Evaluating prediction of short-term tolerability of five type 2 diabetes drug classes using routine clinical features: UK population-based study

TY - JOUR

T1 - Evaluating prediction of short-term tolerability of five type 2 diabetes drug classes using routine clinical features

T2 - UK population-based study

AU - Cardoso, Pedro

AU - Young, Katie G.

AU - Hopkins, Rhian

AU - Mateen, Bilal

AU - Pearson, Ewan

AU - Hattersley, Andrew

AU - McKinley, Trevelyan

AU - Shields, Beverley

AU - Dennis, John M.

AU - the MASTERMIND consortium

N1 - Publisher Copyright: © 2025 The Author(s). Diabetes, Obesity and Metabolism published by John Wiley & Sons Ltd.

PY - 2025/8

Y1 - 2025/8

N2 - Aims: A precision medicine approach in type 2 diabetes (T2D) needs to consider potential treatment risks alongside established benefits for glycaemic and cardiometabolic outcomes. Considering five major T2D drug classes, we aimed to describe variation in short-term discontinuation (a proxy of overall tolerability) by drug and patient routine clinical features and determine whether combining features in a model to predict drug class-specific tolerability has clinical utility. Materials and Methods: UK routine clinical data (Clinical Practice Research Datalink, 2014–2020) of people with T2D initiating glucagon-like peptide-1 receptor agonists (GLP-1RA), dipeptidyl peptidase-4 inhibitors (DPP4i), sodium-glucose co-transporter-2 inhibitors (SGLT2i), thiazolidinediones (TZD) and sulfonylureas (SU) in primary care were studied. We first described the proportions of short-term (3-month) discontinuation by drug class across subgroups stratified by routine clinical features. We then assessed the performance of combining features to predict discontinuation by drug class using a flexible machine learning algorithm (a Bayesian Additive Regression Tree). Results: Amongst 182 194 treatment initiations, discontinuation varied modestly by clinical features. Higher discontinuation on SGLT2i and GLP-1RA was seen for older patients and those with longer diabetes duration. For most other features, discontinuation differences were similar by drug class, with higher discontinuation for patients who had previously discontinued metformin, females and people of South-Asian and Black ethnicities. Lower discontinuation was seen for patients currently taking statins and blood pressure medication. The model combining all sociodemographic and clinical features had a low ability to predict discontinuation (AUC = 0.61). Conclusions: A model-based approach to predict drug-specific discontinuation for individual patients with T2D has low clinical utility. Instead of likely tolerability, prescribing decisions in T2D should focus on drug-specific side-effect risks and differences in the glycaemic and cardiometabolic benefits of available medication classes.

AB - Aims: A precision medicine approach in type 2 diabetes (T2D) needs to consider potential treatment risks alongside established benefits for glycaemic and cardiometabolic outcomes. Considering five major T2D drug classes, we aimed to describe variation in short-term discontinuation (a proxy of overall tolerability) by drug and patient routine clinical features and determine whether combining features in a model to predict drug class-specific tolerability has clinical utility. Materials and Methods: UK routine clinical data (Clinical Practice Research Datalink, 2014–2020) of people with T2D initiating glucagon-like peptide-1 receptor agonists (GLP-1RA), dipeptidyl peptidase-4 inhibitors (DPP4i), sodium-glucose co-transporter-2 inhibitors (SGLT2i), thiazolidinediones (TZD) and sulfonylureas (SU) in primary care were studied. We first described the proportions of short-term (3-month) discontinuation by drug class across subgroups stratified by routine clinical features. We then assessed the performance of combining features to predict discontinuation by drug class using a flexible machine learning algorithm (a Bayesian Additive Regression Tree). Results: Amongst 182 194 treatment initiations, discontinuation varied modestly by clinical features. Higher discontinuation on SGLT2i and GLP-1RA was seen for older patients and those with longer diabetes duration. For most other features, discontinuation differences were similar by drug class, with higher discontinuation for patients who had previously discontinued metformin, females and people of South-Asian and Black ethnicities. Lower discontinuation was seen for patients currently taking statins and blood pressure medication. The model combining all sociodemographic and clinical features had a low ability to predict discontinuation (AUC = 0.61). Conclusions: A model-based approach to predict drug-specific discontinuation for individual patients with T2D has low clinical utility. Instead of likely tolerability, prescribing decisions in T2D should focus on drug-specific side-effect risks and differences in the glycaemic and cardiometabolic benefits of available medication classes.

KW - anti-hyperglycaemic treatment

KW - clinical care

KW - DPP4i

KW - drug tolerability

KW - GLP-1RA

KW - precision medicine

KW - SGLT2i

KW - SU

KW - treatment effect heterogeneity

KW - TZD

UR - https://www.scopus.com/pages/publications/105005847461

U2 - 10.1111/dom.16470

DO - 10.1111/dom.16470

M3 - Article

C2 - 40375735

AN - SCOPUS:105005847461

SN - 1462-8902

VL - 27

SP - 4320

EP - 4329

JO - Diabetes, Obesity and Metabolism

JF - Diabetes, Obesity and Metabolism

IS - 8

ER -