The DEXLIFE study methods: Identifying novel candidate biomarkers that predict progression to type 2 diabetes in high risk individuals

https://doi.org/10.1016/j.diabres.2014.07.025Get rights and content

Abstract

The incidence of type 2 diabetes (T2D) is rapidly increasing worldwide and T2D is likely to affect 592 million people in 2035 if the current rate of progression is continued. Today, patients are diagnosed with T2D based on elevated blood glucose, either directly or indirectly (HbA1c). However, the information on disease progression is limited.

Therefore, there is a need to identify novel early markers of glucose intolerance that reflect the underlying biology and the overall physiological, metabolic and clinical characteristics of progression towards diabetes.

In the DEXLIFE study, several clinical cohorts provide the basis for a series of clinical, physiological and mechanistic investigations in combination with a range of – omic technologies to construct a detailed metabolic profile of high-risk individuals across multiple cohorts.

In addition, an exercise and dietary intervention study is conducted, that will assess the impact on both plasma biomarkers and specific functional tissue-based markers.

The DEXLIFE study will provide novel diagnostic and predictive biomarkers which may not only effectively detect the progression towards diabetes in high risk individuals but also predict responsiveness to lifestyle interventions known to be effective in the prevention of diabetes.

Introduction

The prevalence of diabetes has reached epidemic proportions and it is now recognized as the fastest growing chronic disease. The global prevalence of diabetes increased 730% between 1985 and 2010 and it is predicted that 592 million people will have diabetes by 2035 [1]. If this exponential increase is not slowed or reversed the health and economic consequences, now approaching 10% of all health costs, will not be sustainable. It is estimated that approximately 50% of patients are undiagnosed and may have diabetes for 5–10 years before clinical identification [1]. It is not surprising that many of these patients have diabetes-related complications at diagnosis with significantly greater treatment costs and risk of mortality. Therefore, the long-term solution for the diabetes epidemic is to prevent disease onset by detecting the risk of progression at an earlier stage and implementing effective prevention programmes.

There are a number of reasons why it has been difficult to identify high-risk individuals and effectively track the progression of glucose intolerance. A major limitation is that blood glucose, directly measured or averaged (HbA1c), is the only accepted disease marker for diabetes. However, blood glucose levels may not reflect the impaired β-cell function or insulin resistance [2] and the optimal value for HbA1c to accurately diagnose the pre-diabetic state is still debated [3], [4], [5]. While the 2-h oral glucose tolerance test may better reflect the underlying pathophysiology of diabetes [2], this is a time consuming procedure and does not favour routine use in general practice. Furthermore, impaired fasting glucose (IFG) and impaired glucose tolerance (IGT) represent two distinct pathological states with distinct patterns of progression [6], and there is a high degree of inter-individual variation in the progression towards diabetes. Finally, it is not known whether the deterioration in glucose tolerance and β-cell function is linear or whether there is an accelerated loss of function at some point prior to the onset of diabetes.

Given these challenges there is a need to identify novel markers of glucose intolerance that reflect the physiological, metabolic and clinical characteristics of progression towards diabetes. These markers should ideally be sensitive to subtle physiological changes and track with deteriorating glucose tolerance. The identification of novel biomarkers of this kind would facilitate a completely new approach to disease prevention. In addition to the accurate prognostic phenotyping of high-risk individuals the biomarkers could be used as an intermittent monitoring tool to prevent progression towards diabetes. This tool would benefit healthcare providers, general practitioners and health insurance companies and could be used by the patients themselves to assist behaviour change and adherence. Therefore, the best biomarkers will be those that predict diabetes progression but are responsive to lifestyle changes that improve glucose tolerance.

The purpose of this paper is to describe the experimental design and methodology of DEXLIFE (www.dexlife.eu), an EU FP7 funded project with the objective to identify novel metabolomic and lipidomic biomarkers that predict progression towards type 2 diabetes and are responsive to lifestyle intervention in people at increased risk. This will be achieved by (i) identifying novel biomarkers in longitudinal cohorts where individuals progress from normal glucose tolerance to diabetes; (ii) investigating the mechanistic changes in skeletal muscle in different diabetes populations; (iii) determining the biomarkers that track with improved glucose tolerance following a 12-week lifestyle intervention. The outcomes of this project will lead to the development of new diagnostic and prognostic tools for widespread use in risk assessment and prediction.

Section snippets

Overall strategy

The central theme of the project is to track progression from a state of normal metabolism with normal glucose tolerance through to pre-diabetes and on to frank type 2 diabetes. We have selected a range of carefully phenotyped cohorts that allow us to map natural background progression to diabetes. The METSIM cohort [7] of Finnish men that have been followed over a 5 year period and the Irish Vhi cohort that has tracked health insurance clients for 3 years are being used to identify novel

Ethical considerations

This research project requires ethical approval from a number of different institutions and countries for the use of human data (i) that has been obtained from existing studies (METSIM, RISC, YT2, DMVhi) and (ii) from the intervention trial. Research ethics approval for the existing cohort studies was obtained prior to DEXLIFE and any modifications required for use of samples or data in this research project was subsequently obtained from the local ethics committee. The prospective lifestyle

The DEXLIFE cohorts and the intervention

Table 1 provides a detailed overview of measurements and sample size in each of the included cohorts and the intervention study. Fig. 1 outlines the flow of serum and muscle samples from intervention and cohorts to biomarker analysis.

Metabolomics

Non-targeted metabolic profiling is conducted using three independent platforms: ultrahigh performance liquid chromatography/tandem mass spectrometry (LC/MS) optimized for basic species, LC/MS optimized for acidic species, and gas chromatography/mass spectrometry (GC/MS) as described previously [11], [12]. Metabolites are identified by automated matching to chemical reference library standards on the basis of retention time, molecular weight (m/z), preferred adducts, and in-source fragments as

Transcriptome analysis

RNA isolation is performed using a mix of Trizol and Maxwell® (Promega). The RNA recovery is around 5 mg per sample. Microarray services are provided by the IRB Functional Genomics Core Facility, including quality control tests of total RNA by Agilent Bioanalyzer and Nanodrop spectrophotometry. cDNA preparation, hybridization, and subsequent chip analysis were all performed according to the Affymetrix protocol [16].

Epigenetics

DNA is isolated applying the QIAamp® DNA Mini Kit (Qiagen Iberia, Spain). To

Data integration

In DEXLIFE, a network-based approach for causal modelling will be employed. To distinguish direct and indirect interactions of intermediate phenotypes, we will utilize the QPGRAPH method which has been previously applied, e.g., in the domains of obesity [17] and psychotic disorders [18]. We will use partial correlations as a measure of direct interaction and construct a Gaussian graphical model where the intermediate phenotypes are connected if their partial correlation is significantly

Outcomes and impact

The application of each of these -omics approaches has contributed to our understanding of type 2 diabetes. However, standing alone, each -omics approach captures only one layer of the complexity inherent within human biology. In order to develop biomarkers that are responsive to lifestyle intervention, one has to be able to collect all available information from predisposition to the disease onset. The goal of assessing the effect of intermediate phenotypes and associated biomarkers for type 2

References (18)

  • X. Zhuo et al.

    Alternative HbA1c cutoffs to identify high-risk adults for diabetes prevention: a cost-effectiveness perspective

    Am J Prev Med

    (2012)
  • International Diabetes Federation

    IDF diabetes atlas

    (2013)
  • E. Bonora et al.

    The pros and cons of diagnosing diabetes with A1C

    Diabetes Care

    (2011)
  • American Diabetes Association

    Diagnosis and classification of diabetes mellitus

    Diabetes Care

    (2010)
  • T.I.E. Committee

    International Expert Committee Report on the role of the A1C assay in the diagnosis of diabetes

    Diabetes Care

    (2009)
  • K. Færch et al.

    Trajectories of cardiometabolic risk factors before diagnosis of three subtypes of type 2 diabetes: a post-hoc analysis of the longitudinal Whitehall II cohort study

    Lancet Diabetes Endocrinol

    (2013)
  • A. Stancáková et al.

    Changes in insulin sensitivity and insulin release in relation to glycemia and glucose tolerance in 6,414 Finnish men

    Diabetes

    (2009)
  • S.A. Hills et al.

    The EGIR-RISC STUDY (The European group for the study of insulin resistance: relationship between insulin sensitivity and cardiovascular disease risk): I. Methodology and objectives

    Diabetologia

    (2004)
  • J. Lindström et al.

    The Finnish Diabetes Prevention Study (DPS): lifestyle intervention and 3-year results on diet and physical activity

    Diabetes Care

    (2003)
There are more references available in the full text version of this article.

Cited by (0)

ISRCTN Trial registration number: ISRCTN66987085.

1

See Appendix A for collaboration details.

View full text