EMIF Metabolic

EMIF-Metabolic Project Overview

Accumulating evidence indicates that obesity is closely associated with an increased risk of metabolic complications such as type 2 diabetes, coronary heart disease, non-alcoholic fatty liver disease and cancers.  However, the association between obesity and any of these complications is complex, with high inter-individual variability in susceptibility to specific metabolic complications of obesity. This association is expected to be additionally influenced by different constitutional, environmental and obesity-specific factors, thus further complicating the development of adequate treatments.

EMIF-Metabolic Project Objectives

EMIF-Metabolic will focus on two distinct pathways through which individuals who are obese may vary in their risk of the complications of obesity. One is to identify genetic causes of obesity and their relation to the metabolic complications of obesity. The other pathway is to characterize individuals and identify markers associated with metabolic risk irrespective of degree of obesity based on the knowledge that many obese individuals do not become dysmetabolic and insulin resistant. In both instances the complementary approaches of studying extreme phenotypes and population-based cohorts will be followed.

Placeholder image

 

WPEMIFMET Evaluate epidemiology of obesity-related clinical conditions in large real-world databases WP8 Validate sensitivity/specificity of novel biomarkers vs. currently available biomarkers WP7 Characterise heterogeneity in metabolic consequences of obesity in small, medium and large clinical populations with outcome data WP6 Identify novel biomarkers and mechanisms for obesity-associated complications using extreme phenotypes and unbiased “omics” technologies, as well as targeted approaches WP5

Prof Ulf Smith, University of Gothenburg

The discovery of predictors of the metabolic complications of adult and paediatric obesity will lead to innovative diagnostic tests, pave the way to novel therapeutics targeted to high-risk individuals, and provide the infrastructure to select individuals for such targeted pharmacological interventions (genetic, epigenetic and ‘omics platforms).

Identified novel mechanisms and potential targets for therapy will be characterized in appropriate cell- and animal-based models to facilitate transition to human studies. To do so, EMIF-Metabolic brings together top European experts in genetics, epidemiology, genomics and other ‘omic’ technologies, human and murine physiology and systems approaches and state-of-the-art technologies. 

EMIF-Metabolic Project Achievements

WP 5 Use carefully characterized extreme phenotypes to identify biomarkers of metabolic risk

Peter Arner (KI) – Markku Laakso (UEF) – Dawn Waterworth (GSK)

KI (CO-LEAD), UEF (CO-LEAD), GSK (CO-LEAD), UGOT, UCAM, UH, INSERM, ULEI, UGLA, VKJK, Janssen, PFIZER, NOVO, Amgen, BI

GOALS & OBJECTIVES

The main objective is:

To characterize in detail using “omics” technologies extreme phenotypes related to the complications of obesity (including insulin resistance and liver fat in first degree relatives of diabetics). Response to weight loss will also be characterized. The most interesting biomarkers will subsequently be further evaluated in tissue culture and animal models and taken forward for further validation in medium sized cohorts in WP6. The most interesting markers will be those most correlated with the endpoints of interest and with properties that would lend themselves to being clinically useful; such as robustness, variability, frequency, tissue specificity and ease of detection.

The sub-objectives are:

  1. Detailed omics analysis in well characterized samples with extreme phenotypes or dietary interventions
  2. Comprehensive integrated analysis of detailed omics data using data reduction/mining/statistical techniques to inform marker selection.
  3. Experimental assessment of selected biomarkers in cell-based and animal models to assess validity as drug targets and diagnostic tools.

KEY DISSEMINATION ACTIVITIES

  • Dahlman et al. The epigenetic signature of systemic insulin resistance in obese women, under revision 2016
  • Zhou et al Circulating triacylglycerol signatures and insulin sensitivity in NAFLD associated with the E167K variant in TM6SF2. Journal of Hepatology 2015
  • Hyysalo et al, Circulating triacylglycerol signatures in nonalcoholic fatty liver disease associated with the I148M variant in PNPLA3 and with obesity. Diabetes 2014

KEY ACHIEVEMENTS 2013-2015

Gene expression and epigenetic signature of insulin resistance in obesity and weight reduction (KI/INSERM/GSK)

  • Differential gene expression show directionally consistent results between visceral and subcutaneous adipose, but not with PBMCs
  • DNA methylation—may be important for specific genes
  • Identification of genes related to insulin sensitivity improvement after weight loss

Identification of adipokines related to insulin resistance (KI)

  • Adipose and circulating levels of the chemokine CCL18 associate with insulin resistance and metabolic risk factors in women

Metabolomics/lipidomics data (UGOT/UEF/ULEI/UH/UCAM/ GSK/Pfizer)

  • Integrated network analysis reveals an association between plasma mannose levels and insulin resistance and secretion
  • Markers of insulin secretion validated in METSIM
  • Analyses of metabolomics and lipoprotein flux data in 86 subjects with NAFLD, systems biology approach has identified possible intervention approach (serine supplementation). Hypothesis confirmed in pilot study demonstrating improvements in liver function and reduction in fat
  • Characterization of genetically validated NAFLD/NASH targets: PNPLA3 (I148M) and TM6SF2 (E176K) in liver metabolome of 90 patients with liver biopsy. Extended list of target genes to be characterized in serum metabolome in 8500 subjects

TranSMART (Janssen/GSK/all academic partners)

    • Uploaded genetics and metabolomics data

WP 6 The investigation of heterogeneity in the metabolic consequences of obesity in medium-sized well phenotyped cohort studies

Nick Wareham (UCAM) – Dawn Waterworth (GSK) – Julia Brosnan (PFIZER)

UCAM (CO-LEAD), GSK (CO-LEAD), PFIZER (CO-LEAD), UGOT, UH, UEF, INIPI, VKJK, ULEI, Janssen, Amgen, BI

GOALS & OBJECTIVES

The main objective is:

To use candidate and discovery approaches to identify biomarkers which are associated with the metabolic consequences of obesity and to investigate their causal significance in medium-sized well phenotyped cohorts and large-scale nested case-cohort studies.

The sub-objectives are:

  1. To select quantitative metabolic cohort studies to test biomarkers for the prediction of the metabolic consequences of obesity.
  2. To use candidate and discovery approaches to identify potential biomarkers.
  3. To study the association and likelihood of causality of biomarkers for the prediction of the metabolic consequences of obesity.

KEY DISSEMINATION ACTIVITIES

  • Scientific article in PLoS Medicine: “Associations between potentially modifiable risk factors and Alzheimer’s disease: A Mendelian randomization study”
  • Scientific article in Journal of the National Cancer Institute: “Evidence of a causal association between hyperinsulinaemia and endometrial cancer: A Mendelian randomization analysis”
  • Scientific article in Diabetes Care: “Definitions of Metabolic Health and Risk of Future Type 2 Diabetes in BMI Categories: A Systematic Review and Network Meta-analysis”

KEY ACHIEVEMENTS 2013-2015

  • Completed systematic review of aetiologic and predictive role of biomarkers in type 2 diabetes
  • Published a systematic review of the predictive utility of different binary definitions of metabolically healthy and unhealthy obesity
  • Selected candidate biomarkers based on a review of biomarker to disease and gene to biomarker evidence
  • Completed more than 50% of the candidate biomarker measurements in 23,000 people
  • Completed discovery biomarker measurements (n=1165 biomarkers) in a cohort of 7,500 people
  • Used MR approaches to show that the association of BCAA levels with type 2 diabetes is compatible with an underlying causal mechanism
  • Used MR approaches to show that insulin resistance is likely to be causally linked to endometrial cancer
  • Used MR approaches to show that genetic variants associated with higher blood pressure are associated with dementia

WP 7 Validate novel risk factors in general population in adults and children

Naveed Sattar (UGLA) – Katrina Loomis (PFIZER)

UGLA (CO-LEAD), PFIZER (CO-LEAD), UGOT, UCPH, Janssen, GSK, ROCHE, Amgen, BI

GOALS & OBJECTIVES

The main objective is:

The overall objective of this WP is to measure then test the ability of new metabolic biomarkers to predict obesity-related outcomes using retrospective and prospective studies.

The sub-objectives are:

  1. Identify cohorts from EHR which have available baseline serum / plasma samples, acceptable phenotyping (i.e. conventional vascular risk factors, family history data and level of BMI or waist circumference) and with access to outcomes for vascular disease (i.e. CHD events), incident diabetes, and other obesity-related comorbidities such as fatty liver disease or sleep apnoea.
  2. To measure new biomarkers in established biobanks with available serum or plasma from cohorts with at least conventional vascular risk factors measured at baseline (plus BMI) and with documented outcomes covering at least CHD events but potentially also including incident cancers, diabetes and fatty liver disease.
  3. Compare the ability of biomarkers to predict obesity-related outcomes in comparison to conventionally measured risk factors and simple demographic information.
  4. Test the predictive capacity of the most promising biomarkers by utilising samples taken from trials held by Industry participants or from EHR/biobank studies.

Very limited access to samples for biomarker analysis

  • Confirming new metabolic biomarkers identified in WP5 and WP6 will require EMIF to bring in additional biobanks or cohort resources with available biospecimens.

KEY DISSEMINATION ACTIVITIES

  • Loomis AK, Kabadi S, Preiss D, Hyde C, Bonato V, St Louis M, Desai J, Gill JM, Welsh P, Waterworth D, Sattar N. Body Mass Index and Risk of Nonalcoholic Fatty Liver Disease: Two Electronic Health Record Prospective Studies. J Clin Endocrinol Metab. 2016 Mar;101(3):945-52. doi: 10.1210/jc.2015-3444. Epub 2015 Dec 16.
  • Preiss D, Rankin N, Welsh P, Holman RR, Kangas AJ, Soininen P, Würtz P, Ala-Korpela M, Sattar N. Effect of metformin therapy on circulating amino acids in a randomized trial: the CAMERA study: Metabolism. Diabet Med. 2016 (Epub ahead of Print).
  • Rankin NJ, Preiss D, Welsh P, Burgess KE, Nelson SM, Lawlor DA, Sattar N. The emergence of proton nuclear magnetic resonance metabolomics in the cardiovascular arena as viewed from a clinical perspective. Atherosclerosis. 2014 Nov;237(1):287-300.

KEY ACHIEVEMENTS 2013-2015

Developed algorithms with Platform colleagues to extract BMI and metabolic endpoints from EMIF data providers

  • Type 2 diabetes, BMI, NAFLD/NASH, CVD

Developed protocols for two use cases to test the utility of the platform architecture for data access and analysis

  • Use Case 9: Characterize the relationship between obesity and risk/progression of CVD.
  • Use Case 10: Characterize the relationship between obesity, NAFLD/NASH and metabolic complications. Protocol has been fully approved by 4 data providers, data extraction is complete, and analysis is in progress.

Proof of Concept: Preliminary analysis for UC10 from IPCI

Incidence of NAFLD/NASH in IPCI 1996-2014, Female Patients

Age

Incidence of IHD per 1000 person years

95%CI

NAFLD/NASH Cases

11.84

(10.75-13.01)

Control population

8.09

(8.00-8.18)

 

Defined analysis process within the EMIF framework

Defined analysis process within the EMIF framework
 

WP 8 Identify and select individuals for pharmacological or non-pharmacological interventions

Ele Ferrannini (UNIPI) – Stuart Kendrick (GSK) – Janet Addison (Amgen)

UNIPI (CO-LEAD), GSK (CO-LEAD), Amgen (CO-LEAD), UCPH, Janssen, PFIZER, BI

GOALS & OBJECTIVES

The main objective is:

The pharmaceutical intervention work package is designed to deliver real world observational data and data from randomised controlled studies, validating the extreme phenotype approach for classification of obesity, diseases related to obesity, and biomarkers identified by other work packages. It is also designed to test the feasibility of electronic healthcare record driven recruitment to intervention trials.

The sub-objectives are:

  1. Establish whether or not EHR’s can be used to conduct feasibility for our clinical trials?
  2. Establish whether or not EHR’s can be used to contact the patients for recruitment into our clinical trials
  3. Better understanding relationship (adverse effect & therapeutic effect) between extreme phenotypes and response to prescribed drugs
  4. Validate biomarkers to predict therapeutic response in interventional trials
  5. Utilize clinical trial data generated by EFPIA partners to investigate the liver complications of obesity.

KEY DISSEMINATION ACTIVITIES

  • Presentation delivered at Learning Health System Conference, September 2015: “EMIF Ethical Code of Practice”
  • Poster presented at the EMIF AD-MET Joint General Assembly Meeting, March 2016: "Temporal evolution of liver function tests in nondiabetic and diabetic subjects: association with obesity."

KEY ACHIEVEMENTS 2013-2015

Guidance on Secondary Use of RCT Data:

  • Produced White Paper on Ethics and Privacy best practice for all WPs
  • This has been disseminated across the Consortium as guidance on legal/ethical aspects of re-use and sharing of existing data, as applied to both research cohorts and existing databases.

Research Project Utilising RCT Data:

  • Several Pharma partners initiated a pilot study to evaluate the possibility of pooling placebo-arm RCT data
  • Technical challenges of sharing data are being worked on
  • Individual companies are applying a common protocol to their RCT data, to be followed with meta-analysis of results
  • Protocol has the objective of exploring potential use of liver function measurement as early predictors of development of liver disease, and the relationship with obesity
  • Analysis utilising data from a large CV outcomes trial is close to completion; Analysis utilising data from a large diabetes trial is underway
  • A joint manuscript is in preparation

COLLABORATIONS & OUTREACH

  • Collaborations between Pharma and academic partners are well-established
  • Collaboration with WPs 7 and 15 has facilitated progress in research protocols and in developing an Ethical Code of Practice, respectively