Precision medicine in diabetes care

Purpose of review This review highlights emerging evidence supporting the premise of precision diabetes care including but not limited to monogenic diabetes and discuss potential opportunities, challenges, and limitations for clinical adoption. Recent findings Driven by a single gene mutation, monogenic diabetes remains the best use-case for precision diabetes care. However, the increasing prevalence of diabetes among adolescents and young adults in an obesogenic environment makes triaging potential patients for genetic screening clinically challenging. High-dimensional molecular biomarkers (i.e., multiomics) can improve the risk prediction for incident type 2 diabetes (T2D), over and above a well established prediction model based on clinical variables alone. Machine learning approaches using clinical variable-based clustering methods have generated novel and reproducible T2D subgroups with distinct phenotypic and omics characteristics that are associated with differential long-term outcomes. This stratification-strategy may inform clinical decisions. However, on-going discussion and research will be needed to understand the clinical utility of sub-phenotyping T2D for precision care. Summary Precision diabetes care has extended from uncommon monogenic diabetes to T2D which will need more complex approaches like multiomics and machine-learning methods. The successful clinical translation will require cumulative evidence and close collaboration among the stake holders.