The evolution of drug resistance is a major limiting factor to the efficacy of targeted therapies in cancer treatment. Therefore, Leighow et al. developed an alternative approach to control the fate of this tumour evolution and in doing so, ‘forward engineer’ a therapeutic intervention.

The authors began by testing their strategy using a mathematical model in which multiple parameters were input to determine the criteria necessary for the system to succeed. Next, the authors designed the interchangeable switch modules. For switch 1, the kinase domain of a drug target, in this case EGFR harbouring a mutation conferring resistance to the EGFR inhibitor osimertinib, was fused to a synthetic dimerization domain (FKBP12 F36V) to mimic oncogenic signalling. This fusion protein (S1vEGFRosi) is then activated by adding a small molecule dimerizer. For switch 2, an enzyme, in this case cytosine deaminase (S2vCyD), a so-called suicide gene, which converts a prodrug into a cytotoxic compound (5-fluorouracil), was used. Both switches were then cloned into a single construct (S1vEGFRosi–S2vCyD) and transduced into EGFR+ non-small-cell lung cancer PC9 cells. In the presence of dimerizer, these engineered cells were resistant to the EGFR inhibitor osimertinib and were able to grow out in vitro. Subsequent treatment of these expanded engineered cells with both osimertinib and the prodrug 5-fluorocytosine led to their elimination.