The New Classification Targets for Diabetes Treatment

Data suggest that personalizing therapy and classification targets may ease the burden of diabetic complications and prevent diabetes in the future generation.

Diabetes continues to be a significant global burden due to increased prevalence and complications. Early Identification of diverse subtypes of disease could lead to more tailored medicine. Targeting the right resources to the suitable patients would be a critical step toward precision medicine, increasing patient health, and lowering societal health expenditures. Individualized diabetes therapy is beneficial to prevent diabetes in the future generation and ease the burden of diabetes complications. Before this study, diabetes treatment is based on the traditional classification of diabetes, which includes Type 1 diabetes (T1D), Type 2 diabetes (T2D), latent autoimmune diabetes in adults (LADA), and maturity-onset diabetes in young (MODY). 

This study used six clinical variables to cluster adult-onset diabetes patients into five subgroups. These variables include glutamate decarboxylase autoantibodies, age at onset of diabetes, glycated hemoglobin, BMI, and simple measures of insulin resistance and insulin secretion. These subgroups are related to varied risks of diabetic complications and treatment responses in various populations worldwide. Following the elimination of these groupings, type 1 diabetes is estimated to affect 10–15 percent of the population, with T2D accounting for 75–80 percent. Because autoantibodies are not always measured and a genetic workup is not always possible, people with various undiagnosed diabetes frequently join the T2D group. The All-New Diabetics In Scania (ANDIS) study began in 2008, intending to determine the heterogeneity of T2D. The ANDIS initiative seeks to cover all newly diagnosed diabetes patients in southern Sweden’s Scania (Skne) region within one year after diagnosis. It includes more than 20,000 people, accounting for more than 90% of newly diagnosed patients. Two additional assays are added to standard measurements during registration: glutamate decarboxylase autoantibodies (GADA) and C-peptide. 

The study model believes that various interconnected pathways cause diabetes, but it also considers that each patient has a distinct color preference. As a result, categorizing patients into shades of red, green, and blue is more instructive than just categorizing them as shades of brown, even if some patients have intermediate colors. At registration, two measurements, that is of glutamate decarboxylase autoantibodies (GADA) and C-peptide are added to standard measurements BMI, glycated hemoglobin (HbA1c), age at diabetes onset, and measures of beta-cell function (HOMA2-B) and insulin resistance (HOMA2-IR) estimated from fasting glucose and C-peptide are used to divide adult individuals into subtypes using a data-driven approach using the most relevant, readily available clinical variables for individuals with diabetes. The five clusters examined found different clinical features, disease development, and prognosis. In addition, the subgroups indicated considerable disparities in the development of diabetic complications and differences in clinical features.

Treatment with GLP-1RA or SGLT2i is indicated for patients with established cardiovascular disease or CKD. Still, for others, the focus is on decreasing glucose by gradually adding medications other than metformin, guided by the severity of hyperglycemia, obesity, or cost constraints. In addition, because of the clinical distinctions across the subtypes, it may be helpful to choose the medication that addresses the primary underlying metabolic abnormality, such as insulin insufficiency in SIDD and insulin resistance in SIRD.

Diabetes remains a leading cause of death and a primary global health concern. According to this study, individualized diabetes therapy can be beneficial to prevent diabetes in the future generation and ease the burden of diabetes complications. In addition, this study’s innovative subclassification of T2D, which divides the disease into clinically significant subgroups, lays the groundwork for increased tailored diabetic therapy beyond what is now available for monogenic and, to a lesser extent, type 1 diabetes. Clustering also opens up new research possibilities, allowing for more precise classifications of pathogenic abnormalities. Additional cluster factors, such as biomarkers, genotypes, or genetic risk scores, could be used to refine the ANDIS classification in the future. Artificial intelligence-based technologies offer a possible path to subtypes with even higher resolution and predictive power. 

Practice Pearls:

• Personalizing diabetes treatment and targets is the key to avoiding diabetes in future generations and reducing the burden of diabetes complications.
• Type 2 diabetes patients can be divided into five groups based on clinical features and disease progression using easily detectable clinical indicators.
• This study clustering opens up new research possibilities, allowing for more precise classifications of pathogenic abnormalities.

Reference for “The New Classification Targets for Diabetes Treatment”:
Ahlqvist E, Prasad RB, Groop L. 100 years of insulin: Towards improved precision and a new classification of diabetes mellitus. Journal of Endocrinology. 2022;252(3). doi:10.1530/joe-20-0596

Betsy Adegborioye, PharmD Candidate 2022, South College School of Pharmacy