Integrating DNA Methylation and Drug Abundance at Single-Cell Resolution to Predict Cancer Drug Sensitivity

Ziel

One of the greatest challenges in cancer therapy is treatment resistance, which arises from the selection of pre-existing cell states carrying survival advantages during treatment. Although genetic alterations have been extensively studied, genetics alone accounts for only a subset of relapses observed in clinical settings. EPI-SENSOR proposes that therapy response is driven by epigenetic divergence – the extent to which a tumor cell’s epigenetic state deviates from its cell-of-origin. I hypothesize that pre-existing epigenetic divergence determines whether individual cancer cells are sensitive or resistant to therapy. Testing this hypothesis requires a joint characterization of genetic mutations, differentiation state, epigenetic patterns, and drug abundance within a single cell - a level of integration that has not previously been possible. This underscores the need for novel research methods. We have developed innovative single-cell epigenomic technologies and computational tools that enable us to test our hypothesis. Using acute myeloid leukemia as a cancer model system, EPI-SENSOR will profile longitudinal patient samples at diagnosis and during cancer treatment, utilizing joint profiling of DNA methylation, somatic mutations, and surface protein expression in single cells. This approach will enable us to understand which molecular mechanisms determine sensitive or resistant cell phenotypes. Furthermore, we will develop a novel approach that tags cancer drugs with an oligonucleotide compatible with single-cell readouts. This approach will quantify cancer drug abundance and DNA methylation states at the single-cell level. Finally, we will develop a prognostic tool based on bulk DNA methylation data to implement EPI-SENSOR in clinical practice. All of these approaches combined will reveal non-genetic mechanisms of therapy resistance, define predictive DNA methylation signatures, and provide translational tools for personalized cancer treatment.

Wissenschaftliches Gebiet (EuroSciVoc)

Dieses Projekt wurde noch nicht bei EuroSciVoc klassifiziert.
Schlagen Sie die Wissenschaftsbereiche vor, die Ihrer Einschätzung nach besonders relevant sind, und helfen Sie uns, unseren Klassifizierungsdienst zu verbessern.