Synthetic Lethality in Cancer Research Via Genetic Minimal Cut Sets

Synthetic lethality is a promising approach in precision medicine and cancer as it largely expands the number of possible drug targets and creates an opportunity for selectivity. The increasing evidence of metabolic reprogramming of cancer cells makes it ideal to exploit the concept of synthetic lethality. A number of in-silico tools have been developed to target cancer metabolism using a synthetic lethality approach. In particular, constraint-based modeling (CBM) for genome-scale metabolic networks has received much attention. Here, we present a novel CBM approach to synthetic lethality that is based on the concept of genetic Minimal Cut Sets (gMCSs). With respect to existing methods, our approach avoids the step of network contextualization and integrates –omics data in a more natural and objective manner. In addition, it enables not only the detection metabolic targets but also response biomarkers for them. To illustrate our approach, we first show the results of an experimental proof-of-concept in multiple myeloma (MM), where we validated the therapeutic potential of RRM1 inhibition in different MM cell lines. We also predicted a metabolic signature based on gene expression data that explained the response to RRM1 inhibition in different cancer cell lines. Second, we show preliminary results of a study where response biomarkers for the effectiveness of Methotrexate in different cancer types is predicted following our methodology. This new algorithm, freely available in the COBRA Toolbox, undoubtedly opens new avenues to develop precision medicine strategies in complex and unaddressed clinical questions involving heterogeneous molecular data.