The DIRECT Work Packages

The main goal of WP1 is to support WPs 2 to 9 by carrying out activities of project management and providing administrative oversight.

The principal participants in WP1 are Sanofi-Aventis (coordinator), Eli Lilly (deputy coordinator) and the University of Dundee (Managing Entity of IMI-JU funding for academic partners).

Studies in WP2 focussed on the phenotype of ‘glycaemic deterioration’ – in other words gradual loss of control of the level of sugar in your blood. In this Work Package we followed individuals at high risk of developing Type 2 Diabetes (i.e. those who displayed signs of pre-diabetes, WP2.1), and those with recent onset of type 2 diabetes (WP2.2). The objectives are to discover biomarkers that identify those groups of individuals who are likely to have rapid deterioration of glycaemia, and to identify surrogate biomarkers of diabetes progression, to enable targeted therapy by patient stratification. We utilised the large cohorts available to the consortium to model incident diabetes and diabetes progression in our populations and to complete the phenotyping of these already well-characterised subjects using standard protocols.

WP2 assimilated data from existing prospective cohorts studies and generated new deep-phenotyping data in studies designed to assess the dynamic process underlying glycaemic deterioration and beta-cell degeneration in individuals at high-risk of developing type 2 diabetes and in those with clinically manifest diabetes. The primary extreme phenotypes that arise from WP2 will be rapid or slow glycaemic deterioration in high risk people with pre-diabetes, and in those with recently diagnosed type 2 diabetes. In addition through integration with WP4 and 5, we will identify predictive and surrogate clinical, physiological and genomic / proteomic / metabolomic biomarkers.

This approach will hopefully yield novel biomarkers that will help target current and future medical therapies focused on preventing and treating type 2 diabetes.

Studies of WP3 focus on therapeutic interventions to identify biomarkers that predict response or non-response to metformin, sulphonylureas, GLP-1R agonists, and early remission from obesity surgery. We utilised the large cohorts available to model drug response in our populations to complete the phenotyping of these already well-characterised subjects using standard protocols.

Some patients with type 2 diabetes do not respond to medication and others show an extreme response, or extreme adverse reaction. If biomarkers can be identified that predict response, these could be used to target therapy by patient stratification. This may allow, for example, earlier intervention with drugs currently reserved for second or third-line treatment in those predicted not to respond to traditional first line (metformin) or second-line (sulphonylurea) therapy.

The major objective of this work package is to develop treatment response cohorts for provision to the genomics work package (WP4), and the integrative biology work package (WP5) with a view to identifying biomarkers that predict extreme treatment response (or side effect) to a number of therapeutic interventions.

This approach should yield novel biomarkers that will help target current and future medical therapies focused on preventing and treating type 2 diabetes.

Studies in WP4 provide genomic and genetic analyses in phenotypically well-characterised samples available to the DIRECT Consortium to identify sets of biomarkers of potential clinical value with respect to disease progression and treatment response. This is done by generating and analysing high-quality, large-scale genetic and genomic data from the biological samples collected from patients in the studies described in both WP2 and WP3 and by functional genomics studies designed to provide mechanistic insights into key processes associated with disease progression and treatment response.

WP4 provides genetic and genomic data in genetic discovery, transcriptional and functional genomics, proteomics, metabolomics and metagenomics. Therefore, WP4 cooperates closely with WP2 and WP3 to define the most efficient set of genetic and genomic analyses within the key study cohorts, and with WP5 to incorporate the data generated into the DIRECT database and make them available for multi-omic analysis in WP5.

The best biomarkers will be selected for assay development and validation in WP6.

The overall aim of WP5 is to perform and develop methods for integrative, functional analysis of human variation data which go beyond what can be achieved by each of the groups involved in the DIRECT Consortium. Furthermore, the data generated in the project will be combined with other proprietary and publicly available data sources.

Additionally, within WP5 we developed -for all consortium partners – a secure data warehouse infrastructure needed to collect data input, perform computational data analysis, including phenotype definition and systems level analyses that is compliant with all ethical, legal and regulatory requirements. The data warehouse structure is compatible with the integrative pathway oriented systems biology methodology.

Data from WPs 2 to 4 will target multiple phenotypes, and the variable amounts of genomic and non-genomic data will provide numerous opportunities for performing integrative analysis of core data sets across multiple levels. The overlapping biology of phenotypes and shared levels of data across the cohorts enables an integrated biology approach.

The remit of WP6 is to assess potential biomarkers that arise from WPs 2 to 5 for applicability to development of high-throughput, high-resolution assay, and develop into assays where appropriate. The analyses will be first checked for quality aspects before going on to validation in DIRECT patient samples. Validation will be done in cohorts of type 2 diabetes patients existing within the consortium and in prospective trial data available to EFPIA partners.

In assay development, we will explore novel technologies allowing multiplexing of biomarkers, their quantification and quality assurance. Biomarker assay development and validation will include immunoassays, assays for genotyping and sequencing analytics of biomarkers, metabolomic and lipidomic assays, and antibody-based proteomics assays.

All new sample collection and reuse of existing samples will be carried out in conformity with all necessary ethical, legal and regulatory requirements in consultation with WP9. Furthermore, work on biomarker validation and any analyses in clinical trials (WP7 and WP8) will be in line with biomarker qualification guidelines and with close involvement of the EMA.

The aim of the clinical trial work packages 7&8 will be primarily to validate biomarkers from discovery work packages 2 to 5 as predictors of events in prospective clinical trials. Additionally it may be possible to establish their clinical utility as predictive of response (surrogate) biomarkers for shortening clinical trials.

The clinical trial design will clearly depend upon any biomarkers discovered in the main phase of the project, but could range from several smaller trials to validate biomarkers of therapeutic response to a large trial of an intervention to delay progression of diabetes or pre-diabetes (WP8).

Ultimately there was insufficient time to set up and complete clinical trials within DIRECT.  The WP7 trials were envisaged to be of short duration and to follow on from the outputs of work package 3 (biomarkers predicting response/non-response to metformin, sulphonylureas, GLP-1 receptor agonists, metformin intolerance and non-remission with obesity surgery). The final decision regarding clinical study design will depend on the putative biomarkers identified where we hope to have  a range of markers (genomic, proteomic, metabolomic, lipidomic, physiological and imaging) that predict variation in response or intolerance to diabetes drugs including metformin, sulphonylureas, or GLP-1 receptor analogues, or predict remission from diabetes following obesity surgery.

Possible WP7 clinical trials are:

The aim of the clinical trial work packages 7&8 will be primarily to validate biomarkers from discovery work packages 2 to 5 as predictors of events in prospective clinical trials. Additionally it may be possible to establish their clinical utility as predictive or response (surrogate) biomarkers for shortening clinical trials.

Clinical trial design will clearly depend upon the biomarkers discovered in part 1 of the project, but could range from a large trial of an intervention to delay progression of diabetes or pre-diabetes to several smaller trials to validate biomarkers of therapeutic response (WP7).

Ultimately there was insufficient time to set up and complete clinical trials within DIRECT. The WP8 trial was envisaged to follow on from outputs of work package 2 (predictive and surrogate biomarkers of glycaemic deterioration) where longer larger trials are needed with hard endpoint. Ultimately the final decision was to focus on performing intensive investigations for possible biomarkers using genomic, proteomic, metabolomic, lipidomic, physiological and imaging techniques that could potentially predict deterioration of glycaemic control in patients with type 2 diabetes.

WP9 was established to identify, address and resolve the Ethical Legal and Regulatory (ELR) factors and  issues that emerge as the scientific work programme unfolded and to ensure the development of best practice.

This ensured that:

The purpose of this approach is to ensure that the consortium activities are carried out in compliance with all ethical, legal and regulatory requirements and that these are in tune with patient expectations and can become a benchmark of best practice.