Background: Diabetes mellitus is characterized by elevated blood sugar due to absolute or relative insulin deficiency. Diabetes is classified as type 1 (T1D) or type 2 diabetes (T2D), gestational diabetes, and other types, such as monogenic diabetes, exocrine pancreatic disorders, and medication-induced diabetes. Objectives: This review article provides an overview of diabetes genetics, covering polygenic, monogenic, and syndromic forms of the disorder with emphasis on aspects to help clinicians in diagnosis, management, and counseling, but also to foster valuable knowledge for diabetic researchers in identifying phenotypes that will help inform gene discovery. Key Findings: Most cases of T1D and T2D are polygenic with environmental triggers. T1D results from autoimmune destruction of pancreatic beta cells leading to absolute insulin deficiency. Genetic studies of T1D have focused on the identification of loci associated with increased susceptibility to T1D. Early studies showed a linkage between T1D and several human leukocyte antigen (HLA) susceptibility loci on chromosome 6. Genome-wide association studies (GWAS) have identified more than 100 HLA- and non-HLA loci that increase susceptibility to T1D. It has been well established that a substantial portion of the genetic risk for T1D is encoded in the HLA locus. The non-HLA loci INS, CTLA4, IL2RA, IFIH1, and PTPN22 make moderate contributions to T1D risk. Many other non-HLA loci have small effects to the phenotype and are relevant to autoimmunity, but they are yet to be identified. T2D, on the other hand, is associated with obesity and insulin resistance with relative insulin deficiency. Thousands of gene variants that are common and contribute small effects have also been identified through GWAS to contribute to T2D risk, but the rarer variants may confer significant risk to an individual’s risk. Common variants in the TCF7L2 locus consistently carry one of the largest risks associated with T2D with a reported 1.7-fold disease odds for homozygous carriers. The usefulness of individual variants for genetic counseling in the common forms of diabetes has been limited in clinical settings in the past. The development of polygenic risk scores (PRS) and partitioned polygenic risk scores (PPRS), statistics derived from GWAS, are being used to predict and classify diabetes. The performance of PRS and PPRS varies by ancestry and type of diabetes. The PRS performs better with T1D, with an area under the curve and receiver operating characteristics (AUC-ROC) ranging from 0.87 to 0.93, compared to 0.72–0.75 for T2D. The genetic architecture of T2D is markedly more polygenic than T1D, and the PPRS has been useful in assessing risk in that setting. Monogenic diabetes comprises several dysglycemic disorders that include neonatal diabetes, maturity-onset diabetes of the young (MODY), and other genetic syndromes that have diabetes either as an associated finding and/or as a complication. Some of the monogenic diabetes gene variants have incomplete penetrance and variable expressivity leading to different ages of onset and variable presentation even within the same family. Hence some patients with these conditions have been previously diagnosed as having T1D or T2D. Many monogenic disorders follow Mendelian inheritance patterns, so genetic counseling is relatively straightforward if pathogenic variants are found to be inherited from a parent. Counseling for forms of diabetes due to maternally inherited mitochondrial cytopathies, such as MELAS and Kearns–Sayres syndrome, is not straightforward due to the occurrence of two or more populations of genetically distinct mitochondrial DNAs in the cells (heteroplasmy); the higher the percent of pathogenic variants in a cell or tissue, the greater the chance for affectation of disorder. Implications: Early stages of diabetes may be asymptomatic, and improvement in methodologies to identify individuals at high risk is important so prevention strategies can be targeted to susceptible individuals to slow or obviate the onset of disease and to minimize complications. Conclusions: Diabetes is a heterogeneous disorder, and accurate definition of phenotypes in the setting of non-syndromic and syndromic forms, development of powerful statistical methodologies, use of next-generation sequencing applications to interrogate the genome, incorporation of epigenetic mechanisms in statistical modeling and accurate curation of gene variants, will help us to realize application of genomic medicine and to inform diabetes care.
Asamoah et al. (Tue,) studied this question.