Cancer biology has relied on genetic variants as a roadmap for therapy. Linking them to drugs is complicated by the non-linear nature of biological systems. Cells can adapt their phenotype during treatment, making treatment strategies focused only on their initial makeup flawed. We propose a different approach that considers this cellular dynamics. The focus in this project is on acute lymphoid leukemia which is the most common pediatric cancer. Using patient time series profiles, we have identified distinct drug-tolerant cell states, creating opportunities for drug targeting. A cell state involving expression of genes that function in cellular memory regulation in blood cells is the focus of this study. Our goal is to develop strategies to block this mechanism through systems biology and artificial intelligence approaches to reduce toxic chemotherapy burden in children. The technological approaches developed can be used beyond ALL in predictive and preventive medicine.

2026

2030