Heterogeneity of Diabetes: β-Cells, Phenotypes, and Precision Medicine: Proceedings of an International Symposium of the Canadian Institutes of Health Research's Institute of Nutrition, Metabolism and Diabetes and the U.S. National Institutes of Health's National Institute of Diabetes and Digestive and Kidney Diseases

Inaccurate diagnosis of diabetes type in youth: prevalence, characteristics, and implications

Classifying diabetes at diagnosis is crucial for disease management but increasingly difficult due to overlaps in characteristics between the commonly encountered diabetes types. We evaluated the prevalence and characteristics of youth with diabetes type that was unknown at diagnosis or was revised over time. We studied 2073 youth with new-onset diabetes (median age [IQR] = 11.4 [6.2] years; 50% male; 75% White, 21% Black, 4% other race; overall, 37% Hispanic) and compared youth with unknown versus known diabetes type, per pediatric endocrinologist diagnosis. In a longitudinal subcohort of patients with data for ≥ 3 years post-diabetes diagnosis (n = 1019), we compared youth with steady versus reclassified diabetes type. In the entire cohort, after adjustment for confounders, diabetes type was unknown in 62 youth (3%), associated with older age, negative IA-2 autoantibody, lower C-peptide, and no diabetic ketoacidosis (all, p < 0.05). In the longitudinal subcohort, diabetes type was reclassified in 35 youth (3.4%); this was not statistically associated with any single characteristic. In sum, among racially/ethnically diverse youth with diabetes, 6.4% had inaccurate diabetes classification at diagnosis. Further research is warranted to improve accurate diagnosis of pediatric diabetes type.

Scaling Insulin-Producing Cells by Multiple Strategies

In the quest to combat insulin-dependent diabetes mellitus (IDDM), allogenic pancreatic islet cell therapy sourced from deceased donors represents a significant therapeutic advance. However, the applicability of this approach is hampered by donor scarcity and the demand for sustained immunosuppression. Human induced pluripotent stem cells are a game-changing resource for generating synthetic functional insulin-producing β cells. In addition, novel methodologies allow the direct expansion of pancreatic progenitors and mature β cells, thereby circumventing prolonged differentiation. Nevertheless, achieving practical reproducibility and scalability presents a substantial challenge for this technology. As these innovative approaches become more prominent, it is crucial to thoroughly evaluate existing expansion techniques with an emphasis on their optimization and scalability. This manuscript delineates these cutting-edge advancements, offers a critical analysis of the prevailing strategies, and underscores pivotal challenges, including cost-efficiency and logistical issues. Our insights provide a roadmap, elucidating both the promises and the imperatives in harnessing the potential of these cellular therapies for IDDM.

Machine learning in precision diabetes care and cardiovascular risk prediction

Artificial intelligence and machine learning are driving a paradigm shift in medicine, promising data-driven, personalized solutions for managing diabetes and the excess cardiovascular risk it poses. In this comprehensive review of machine learning applications in the care of patients with diabetes at increased cardiovascular risk, we offer a broad overview of various data-driven methods and how they may be leveraged in developing predictive models for personalized care. We review existing as well as expected artificial intelligence solutions in the context of diagnosis, prognostication, phenotyping, and treatment of diabetes and its cardiovascular complications. In addition to discussing the key properties of such models that enable their successful application in complex risk prediction, we define challenges that arise from their misuse and the role of methodological standards in overcoming these limitations. We also identify key issues in equity and bias mitigation in healthcare and discuss how the current regulatory framework should ensure the efficacy and safety of medical artificial intelligence products in transforming cardiovascular care and outcomes in diabetes.

Imprecise Diagnosis of Diabetes Type in Youth: Prevalence, Characteristics, and Implications

Classifying diabetes at diagnosis is crucial for disease management but increasingly difficult due to overlaps in characteristics between the commonly encountered diabetes types. We evaluated the prevalence and characteristics of youth with diabetes type that was unknown at diagnosis or was revised over time. We studied 2073 youth with new-onset diabetes (median age [IQR]=11.4 [6.2] years; 50% male; 75% White, 21% Black, 4% other race; overall, 37% Hispanic) and compared youth with unknown versus known diabetes type, per pediatric endocrinologist diagnosis. In a longitudinal subcohort of patients with data for ≥3 years post-diabetes diagnosis (n=1019), we compared youth with unchanged versus changed diabetes classification. In the entire cohort, after adjustment for confounders, diabetes type was unknown in 62 youth (3%), associated with older age, negative IA-2 autoantibody, lower C-peptide, and no diabetic ketoacidosis (all, p<0.05). In the longitudinal subcohort, diabetes classification changed in 35 youth (3.4%); this was not statistically associated with any single characteristic. Having unknown or revised diabetes type was associated with less continuous glucose monitor use on follow-up (both, p<0.004). In sum, among racially/ethnically diverse youth with diabetes, 6.5% had imprecise diabetes classification at diagnosis. Further research is warranted to improve accurate diagnosis of pediatric diabetes type.

Precision Medicine in Type 1 Diabetes

Type 1 diabetes is a complex, chronic disease in which the insulin-producing beta cells in the pancreas are sufficiently altered or impaired to result in requirement of exogenous insulin for survival. The development of type 1 diabetes is thought to be an autoimmune process, in which an environmental (unknown) trigger initiates a T cell-mediated immune response in genetically susceptible individuals. The presence of islet autoantibodies in the blood are signs of type 1 diabetes development, and risk of progressing to clinical type 1 diabetes is correlated with the presence of multiple islet autoantibodies. Currently, a "staging" model of type 1 diabetes proposes discrete components consisting of normal blood glucose but at least two islet autoantibodies (Stage 1), abnormal blood glucose with at least two islet autoantibodies (Stage 2), and clinical diagnosis (Stage 3). While these stages may, in fact, not be discrete and vary by individual, the format suggests important applications of precision medicine to diagnosis, prevention, prognosis, treatment and monitoring. In this paper, applications of precision medicine in type 1 diabetes are discussed, with both opportunities and barriers to global implementation highlighted. Several groups have implemented components of precision medicine, yet the integration of the necessary steps to achieve both short- and long-term solutions will need to involve researchers, patients, families, and healthcare providers to fully impact and reduce the burden of type 1 diabetes.

Dysregulated UPR and ER Stress Related to a Mutation in the Sdf2l1 Gene Are Involved in the Pathophysiology of Diet-Induced Diabetes in the Cohen Diabetic Rat

The Cohen Diabetic rat is a model of type 2 diabetes mellitus that consists of the susceptible (CDs/y) and resistant (CDr/y) strains. Diabetes develops in CDs/y provided diabetogenic diet (DD) but not when fed regular diet (RD) nor in CDr/y given either diet. We recently identified in CDs/y a deletion in Sdf2l1, a gene that has been attributed a role in the unfolded protein response (UPR) and in the prevention of endoplasmic reticulum (ER) stress. We hypothesized that this deletion prevents expression of SDF2L1 and contributes to the pathophysiology of diabetes in CDs/y by impairing UPR, enhancing ER stress, and preventing CDs/y from secreting sufficient insulin upon demand. We studied SDF2L1 expression in CDs/y and CDr/y. We evaluated UPR by examining expression of key proteins involved in both strains fed either RD or DD. We assessed the ability of all groups of animals to secrete insulin during an oral glucose tolerance test (OGTT) over 4 weeks, and after overnight feeding (postprandial) over 4 months. We found that SDF2L1 was expressed in CDr/y but not in CDs/y. The pattern of expression of proteins involved in UPR, namely the PERK (EIF2α, ATF4 and CHOP) and IRE1 (XBP-1) pathways, was different in CDs/y DD from all other groups, with consistently lower levels of expression at 4 weeks after initiation of DD and coinciding with the development of diabetes. In CDs/y RD, insulin secretion was mildly impaired, whereas in CDs/y DD, the ability to secrete insulin decreased over time, leading to the development of the diabetic phenotype. We conclude that in CDs/y DD, UPR participating proteins were dysregulated and under-expressed at the time point when the diabetic phenotype became overt. In parallel, insulin secretion in CDs/y DD became markedly impaired. Our findings suggest that under conditions of metabolic load with DD and increased demand for insulin secretion, the lack of SDF2L1 expression in CDs/y is associated with UPR dysregulation and ER stress which, combined with oxidative stress previously attributed to the concurrent Ndufa4 mutation, are highly likely to contribute to the pathophysiology of diabetes in this model.

Distribution and characteristics of newly-defined subgroups of type 2 diabetes in randomised clinical trials: Post hoc cluster assignment analysis of over 12,000 study participants

AIMS

Newly-defined subgroups of type 2 diabetes mellitus (T2DM) have been reported from real-world cohorts but not in detail from randomised clinical trials (RCTs).

METHODS

T2DM participants, uncontrolled on different pre-study therapies (n = 12.738; 82 % Caucasian; 44 % with diabetes duration > 10 years) from 14 RCTs, were assigned to new subgroups according to age at onset of diabetes, HbA1c, BMI, and fasting C-peptide using the nearest centroid approach. Subgroup distribution, characteristics and influencing factors were analysed.

RESULTS

In both, pooled and single RCTs, "mild-obesity related diabetes" predominated (45 %) with mean BMI of 35 kg/m². "Severe insulin-resistant diabetes" was found least often (4.6 %) and prevalence of "mild age-related diabetes" (23.9 %) was mainly influenced by age at onset of diabetes and age cut-offs. Subgroup characteristics were widely comparable to those from real-world cohorts, but all subgroups showed higher frequencies of diabetes-related complications which were associated with longer diabetes duration. A high proportion of "severe insulin-deficient diabetes" (25.4 %) was identified with poor pre-study glycaemic control.

CONCLUSIONS

Classification of RCT participants into newly-defined diabetes subgroups revealed the existence of a heterogeneous population of T2DM. For future RCTs, subgroup-based randomisation of T2DM will better define the target population and relevance of the outcomes by avoiding clinical heterogeneity.

Mathematical model of insulin kinetics accounting for the amino acids effect during a mixed meal tolerance test

Amino acids (AAs) are well known to be involved in the regulation of glucose metabolism and, in particular, of insulin secretion. However, the effects of different AAs on insulin release and kinetics have not been completely elucidated. The aim of this study was to propose a mathematical model that includes the effect of AAs on insulin kinetics during a mixed meal tolerance test. To this aim, five different models were proposed and compared. Validation was performed using average data, derived from the scientific literature, regarding subjects with normal glucose tolerance (CNT) and with type 2 diabetes (T2D). From the average data of the CNT and T2D people, data for two virtual populations (100 for each group) were generated for further model validation. Among the five proposed models, a simple model including one first-order differential equation showed the best results in terms of model performance (best compromise between model structure parsimony, estimated parameters plausibility, and data fit accuracy). With regard to the contribution of AAs to insulin appearance/disappearance (k_AA model parameter), model analysis of the average data from the literature yielded 0.0247 (confidence interval, CI: 0.0168 - 0.0325) and -0.0048 (CI: -0.0281 - 0.0185) μU·ml^-1/(μmol·l^-1·min), for CNT and T2D, respectively. This suggests a positive effect of AAs on insulin secretion in CNT, and negligible effect in T2D. In conclusion, a simple model, including single first-order differential equation, may help to describe the possible AAs effects on insulin kinetics during a physiological metabolic test, and provide parameters that can be assessed in the single individuals.