Clinical predictors of therapeutic response to SGLT-2 inhibitors

Carolina Garcia-Figueras-Mateos; Manuel Cayon-Blanco

doi:10.1530/endoabs.56.P494

P494

Prev Next

Section Contents Cite

Endocrine Abstracts (2018) 56 P494 | DOI: 10.1530/endoabs.56.P494

ECE2018 Poster Presentations: Diabetes, Obesity and Metabolism Diabetes therapy (43 abstracts)

Clinical predictors of therapeutic response to SGLT-2 inhibitors

Carolina García-Figueras-Mateos & Manuel Cayón-Blanco

Err

Author affiliations

Hospital SAS Jerez de la Frontera, Jerez de la Frontera, Spain.

Introduction: Predictors for therapeutical response of SGLT-2 inhibitors (SGLT2i) is scarcely studied. The aim of the study was to retrospectively analyse the clinical parameters that contribute to the therapeutic outcomes of SGLT2i.

Methods/design: We enrolled 75 patients with type 2 diabetes mellitus (T2DM), treated with SGLT2i for longer than three months. The patients were divided into two groups: good responders (if HbA1c-lowering effects were ≥0.5%) and poor responders (if <0.5%), based on pretreatment and post-treatment differences in HbA1c levels. Good responders were also divided into three groups according to decreases of HbA1c: intermediate responders (0.5–0.9%) high responders (1–1.9%), very high responders (≥2%). Patients receiving GLP-1 analogues, with glomerular filtration rate below 60 ml/min/1.73 m² or any contraindication for the use of SGLT2i were excluded. We used univariate and multivariate analyses to assess pretreatment parameters between the two groups and to identify predictors for response.

Results: 78.7% of the patients were good responders. According to decreases of HbA1c, 21.3% were classified as poor responders, 18.7% as intermediate responders, 28% as high responders and 32% as very high responders. Responders had a higher HbA1c at baseline (9.2±1.3% vs 7.5±1%; P<0.001) but no differences according to age, gender, time of evolution of T2DM, body mass index, weight, weight loss, use of other antidiabetic oral agents or insulin were found as compared to poor responders in an univariate analyses. Very high responders had significant (P<0.001) higher HbA1c at baseline (10.4±1.1%) as compared to high (8.5±0.7%) and intermediate responders (8.3±0.6%). Multivariate analyses showed that HbA1c at baseline was the only independent predictor for good response to SGLT2i (OR 5.3; 95%CI: 2.1 – 13.1).

Conclusions: Our study showed that level of HbA1c at baseline, but no other parameter, is linked to the therapeutic effect of SGLT2i on T2DM patients and also, related to the magnitude of the response.