•  EN 
  • Employee login

Research Projects

Print 

Please find below a description of the Research Projects.

Disease Mechanisms

Project 1.1 Genomics guided drug repositioning in asthma  

Genomics guided drug repositioning in asthma

Name Prominent PI / Promotor: Prof. Gerard Koppelman
Name 2nd Promotor: Prof. Reinoud Gosens
Other project members: Prof. Dr. Frank Dekker, Dr. Judith Vonk, Dr. Martijn Nawijn, Dr. Maarten Van den Berge, Dr. Loes Kistemaker

Background

Asthma is a chronic respiratory disease that affects approximately 300 million people worldwide [1]. It is characterized by recurrent respiratory symptoms and variable airflow limitation. Due to its significant health impact and continuously high prevalence (~6 – 8 %) both in (early) childhood as well as adulthood, asthma significantly impairs healthy ageing throughout the lifespan. Despite progress in understanding this chronic disease, there has been no success in developing new drugs that target the underlying mechanisms rather than suppressing symptoms. Genomics-directed drug repositioning may provide a solution to this problem [2]. Based on our increasing understanding of asthma genetics, druggable targets may be identified that could allow for improved treatment of asthma by existing drugs or drugs under development for unrelated diseases.

In 2017, we anticipate our publication of the largest genetic consortium on allergic disease to date, the SHARE consortium (Fereira et al, Nature Genetics, accepted for publication). This analysis was performed in 360,838 subjects worldwide, including the Dutch LifeLines cohort (Koppelman and Vonk, co-PIs). We identified 136 genetic risk variants in 99 loci contributing to allergic disease (asthma, rhinitis, eczema). Functional analysis of these 136 risk variants revealed that they regulate expression of 244 plausible target genes in for allergy relevant cells and tissues. Following the approach as described by Sanseu et al. [2], we identified that 61 of these target genes are being considered as drug targets in human disease; 15 of which for allergic disease (for example JAK2, STAT6, IL33, IL1RL1, TSLP), 11 for autoimmune diseases and 35 for other diseases . This opens up a great opportunity to reposition drugs developed or in clinical trials for other diseases for the treatment of asthma and allergies.

Overall aim

In this project, we aim to identify the relevant targets for repositioned drugs for asthma intervention, validate these in in vitro model systems for preclinical studies assessing the suitability of selected candidate compounds and validate their efficacy in relevant in vivo disease models. Our ultimate aim of this project is to validate a novel drug for asthma, that can be used in human phase II studies.

Methods

We will accomplish this by taking the following stepwise approach.

(1) We will first validate the risk SNPs associated to the 61 druggable genes for asthma, lung function and blood eosinophils (LifeLines cohort) and then determine which specific asthma subtype is regulated by these SNPs, by relating risk SNPs to asthma and its related subtypes (total IgE, IgE sensitization, blood eosinophils, airway hyperresponsiveness (AHR), lung function, asthma onset) in three cohorts with rich asthma phenotypes. (Dutch Asthma GWAS study, PIAMA). We will use this information to understand which asthma related trait is specifically regulated by this SNP. This may help to define readouts in any further drug study. For example, a SNP may regulate a transcript involved in AHR. A drug developed to counteract this transcript may then first be studied for any effect on AHR. 

(2) In order to select the targets to be considered for drug repositioning, we will employ a staged approach: We will first validate if the gene transcripts display the expected differential expression in RNA Seq datasets from lung tissue or bronchial epithelial cells obtained from asthma patients and controls (biopsy and single-cell RNA sequencing datasets that are already available). Next, we will interrogate the expression of these genes in human primary cell culture models such as lung epithelial organoid cultures, air-liquid interface cultured human airway epithelial cells and human airway smooth muscle cells to evaluate the suitability of these models for step (3). We will compare expression levels between cells obtained from asthma patients and controls, as well as generate an asthma-like phenotype in these primary cells by exposing them to IL-13 and/or TGF-β. These experiments will not only provide additional support for the gene transcripts to be associated with asthma, but also will help prioritize the validation studies under (4). We will prioritize druggable genes, that (1) show consistent association with asthma (related traits); (2) and show differential expression in human asthma versus controls datasets.

(3) Drugs that have been developed to target validated genes will then be prioritized based on type (where we will select small molecules over biologicals), availability and ability to manufacture. These drugs will then be tested in proof of concept studies in vitro, for which the selection of the models will be based on the validation studies under 2. The  most promising drugs will then be evaluated in vivo. To direct the choice of in vivo model, we will first test which genes display similar expression changes in lung tissue in our in vivo models of asthma, including house dust mite induced asthma as well as ovalbumin induced asthma in the mouse and guinea pig by analysing existing gene expression datasets (whole lung RNA seq, airway epithelial RNA seq). We anticipate to test at least 3 drugs in the in vivo models that have been shown relevant for the specific target in the validation studies.

Tentative thesis Chapter outline

  1. Identification of the relevant targets for repositioned drugs for asthma
  2. Validation of targets in in vitro model systems for preclinical studies
  3. Detection methods and investigation of the suitability of candidate compounds
  4. Validation of the selected compounds and detection methods in relevant in vitro and vivo models.

 

How does this project contribute to Precision Medicine

Asthma genetics has been a key strength of GRIAC since the mid-90s, when the chromosome 5q locus was first identified as harbouring asthma and hyperresponsiveness genes. Since then, technological advances have allowed for more rapid, cost-effective and unbiased screening for asthma susceptibility genes. Despite of the increasing understanding of asthma genetics and mechanisms, these successes have not yet lead to new drugs that target the underpinnings of asthma rather than its symptoms. We will use a genomics-directed drug repositioning strategy to identify potential new drug candidates for the treatment of asthma based on drugs on the market or under development for unrelated diseases. This drug repositioning strategy has proven efficacy [2] and takes full advantage of the opportunities that the diversity in expertise and  the existing interrelationships between GRIP and UMCG have to offer.  Functional genomics and genomics-directed drug discovery will likely take center stage in (personalized) medicine.

Possible links to other ProminenT domains

Obesity, diabetes, and asthma have attained global epidemic proportions. Multiple studies have shown a strong epidemiological and experimental link between obesity and asthma that further relates to a diverse set of etiologic factors including altered lung mechanics, adipose hormones, and inflammatory cytokines. There is strong evidence that overt diabetes and insulin resistance are strongly associated with reduced lung function characterized by low forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC). Given these epidemiological findings, common mechanistic links between asthma and diabetes may exist, which positions this proposal in an unconventional yet unique place in the PROMINENT program, by developing disease-overarching ideas and personalized medicine approaches.

Project 1.2 The bile acid receptor as treatment in liver disease  

The bile acid receptor as treatment in liver disease

Name Prominent PI / Promotor: Prof. dr. F. Kuipers/Prof. dr. H.J. Verkade
Name Co-Promotor:
Other project members: dr. J.F. de Boer/Prof. dr. B. van der Sluis

Background

Over the past decade, it has become apparent that bile acids are involved in a host of activities beyond their classic functions in bile formation and fat absorption. The identification of the farnesoid X receptor (FXR) as a nuclear receptor directly activated by bile acids has opened new avenues of research. FXR regulates various elements of glucose, lipid and energy metabolism. Consequently, a picture has emerged of bile acids acting as modulators of (postprandial) metabolism. Therefore, strategies that interfere with either bile acid metabolism or signaling cascades mediated by bile acids represent novel therapeutic approaches for metabolic diseases, in addition to more evident applications in (cholestatic) liver disease. Synthetic modulators of FXR have been designed and the first, i.e., obeticholic acid (OCALIVA) has been FDA-approved in 2016 for the treatment of the rare cholestatic liver disease Primary Biliary Cholangitis.

Disorders associated with disrupted nutrient/energy homeostasis, e.g., obesity, metabolic syndrome, and non-alcoholic fatty liver disease (NAFLD), are increasing worldwide. Susceptibility to develop these diseases may originate from early life (programming).  In NAFLD, a progressive subtype exists, designated as non-alcoholic steatohepatitis (NASH), that is recognized as major cause liver cirrhosis and hepatocellular carcinoma. Disturbed signaling of FXR appears to contribute to the pathogenesis of NAFLD. Standard therapeutic interventions have not been established for NAFLD, but some new agents that activate FXR signaling have shown promise as possible therapeutics. Yet, many steps, involving mechanistic studies in relevant animal models, are still required for tailoring pharmacotherapy to the dominant pathogenic pathways in a given patient, possibly with use of combination therapy. The current project will contribute to the future direction in (personalized) treatment of patients with NASH, through application of a newly developed mouse model with humanized bile acid metabolism to allow rapid and more accurate  translation to the human situation.

Overall aim

Evaluation of novel FXR modulators in mice for potential application in humans is hampered by the presence of mouse-specific pathways in bile acid metabolism. We have recently generated a mouse model with humanized bile acid metabolism.

Overall aims:

  1. To determine the role of FXR in the context of a humanized bile acid pool in the etiology of diet-induced NAFLD/NASH in young and adult mice, with focus on lipogenic, gluco-regulatory, cholesterogenic and inflammatory pathways
  2. To determine the efficacy of) pharmacological FXR modulators in the context of a humanized bile acid pool for treatment of diet-induced NAFLD/NASH

Methods

Genetically-modified mouse models have greatly contributed to our understanding of the physiological functions of FXR. However, the presence of mouse-specific pathways in bile acid metabolism, leading to the formation of very hydrophilic muricholic acids with FXR antagonistic rather than agonistic actions, compromizes the use of mouse models for translational studies in the FXR field. Using CRISPR/Cas9-mediated somatic gene editing, we silenced the Cyp2c70 gene specifically in hepatocytes, resulting in a novel mouse model with a human-like bile acid metabolism that overcomes these drawbacks while still allowing methods developed in our laboratory for in vivo evaluation of NAFLD/NASH development.

Aim 1

To silence Cyp2c70 in liver, young and adult mice expressing Cas9 specifically in hepatocytes will be injected with adenovirus containing sgRNA targeting the Cyp2c70 gene. Three weeks after virus injection the bile acid composition will analyzed by LC-MS to evaluate effectivity of the procedure. In subgroups of mice, hepatic Fxr and Cyp2c70 will be deleted simultaneously. Mice will be fed chow diet or high-fat/high-fat + cholesterol (to induce inflammation)/fructose (to induce lipogenesis) diets up to 12 weeks to induce varying degrees of NAFLD/NASH.

At specific points in time, relevant physiological pathways will be quantified by stable isotope-based methods established in our laboratory (lipogenesis, lipolysis, beta-oxidation, cholesterogenesis and turnover, bile acid synthesis, glucose metabolism). Upon termination, liver tissue will be harvested for histological evaluation (with prof Alain de Bruin, Utrecht/UMCG), as well as for gene expression and lipid analyses. Simultaneously, other relevant organs (intestine, fat depots) will be harvested for evaluation.

Aim 2

Chow-fed mice will, upon deletion of Cyp2c70, be treated with available (GW6046, PX-20606) and/or novel (collaboration industrial partners) FXR agonist for periods up to 2 weeks to establish their effects on bile formation, bile acid metabolism and cholesterol turnover, as described (De Boer et al. Gastroenterology, 2017).

Next, the metabolic effects of FXR agonists will be evaluated in ‘humanized mice’ with diet-induced NAFLD, chosen on the basis of outcome Aim 1.

Tentative thesis Chapter outline

  1. Introduction
  2. Cholesterol and bile acid metabolism and bile formation in a mouse model with a humanized bile acid pool
  3. Development of diet-induced NAFLD in a mouse model with a humanized bile acid pool-role of FXR – controlled pathways in lipid and cholesterol metabolism
  4. Does a humanized bile acid pool impact on the development of insulin resistance and disturbed glucose metabolism in diet-induced NAFLD in mice?
  5. Does a humanized bile acid pool contribute to the development of inflammation in diet-induced NAFLD in mice?
  6. Effects of pharmacological FXR activation on cholesterol and bile acid metabolism in a mouse model with a humanized bile acid pool
  7. Effects of pharmacological FXR activation on diet-induced NAFLD pathologies in a mouse model with a humanized bile acid pool
  8. Combination therapies?
  9. General discussion: is there a place for FXR modulation in the treatment of (specific aspects) of NAFLD/NASH?

How does this project contribute to Precision Medicine

Obesity, metabolic syndrome, and non-alcoholic fatty liver disease (NAFLD) are increasing worldwide. Standard therapeutic interventions have not been established for NAFLD and its sequel NASH. Apart from genetic components, environmental factors – e.g., by exposure during early life - such as dietary intake, diet composition but also composition of the microbiome have been implicated in the etiologies of NAFLD/NASH. Specific metabolic and/or inflammatory pathways may be involved in development and progress of the disease, hence, a one-size-fits-all preventive/therapeutic approach is deemed to be unsuccessful. New agents that activate FXR signaling have shown promise, yet, are likely to be more effective in combination with dietary recommendations and/or other pharmacological means (e.g., inhibitors of lipogenesis, bile acid sequestrants) depending on the individual’s specific features. This project will contribute to our understanding of the etiology of (humanized) NAFLD/NASH and may contribute identification of biomarkers to allow personalized tailoring of therapies for NAFLD/NASH in humans.

Possible links to other ProminenT domains

  1. This project will increase insight into the mechanisms of NAFLD development and its progression to NASH (Disease Mechanisms)
  2. Through interactions with pharma partners, this project may contribute to evaluation/design of novel FXR agonists cq design of novel drug combinations for treatment of NAFLD/NASH (Drug Development)
Project 1.3 Epigenetic regulation of the inflammatory response in type 2 Diabetes  

Epigenetic regulation of the inflammatory response in Diabetes Mellitus Type II (DMT-II) through protein acetylation

Name Prominent PI / Promotor: Rainer Bischoff
Name Co-Promotor: Sabeth Verpoorte
Other project members: Marcel Kwiatkowski, Alienke van Pijkeren, Barbro Melgert, Frank Dekker

Background

Diabetes mellitus typ-2 (DMT-II) is a major cause of morbidity and mortality worldwide. Diabetes belongs to age-related diseases and is associated with chronic inflammation. The underlying molecular mechanisms responsible for the chronic inflammation are still largely unknown and patients suffer from little therapeutic possibilities to effectively treat chronic inflammation. Recent research indicates that chronic inflammation is closely linked to changes in energy metabolism via lysine acetylation of both histone and none-histone proteins. Protein acetylation is carefully balanced by a network of lysine acetyl transferases (KAT, also known as histone acetyl transferases (HAT)) and lysine deacetylases (KDACs, also known as histone deacetylases (HDACs)). Reversible protein acetylation regulates a number of important cellular processes including gene expression, via acetylation of histones and transcription factors such as NFkB. Acetyl-CoA represents the substrate for KATs to acetylate lysine residues. Changes in glycolysis, tricarboxylic acid (TCA) cycle and beta-oxidation of fatty acids induce changes in cellular acetyl-CoA concentrations and may represent an important link between changes in energy metabolism and gene expression. Since DMT-II is associated with changes in energy metabolism, we hypothesize that regulation of protein acetylation and deacetylation is misbalanced resulting in disease-specific gene expression changes. To investigate how changes in energy metabolism affect protein acetylation profiles, a methodology will be developed to monitor changes in cellular metabolites (e.g. cellular acetyl-CoA) on the one hand and dynamics of protein acetylation patterns on the other hand and correlate them to changes in gene expression. Experiments will be carried out in perfused tissue slices, notably precision cut liver slices, in a dedicated microfluidics device allowing precise control over nutrient delivery as well as sample collection. We will use this method to study the anti-inflammatory effects of KDAC inhibitors by investigating changes in protein acetylation and deacetylation dynamics.

Overall aim

The aim of this research project is to identify molecular mechanisms linking changes in energy metabolism to pro-inflammatory gene expression in DMT-II. The project follows the hypothesis that aberrant changes in the levels of cellular energy metabolites, notably acetyl-CoA, lead to a change in the relative levels of acetylated protein species driving pro-inflammatory gene expression. To this end we will establish methodology based on liquid chromatography – mass spectrometry (LC-MS) in conjunction with microfluidics and stable isotope labelling.

Methods

Experiments will be carried out in microfluidic perfusion tissue slice culture in collaboration with the group of Sabeth Verpoorte. For microfluidic perfusion tissue slice culture different tissues relevant to DMT-II, such as precision cut liver slices, will be used. To investigate how changes in energy metabolism affect protein acetylation profiles, a methodology will be developed to monitor changes in cellular metabolites and protein acetylation patterns using metabolic and chemical labeling with stable isotopes. This methodology will include analysis of cellular metabolites (e.g. citrate and acetyl-CoA) by mass spectrometry (MS) and NMR on the one hand and MS analysis of acetylation dynamics of histones and none-histone proteins on the other hand.
To investigate histone acetylation dynamics, a combination of metabolic (tracer molecules of energy metabolism such as glucose, glutamine, octanoic acid) and chemical stable isotope labeling (acetic-anhydride) will be used. This combination will enable us to investigate changes in histone acetylation dynamics upon treatment with high glucose, high lipid and high insulin concentrations with or without KDAC inhibitors. The kinetics of histone acetylation and deacetylation will be correlated to changes in gene expression of both pro- and anti-inflammatory cytokines, and to changes in cellular metabolite concentrations.
Non-histone protein-acetylation dynamics will be investigated by differential proteomics/acetylomics using isobaric tandem mass tags (TMT). Upon treatment with high glucose, high lipid and high insulin concentrations with or without KDAC inhibitors, proteins will be isolated from tissue slices at different time points. Differential proteomics/acetylomics experiment will be carried out in collaboration with the group of Hartmut Schlüter (University Medical Center Hamburg-Eppendorf).

Tentative thesis Chapter outline

  1. Current knowledge about the link between energy metabolism and chronic inflammation
  2. New methodology to study histone acetylation dynamics in microfluidic perfusion tissue slice culture
  3. New methodology to study protein acetylation dynamics in microfluidic perfusion tissue slice culture
  4. Effects of nutrient supply, stimulation (insulin) and KDAC inhibitors on histone acetylation dynamics and regulation of pro- and anti-inflammatory gene expression in diabetes
  5. Effects of nutrient supply, stimulation (insulin) and KDAC inhibitors on protein acetylation dynamics and regulation of pro- and anti-inflammatory gene expression in diabetes

How dows this project contribute to Precision Medicine

The proposed research project will make a significant contribution to our understanding of how changes in energy metabolism are linked to chronic inflammation in DMT-II. The research project will elucidate molecular mechanisms at the protein species level that can be ultimately exploited as drug targets for development of therapeutics that can be addressed by KAT and/or KDAC inhibitors. Responders/Non-Responders based on changes in acetylation patterns, which may serve as novel, mechanism-based biomarkers.

Possible links to other ProminenT domains

The proposed research will elucidate disease mechanisms linking aberrant energy metabolism to chronic inflammation in DMT-II by assessing the balance between protein acetylation and deacetylation protein. The proposed research will identify key protein species that may ultimately serve as starting points for the development of future therapeutics in conjunction with diagnostic assays. The project is therefore linked to the ProminenT domains “drug development” and “drug application” via the domain ''disease mechanisms''.

 

Project 1.4 Immune signatures in chronic inflammatory diseases: focus on Interleukin-6  

Common immune signatures in chronic inflammatory diseases for prediction of disease progression. Focus on Interleukin-6

Name Prominent PI / Promotor: E.Brouwer / A.M.H.Boots*, H. Lambers Heerspink , P.Heeringa
Name Co-Promotores: * E.Brouwer will be first promotor when she is professor, W.H. Abdulahad .
Other project members: K.S.M van der Geest, Y van Sleen

Background

Chronic inflammation is a hallmark of systemic autoimmune diseases like Giant cell arteritis (GCA) and rheumatoid arthritis (RA) and systemic inflammatory diseases like type 2 diabetes. For both systemic autoimmunity and systemic inflammation, advancing age is considered a risk factor which is associated with a chronic smoldering inflammatory state known as inflamm-aging. Hallmarks of Inflamm-aging are increased serum levels of inflammatory cytokines and acute phase proteins. One of the key players in chronic inflammation and inflamm-aging is Interleukin 6 (IL-6). Dysregulation of the IL-6 axis plays a central role in GCA, RA and type 2 diabetes. Interleukin 6 (IL-6) is a pleiotropic cytokine involved in many biologic functions that affect the immune system, the vasculature and organs. In general, classic IL-6 signaling via the transmembrane IL-6 receptor (IL- 6R) is thought to be responsible for the anti-inflammatory properties of IL-6, whereas trans-signaling via the trans-signaling pathway upon binding to the soluble form of IL-6R (sIL-6R) is responsible for the proinflammatory actions of IL-6.
It is generally accepted that inflamm-ageing of both the immune system and the vascular wall of large vessels are triggers for disease development in GCA and possibly also in type 2 diabetes.
Vascular dendritic cells and CD4+ T cells play an important role in the initiation of vessel inflammation and in granuloma formation. Moreover, two dominant pro-inflammatory cytokine clusters have been linked to disease activity in GCA: the Interleukin(IL)-12-IFNγ axis promoting T helper (h)1 responses and the IL-6-IL-17 axis promoting Th17 responses. Recently, data from the GiACTA trial demonstrated that Tocilizumab, a blocker of the IL-6 Receptor (IL-6R; CD126), is effective at achieving sustained, glucocorticoid-free remission in GCA patients. The cellular mechanisms, however, by which IL-6 drives the disease remain to be resolved.

Overall aim

The overarching hypothesis of the current proposal is that dysregulation of interleukin 6 (IL-6) axis is a common mechanism responsible for the individual variety in disease course and treatment response in GCA, RA and type 2 diabetes.
The current proposal aims to establish the consequences of chronic IL-6 exposure on survival, expansion and commitment of especially naïve CD4+ T cells towards autoimmune effector cells in the development of GCA. In addition, the consequences of circulating (physiological) levels of IL-6 on regulatory T cells will be assessed in GCA, RA and type 2 diabetes.

Methods

In order to explore the modulating role of the IL-6 axis on CD4+ T cell subsets we will investigate total CD4 and CD4 subsets, examine expression of IL-6R and gp130 on CD4+ T cells and subsets and assess IL-6, soluble IL-6R and soluble gp130 (inhibitor of IL-6 signaling) in sera of the participating cohorts. We will assess the effect of both the classical IL-6R signaling and IL-6 trans signaling pathways (IL-6+sIL-6R) on naïve T cell activation, proliferation and Th1/Th17 skewing potential and on Treg dysregulation.
Bio repositories and patient cohorts
GCA, healthy control and RA cohort
Unique cohorts of newly-diagnosed, untreated GCA patients have been built at the UMCG Vasculitis Expertise Centre. From these patients serial blood samples/ peripheral blood mononuclear cells PBMCs at the time of diagnosis and at regular intervals at follow up have been prospectively collected and bio-banked (total cohort contains more than 100 patients and inclusion and follow up is ongoing) . Treatment is done according to a fixed protocol including tocilizumab which has recently been registered for GCA.
Our SENEX cohort including around 100 aged individuals serves as a healthy, age-matched control group.
For RA we have a unique follow up cohort of patients at risk for RA defined as autoantibody ( ACPA+ and/or RF+) positive patients without arthritis. During follow up 40% of RA risk patients developed RA. We also have stored PBMC samples from seropositive RA (ACPA+ and/or RF+) patients and
recently diagnosed seronegative ( ACPA- and RF-) RA patients (more than 150 patients in total).
Diabetes cohort
In the second part of this project the discovered inflammatory signatures will be measured in patients with type 2 diabetes. We manage a large observational study of patients with type 2 diabetes in primary care in Groningen. A total of 903 patients are participating in this study. Blood and urine samples are collected at baseline and at annual visits. In a sub group of patients also PBMCs will be collected. The follow-up duration ranges from 3 to 6 years. This cohort provides a unique opportunity to translate the inflammation signatures from GCA and RA cohorts to diabetes and to test to what extend they are common in both chronic conditions.

Tentative thesis Chapter outline

  1. Evidence for elevated IL-6 signaling in CD4+ T cell differentiation subsets (with emphasis on naïve CD4+ T cells) in GCA/PMR?
  2. What is the contribution of IL-6 to naïve CD4 T cell survival, expansion and differentiation?
  3. What is the effect of chronic IL-6 stimulation on Treg functionality?
  4. Can CD4+ T cell IL-6 signatures help stratify GCA patients for personalized medicine?
  5. Is there a differential treatment response between GCA patients with high IL-6 levels versus those with low levels?
  6. Do CD4+ T cell IL-6 signatures translate to type 2 diabetes and can then be used to select subgroup of patients with a pro-inflammatory phenotype more likely to develop atherosclerotic / vascular disease?

How dows this project contribute to Precision Medicine

A better understanding of the factors governing CD4 T cell driven autoimmune diseases is highly awaited as it is likely to provide a rational for novel targets of treatment. Interleukin 6 (IL-6)  being a pleiotropic cytokine  is hypothesized to be responsible for the individual  variety in early disease course and early  treatment response in  GCA, RA and type 2 diabetes. Insight into the individual response within the different disease backgrounds will contribute to  identifying serum biomarkers reflecting the immune signatures in these chronic inflammatory diseases which will ultimately form the foundation for stratification of patients for personalized medicine.

Possible links to other ProminenT domains

  • Biologicals targeting immune subsets and chemical drugs targeting signaling pathways can be designed based on insights into common immune signatures in chronic inflammatory diseases for prediction of disease progression.
  • Also especially in RA there is a wealth of drugs targeting specific pathways which can be used and translated to GCA and type 2 diabetes
  • Lastly most drugs still come at a price of immune suppression and subsequent vulnerability to infectious diseases. Creating an urgent need for precision medicine.
  • The project links to various other research projects within ProminenT. Bischoff investigates chronic inflammatory phenotypes in obese and diabetic individuals and aims to develop new methodologies to measure inflammatory metabolites. We will discuss with Bischoff if we can use his methodology in our cohorts as well. Poelarends will probably focus on inflammatory drug development and it may be of interest to discuss if the drugs that he will synthesize affect IL-6 pathways and apply to GCA/RA/Diabetes cohorts as well.
Project 1.5 Microfluidic endothelial disease models for precision medicine  

Microfluidic endothelial disease models for precision medicine

Name Prominent PI / Promotor: Sabeth Verpoorte
Name Co-Promotor: Hiddo Lambers Heerspink
Other project members: Gerrit Poelarends, Alex Dömling

Background

The endothelium is a thin layer of cells lining all blood and lymphatic vessels in the body. This dynamic layer controls to a large extent which compounds cross over from the blood in the circulatory system into tissue. The barrier function is dictated largely by tight junctions formed between cells. However, in many regions the endothelium is fenestrated – that is, contains tiny holes with diameters on the order of 60 to 80 nm – in order to accommodate the passage of water and proteins. Blood vessels in the glomerulus, for example, are characterized by a high degree of fenestration. A polysaccharide gel layer known as the glycocalyx covers the luminal surface of the endothelium and acts as a filtration barrier in fenestrated regions to control transport of proteins such as albumin out of blood vessels.

The role of the endothelium extends beyond acting just as a barrier, and vascular endothelial cells are actively involved in almost all disease pathophysiologies [Aird, 2007]. The subject of this proposal is related to endothelial damage brought on as a result of persistent activation of the endothelium by cardiovascular risk factors such as diabetes mellitus [Rabelink 2015]. Glycocalyx function in patients with diabetes, obesity, or chronic kidney disease is diminished, leading to enhanced albumin excretion from blood vessels, which in turn drives vascular inflammation and progressive organ function loss. Experimental studies have suggested that glycocalyx dysfunction is reversible, rendering the glycocalyx a promising therapeutic target.

We propose to develop new in vitro models for the glycocalyx to better be able to investigate this component of the vascular system as a target for potential therapeutic intervention in renal and cardiovascular disease. These models will be based on microfluidics, as this technology enables ultra-small-volume liquid handling in the femtoliter to microliter range in tiny channels having effective diameters from 1 to 1000 micrometers [Whitesides 2006]. Micofluidics has been applied to a multitude of life-science research questions, including those involving biochemical reactions and bioanalysis in the absence of cells (cell-free) [Mross 2015], as well as microperfusion cell culture (e.g. organs-on-a-chip). Organs-on-a-chip are microfabricated devices incorporating microchambers in which cells or tissue are cultured under continuous perfusion and other biochemical and physical stimuli to simulate tissue or organ physiology [Bhatia 2014]. These devices, currently under development for many different organs, enable the engineering of in vivo-like cellular microenvironments. It is also possible to implement dynamic experimental conditions to mimic the temporal changes to these microenvironments experienced by tissue in vivo. Microfluidic devices designed for both cell-free and cell-culture applications can provide real-time insight into physiological processes under different conditions, and the opportunity to test the efficacy of pharmacological interventions under varying circumstances.

Overall aim

In this project, we propose to develop microfluidic models for the glycocalyx with or without endothelial cells present to investigate the effects of glycocalyx degradation on the development of albuminuria and subsequent progression of renal and cardiovascular disease. We envision first the realization of a robust, cell-free model of the gel layer making up the glycocalyx. The degradation of this layer under different circumstances and resulting changes to glycocalyx permeability will be studied using this model. A second endothelium-on-a-chip model will combine cells with the glycocalyx layer under controlled flow and biochemical conditions. It will incorporate a means to directly probe the effect of albumin leakage (albuminuria) on vascular cell health to improve our understanding of the downstream consequences of glycocalyx dysfunction in diabetes.  

Methods

Microfluidic devices will first be implemented for the development of a robust glyocalyx layer. A second generation of devices will combine the glycocalyx with perfusion culture of endothelial cells (primary human umbilical vein endothelial cells (HUVEC) will be used). Devices will be fabricated in the Verpoorte rapid prototyping lab, which will allow design-to-test cycles of less than a week. A protocol will be developed to introduce hydrogel solutions to microchambers. HUVEC will be cultured in microchannels as per protocols established in the Verpoorte lab for previous HUVEC work. The use of microfluidic devices will enable experiments in which the concentrations of various compounds are changed continuously in controlled spatiotemporal gradients. Cell cultures can be monitored in real time using fluorescence microscopy using specially adapted microscope stages designed in the Verpoorte group. An assay for albumin leakage in an epithelium organ-chip device will most likely be based on fluorescence.

Technical objectives:

1) a) robust, microperfusion glycocalyx model in-chip b) combined endothelial cell-glycocalyx model

2) investigate effect of various parameters on induction of inflammation in combined model to gain
more insight into pathophysiology of albuminuria in diabetic patients

3) test potential drugs targeting the glycocalyx

4) identify the circumstances under which these drugs offer the most benefit

Tentative thesis Chapter outline

  1. Establishment and characterization of a microfluidic, cell-free glycocalyx model
  2. Establishment of a glycocalyx-endthelium organ-on-a-chip (GEOoaC) model
  3. Design of an albumin leakage assay in the GEOoaC
  4. Characteristics of endogenous stimulation of albumin leakage effects. Consideration of glucose concentrations and profiles; inflammation by LPS; influence of cholesterol; etc.
  5. Investigation of the pharmacological inhibition of albumin leakage (inhibition of heparinase, restoration of glycocalyx layer, etc.)

How dows this project contribute to Precision Medicine

Though in vitro drug screening is well established, it is often difficult to accurately extrapolate test results back to possible therapeutic or pathological effects in the human being. This is often due to species differences in drug response (when animal cells or models are used) or human cell models being too simplistic. As a result, a large majority of drug candidates fail in clinical trials. Using microfluidics, nanoliter cellular environments can be created that better mimic the in vivo situation with respect to both chemical and physical (e.g. flow) cues, allowing the behaviour of organs to be better reproduced ex vivo (Verpoorte, Pharmaceutical Analysis), irrespective of whether they are animal- or human-based. These more accurate “organ-on-a-chip” models will enable much earlier preclinical detection of the pharmacological efficacy of new chemical entities. Moreover, microfluidics lends itself very well to the culture and analysis of very small numbers of cells (few thousand) or small amounts (mg) of tissue, which allows the efficient use of often-scarce patient material and opens a route to precision medicine

Possible links to other ProminenT domains

This project in Disease Mechanisms links well to collaborators working in Drug Development, as the model, once developed, can be used for testing potential pharmaceutical compounds. Colleagues Dömling and Poelarends in GRIP are foreseen as collaborators with respect to the drug testing aspects of this project.

Project 1.6 Exocrine-endocrine interactions   

A human cell model to define the role of exocrine-endocrine interactions at the onset of type 1 diabetes

Name Prominent PI / Promotor: Prof. dr. Sven C. D. van IJzendoorn
Name Co-Promotor: dr. Ben N.G. Giepmans
Other project members: Dr. Daniel O. Warmerdam (iPSC facility)

Background

Type 1 diabetes (T1D) is characterized by elevated glucose levels as a consequence of loss of insulin secretion by endocrine beta cells in the islets of Langerhans in the pancreas. In T1D it is well established that the insulin-producing beta cells are destroyed by an auto-immune attack. Early T1D beta cells are characterized by upregulation of HLA class 1, endoplasmic reticulum stress and apoptosis, which can all be evoked within an inflammatory environment (Kracht et al. 2017; Marroqui et al. 2017). However, the trigger for these events remains elusive. Pilot electron microscopy studies revealed the cellular mixing of acinar cells from the exocrine pancreas, which produce digestive enzymes, with the endocrine cells within the islets of Langerhans in the early stages of T1D in both human patient material (manuscript in preparation) and a T1D rat model (Scotuzzi et al. 2017). The cause of this endocrine-exocrine cell mixing is unknown. Moreover, T1D patients have a reduced pancreas organ weight, which cannot be explained only by the loss of islets, since islets comprise just a few percent of the total pancreas. Furthermore, increased immune cell infiltrates have been found in the exocrine pancreas, suggesting a role for the exocrine pancreas in the pathogenesis of T1D.

Interactions between the exocrine and endocrine pancreas in the development of T1D have gained increasing attention, and include reports on immune infiltrates in the exocrine pancreas during T1D and a reduced total pancreatic weight in T1D patients compared to controls (Campbell-Thompson, Rodriguez-Calvo, and Battaglia 2015; Campbell-Thompson et al. 2016). We hypothesize that initial damage to exocrine cells in the pancreas causes the release of (auto)digestive proteins, which subsequently trigger beta cell stress that ultimately result in an autoimmune response that is characteristic for T1D.

Overall aim

The overall aim of the project is to investigate the interactions between exocrine acinar cells and endocrine beta cells in the early pathogenesis of type 1 diabetes and the potential role in triggering T1D. 

Methods

Advanced electron microscopy (EM), color EM that will allow high throughput analysis of Islet morphology up to the ultrastructural level and that allows discrimination of cellular hormones versus enzymes will be applied as described before (Scotuzzi, 2017; Ravelli 2013).

A patient-specific human pancreas cell culture model for the study of exocrine-endocrine interactions will be developed. For this Induced pluripotent stem cells (iPSC) will be generated from fresh urine-derived cells donated by healthy control individuals and (HLA genotyped) patients diagnosed with Type 1 diabetes. The iPSC will be differentiated to exocrine acinar cells and endocrine beta cells according to published protocols. 3D co-culture systems will be used to study exocrine-endocrine cell interactions. The responses of patient- or control-derived beta cells on exocrine lysate will be studied using cytokine and chemokine production (ELISA, RT-PCR).

Tentative thesis Chapter outline

  1. Introduction to thesis
  2. Introduction to type 1 diabetes and potential triggers
  3. Mechanisms of exocrine-endocrine cell mixing in type 1 diabetes
  4. Development of a patient-specific 3D pancreatic cell culture system for the study of exocrine-endocrine interactions
  5. The interaction between exocrine and endocrine cells in the pathogenesis of type 1 diabetes
  6. The role of type 1 diabetes patient-specific HLA haplotypes in exocrine-endocrine cell interactions. 
  7. Discussion and future perspectives

How dows this project contribute to Precision Medicine

Patient-derived induced pluripotent stem cells and thereof derived exocrine and endocrine pancreatic cells allow the investigation of patient-specific disease pathogenesis and responsiveness to therapeutic interventions.  

Possible links to other ProminenT domains

Patient-derived induced pluripotent stem cells and thereof derived exocrine and endocrine pancreatic cells allow:

  • The investigation of patient-specific disease pathogenesis (pilar Disease Mechanisms)
  • The investigation of responsiveness to (potential) therapeutic interventions (pillars Drug development and Drug application)

Drug Development

Project 2.1 Design and synthesis of sulfotransferase inhibitors for diabetes  

Design and synthesis of sulfotransferase inhibitors for diabetes

Name Prominent PI / Promotor: Alexander Dömling
Name Co-Promotor: HJ Lambers Heerspink
Other project members: Matthew Groves, Gerrit Poelarends, Dick v.Zeeuw

Background

To retain albumin, one of the main circulating blood proteins, in the vasculature the vessel is aligned with a polysaccharide layer called the glycocalyx that repels albumin. Small glycocalyx defects lead to albumin leakage into tissues, which in turn leads to macrophage recruitment, chronic inflammation (to clear the albumin) and ultimately loss of organ function. In diabetes (an ever-growing global disease) albumin leakage occurs often,is associated with glycocalyx loss, and with high (50%) morbidity and mortality (renal and cardiovascular). Drugs that lower albuminuria protect the patient from renal (and cardiovascular) disease progression. However, the efficacy to slow disease progression is low and novel interventions are highly desired. Drugs that target the glycocalyx repair are not available and if successful would be a huge step forward in global diabetes health care. Glycocalyx breakdown is associated with heparanase I over-activity. Thus inhibition of heparanase would be a potential goal for drug development. In addition to diabetes, heparanase inhibition is interesting in other areas such as cancer in particular to mitigate the course of inflammation induced cancer.
In the Dömling laboratory different tools have been developed for the rational structure based design of drugs against biochemical protein targets. These include the pharmacophore suite AnchorQuery™, MCR chemistries, a ~700-compound fragment library, a >2000 compound screening collection, a large collection of unique building blocks, protein crystallization (with monthly 24h access to major European synchrotrons) and a manifold of biophysical screening methods.
The interplay of these technologies was very valuable in the past in the Dömling laboratory and has resulted in small molecule leads for multiple protein targets including PD1-PDL1, p53-MDM2, p53-MDMX, arginase, caspase-1, IL17, CD44, endothiapepsin, RAS, 15-lipoxygenase-1, telomerase, Hsp-70, etc. Of special value in the proposed project will be the AnchorQuery™ tool. AnchorQuery™ is a specialized pharmacophore search technology that brings interactive virtual screening of novel protein-protein inhibitors to the desktop.

Overall aim

The hypothesis is that heparanase inhibitors will have profound effects on the course of the diabetes disease.
Therefore, we propose three aims:

  1. Expression, purification, crystallization, crystal structure analysis and screening method establishment of human heparanase;
  2. Design, synthesis and optimization of novel heparanase inhibitors;
  3. Biophysical and in vitro characterization of the inhibitors.

Methods

Aim 1: Expression, purification, crystallization, crystal structure analysis and screening method establishment of human heparanase:
For the envisioned drug discovery program recombinant heparanase is needed. In the fully equipped protein laboratory (Groves/Domling) human heparanase will be expressed, purified and crystallized and its apo structure will be solved at a European synchrotron site. The resulting structure will be compared with already existing structures (5E8M, 5E98, 5E9B, 5E9C, 5L9Y, 5L9Z, 5E97, 5LA4, 5LA7). Different constructs (e.g. cysteine mutants, His-tag) will be generated to be suitable for labeling for MST experiments. Next an in-house fragment library will be soaked into apo heparanase crystals and their structure elucidated. In parallel an inhouse library of MCR compounds will be screened for affinity to heparanase. Binders will be cocrystallized and structures will be obtained. A biochemical activity based assay will be established (collaboration Polarends).

Aim 2: Design, synthesis and optimization of novel heparanase inhibitors:
The challenge of the project lies in the discovery and design of drug-like compounds despite the highly hydrophobic heparine binding pocket of heparanase. So far only sugar derived molecule have been described as heparanase inhibitors which proved to be challenging during development. We will use the currently available and our own crystal structures to build pharmacophore models. These will be screened with the inhouse technology platform AnchorQuery™.

Aim 3: Biophysical and in vitro characterization of the inhibitors.
The resulting virtual compounds will be resynthesized and profiled biochemically and biophysically. If they exhibit a suitable profile they will be optimized for affinity and biochemical activity, while keeping a close eye on drug-like properties.

Tentative thesis Chapter outline

  1. Review on disease and structural biology of heparanase
  2. Discovery of heparanase inhibitors
  3. Biophysical and structural characterization of heparanase inhibitors
  4. Optimisation of heparanase inhibitors: form H-2-L
  5. Biological evaluation of heparanase inhibitors
  6. Summary and outlook

How does this project contribute to Precision Medicine

This project deals with the synthesis of new drugs that inhibit heparanase I. Ultimately, these drugs would only be used in patients in whom heparanase is overexpressed. The PIs / promotors of the project will continue to develop specific urinary heparanase assays for clinical use. To this end, large data and sample repositories of patients with type 2 diabetes can be used which are managed by Lambers Heerspink and De Zeeuw. This offers the opportunity that once a heparanase inhibitor is successfully synthesized in this project, the inhibitor can be targeted to the right patient population.

Possible links to other ProminenT domains

This project links to disease mechanisms and in particular to the work carried out by Sabeth Verpoorte who will develop a disease on a chip model for dysfunction of the endothelial surface layer and albumin permeability. During the course of the project, the drugs synthesized in this program could potentially be tested on the chip developed by Verpoorte et.al.

Project 2.2 Development of antibacterial drugs for diabetic infections  

Development of antibacterial drugs for diabetic infections

Name Prominent PI / Promotor: Prof.dr. G.J. Poelarends
Name Co-Promotor: Prof.dr. F.J. Dekker
Other project members: P.G. Tepper (technician), Haigen Fu (PhD student)

Background

The immunocompromised state and diabetic complications lead to high risk for infections in diabetic patients. Also, higher antibiotic resistance rates in diabetic patients compared to those without diabetes have been reported, further complicating the treatment of the infectious disease. The β-lactam antibiotics, including penicillins, cephalosporins, carbapenems and monobactams, are the most commonly prescribed drugs for the treatment of infections caused by Gram-negative bacteria. However, the efficacy of β-lactam antibiotics is severely impaired by several bacterial resistance mechanisms, most importantly hydrolytic inactivation by β-lactamases. Based on the β-lactam ring-opening mechanism, β-lactamases can be subdivided into two major groups. Serine β-lactamases (SBLs) use an active-site serine residue to covalently attack the β-lactam ring resulting in an inactive ring-opened product. Fortunately, several SBL inhibitors (such as clavulanic acid, sulbactam, tazobactam and avibactam) are clinically available as codrugs that are coadministered with β-lactam antibiotics to overcome SBL-related resistance. Metallo-β-lactamases (MBLs), typified by recently emerged New Delhi metallo-β-lactamase-1 (NDM-1), employ an active-site zinc-stabilized OH anion that acts as a nucleophile in β-lactam ring opening. MBL-producing pathogens are resistant to virtually all clinically used β-lactam antibiotics, including the last-resort antibiotics carbapenems. In fact, the first NDM-1-producing pathogen has been found in a diabetic patient. Despite the rapid spread of MBL-producing bacteria, which is a major threat to public health, no inhibitors of NDM-1 or other MBLs are clinically approved so far. Thus, there is a great need to develop selective MBL inhibitors as clinically relevant codrugs to restore the activity of β-lactam antibiotics.

The fungal natural product aspergillomarasmine A (AMA) has recently been identified as a selective and potent NDM-1 inhibitor and a promising codrug candidate both in vitro and in vivo (King et al., 2014, Nature 510:503-506). We have recently developed a simple biocatalytic method to prepare AMA and related aminopolycarboxylic acids (Poelarends et al., manuscript under review in Nature Catalysis). In this project, we will synthesize various aminopolycarboxylic acids, including novel analogues of AMA, as well as smart pro-drug versions of these compounds, and explore their bioactivity, target selectivity and therapeutic potential. As such, the proposed project will contribute to the development of new chemotherapeutic strategies to fight antibiotic-resistant pathogens, which are more prevalent in individuals with diabetes.

Overall aim

The overall aim of this project is to design and develop new and smart antibacterial drugs, based on existing and novel aminopolycarboxylic acid inhibitors of metallo-β-lactamases, to combat antibiotic-resistant pathogens, which are frequently encountered in diabetic patients

Methods

In this project, the following methods are used:

  1. Chemical and chemoenzymatic synthesis of new aminopolycarboxylic acids, including analogues of the natural products Aspergillomarasmine A & B and ethylenediamine-N,N-discuccinic acid, as well as smart and potentially selective pro-drug versions of these compounds
  2. Characterization (structural identity and stereochemistry) of the newly synthesized compounds by mass spectrometry, NMR spectroscopy, chiral HPLC, and optical rotation measurements
  3. Evaluation of the inhibitory properties of the newly synthesized aminopolycarboxylic acids and corresponding pro-drugs against NDM-1, VIM-2 and other metallo-β-lactamases, as well as possible off-target metallo-enzymes, using in vitro dose-dependent enzyme inhibition assays and inactivation kinetics
  4. The most potent and selective aminopolycarboxylic acids and pro-drugs will be studied for their ability to rescue the activity of existing β-lactam antibiotics in NDM-1- or VIM-2-positive clinical isolates
  5. Selected compounds will be tested for their ability to restore the activity of β-lactam antibiotics (e.g. meropenem) in worms, insect larvea and/or mice infected with a NDM-1 or VIM-2 expressing bacterial strain
  6. During the project we will continuously assess opportunities for patent applications as well as for collaborations with academic and industrial partners to further develop the products

Please note that Prof. Poelarends is also participating in the Centre for Sustainable Antimicrobials (CeSAM), which comprises advanced facilities for fundamental research to develop novel antibiotics and therapeutic concepts to fight resistant bacterial strains as well as high-throughput (animal) testing and adequate patient-screening facilities. This existing network of collaborators will further support the proposed project.

Tentative thesis Chapter outline

  1. Recent advances in the synthesis of the fungal natural products Aspergillomarasmine A and B and related aminopolycarboxylic acids
  2. Chemoenzymatic synthesis and evaluation of novel aminopolycarboxylic acids as clinically relevant metallo-β-lactamase inhibitors
  3. Chemical and enzymatic synthesis of aminopolycarboxylic acid-based metallo-β-lactamase inhibitors: Exploration of structure-activity relationships
  4. Development of aminopolycarboxylic acid-based pro-drugs: Exploration of activity, target selectivity and therapeutic potential
  5. Recent progress in the development of β-lactamase inhibitors

How does this project contribute to Precision Medicine

The immunocompromised state and diabetic complications lead to high risk for infections in diabetic patients. Indeed, infectious diseases are more prevelant and often more serious in diabetic patients, which potentially increases their morbimortality. Furthermore, higher antibiotic resistance rates in diabetic patients compared to those without diabetes have been reported, further complicating the treatment of the infectious disease. Due to common and recurrent infections, diabetic patients need more antibiotic treatments, which may further select for the presence and increase in antibiotic-resistant pathogens. New chemotherapeutic strategies to fight antibiotic-resistant pathogens, particularly in patients with diabetes, are therefore urgently needed. The proposed project contributes to precision medicine because it aims to develop a clinical inhibitor of metallo-β-lactamases, which could be used as a codrug to rescue or potentiate β-lactam antibiotics in combination therapies. In this way resistance can be overcome, allowing more efficient treatment of infectious diseases in diabetic as well as other patients.

Possible links to other ProminenT domains

The proposed project fits in the Drug Development domain and will contribute to the development of new chemotherapeutic strategies to fight antibiotic-resistant pathogens, which are more prevalent in individuals with diabetes. As such, it may also become linked to drug application.

Project 2.3 In-vitro disease modeling in diabetic cardiomyopathy  

Diabetes In Heart Failure with a Reduced and Preserved Ejection Fraction

Name Prominent PI / Promotor: Peter van der Meer
Name Co-Promotor: Jasper Tromp

Background

Diabetes mellitus type II (DM) is an important risk factor for developing heart failure (HF). Within the HF syndrome, we distinguish HF patients with reduced (HFrEF) and preserved (HFpEF) ejection fraction. Overall, DM is equally prevalent among patients with HFrEF and HFpEF. Yet, there is evidence that DM affects cardiac geometry differently in HFrEF and HFpEF. Indeed, diabetic cardiomyopathy (DMCMP) with HFrEF is associated with more eccentric cardiac hypertrophy, while DMCMP with HFpEF is associated with more concentric cardiac hypertrophy.
It has become increasingly clear that also on a cellular level, DMCMP presents as two distinct diseases with a HFrEF and HFpEF phenotype. Early evidence suggests that in the DMCMP HFpEF phenotype collagen and advanced glycation end-products (AGEs) deposition are increased. In contrast, cardiomyocytes in patients with the DMCMP HFrEF phenotype show an overall loss of sarcomeres.
Unfortunately, limited information is available on differences between DMCMP with HFrEF and HFpEF. Elucidating these differences will aid in our understanding of the underlying pathophysiological differences of DMCMP with HFrEF and HFpEF and provide for novel treatment targets. Additionally, it has become increasingly clear that particular DM medications might have beneficial effects in patients with HF. Results from the EMPA-REG trial showed that sGLT2 inhibitors might possible aid in reducing the number of HF hospital readmissions. Novel targets on DMCMP patients with HFrEF and HFpEF will aid in identifying HF patients potentially benefitting from existing treatment for DM.

Overall aim

The overall aim of this proposal is to elucidate the differences between patients with DMCMP presenting with a HFrEF or HFpEF phenotype. We will then investigate mechanistic implications of patient specific treatment targets in in-vitro and in-vivo models. 

Methods

To approach the above stated aim, we propose four sub aims using a translational approach:

  1. We will perform a full blood proteomics screening on age- and sex- matched patients with DMCMP and HFrEF (N=20)/HFpEF(N=20) and compared to age- and sex- matched healthy (absence of coronary artery disease, hypertension, DM and HF) controls (N=20). We will validate our findings in a larger cohort of healthy control patients (N=200) vs. DMCMP patients with HFrEF (N=100) and HFpEF (N=100). 
  2. Using the top targets found in (1), we will examine the effect of patient specific targets on cardiomyocyte phenotype and function using an in-vitro approach. Here, we will employ human embryonic stem cell derived cardiomyocytes (hESC-CM) as a model using recombinant proteins of targets found in (1) and/or antagonists of targets found in (1) for HFrEF and HFpEF). Following, we will phenotype the hESC-CM and assess response to stimulants (e.g. High glucose; cardiac stretch; oxidative stress; hypoxia) on both RNA (RT-qPCR) and Protein (Western Blot) level using a panel of genes and proteins known to be associated in eccentric and concentric hypertrophy, and extracellular matrix and sarcomere structure. Additionally, we will assess the effect on metabolism of hESC-CM using the Seahorse assay. Lastly, we will study contractility and stiffness of the targeted hESC-CM using an in-house assay. 
  3. Following our mechanistic studies in (2), we will perform an in-vivo study of the most promising individual targets in HFrEF and HFpEF from (1) & (2). Here, we will produce total knock-out and overexpression mouse models and phenotype mice using echocardiography and study of cardiac tissue. In case of a less-severe phenotype, we will investigate the role of said targets in a pathological setting using a trans-aortic constriction (TAC) model to produce LV dysfunction. 
  4. Lastly, we will investigate whether targets found in (1), (2) and (3), are associated with new-onset HFrEF and HFpEF. We will measure targets found in individuals from the LIFELINES cohort and investigate a potential predictive association with new-onset HFrEF and HFpEF. Furthermore, we will measure said targets in HF patients with HFrEF and HFpEF (with and without DM) from the BIOSTAT-CHF cohort and investigate whether these novel targets will predict outcomes in these patients. 

Tentative thesis Chapter outline

  1. Target identification and validation in patients with DM and HFrEF and HFpEF
  2. In-vitro approach to most important targets found in Chapter 1 using hESC derived cardiomyocyte models
  3. In-vivo mouse study most promising target in HFrEF and HFpEF
  4. Measurement of targets found in patients with HF and association with outcome. 

How does this project contribute to Precision Medicine

Creating a better understanding of DMCMP with HFrEF and HFpEF will lead to potential novel patient specific treatment targets for these patients. Furthermore, by measuring novel targets in patients with HF, we will identify patients benefiting from a personalized medicine approach (sub aim 4) using existing DM treatments. 

Possible links to other ProminenT domains

This proposal is linked to the following ProminenT domains:

  • Disease mechanisms (sub aim [1], [2] and [3])
  • Drug development (sub aim [1], [2] and [3])
  • Drug application (sub aim [1] and [4])

Drug Registration and Evaluation

Project 3.1 Personalised diabetes treatment decisions in primary care  

Personalised diabetes treatment decisions in primary care – current practice and needs for change

Name Prominent PI / Promotor: Petra Denig

Background

Current clinical guidelines allow already for some degree of personalized treatment for people with type 2 diabetes (T2D). Factors such as age and comorbidity as well as patient preferences and goals should be taken into account when prescribing glucose-regulating and cardiovascular medication treatment. Recent cross-sectional research indicates that this personalised approach is not yet common practice, and that undertreatment occurs in younger patients (<50 years) and overtreatment occurs in older patients (>70 years). This apparent lack of implementing personalized treatment needs further study in order to develop support strategies for both healthcare providers and patients to adopt more personalised treatment decision making. Since the majority of people with type 2 diabetes treatment are managed in primary care, this project will focus on the general practice setting where both general practitioners and practice support staff are responsible for routine diabetes care.

Questions may include:

  • What was the impact of the guideline recommendations regarding personalized treatment on potential under- and overtreatment in specific subgroups of T2D patients?
  • What are individual trends in risk factor control over time for T2D patients in relation to treatment changes (e.g. initiation of novel drugs)?
  • Which patients do or do not receive intensification of treatment when indicated?
  • How can patient preferences be assessed?
  • Which tools are needed to support the process of personalized treatment in T2D from the healthcare provider and the patient perspective?
  • What is the feasibility and impact of implementing these tools in practice?

Overall aim

Develop patient-oriented strategies and tools to support personalized decision making for people with type 2 diabetes.

Methods

This project includes the essential steps for developing a new practice-oriented intervention strategy before it can be implemented in practice: (1) describe the current situation and its determinants, (2) develop a toolbox to support healthcare professionals and patients in making personalized decisions, (3) evaluate the feasibility and potential effects of implementing the toolbox in general practice.

For part 1: cohort studies making use of existing data from the GIANTT cohort.
For part 2: literature/interview/survey studies focusing on healthcare providers and patients in primary care.
For part 3: pilot studies in general practice

Tentative thesis Chapter outline

  1. Changing trends in medication treatment with glucose, blood pressure and lipid lowering agents with respect to relevant patient characteristics (e.g. age, comorbidity, disease markers).
  2. Individual treatment trajectories leading to initiation of novel drug treatments.
  3. Which patients with type 2 diabetes do not receive optimal treatment? 
  4. The design of a tool for assessing patient preferences regarding medication treatment.
  5. The design of an algorithm to support personalized treatment decisions in clinical practice. 
  6. Pilot study with a toolbox for personalized treatment decision making.

How does this project contribute to Precision Medicine

This project will give insight in the current status and needs for implementing patient-oriented personalized treatment decision making in diabetes care, and provide a toolbox to support this implementation process in clinical practice. 

Possible links to other ProminenT domains

Important links can be made within the Drug Regulation Domain and the Drug Application Domain.
This may lead to expanding the patient-level factors that will be studied; developing tools together that can be used for assessing patient preferences; developing algorithms together that can be used to support healthcare professionals in making personalized decisions.

Project 3.2 Incorporating patient preferences in regulatory decisions for antidiabetic drugs  

Personalised diabetes treatment decisions and regulatory decision – making. Current practices and suggestions for change

Name Prominent PI / Promotor: Hans Hillege & Peter Mol
Name Co-Promotor: Douwe Postmus

Background

Treatments for type 2 diabetes (T2D) are approved based on clinical trial data. Favourable and unfavourable drug effects are weighed on a population level, and if the first outweigh the latter a positive ratio of benefit and risk can be assumed and the drug is approved. In diabetes, drugs are approved based on its effects on a validated surrogate endpoint, the biomarker glycosylated haemoglobin: HbA1c, and when there is sufficient reassurance the drugs are not harmful (cardiovascular safety). A thorough assessment of other, on- and off-target, effects, extensive subgroup analyses and available non clinical data further shapes the conditions – labeling and postmarketing surveillance – under which the drug is to be used in daily practice. Regulators translate this drug knowledge to healthcare providers (HCPs) and patients primarily through the Summary of Product Characteristics (SmPC) respectively Patient Information Leaflet (PIL). HCPs then have to translate this population level information to an individual patient. Importantly, reimbursement authorities, HCPs and patients may take different views when compared to regulators and may value drug effects and the ratio of benefits and risks differently.

Various classes of antidiabetes drugs are currently available that have differential drug effects. Current clinical T2D guidelines favour drugs with demonstrated clinical benefit and well known safety profiles, but do allow some degree of personalized treatment. The main goal of antidiabetic treatment is to achieve normo-glycemia without serious high or low blood glucose levels to prevent or delay the onset and progression of diabetic complications. Factors to be considered include efficacy, age, potential side effects, weight gain, comorbidities, hypoglycemia risk, costs and patient preferences (see also ProminenT Denig program). Further, recent research shows that collecting evidence-based information based on individual patient preferences in the context of favourable and unfavourable effects is feasible and useful and could guide to a more patient-centered value judgement of pharmacological agents and thus contribute to a more transparent communication on how the patient views have been incorporated in the regulatory decision making. However, these findings have not been systematically evaluated and translated back into the drug approval process. Moreover, while regulators evaluate carefully drugs’ various effects these are not quantitatively weighted. Finally, how evidence on individualized / personalized treatments should translate back into the regulatory decision-making is largely unknown.

This project aims to study, from a regulatory perspective, how elicited patient preference information can be combined with clinical trial data to estimate the acceptability of various classes of antidiabetes drugs currently available that have differential drug effects.
Questions may include:

  • What are the most relevant treatment related attributes with regards to glucose-regulating medicines’ favourable and unfavourable effects from a patient, HCP and regulator perspective?
  • What is the current available knowledge in terms of comparative efficacy and safety of various antidiabetic medication therapies in patients with Type 2 DM not adequately controlled on stable and optimized metformin monotherapy.
  • What glucose-regulating medicines meet which stakeholder’s preferences?
  • How can known (or new) drug effects be translated more efficiently to patients and HCPs?
  • How can patient preferences and individual patients’ needs be fed back into the regulatory decision-making? And is this needed?

Overall aim

To improve regulatory decision-making for glucose-regulating medicines incorporating patient preferences and individual patient needs.

Methods

This project includes four main parts: 1) evaluation of current practices (review of drug approval decision-making based on publically available data [European Public Assessment Reports] and systematic review of studies into patient preference elicitation of glucose-regulating medication, 2) collecting information about comparative efficacy and safety of various antidiabetic medication therapies in patients with Type 2 DM not adequately controlled on stable and optimized metformin monotherapy 3) applying a model to weigh quantitative comparative drug favourable and unfavourable effect data considering patient preferences, 4) develop drug effects materials (PILs) that meet patient demands for information, based on the preference elicitation (test these in Denig’s program), and 5) feed this information back to the regulatory process.

For part 1: review regulator’s drug dossiers /systematic review of public literature on patient preferences for glucose-regulating medication
For part 2: a list of candidate attributes will be determined by reviewing existing commonly reported study outcomes, patient reported outcome questionnaires, preference literature concerning type 2 DM and the patient information leaflets.
For part 3: SMAA based MCDA will be applied on glucose-regulating effects and patient preferences
For part 4: develop drug facts boxes [Shwartz&Woloshin] for glucose-regulating medicines
For part 5: organize stakeholder conference with regulators (and payers) on how to collect and weigh and drug effects information relevant to individual patients

 

Tentative thesis Chapter outline

  1. Effects that determine regulatory decision-making of antidiabetes drugs.
  2. How do T2DM value their antidiabetes drugs? A systematic review of patient preference elicitation studies.
  3. Measuring up antidiabetes agents, on and off target favourable and adverse effects. 
  4. Ranking antidiabetes drugs utilising a SMAA / MCDA approach.
  5. Drug fact boxes of glucose-regulating medicines. 
  6. A stakeholder conference report on collecting, weighing and presenting drug information that matters to patients.

How does this project contribute to Precision Medicine

This project will give insight in the current regulatory decision-making for diabetes medication and aims to more clearly display differential drug effects important to individual patients. It will feedback to regulators (and HTAs) how information relevant for personalized treatment should be weighted and disclosed to a wider patient and HCP audience. 

Possible links to other ProminenT domains

Important links can be made within the Drug Regulation Domain and the Drug Application Domain.
This may lead to expanding the patient-level factors that will be studied; developing tools together that can be used for assessing patient preferences; developing algorithms together that can be used to support healthcare professionals in making personalized decisions.

Project 3.3 Cost-effectiveness of personalised diabetes treatment  

Cost-effectiveness of personalised diabetes treatment

Name Prominent PI / Promotor: Maarten J Postma
Name Co-Promotor: Job FM van Boven

Background

This project aims at enlightening issues on health economics and precision medicine in diabetes treatment and prevention, with specific reference to reimbursement of new drugs. Notably, diabetes is involved with various risk factors for onset and complications, including genetics (family history), life style and BMI. Additionally, situational issues come in concerning comorbidities, polypharmacy and adherence to therapies. Known biomarkers concern blood sugar levels and Hb1Ac, but also nephropathic complication-related biomarkers such as albumin. Associations of biomarkers, risk factors and situational issues with hard outcomes such as hospitalizations (that can be economically valued) will be investigated using secondary analysis on clinical trials and analysis of real-world data to identify patient subgroups most benefiting or being minimally harmed with targeted therapies.

The project is in the context of a PhD with building blocks being

  1. Reviews on current health economic models in diabetes precision medicine;
  2. Analyses of burden of diabetes comorbidities and non-adherence;
  3. Associations of diabetes comorbidities with clinical & economic outcomes;
  4. Consideration of risk factors in secondary analyses of clinical trials; and
  5. Integration in a health-economic model, potentially building on existing models (such as the CORE-model).

Overall aim

To pave the way for cost-effectiveness analyses of personalized diabetes treatment

Methods

This project includes the essential steps for developing a new health economic diabetes model that is able to assess personalized medicine (1) describe the currently available models and their fitness to assess personalized medicine, (2) assess associations of real-life patient characteristics (non-adherence, comorbidities, biomarkers) with morbidity and mortality, (3) incorporate these personalized patient profiles and develop and validate a new health economic model

For chapters 1-2: literature reviews
For chapter 3-4: GIANTT real-world database analyses
For chapter 5-6: modelling using R and/or Excel

Tentative thesis Chapter outline

  1. State-of-the-art review of current health economic diabetes models: are they with for personalized medicine?
  2. The clinical and economic burden of comorbidities in diabetes
  3. Associations of diabetes comorbidities with hospitalizations and mortality
  4. Associations of non-adherence with hospitalizations and mortality
  5. The design and validation of a health economic diabetes model to assess personalized medicine
  6. Cost-effectiveness analyses of several personalized diabetes treatments

How does this project contribute to Precision Medicine

This project will give insight in the current status of the health economic assessments of personalized diabetes treatment, and provide an optimized model taking into account patient heterogeneity. 

Possible links to other ProminenT domains

Important links can be made within the Drug Regulation Domain and the Drug Application Domain.
This may lead to expanding the patient-level factors that will be assessed in reimbursement decision making; as well as support healthcare professionals in making personalized decisions.

Drug Application

Project 4.1 Integrating multiple effects of single drugs to tailor type 2 diabetes  

A novel approach to improve and tailor treatment of type 2 diabetes: Integrating multiple effects of single drugs

Name Prominent PI / Promotor: Hiddo Lambers Heerspink
Name Co-Promotor: Petra Denig
Other project members: Floriaan Schmidt

Background

Current guidelines of type 2 diabetes management advocate targeting multiple renal and cardiovascular risk markers, including HbA1c, blood pressure, albuminuria and lipid levels, in order to mitigate long-term health risks. Despite targeting these risk markers with drug treatment, patients with type 2 diabetes remain at high residual risk for CV and renal disease.

Part of this high residual risk is due to the fact that not all patients beneficially respond to the drugs that target the abovementioned cardiovascular risk factors. Indeed, previous studies from our group have shown that some patients benefit from a given treatment, but many others do not. In addition to this between individual variability in drug response in a single risk factor, we have shown that a single drug affects many more risk markers than the one intended. For example, the antihypertensive angiotensin receptor blocker (ARB) losartan decreases uric acid, hemoglobin, albuminuria and increases serum potassium. Some of these effects may be beneficial for renal and cardiovascular outcomes, such as a reduction in blood pressure, albuminuria, or uric acid. Yet, other effects, such as an increase in potassium may increase renal and cardiovascular risk. We have shown that the multiple effects of a drug on multiple renal and cardiovascular risk factors vary within individuals indicating that some patients show a reduction in blood pressure in response to ARB treatment but no change in albuminuria or vice versa.

Given the large variation in drug response in multiple cardiovascular risk factors one should combine the short term effects of a single drug in each individual to obtain a more accurate estimate of the ultimate drug effect per patient. We therefore developed an algorithm, a so-called multiple risk Parameter Response Efficacy (PRE) score, to predict the potential long renal effect of a drug based on the composite of short term drug effects in individual patients. In previous work we showed that integrating the short-term changes in all measured cardiovascular risk markers following ARB treatment gave a much better prediction who would benefit from the ARB losartan compared to when based on blood pressure alone (the on-target risk factor).

Clinical practice currently lacks an holistic and patient-centered treatment approach, which can be attributed to several reasons:

  1. Current treatment guideline recommendations focus mostly on the treatment of single risk factors and give little support on how to provide personalized care.
  2. There is insufficient guidance to make an integrated assessment of the renal and cardiovascular risk changes of patients after start of an intervention.
  3. Translation of risk factors to individual risk scores and subsequent treatment changes is difficult without (computer-based) decision support systems.
  4. Involving patients and making patient-centered decisions is difficult without decision support aids.

Thus, a novel tool that is both accurate in predicting long-term outcomes and feasible to use in clinical practice to optimize pharmacotherapy remains urgently needed in order to maximize renal and cardiovascular prognosis of patients with type 2 diabetes.

Overall aim

The aim of this project is to design and test a decision support system based on a personalized multiple parameter efficacy (PRE) score that translates the short term response in multiple renal and cardiovascular risk markers into a predicted long term drug effect in type 2 diabetes care. Ultimately this should lead to a more personalized treatment approach and improved outcomes.

In this project a prospective study will be conducted to determine if the PRE score website (web app) can improve risk marker management for individual patients with type 2 diabetes. The presentation of the PRE score will be tailored to be informative for both primary care physicians as well as patients, and will include the following:

  1. Personalized risk information by integrating drug effects on all relevant risk markers and translating this to long-term risk changes of renal and cardiovascular outcomes.
  2. Tailor-made treatment guidance for individual patients indicating which guideline recommended targets (e.g. cholesterol) are not achieved, and recommendations for additional therapy, including possible risk reductions attained by optimal treatment.

Methods

Firstly, we will develop a version of the PRE score that is suitable for use in primary care, with the ability to predict long-term renal and cardiovascular outcomes for a diverse type 2 diabetes population. The PRE score will be tested and validated in independent datasets including the GIANTT study. Secondly, we will determine the feasibility of using the score in practice by performing qualitative and semi-quantitative studies. Thirdly, the PRE score will be pilot tested in primary care practice in a pragmatic trial design (comparing PRE score decision support guided therapy vs. standard of care).

Ad 1: The study will be performed by using the already prepared GIANTT database that consists of data from over 25.000 patients with type 2 diabetes that are managed in primary care. The database contains individualized patient data on demographics, drug treatment, risk marker responses and clinical outcomes during long-term follow-up. Data from the following risk markers will be included in the PRE score: systolic blood pressure, albuminuria, potassium, cholesterol, body weight, HbA1c. From the databases we will use all patients that initiated either treatment with ACEi/ARBs, metformin, or statins starting from 2007 onwards. We will measure change in the markers and assess if we are more accurate in predicting outcomes than single marker changes.
Ad 2: In this study we will determine the acceptability and usefulness of the validated PRE score in combination with treatment decision support for primary care. This will be done by performing a mixed method approach of semi-qualitative and quantitative methods in small groups of primary care physicians and practice assistants.
Ad 3: A pragmatic controlled pilot study will be performed comparing the PRE score with care as usual. The population consists of primary care practices in the northern part of the Netherlands, with possible extension to other parts of the Netherlands. Outcomes are tailoring of treatment and risk factor control.

Tentative thesis Chapter outline

  1. Introduction and review on decision support tools and personalized medicine
  2. GIANTT retrospective analysis on PRE score prediction in clinical practice
  3. Focus report on applicability and usefulness of PRE score in practice
  4. Results of pragmatic controlled study on PRE score
  5. TBD
  6. Conclusions and future perspective

How does this project contribute to Precision Medicine

This project addresses Precision Medicine in its purest form. It deals with the variable drug effects in individual patients, aims to optimize treatment based on the individual patient response and integrates a patient centered approach taking involving the patient’s voice in decision making.

The project thus addresses the 4-Ps (Prevention, Prediction, Personalized and Patient) of precision medicine:

  1. Prevention: the project focuses on prevention of cardiovascular disease
  2. Prediction: the project aims to validate a novel prediction algorithm in clinical practice
  3. Personalized: The algorithm takes into account the individual drug response
  4. Patient: The decision support tool is patient centered

Possible links to other ProminenT domains

This project links to the research program intended by Prof. P Denig focusing on how to involve patient and physician in implementing personalized therapy approaches. In addition, it also links to the project of Prof. Hillege and Mol which is focused on how to interpret drug efficacy for regulatory approaches based on multiple effects of single drugs. 

Project 4.2 Biomarker signatures for disease and drug response prediction  

Biomarker signatures for disease and drug response prediction

Name Prominent PI / Promotor: Prof. dr. S.J.L. Bakker
Name Co-Promotor: Prof. dr. H.J. Lambers Heerspink / Prof. G.J. Navis
Other project members: To be determined.

Background

Diabetes is a devastating disease, with numerous complications, due to its adverse effects on the micro- and macrovascular system. This is not only true in the general population, but particularly so in patient populations, where diabetes is often not only more frequent, but – likely due to impaired homeostatic capacity – often also gives rise to more severe complications. We hypothesize that patient populations may serve as a good model for the general population, with stronger relations between biomarkers and higher numbers and in time more nearby end-points,  leading to higher efficiency of proof-of-principle studies for biomarkers. Omic-technologies, like NMR-metabolimics provide large amounts of data of which the predictive value for development of diabetes and for complications of diabetes is still in the beginning of its evaluation. We are currently generating such data in large cohorts of the general population, patient cohorts and in intervention studies in patients with diabetes. 

The aim of this project is to evaluate omics biomarkers for:

  1. Prediction of the development of diabetes
  2. For the development of complications of diabetes in general populations cohorts, patients population cohorts and intervention studies and to compare their performance
  3. Prediction of drug response to commonly used and novel treatments

Collectively these studies should foster individualization of treatment in patients with diabetes.

Overall aim

To assess biomarkers in general population and patients cohorts and to compare predictive capacity for development of diabetes, complications of diabetes, and drug response. For the latter we use intervention studies performed in patients with diabetes.

Methods

Assessments of single biomarkers
Assessments of omic profiles (metabolomics, proteomic, peptidomic)

The omic profiles in general population and patients with diabetes will be compared to assess if there are disease specific profiles associated with diease progression or whether common omic profiles exists independent of background population, type of disease, or severity of the disease.

The population will be divided in test and validation cohorts to ensure external validation of the discovered profiles

The omic profiles will be assessed in alreadly collected samples. Different matrices are available and will be used (plasma, serum, urine).

Tentative thesis Chapter outline

  1. Omics predictive biomarkers for the development of diabetes in the general population
  2. Omics predictive biomarkers for the development of diabetes in patient populations
  3. Omics predictive biomarkers for the development of complications from diabetes
  4. Omics predictive biomarkers for treatment response in patients with diabetes
  5. Review on the utility and analysis of Omics Biomarkers in the field of diabetes

How does this project contribute to Precision Medicine

We will identify biomarkers that allow for early detection of diabetes, its complications and response to treatment. This will allow for better personalization of treatment.

Possible links to other ProminenT domains

This project links with various other ProminenT projects on biomarkers and response to treatment in diabetes within ProminenT (Heerspink). It is likely that ‘inflammatory’ biomarkers are linked to disease progression and as such links between the projects from Boots, Koppelman, Bischof are envisioned. 

Project 4.3 Analysis of Lifestyle Patterns for Improvement of Diabetes  

Innovative Lifestyle Pattern Analysis for Improvement of Diabetes Management

Name Prominent PI / Promotor: GJ Navis, Dep Medicine UMCG
Name Co-Promotor: Dr Juan-Jesus Carrero, Karolinska Institutet, Stockholm
Other project members: Dr LH Dekker (UMCG) ; JO Dijkstra (De Friesland Zorgverzekeraars)

Background

Lifestyle is a main driving force of diabetes and its complications, with a main role for diet and physical (in-) activity. Accordingly, diabetes management includes both lifestyle management and pharmacotherapy, as stated in (inter-)national guidelines. Better prevention and treatment of diabetes requires improvement of both lifestyle management and pharmacotherapy, preferably in an integrated way.

Considering the overall quality of management of diabetes, the efficacy of lifestyle management is lagging behind considerably relative to pharmacological treatment as the large majority of patient fails to achieve any lifestyle treatment aim, despite substantial efforts and costs for lifestyle management. Hence, there is a large unmet need for better lifestyle management.

Recent studies on determinants of effective lifestyle management underline the importance of considering personal preferences and habits, and environmental factors as success factors for sustained efficacy. In diabetes however, little attention has been given to identifying habits and preferences, assessment of environmental factors, or assessing their associations with morbidity and their consequences for (better) management. Assessment of robust lifestyle patterns provides a relevant strategy to map habits and preferences at the aggregate level, and analyze for relevant environmental determinants that can guide better intervention strategies.

In the Lifelines cohort (n=160,000) we identified several dietary patterns, robust after adjustment for confounders. These patterns strongly associate with (multi-) morbidity, demonstrating their clinical relevance. Moreover, their marked regional distribution supports the role of (socio-cultural) environmental factors. This provides an excellent starting point to analyze the role of lifestyle patterns as determinants of morbidity in diabetes, their consequences for current management, and design of new strategies that effectively account for preferences, habits and environmental factors .

By this innovative approach of analyzing personal/environmental characteristics at the aggregate level, the project creates an intermediate level between generic approaches (one size fits all) and strictly individual approaches ( time consuming, expensive). As such it is uniquely fitted to provide an empirical basis for new approaches towards better personalization of diabetes management that are within reach with currently available technology at affordable costs.

Our underlying assumption is that better lifestyle management will facilitate overall management of diabetes, lead to lower requirements for pharmacotherapy, better outcomes and lower costs.

Overall aim

Identify the role of differences in dietary and lifestyle habits as a determinant of (multi-) morbidity and treatment quality in diabetes, as a basis for better personalized prevention and management strategies.

Methods

The main approach of the project is epidemiological. The strategy is to translate advanced big data analysis into practical guidelines for better personalization of diabetes management.

Data from available cohorts (general population including diabetes, and diabetes-only, respectively) are available for analysis of dietary patterns, their association with (multi-) morbidity, with patient management (i.e: pharmacological treatment, dietary counseling and the consequent health care expenditure) and with medical outcomes.

Dietary patterns will be analyzed from Food Frequency Questionnaires by Principal component analysis, and related to (multi-morbidity) by multivariate modeling. Data on physical activity (SQUASH) and on smoking, alcohol and substance use will be integrated into an overall lifestyle score. Combination with data on pharmacotherapy will reveal interaction between lifestyle pattern and (need for) pharmacotherapy, and modification of efficacy of pharmacotherapy by lifestyle pattern. Health economic aspects will be analyzed to assess the costs of inadequate lifestyle management on overall health care costs in diabetes (i.e need for more medication, higher complication rate etc) and support the business case for better lifestyle intervention approaches.

Role of environmental determinants for lifestyle habits will be assessed by applying geo-mapping analyses (Global Moran’s I spatial statistic) to identify regional differences in lifestyle patterns. Subsequent adjustment for relevant factors (age, gender, income, education etc) will serve to identify confounders and modifiable environmental factors are new targets for more effective lifestyle intervention .

Tentative thesis Chapter outline

  1. Association of lifestyle patterns with (multi-)morbidity in diabetes
  2. Association of lifestyle patterns with overall management of diabetes: consequences for medical outcomes.
  3. Association of lifestyle patterns with overall management of diabetes: consequences for health care expenditure
  4. Regional differences in lifestyle patterns as a dissection tool in diabetes: identification of new targets for intervention
  5. Regional differences in lifestyle patterns as a dissection tool for the prevention of diabetes: identification of new targets for intervention 

How does this project contribute to Precision Medicine

By its highly innovative approach of analyzing personal/environmental characteristics at the aggregate level, this project creates an intermediate level between generic approaches (one size fits all) and the strictly individual approaches that are time consuming and expensive.
As such it is uniquely fitted to provide an empirical basis for readily feasible approaches towards better personalization of diabetes management that are within reach with currently available technology, at affordable costs. Implementation and success of such approaches will greatly contribute to the credibility of personalized medicine as a new paradigm.

Possible links to other ProminenT domains

The project contributes to:

  • Identification of disease mechanisms (epidemiological analysis as a dissection tool);
  • Drug application (better alignment of lifestyle management and pharmacotherapy, better targeting of drug use)
  • Health Technology assessment (health economic analysis of different management strategies)
Project 4.4 Is atrial fibrillation a mechanism or a bystander in heart failure?  

Is atrial fibrillation a mechanism or a bystander in heart failure – the role of risk personalized factor management

Name Prominent PI / Promotor: Prof.Dr. I.C. van Gelder
Name Co-Promotors: Dr. M. Rienstra and Prof.Dr. A. Voors

Background

Atrial fibrillation (AF) is a vascular disease being associated with cardiovascular diseases and risk factors like hypertension, diabetes, obesity, and heart failure with a preserved ejection fraction (HFpEF). AF is not benign. It is progressive due to atrial structural remodelling caused by AF risk factors, cardiovascular diseases, and AF itself, i.e. due to an atrial myopathy. AF progression is associated with impaired prognosis.
AF and HFpEF are vicious twins. Both AF and HFpEF are increasing in prevalence. Patients with AF and HFpEF are heterogeneous and share clinical risk factors, like hypertension, diabetes and obesity. These factors are linked, both to each other and to adverse cardiovascular outcomes. AF is an independent prognostic factor in patients with HFpEF. It is questioned whether it is AF itself that contributes to worse prognosis, or, instead, whether AF is just a bystander being a marker of more severe atrial and ventricular diseases.
There are many unanswered questions about the pathophysiology, risk factors, symptomatology, diagnosis, and prognosis of AF and HFpEF. The diagnosis of HFpEF in AF, however, is challenging because risk factors, symptoms, and natriuretic peptides overlap, and diastolic dysfunction, a hallmark of HFpEF, is difficult to determine in AF. Additionally, treatment is cumbersome. Although it is generally assumed that eliminating AF is associated with improved outcome, so far, however, the trials did not show any benefit of attempts to abolish AF. Recent data, though, demonstrated that in patients with AF and HFpEF a strategy focusing on risk factor management, i.e. optimal therapy of HFpEF, hypertension, diabetes and obesity, instead of antiarrhythmic therapies, was associated with a favourable effect on sinus rhythm maintenance, in addition to risk factor reduction.
Therefore, more systematic research is needed to answer these issues and to provide treatments that improve quality of life and reduce adverse outcomes. For that, extensive phenotyping to assess the presence of risk factors using (new) imaging techniques, measures of atrial myopathy, and of diastolic dysfunction, are essential.
The central hypothesis of our proposal links AF (progression) with HFpEF (progression), and risk factors. It is our hypothesis that 1) the pathophysiology and prognostic significance of AF depends on severity of risk factors; 2) atrial myopathy develops in association with and as marker of ventricular myopathy; 3) diabetes, hypertension and obesity play a pivotal role as risk factors for atrial and ventricular myopathy; and finally 4) personalized risk factor reduction reduces AF progression, atrial myopathy and severity of HFpEF. AF, thus, is just a bystander being a marker of severity of disease.

Overall aim

  1. To link AF (progression) with HFpEF (progression), and (borderline) risk factors including hypertension, diabetes and obesity
  2. To show that the pathophysiology and prognostic significance of AF differs depending on severity risk factors.
  3. To show that atrial myopathy develops in association with a ventricular myopathy, due to shared mechanisms.
  4. To assess the role of personalized risk factor reduction on AF progression, atrial myopathy and severity of HFpEF

Methods

A total of 200 patients will be included: 50 patients with early AF and early HFpEF; 50 with early AF and more progressive HFpEF; 50 patients without (a history of) AF and early HFpEF; and 50 patients without (a history of) AF and more progressive HFpEF. An implantable loop recorder is implanted to monitor AF burden and extensive phenotyping at baseline is performed as depicted in the Table.

  Inclusion T=1 yr T=2.5 yr
Informed consent X    
Clinical history X X X
Current medication X X X
ECG X X X
Blood samples and biomarkers X   X
24-hour urine X   X
Echocardiography X   X
Cardiac catheterisation X   X
Vascular assessment incl. PET vascular imaging X   X
Bicycle exercise test X   X
Questionnaires X   X
Implantable Loop Recorder X X X
Personalised risk factor management X X X

Blood samples for analysis of biomarkers will be collected at inclusion and at 2.5 years of follow-up. During the course of the study, additional biomarkers of interest can be added to the list of biomarkers including now haemodynamic parameters, inflammation, fibrosis, adipose tissue, hypercoagulability, ischemia, metabolic stress.
Echocardiographic image acquisition will be performed to assess atrial and ventricular sizes and function in accordance to the recommendations of the American Society of Echocardiography and European Association of Cardiovascular Imaging.
Cardiac catheterization will be done to asses pulmonary wedge pressure as a surrogate measure of the left ventricular end diastolic pressure (LVEDP). Left heart catheterization will be performed to assess the presence of significant coronary artery disease or left-sided valve disease, and LVEDP will be assessed.

Tentative thesis Chapter outline

  1. Personalised risk factor management in AF - review
  2. Shared underlying mechanisms of AF (progression), HFpEF (progression), due to shared associated risk factors
  3. Pathophysiology and prognostic significance of AF differs depending on severity risk factors.
  4. Atrial myopathy develops in association with a ventricular myopathy, due to shared mechanisms.
  5. Personalised risk factor reduction has a benefical outcome on AF progression, atrial myopathy and severity of HFpEF
  6. Discussion and future perspectives

How does this project contribute to Precision Medicine

Extensive phenotyping of patients with AF and HFpEF will more than before reveal which risk factors need attention and therapy. These will differ among patients and contribute to precision medicine
In this way, medical decisions and therapies are tailored to the individual patient. We use diagnostic testing to select appropriate and optimal therapies based on the context of a patient’s phenotype including risk factors, blood biomarkers and imaging. In this way interventions are concentrated on those who will benefit, sparing expense and side effects for those who will not.

Possible links to other ProminenT domains

Disease mechanisms

  • Bootsma – immunology
  • Bisschof - biomarkers
  • Slart - vascular imaging
  • Navis - lifestye