Massive-scale genomic risk assessment to inform precision medicine across the spectrum of monogenic and common forms of diabetes

Diabetes represents one of the biggest global health problems of the 21st century. Genome wide association studies (GWAS) have identified hundreds of common genetic variants associated with the most prevalent forms of diabetes, type 1 (T1D) and type 2 (T2D). Additionally, monogenic diabetes accounts for approximately 2-6% of pediatric-onset diabetes cases and is caused by rare variants with large effect. Monogenic diabetes is often undiagnosed, particularly in people of non-European genetic ancestry. Determining whether a given rare variant causes monogenic diabetes remains challenging in people of all ancestries but is particularly problematic for populations of non-European ancestry. An important class of genetic variants increasingly being recognized are rare variants of intermediate penetrance (VIP), which are associated with an intermediate diabetes subtype between common and monogenic forms of diabetes. It is currently unknown whether or how to document VIPs in clinical reports, and as a result, reporting approaches by clinical labs vary widely, creating confusion for providers and patients. This group of genetic variants highlights the continuum of diabetes subtypes that exists and underscores how the current framework for categorizing diabetes into monogenic vs common diabetes is likely over-simplified. A unified framework and updated guidelines for interpretation, classification, and reporting, i.e. for diagnostic and screening purposes, of variants across the spectrum of penetrance are urgently needed. To address these limitations, we will analyze the largest ever dataset of rare variation linked to diabetes phenotyping, comprising > 1.1M participants (33% non-European) to assess variant penetrance (Aim 1). We will integrate this resource with i) high-throughput cellular assays to allow functional validation rare variants at scale and with ii) deep phenotypic characterization of variant carriers in the world largest monogenic diabetes registries (Aim 2). We will work with the Clinical Genome Resource (ClinGen) Monogenic Diabetes Expert Panel (MDEP) and ClinGen Low Penetrance/Risk allele working groups to improve variant interpretation, develop a new variant classification framework, and create a prototype of a more comprehensive report that incorporates genetic variation across the allele frequency spectrum (Aim 3). Accomplishing the aims of this proposal will demonstrate how genomic data from ancestrally diverse populations can improve diagnosis of diabetes subtypes, thereby advancing precision medicine for all. Additionally, the framework we develop for classifying and reporting findings will serve as a paradigm for many other diseases with complex and monogenic forms.