Ajouter des agonistes des récepteurs du GLP-1 quand l'efficience de l'insulinothérapie basale est en train de chuter ou d'échouer

Achievement of satisfactory glycemic control implies to reduce both basal and postprandial glucose concentrations with low risk of hypoglycemia and weight gain. In those persons with type 2 diabetes, who are treated with a once-daily injection of basal insulin and who remain poorly controlled, dual therapy combining a GLP-1 receptor agonist (GLP-1 RA) with basal insulin represents an alternative to the implementation of more complex treatments such as basal-plus or basal-bolus insulin regimens. Consecutively to the seminal study of Buse et al. in 2011, several randomized interventional trials have validated this approach. A particular mention should be given to three of them: the 4B, GetGoal-Duo 2, and FLAT-SUGAR studies, even though the results are yet preliminary for the latter. At present and according to the recommendations published in the latest updated version of the ADA/EASD Position Statement, the addition of GLP-1 RAs to basal insulin appears as an intermediary/transient option and safer strategy than intensified insulin regimens. This is particularly true in those who, after up-titration of basal insulin, have near normal fasting values but are still exhibiting excessive postmeal glucose excursions. Such patients usually have HbA1c levels ranging from 7% to 8%. However we should bear in mind that the HbA1c-lowering effect of GLP-1 RAs remains limited (-0.32% to -1.2% of percentage points) when assessed against a placebo as comparator. Consequently, the following conclusions can be drawn. GLP-1 RAs as add-on therapy to an ongoing basal insulin regimen should more likely be considered the “new strong intermediary link” in obese and vulnerable patients when treatments with basal insulin alone are failing (HbA1c between 7% and 8%) but are still capable to avoid glucose deterioration towards more severe stages (HbA1c >8%). By contrast, in the latter situation, insulin regimens would be a more appropriate option. However, further studies or additional analyses of the data sets of previous controlled trials, using stratification of patients by HbA1c or body weight at baseline, would be welcome /useful for better defining the therapeutic strategies to be used.