Model-Based Systems Engineering for Tracing and Steering Clonal Dynamics
Cancer cell lines have been used for almost 70 years as the workhorse of cancer research. Recent studies have highlighted the contributions of continuous in-vitro evolution to the reproducibility crisis in life sciences, with implications for preclinical drug screening studies and failure rates well above 90% in oncology clinical trials. Extensive and dynamic, genetic and transcriptional heterogeneity within cell lines call for the development of a new framework that connects biological data analysis, next-generation sequencing assays and wet-lab experiments over multiple spatial and temporal scales. We propose Model-Based Systems Engineering as a solution to address these challenges. We present a UML model of an integrated system for identification and characterization of extrinsic and intrinsic factors influencing in-vitro evolution. By modeling steps spanning across experimental and computational domains we identify variations in experimental conditions considered neutral as potential causes for unintended clonal dynamics. Based on these findings we formulate a mathematical model of how small variations in the timing of harvest and culture conditions can shift the clonal composition of a cell line. Our framework represents an early step towards leveraging ongoing in-vitro evolution as an opportunity, not a burden. Because durable success in the clinical setting cannot be achieved without an experimental setting in which our solutions are also constantly challenged with the adaptive nature of cancer.