Fig. 2

A proposed individualized dynamic study in colorectal cancer. Multiple hypotheses are tested in this parallel individualized dynamic design. This hypothetical example is in colorectal cancer patients after progression on standard therapies. Baseline tumor characterization includes whole-genome sequencing (WGS)/whole-exome sequencing (WES) and transcriptome sequencing from fresh tumor biopsies, circulating tumor DNA (ctDNA) sampling, immune profiling, and radiomics analysis. Patient-derived xenografts (PDXs)/patient-derived organoids (PDOs) are also generated. Drug therapy is then tailored to each patient’s mutational and immune profile. While on treatment, serial ctDNA sampling occurs 4 weekly and radiomics is performed every 8 weeks to guide therapeutic decisions. Patient one is used as an example: (1) at week 0, started on programmed cell death protein-1 (PD-1) inhibitor and MEK inhibitor; (2) at week 12, treatment is changed to phosphoinositide 3-kinase (PI3K) inhibitor and MEK inhibitor due to the increase in the allele frequency of a PIK3CA mutation; and (3) at week 20, the allele frequencies of both PIK3CA and KRAS mutations continue to rise and treatment is changed to therapy informed by PDX/PDO data. CT computed tomography, mut mutation, PD progressive disease, PR partial response, SD stable disease, wt wild type, MSI microsatellite instability, inh inhibitor