entry into cells. Using the DSM, we investigate the difference in the fraction of cells consumed by so-called extinct versus established virus infections, and find that this distinction becomes biologically meaningless as the probability of extinction approaches 100%. We show that infections wherein virus is release by an infected cell as a single burst, rather than at a constant rate over the cell's infectious lifespan, has the same probability of infection extinction, despite previous claims to this effect [Pearson 2011, doi:10.1371/journal.pcbi.1001058]. Instead, extending previous work by others [Yan 2016, doi:10.1007/s00285-015-0961-5], we show how the assumed distribution for the stochastic virus burst size, affects the extinction probability and associated critical antiviral efficacy.

q-bio.CB↗

arXiv2013

Modeling the Dichotomy of the Immune Response to Cancer: Cytotoxic Effects and Tumor-Promoting Inflammation

Although the immune response is often regarded as acting to suppress tumor growth, it is now clear that it can be both stimulatory and inhibitory. The interplay between these competing influences has complex implications for tumor development and cancer dormancy. To study this biological phenomenon theoretically we construct a minimally parameterized framework that incorporates all aspects of the immune response. We combine the effects of all immune cell types, general principles of self-limited logistic growth, and the physical process of inflammation into one quantitative setting. Simulations suggest that while there are pro-tumor or antitumor immunogenic responses characterized by larger or smaller final tumor volumes, respectively, each response involves an initial period where tumor growth is stimulated beyond that of growth without an immune response. The mathematical description is non-identifiable which allows us to capture inherent biological variability in tumor growth that can significantly alter tumor-immune dynamics and thus treatment success rates. The ability of this model to predict immunomodulation of tumor growth may offer a template for the design of novel treatment approaches that exploit immune response to improve tumor suppression, including the potential attainment of an immune-induced dormant state.

q-bio.CB↗