Towards optimal treatment in elderly type 2 diabetes patients.

Type 2 diabetes is a common chronic disease in the elderly and although a variety of drugs is available a substantial number of patients is unable to reach treatment targets putting them at increased risk to develop complications. Treatment response and adverse effects are influenced by a combination of metabolic and genetic factors of which the effects might be different at different ages. In this proposal we intend to complement our ongoing studies to identify the responsible metabolic factors in a longitudinal cohort of type 2 diabetes patients (n=7500+) with a detailed study of genetic factors. The main aim of this proposal is to identify (bio)markers allowing patient stratification for their response to drug treatment, development of adverse effects and diabetic complications during treatment with glucose lowering medication. Metformin is the first choice drug used in the treatment of type 2 diabetes. However, gastro intestinal (GI) problems are a common cause for dose-limitation and discontinuation. Especially among elderly patients prolonged GI problems might cause additional health risks due to malnutrition and dehydration. On the other hand sulphonylurea’s (SU), the second line treatment, are known for causing hypoglycemia which could lead to coma and in the worst case death. Furthermore it is known that failure to reach HbA1c treatment targets is associated with increased risk of development of diabetes related complications and excess mortality. Certain subgroups however, including elderly individuals with multimorbidity, show increased mortality at currently accepted target values suggesting that individualized targets are needed. For this study we will use data from the Diabetes Care System West-Friesland (DCS). The DCS started in 1998 and its main aim is to provide and improve care for all type 2 diabetes patients living in the region West-Friesland, The Netherlands. From their annual visits to the research center we have collected longitudinal data about health, anthropometry, relevant biochemical parameters, patterns of drug use and treatment success. Data about co-morbidities and mortality will be collected from the relevant registries as part of this proposal. From our patient cohort we will select elderly subjects (above 65 years) who fail to reach treatment targets on metformin mono-therapy and need additional or other drugs within the first year after metformin initiation (n=200+). Within this sub-cohort of metformin non-responders we will also have the opportunity to study response to SU’s and other second line drugs. We will use exome sequencing to identify genetic variation in the genomes of the participants associated with drug response. Exome sequencing is a cost efficient and feasible alternative to whole genome sequencing focusing on the 1 to 2% percent of genome which is known to encode genes. We will further enrich our DCS data set with 220k Illumina exome chip genotyping data, containing known exome variations with a minor allele frequency >0.02%, in another 800 subjects at the extremes of metformin response, i.e. those with extreme good or bad response to metformin treatment. Data from the genome of the Netherlands (GoNL) project and the Leiden Longevity Study (LLS) in which respectively 250 trios and 220 nonagenarians are sequenced using a whole genome sequencing approach will be used as a reference dataset of genetic variation in the Dutch and for imputation in cohorts with GWAS. The LUMC is involved, either as co-PI or first author, in both projects. For the replication phase we will use data from several large-scale studies (n=20000+) which are all designed to study diabetes, pharmacogenetics and or healthy aging. Primary analysis of the data will focus on direct single gene effects on the primary endpoints response or non-response to glucose lowering drugs. Other endpoints include age stratified analyses and analyses of continuous variables related to treatment like glycosylated hemoglobin (HbA1c) and glucose levels, response to other drugs and morbidity and mortality scores. Another important aspect of the proposal will be the integrative analysis of the data as generated from this and the metabolic, pharmacoepidemiologic and aging research that is already performed within this and the other cohorts. Within the participating institutes expert knowledge is available to perform such an integrative analysis. We further like to emphasize that this proposal is build upon already established and ongoing collaborative research between the participating universities including experts in the fields of genetics, pharmacology, bioinformatics, epidemiology, aging, diabetes and clinical diabetology. We feel that the combination of pharmacoepidemiologic, metabolic and genetic research has the potential to identify markers associating with drug response at different ages which will lead to personalized and better treatment opportunities.