Personalized Therapy Selection Algorithm Improves Diabetes Treatment: Study Finds Promise in Tailored Approach
Researchers have developed a personalized therapy selection algorithm that shows promise in improving diabetes treatment, according to a recent study posted on the medRxiv preprint server. The algorithm focuses on two drug classes commonly used for type 2 diabetes (T2D) treatment: sodium-glucose cotransporter 2 (SGLT2) inhibitors and glucagon-like peptide-1 (GLP1) receptor agonist medications.
In the study titled Phenotype-based targeted treatment of SGLT2 inhibitors and GLP-1 receptor agonists in type 2 diabetes, researchers aimed to validate an estimation model that provides individualized estimates of glycemic outcomes for these two therapies over a one-year period. The model uses routine clinical features from a large population of T2D patients in England and Scotland.
The model, built using the Bayesian Causal Forest (BCF) framework, helps identify and estimate conditional average treatment effects (CATEs) on glycated hemoglobin (HbA1c) outcomes based on patients’ clinical characteristics. It also assesses secondary outcomes such as tolerability, weight change, and the risks of adverse renal events and microvascular and macrovascular problems.
Among the 112,274 T2D patients included in the study, the mean uncorrected one-year glycemic responses for GLP1 receptor agonists and SGLT2 inhibitors were -11.7 and -12 mmol/mol, respectively. The BCF framework model identified several clinical characteristics that predicted glycemic responses to these therapies.
The study found that both therapies had comparable average effectiveness but highlighted significant variations in estimated CATEs among individuals. About 48% of participants showed a mean glycemic advantage on SGLT2 inhibitor treatment, while 52% displayed an advantage on GLP1 receptor agonist treatment.
The study further identified specific subgroups that may benefit more from one therapy over the other. Individuals who were older and female, with lower initial HbA1c, body mass index (BMI), and estimated glomerular filtration rate (eGFR), were more likely to show a greater glycemic advantage with GLP1 receptor agonists. On the other hand, individuals with an initial HbA1c of 5.0 mmol/mol were expected to have a larger glycemic advantage with SGLT2 inhibitors.
The study’s findings indicate that a precision medicine approach to type 2 diabetes can help tailor therapy selection. By utilizing routinely obtained clinical data, this approach could be cost-efficient and implemented in various countries.
It’s important to note that the study’s findings are preliminary and have not yet undergone peer review. Therefore, they should not be regarded as conclusive or used to guide clinical practice or health-related behavior.
In conclusion, the development of a personalized therapy selection algorithm for type 2 diabetes treatment shows promise in improving glycemic outcomes. Further research and validation are necessary to establish the algorithm’s efficacy and to determine its applicability in clinical settings.