Abstract
AbstractA precision medicine approach in type 2 diabetes (T2D) could enhance targeting specific glucose-lowering therapies to individual patients most likely to benefit. We utilised Bayesian non-parametric modelling to develop and validate an individualised treatment selection algorithm for two major T2D drug classes, SGLT2-inhibitors (SGLT2i) and GLP1-receptor agonists (GLP1-RA). The algorithm is designed to predict differences in 12-month glycaemic outcome (HbA1c) between the 2 therapies, based on routine clinical features of 46,394 people with T2D in England (27,319 for model development, 19,075 for hold-out validation), with additional external validation in 2,252 people with T2D from Scotland. Routine clinical features, including sex (with females markedly more responsive to GLP1-RA), were associated with differences in glycaemic outcomes. Our algorithm identifies clearly delineable subgroups with reproducible ≥5mmol/mol HbA1cbenefits associated with each drug class. Moreover, we demonstrate that targeting the therapies based on predicted glycaemic response is associated with improvements in short-term tolerability and long-term risk of new-onset microvascular complications. These results show that precision medicine approaches to T2D can facilitate effective individualised treatment selection, and that use of routinely collected clinical features could support low-cost deployment in many countries.
Publisher
Cold Spring Harbor Laboratory