Machine Learning to Identify Predictors of Glycemic Control in Type 2 Diabetes: An Analysis of Target HbA1c Reduction Using Empagliflozin/Linagliptin Data

Precision medicine in type 2 diabetes: A systematic review of treatment effect heterogeneity for GLP1-receptor agonists and SGLT2-inhibitors

Background

A precision medicine approach in type 2 diabetes requires identification of clinical and biological features that are reproducibly associated with differences in clinical outcomes with specific anti-hyperglycaemic therapies. Robust evidence of such treatment effect heterogeneity could support more individualized clinical decisions on optimal type 2 diabetes therapy.

Methods

We performed a pre-registered systematic review of meta-analysis studies, randomized control trials, and observational studies evaluating clinical and biological features associated with heterogenous treatment effects for SGLT2-inhibitor and GLP1-receptor agonist therapies, considering glycaemic, cardiovascular, and renal outcomes.

Results

After screening 5,686 studies, we included 101 studies of SGLT2-inhibitors and 75 studies of GLP1-receptor agonists in the final systematic review. The majority of papers had methodological limitations precluding robust assessment of treatment effect heterogeneity. For glycaemic outcomes, most cohorts were observational, with multiple analyses identifying lower renal function as a predictor of lesser glycaemic response with SGLT2-inhibitors and markers of reduced insulin secretion as predictors of lesser response with GLP1-receptor agonists. For cardiovascular and renal outcomes, the majority of included studies were post-hoc analyses of randomized control trials (including meta-analysis studies) which identified limited clinically relevant treatment effect heterogeneity.

Conclusions

Current evidence on treatment effect heterogeneity for SGLT2-inhibitor and GLP1-receptor agonist therapies is limited, likely reflecting the methodological limitations of published studies. Robust and appropriately powered studies are required to understand type 2 diabetes treatment effect heterogeneity and evaluate the potential for precision medicine to inform future clinical care.

Plain language summary

This review identifies research that helps understand which clinical and biological factors that are associated with different outcomes for specific type 2 diabetes treatments. This information could help clinical providers and patients make better informed personalized decisions about type 2 diabetes treatments. We focused on two common type 2 diabetes treatments: SGLT2-inhibitors and GLP1-receptor agonists, and three outcomes: blood glucose control, heart disease, and kidney disease. We identified some potential factors that are likely to lessen blood glucose control including lower kidney function for SGLT2-inhibitors and lower insulin secretion for GLP1-receptor agonists. We did not identify clear factors that alter heart and renal disease outcomes for either treatment. Most of the studies had limitations, meaning more research is needed to fully understand the factors that influence treatment outcomes in type 2 diabetes.

Application of Machine Learning Algorithms to Predict Uncontrolled Diabetes Using the All of Us Research Program Data

There is a paucity of predictive models for uncontrolled diabetes mellitus. The present study applied different machine learning algorithms on multiple patient characteristics to predict uncontrolled diabetes. Patients with diabetes above the age of 18 from the All of Us Research Program were included. Random forest, extreme gradient boost, logistic regression, and weighted ensemble model algorithms were employed. Patients who had a record of uncontrolled diabetes based on the international classification of diseases code were identified as cases. A set of features including basic demographic, biomarkers and hematological indices were included in the model. The random forest model demonstrated high performance in predicting uncontrolled diabetes, yielding an accuracy of 0.80 (95% CI: 0.79-0.81) as compared to the extreme gradient boost 0.74 (95% CI: 0.73-0.75), the logistic regression 0.64 (95% CI: 0.63-0.65) and the weighted ensemble model 0.77 (95% CI: 0.76-0.79). The maximum area under the receiver characteristics curve value was 0.77 (random forest model), while the minimum value was 0.7 (logistic regression model). Potassium levels, body weight, aspartate aminotransferase, height, and heart rate were important predictors of uncontrolled diabetes. The random forest model demonstrated a high performance in predicting uncontrolled diabetes. Serum electrolytes and physical measurements were important features in predicting uncontrolled diabetes. Machine learning techniques may be used to predict uncontrolled diabetes by incorporating these clinical characteristics.

Review: Machine learning in precision pharmacotherapy of type 2 diabetes-A promising future or a glimpse of hope?

Precision pharmacotherapy of diabetes requires judicious selection of the optimal therapeutic agent for individual patients. Artificial intelligence (AI), a swiftly expanding discipline, holds substantial potential to transform current practices in diabetes diagnosis and management. This manuscript provides a comprehensive review of contemporary research investigating drug responses in patient subgroups, stratified via either supervised or unsupervised machine learning approaches. The prevalent algorithmic workflow for investigating drug responses using machine learning involves cohort selection, data processing, predictor selection, development and validation of machine learning methods, subgroup allocation, and subsequent analysis of drug response. Despite the promising feature, current research does not yet provide sufficient evidence to implement machine learning algorithms into routine clinical practice, due to a lack of simplicity, validation, or demonstrated efficacy. Nevertheless, we anticipate that the evolving evidence base will increasingly substantiate the role of machine learning in molding precision pharmacotherapy for diabetes.

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

Diabetes as a metabolic illness can be characterized by increased amounts of blood glucose. This abnormal increase can lead to critical detriment to the other organs such as the kidneys, eyes, heart, nerves, and blood vessels. Therefore, its prediction, prognosis, and management are essential to prevent harmful effects and also recommend more useful treatments. For these goals, machine learning algorithms have found considerable attention and have been developed successfully. This review surveys the recently proposed machine learning (ML) and deep learning (DL) models for the objectives mentioned earlier. The reported results disclose that the ML and DL algorithms are promising approaches for controlling blood glucose and diabetes. However, they should be improved and employed in large datasets to affirm their applicability.

Rationale and design of the Brazilian diabetes study: a prospective cohort of type 2 diabetes

BACKGROUND

Optimal control of traditional risk factors only partially attenuates the exceeding cardiovascular mortality of individuals with diabetes. Employment of machine learning (ML) techniques aimed at the identification of novel features of risk prediction is a compelling target to tackle residual cardiovascular risk. The objective of this study is to identify clinical phenotypes of T2D which are more prone to developing cardiovascular disease.

METHODS

The Brazilian Diabetes Study is a single-center, ongoing, prospective registry of T2D individuals. Eligible patients are 30 years old or older, with a confirmed T2D diagnosis. After an initial visit for the signature of the informed consent form and medical history registration, all volunteers undergo biochemical analysis, echocardiography, carotid ultrasound, ophthalmologist visit, dual x-ray absorptiometry, coronary artery calcium score, polyneuropathy assessment, advanced glycation end-products reader, and ambulatory blood pressure monitoring. A 5-year follow-up will be conducted by yearly phone interviews for endpoints disclosure. The primary endpoint is the difference between ML-based clinical phenotypes in the incidence of a composite of death, myocardial infarction, revascularization, and stroke. Since June/2016, 1030 patients (mean age: 57 years, diabetes duration of 9.7 years, 58% male) were enrolled in our study. The mean follow-up time was 3.7 years in October/2021.

CONCLUSION

The BDS will be the first large population-based cohort dedicated to the identification of clinical phenotypes of T2D at higher risk of cardiovascular events. Data derived from this study will provide valuable information on risk estimation and prevention of cardiovascular and other diabetes-related events.

CLINICALTRIALS.GOV IDENTIFIER

NCT04949152.

Potential applications and performance of machine learning techniques and algorithms in clinical practice: A systematic review

PURPOSE

The advent of clinically adapted machine learning algorithms can solve numerous problems ranging from disease diagnosis and prognosis to therapy recommendations. This systematic review examines the performance of machine learning (ML) algorithms and evaluates the progress made to date towards their implementation in clinical practice.

METHODS

Systematic searching of databases (PubMed, MEDLINE, Scopus, Google Scholar, Cochrane Library and WHO Covid-19 database) to identify original articles published between January 2011 and October 2021. Studies reporting ML techniques in clinical practice involving humans and ML algorithms with a performance metric were considered.

RESULTS

Of 873 unique articles identified, 36 studies were eligible for inclusion. The XGBoost (extreme gradient boosting) algorithm showed the highest potential for clinical applications (n = 7 studies); this was followed jointly by random forest algorithm, logistic regression, and the support vector machine, respectively (n = 5 studies). Prediction of outcomes (n = 33), in particular Inflammatory diseases (n = 7) received the most attention followed by cancer and neuropsychiatric disorders (n = 5 for each) and Covid-19 (n = 4). Thirty-three out of the thirty-six included studies passed more than 50% of the selected quality assessment criteria in the TRIPOD checklist. In contrast, none of the studies could achieve an ideal overall bias rating of 'low' based on the PROBAST checklist. In contrast, only three studies showed evidence of the deployment of ML algorithm(s) in clinical practice.

CONCLUSIONS

ML is potentially a reliable tool for clinical decision support. Although advocated widely in clinical practice, work is still in progress to validate clinically adapted ML algorithms. Improving quality standards, transparency, and interpretability of ML models will further lower the barriers to acceptability.