diabetes research and clinical practice 106 (2014) 383–389
Contents available at ScienceDirect
Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres
The DEXLIFE study methods: Identifying novel candidate biomarkers that predict progression to type 2 diabetes in high risk individuals§ G.S. Andersen a, T. Thybo a, H. Cederberg b, M. Oresˇicˇ a, M. Esteller c, A. Zorzano d,e,f, B. Carr g, M. Walker h, J. Cobb i, C. Clissmann j, D.J. O’Gorman k, J.J. Nolan a,* on behalf of the DEXLIFE Consortium1 a
Steno Diabetes Center, Gentofte, Denmark Department of Medicine, University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland c Cancer Epigenetics and Biology Program, Spanish Biomedical Research Centre Network for Epidemiology and Public Health, Catalan Institute of Oncology, Bellvitge Biomedical Research Institute, Barcelona, Spain d Institute for Research in Biomedicine, Barcelona, Spain e Departament de Bioquı´mica I Biologia Molecular, Facultat de Biologia, Universitat de Barcelona, Spain f CIBER de Diabetes y Enfermedades Metabo´licas Asociadas (CIBERDEM), Instituto de Salud Carlos III, Spain g Voluntary Health Insurance Board, Dublin, Ireland h University of Newcastle-on-Tyne, Newcastle, UK i Metabolon Inc., Durham, NC, USA j Pintail Ltd., Blackrock, Co., Dublin, Ireland k Centre for Preventive Medicine, School of Health and Human Performance, Dublin City University, Dublin, Ireland b
article info
abstract
Article history:
The incidence of type 2 diabetes (T2D) is rapidly increasing worldwide and T2D is likely to
Received 1 May 2014
affect 592 million people in 2035 if the current rate of progression is continued. Today,
Accepted 20 July 2014
patients are diagnosed with T2D based on elevated blood glucose, either directly or
Available online 27 July 2014
indirectly (HbA1c). However, the information on disease progression is limited.
Keywords:
reflect the underlying biology and the overall physiological, metabolic and clinical char-
Biomarkers
acteristics of progression towards diabetes.
Therefore, there is a need to identify novel early markers of glucose intolerance that
Type 2 diabetes
In the DEXLIFE study, several clinical cohorts provide the basis for a series of clinical,
Prevention
physiological and mechanistic investigations in combination with a range of – omic technolo-
Lifestyle intervention
gies 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. # 2014 Elsevier Ireland Ltd. All rights reserved.
§
ISRCTN Trial registration number: ISRCTN66987085. * Corresponding author. Tel.: +45 4443 5003. E-mail address: [email protected] (J.J. Nolan). 1 See Appendix A for collaboration details. http://dx.doi.org/10.1016/j.diabres.2014.07.025 0168-8227/# 2014 Elsevier Ireland Ltd. All rights reserved.
384
1.
diabetes research and clinical practice 106 (2014) 383–389
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 b-cell function or insulin resistance [2] and the optimal value for HbA1c to accurately diagnose the pre-diabetic state is still debated [3–5]. 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 b-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.
2.
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 circulating metabolomic and lipidomic profiles of progression to diabetes. A small cohort of young people with type 2 diabetes, matched controls and older people with type 2 diabetes (YT2 cohort), will be used to identify biomarkers of early onset diabetes. Our research strategy thus seeks to map as completely as possible the full range of the evolution of diabetes in the general population, from young adults through to middle-aged and older people at various stages of progression of metabolic disease. Finally, the RISC cohort [8] of healthy European people from all over the continent, will be used to validate the new candidate biomarkers. A specific strength of DEXLIFE is the depth to which each of these highly specialized cohorts has been phenotyped, thus enabling the deeper basic science investigations to be contextualized and integrated into a new level of understanding of diabetes progression and prevention. The specificity of candidate biomarkers for progression to diabetes is critical and DEXLIFE has adopted two additional strands to strengthen the outcomes. Firstly, the most robust set of biomarkers will not only track with progression to diabetes but also with improvements in insulin sensitivity and a decreased risk of type 2 diabetes. DEXLIFE will conduct a diet and exercise intervention, based on an approach similar to the Finnish Diabetes Prevention Study [9], which has been shown to reduce the incidence of type 2 diabetes. Secondly, based on the results of mechanistic studies in skeletal muscle we will identify novel circulating biomarkers that are reflective of cellular processes that have become altered in the aetiology of the disease. Skeletal muscle is the major site of glucose uptake, and has been known for many years to be the most important site of peripheral insulin resistance, one of the central components in the pathogenesis of type 2 diabetes. In DEXLIFE, transcriptomic, lipidomic and epigenetic analysis of muscle tissue will be performed during the lifestyle intervention and for the young onset type 2 diabetes groups. All of these modalities are centrally relevant to the influence of environmental change on the pathophysiology of diabetes. The analysis of these mechanistic parameters in muscle will greatly enhance and complement the wider
diabetes research and clinical practice 106 (2014) 383–389
findings in circulating plasma to provide a multi-dimensional metabolic phenotype for correlation with clinical status and progression. The phenotypic, biomarker, skeletal muscle and clinical/ physiological end points from all of these investigations will be integrated using bio-informatic tools to develop models of progression to diabetes. The results of the analyses will be compiled as a set of tools that can be used by healthcare professionals to design personalized regimens based on the biomarker profile of the patient.
3.
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 intervention project and the recruitment of participants from the DMVhi cohort were approved by their respective ethics committees. Ethical approval for the individual projects feeding into DEXLIFE comply with the Declaration of Helsinki (2008), EU FP7 regulations (decision no. 1982/2006/EC) and respective national regulations. Given the complexity of the project it is necessary to protect the confidentiality of the participants with regard to the processing of personal data and on the free movement of such data in accordance with Directive 95/46/EC and for the processing of personal data and the protection of privacy in the electronic communication of data by Directive 2002/58/EC. Finally, as DEXLIFE involves the procurement, testing and storage of human tissue Directive 2004/23/EC also applies. Ethical issues are discussed at the bi-annual DEXLIFE meetings.
4.
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.
4.1.
4.2.
385
DMVhi
The diabetes mellitus and vascular health initiative (DMVhi) enrolled 30,000 policy holders from a large Irish health insurance company, Vhi Healthcare, Ireland. From 2009 to 2012, policy holders aged 45–75 years from two large urban areas, were screened for diabetes risk based on fasting blood glucose and/or a 2-h oral glucose tolerance test in addition to the FINDRISC score [10]. Three year rescreening of the cohort commenced in 2012. A subcohort of 700 participants with serum samples taken at baseline were identified for participation in DEXLIFE. The subcohort was sampled to include all participants with baseline IFG (13%) and/or IGT (7%), in addition to participants with normal glycaemia but elevated diabetes risk based on FINDRISC at baseline (80%). As part of DEXLIFE, baseline and 3-year follow-up samples from this subcohort will be used for metabolomic and lipidomic analysis (Fig. 1).
4.3.
Young T2D study
This cohort consists of 26 young (20–34 years) and 47 older (36– 74 years) people with type 2 diabetes and age-matched obese normal glucose tolerance young (n = 30) and older (n = 32) controls. They were included in a crossover intervention trial comprising a 12 week supervised exercise intervention at 70% maximum aerobic capacity, preceded or followed by a 12 week dietary intervention matched for energy deficit, with a randomized order of intervention. Blood samples from all participants in the YT2D study are being screened for metabolites and a subset of 74 muscle samples are undergoing lipidomic analysis and the transcriptomic and epi-genetic profile will be elucidated (Fig. 1).
4.4.
RISC
The longitudinal multicenter RISC (relationship between insulin sensitivity and cardiovascular disease) cohort examines the relationship between insulin resistance and cardiovascular disease and diabetes risk and includes 1500 Europeans, recruited in 2003 and followed-up in 2008 and 2014, as described in the RISC study protocol [8]. As part of the DEXLIFE project, 200 participants from the RISC cohort will have a reassessment of beta-cell glucose sensitivity and whole body insulin sensitivity by an oral glucose tolerance test and physical activity by accelerometry, to examine biomarker progression in response to 10 year changes in objectively measured physical activity.
METSIM 4.5.
The METSIM (metabolic syndrome in men) study is a longitudinal study following 10,197 men aged 45–73 years, randomly selected from the population register of the city of Kuopio, Eastern Finland. Baseline examinations were completed in 2005–2010, and the follow-up study has been ongoing since 2010 [7]. Serum samples from baseline and followup of 220 participants who developed T2D during the 5-year follow-up period and 440 matched normoglycemic controls that did not develop T2D, are included in the DEXLIFE project for metabolomic and lipidomic analysis (Fig. 1)
Lifestyle intervention
The DEXLIFE lifestyle intervention is a 12-week partially supervised exercise training programme accompanied with dietary advice to improve insulin sensitivity and assist with body fat reduction. The participants will be randomly assigned at a 3:1 ratio to participate in the intervention group (n = 300) or a control group (n = 100) that will be given lifestyle advice to increase daily physical activity and improve their diet. Participants will be recruited from the Vhi Healthcare cohort and the general public who meet one of the following criteria:
386
diabetes research and clinical practice 106 (2014) 383–389
Table 1 – Parameters used in the DEXLIFE cohort description.
Age (in years) Sex Samples taken Sample type Cohort size (N) Included in DEXLIFE (n) Biochemistry Fasting capillary blood glucose 2-h OGTT HbA1c Insulin C-peptide Total cholesterol LDL-cholesterol HDL-cholesterol Triglycerides Plasma adiponectin Urine albumin DNA Clamp study Clinical measures Blood pressure Anthropometric measures Height Weight Waist circumference Hip circumference Body fat percentage Abdominal fat distribution Physical activity, fitness Heart rate Accelerometry Physical act. questionnaire VO2 max Questionnaires Diabetes risk score Family history diabetes Health, lifestyle, risk factor Quality of life, self-efficacy Physical activity score Registry information a b c
METSIM
DMVhi
YT2
Intervention
RISC
45–73 M Baseline 5 years Serum 10,197 660
45–75 M/F Baseline 3 years Plasma (EDTA) 29,548 700
16–74 (35) M/F Baseline 12 weeks Serum/muscle 135 135
M/F Baseline 12 weeks Serum/muscle 400 400
29–63 M/F Baseline 5, 10 years Serum 1538 200
x x x x x x x x x x
xa (x)b,c (x)c
x x x x
x x
x x
x x x x
x x x x x x
x x x x
x x
x
x x
x
x
x x x x x
x x x x (x)c
x
x
x x x
x x x
x x
x
x x x x x
x
x
x x x x x x
x x x x x
x x x x
x
x
x x
x x x
Venous blood sample. Only on those with initial FBG in IFG range at baseline screening. At 3 year follow up only.
(i) impaired fasting glucose or impaired glucose tolerance; (ii) normal glucose tolerance but a score >12 on the FINDRISC questionnaire (suggesting a 1 in 6 chance of developing type 2 diabetes in the next 10 years). Participants who provide consent will be invited to attend for medical screening and baseline measures of aerobic capacity with ECG monitoring, an oral glucose tolerance test, heart rate variability and body composition (DEXA and abdominal ultrasound). They will complete a series of questionnaires to assess quality of life, physical activity, health and lifestyle. A sub-group of the control (n = 50) and intervention (n = 100) participants will have a skeletal muscle biopsy from the vastus lateralis. After the baseline measures have been completed participants will be randomly assigned to the control and intervention groups.
The control group will be given healthy eating and physical activity advice in accordance with national recommendations and will be contacted by telephone during the 12-week followup. The intervention group will be given dietary advice and support to (i) reduce body total caloric intake by 600 kcal/day, (ii)
Contents available at ScienceDirect
Diabetes Research and Clinical Practice journ al h ome pa ge : www .elsevier.co m/lo cate/diabres
The DEXLIFE study methods: Identifying novel candidate biomarkers that predict progression to type 2 diabetes in high risk individuals§ G.S. Andersen a, T. Thybo a, H. Cederberg b, M. Oresˇicˇ a, M. Esteller c, A. Zorzano d,e,f, B. Carr g, M. Walker h, J. Cobb i, C. Clissmann j, D.J. O’Gorman k, J.J. Nolan a,* on behalf of the DEXLIFE Consortium1 a
Steno Diabetes Center, Gentofte, Denmark Department of Medicine, University of Eastern Finland and Kuopio University Hospital, Kuopio, Finland c Cancer Epigenetics and Biology Program, Spanish Biomedical Research Centre Network for Epidemiology and Public Health, Catalan Institute of Oncology, Bellvitge Biomedical Research Institute, Barcelona, Spain d Institute for Research in Biomedicine, Barcelona, Spain e Departament de Bioquı´mica I Biologia Molecular, Facultat de Biologia, Universitat de Barcelona, Spain f CIBER de Diabetes y Enfermedades Metabo´licas Asociadas (CIBERDEM), Instituto de Salud Carlos III, Spain g Voluntary Health Insurance Board, Dublin, Ireland h University of Newcastle-on-Tyne, Newcastle, UK i Metabolon Inc., Durham, NC, USA j Pintail Ltd., Blackrock, Co., Dublin, Ireland k Centre for Preventive Medicine, School of Health and Human Performance, Dublin City University, Dublin, Ireland b
article info
abstract
Article history:
The incidence of type 2 diabetes (T2D) is rapidly increasing worldwide and T2D is likely to
Received 1 May 2014
affect 592 million people in 2035 if the current rate of progression is continued. Today,
Accepted 20 July 2014
patients are diagnosed with T2D based on elevated blood glucose, either directly or
Available online 27 July 2014
indirectly (HbA1c). However, the information on disease progression is limited.
Keywords:
reflect the underlying biology and the overall physiological, metabolic and clinical char-
Biomarkers
acteristics of progression towards diabetes.
Therefore, there is a need to identify novel early markers of glucose intolerance that
Type 2 diabetes
In the DEXLIFE study, several clinical cohorts provide the basis for a series of clinical,
Prevention
physiological and mechanistic investigations in combination with a range of – omic technolo-
Lifestyle intervention
gies 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. # 2014 Elsevier Ireland Ltd. All rights reserved.
§
ISRCTN Trial registration number: ISRCTN66987085. * Corresponding author. Tel.: +45 4443 5003. E-mail address: [email protected] (J.J. Nolan). 1 See Appendix A for collaboration details. http://dx.doi.org/10.1016/j.diabres.2014.07.025 0168-8227/# 2014 Elsevier Ireland Ltd. All rights reserved.
384
1.
diabetes research and clinical practice 106 (2014) 383–389
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 b-cell function or insulin resistance [2] and the optimal value for HbA1c to accurately diagnose the pre-diabetic state is still debated [3–5]. 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 b-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.
2.
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 circulating metabolomic and lipidomic profiles of progression to diabetes. A small cohort of young people with type 2 diabetes, matched controls and older people with type 2 diabetes (YT2 cohort), will be used to identify biomarkers of early onset diabetes. Our research strategy thus seeks to map as completely as possible the full range of the evolution of diabetes in the general population, from young adults through to middle-aged and older people at various stages of progression of metabolic disease. Finally, the RISC cohort [8] of healthy European people from all over the continent, will be used to validate the new candidate biomarkers. A specific strength of DEXLIFE is the depth to which each of these highly specialized cohorts has been phenotyped, thus enabling the deeper basic science investigations to be contextualized and integrated into a new level of understanding of diabetes progression and prevention. The specificity of candidate biomarkers for progression to diabetes is critical and DEXLIFE has adopted two additional strands to strengthen the outcomes. Firstly, the most robust set of biomarkers will not only track with progression to diabetes but also with improvements in insulin sensitivity and a decreased risk of type 2 diabetes. DEXLIFE will conduct a diet and exercise intervention, based on an approach similar to the Finnish Diabetes Prevention Study [9], which has been shown to reduce the incidence of type 2 diabetes. Secondly, based on the results of mechanistic studies in skeletal muscle we will identify novel circulating biomarkers that are reflective of cellular processes that have become altered in the aetiology of the disease. Skeletal muscle is the major site of glucose uptake, and has been known for many years to be the most important site of peripheral insulin resistance, one of the central components in the pathogenesis of type 2 diabetes. In DEXLIFE, transcriptomic, lipidomic and epigenetic analysis of muscle tissue will be performed during the lifestyle intervention and for the young onset type 2 diabetes groups. All of these modalities are centrally relevant to the influence of environmental change on the pathophysiology of diabetes. The analysis of these mechanistic parameters in muscle will greatly enhance and complement the wider
diabetes research and clinical practice 106 (2014) 383–389
findings in circulating plasma to provide a multi-dimensional metabolic phenotype for correlation with clinical status and progression. The phenotypic, biomarker, skeletal muscle and clinical/ physiological end points from all of these investigations will be integrated using bio-informatic tools to develop models of progression to diabetes. The results of the analyses will be compiled as a set of tools that can be used by healthcare professionals to design personalized regimens based on the biomarker profile of the patient.
3.
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 intervention project and the recruitment of participants from the DMVhi cohort were approved by their respective ethics committees. Ethical approval for the individual projects feeding into DEXLIFE comply with the Declaration of Helsinki (2008), EU FP7 regulations (decision no. 1982/2006/EC) and respective national regulations. Given the complexity of the project it is necessary to protect the confidentiality of the participants with regard to the processing of personal data and on the free movement of such data in accordance with Directive 95/46/EC and for the processing of personal data and the protection of privacy in the electronic communication of data by Directive 2002/58/EC. Finally, as DEXLIFE involves the procurement, testing and storage of human tissue Directive 2004/23/EC also applies. Ethical issues are discussed at the bi-annual DEXLIFE meetings.
4.
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.
4.1.
4.2.
385
DMVhi
The diabetes mellitus and vascular health initiative (DMVhi) enrolled 30,000 policy holders from a large Irish health insurance company, Vhi Healthcare, Ireland. From 2009 to 2012, policy holders aged 45–75 years from two large urban areas, were screened for diabetes risk based on fasting blood glucose and/or a 2-h oral glucose tolerance test in addition to the FINDRISC score [10]. Three year rescreening of the cohort commenced in 2012. A subcohort of 700 participants with serum samples taken at baseline were identified for participation in DEXLIFE. The subcohort was sampled to include all participants with baseline IFG (13%) and/or IGT (7%), in addition to participants with normal glycaemia but elevated diabetes risk based on FINDRISC at baseline (80%). As part of DEXLIFE, baseline and 3-year follow-up samples from this subcohort will be used for metabolomic and lipidomic analysis (Fig. 1).
4.3.
Young T2D study
This cohort consists of 26 young (20–34 years) and 47 older (36– 74 years) people with type 2 diabetes and age-matched obese normal glucose tolerance young (n = 30) and older (n = 32) controls. They were included in a crossover intervention trial comprising a 12 week supervised exercise intervention at 70% maximum aerobic capacity, preceded or followed by a 12 week dietary intervention matched for energy deficit, with a randomized order of intervention. Blood samples from all participants in the YT2D study are being screened for metabolites and a subset of 74 muscle samples are undergoing lipidomic analysis and the transcriptomic and epi-genetic profile will be elucidated (Fig. 1).
4.4.
RISC
The longitudinal multicenter RISC (relationship between insulin sensitivity and cardiovascular disease) cohort examines the relationship between insulin resistance and cardiovascular disease and diabetes risk and includes 1500 Europeans, recruited in 2003 and followed-up in 2008 and 2014, as described in the RISC study protocol [8]. As part of the DEXLIFE project, 200 participants from the RISC cohort will have a reassessment of beta-cell glucose sensitivity and whole body insulin sensitivity by an oral glucose tolerance test and physical activity by accelerometry, to examine biomarker progression in response to 10 year changes in objectively measured physical activity.
METSIM 4.5.
The METSIM (metabolic syndrome in men) study is a longitudinal study following 10,197 men aged 45–73 years, randomly selected from the population register of the city of Kuopio, Eastern Finland. Baseline examinations were completed in 2005–2010, and the follow-up study has been ongoing since 2010 [7]. Serum samples from baseline and followup of 220 participants who developed T2D during the 5-year follow-up period and 440 matched normoglycemic controls that did not develop T2D, are included in the DEXLIFE project for metabolomic and lipidomic analysis (Fig. 1)
Lifestyle intervention
The DEXLIFE lifestyle intervention is a 12-week partially supervised exercise training programme accompanied with dietary advice to improve insulin sensitivity and assist with body fat reduction. The participants will be randomly assigned at a 3:1 ratio to participate in the intervention group (n = 300) or a control group (n = 100) that will be given lifestyle advice to increase daily physical activity and improve their diet. Participants will be recruited from the Vhi Healthcare cohort and the general public who meet one of the following criteria:
386
diabetes research and clinical practice 106 (2014) 383–389
Table 1 – Parameters used in the DEXLIFE cohort description.
Age (in years) Sex Samples taken Sample type Cohort size (N) Included in DEXLIFE (n) Biochemistry Fasting capillary blood glucose 2-h OGTT HbA1c Insulin C-peptide Total cholesterol LDL-cholesterol HDL-cholesterol Triglycerides Plasma adiponectin Urine albumin DNA Clamp study Clinical measures Blood pressure Anthropometric measures Height Weight Waist circumference Hip circumference Body fat percentage Abdominal fat distribution Physical activity, fitness Heart rate Accelerometry Physical act. questionnaire VO2 max Questionnaires Diabetes risk score Family history diabetes Health, lifestyle, risk factor Quality of life, self-efficacy Physical activity score Registry information a b c
METSIM
DMVhi
YT2
Intervention
RISC
45–73 M Baseline 5 years Serum 10,197 660
45–75 M/F Baseline 3 years Plasma (EDTA) 29,548 700
16–74 (35) M/F Baseline 12 weeks Serum/muscle 135 135
M/F Baseline 12 weeks Serum/muscle 400 400
29–63 M/F Baseline 5, 10 years Serum 1538 200
x x x x x x x x x x
xa (x)b,c (x)c
x x x x
x x
x x
x x x x
x x x x x x
x x x x
x x
x
x x
x
x
x x x x x
x x x x (x)c
x
x
x x x
x x x
x x
x
x x x x x
x
x
x x x x x x
x x x x x
x x x x
x
x
x x
x x x
Venous blood sample. Only on those with initial FBG in IFG range at baseline screening. At 3 year follow up only.
(i) impaired fasting glucose or impaired glucose tolerance; (ii) normal glucose tolerance but a score >12 on the FINDRISC questionnaire (suggesting a 1 in 6 chance of developing type 2 diabetes in the next 10 years). Participants who provide consent will be invited to attend for medical screening and baseline measures of aerobic capacity with ECG monitoring, an oral glucose tolerance test, heart rate variability and body composition (DEXA and abdominal ultrasound). They will complete a series of questionnaires to assess quality of life, physical activity, health and lifestyle. A sub-group of the control (n = 50) and intervention (n = 100) participants will have a skeletal muscle biopsy from the vastus lateralis. After the baseline measures have been completed participants will be randomly assigned to the control and intervention groups.
The control group will be given healthy eating and physical activity advice in accordance with national recommendations and will be contacted by telephone during the 12-week followup. The intervention group will be given dietary advice and support to (i) reduce body total caloric intake by 600 kcal/day, (ii)
