Final Report Summary - PHASE (A Pharmacogenomic study of statins in the elderly at risk for cardiovascular disease)
Project context and objectives:
Cardiovascular disease is the leading cause of death in industrialised countries. Advancing age is an important risk factor for cardiovascular disease. With the rising number of elderly people in our society, cardiovascular disease has a major impact on healthcare. The prevention of cardiovascular disease is critically dependent on lipid lowering therapy including the 3-hydroxymethyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins). Statins are the most prescribed class of drugs worldwide and therapy is generally associated with a reduction of cardiovascular events by 20-30 %. However, clinical response is highly variable and adverse effects occur in a minority of patients. Recent research provides evidence that genetic variation contributes importantly to this variable drug response.
Pharmacogenomics focuses on unravelling the genetic determinants of such variable drug responses, both in intended, beneficial effects and unintended, adverse effects. The overall goal of the PHASE study is to perform a genome-wide scan in all 5804 participants of the 'Prospective study of pravastatin in the elderly at risk' (PROSPER), a multinational, randomised, placebo-controlled trial in order to develop personalised cholesterol lowering drug therapy based on an individual's genetic make-up to obtain optimal cardiovascular event reduction, minimal side effects, and major cost reduction for society.
The objectives for the first reporting period were the following:
- management office in place;
- kick-off meeting with all members of the PHASE study;
- completion of the PHASE website;
- execution of the PHASE genome wide scan in 5804 participants of the PROSPER study;
- contact other studies for collaborative activities.
The main objective for the first reporting period of the PHASE study was to make a good start with the project by setting up the management office to ensure efficient management and administration of the project. Furthermore, it is essential to ensure efficient communication between the participants of the project. Before the official start of the study, we organised a get-together with all participants of the PHASE project. After receiving the signed grant agreement of the European Commission (EC), we had an official kick-off meeting in Leiden, the Netherlands, all members of the PHASE study came to the Netherlands to attend this important meeting. Furthermore we have set-upped the official PHASE website (please see http://www.lumc.nl/con/9099 online).
Our main aim for this reporting period was to execute the genome wide scan in all 5804 participants of the PROSPER study and we succeeded in reaching that goal. In the first period of the PHASE study, we have performed the genome-wide scan in all 5804 participants of the PROSPER study according to the proposal. The genotyping of all these participants has been performed with the Sentrix Humanhap660K-Quad DNA analysis beadchip from the Illumina platform. With the Illumina 660Kb beadchip we have genotyped 557 192 genetic variants, which we have imputed up to 2.5 million single-nucleotide polymorphisms (SNPs) within the whole genome with MACH imputation software. Quality control has been implemented on all the 5763 successfully genotyped samples of the participants of the PHASE study. This resulted in 5244 samples that are investigated in the PHASE study.
The objectives for the second reporting period were the following:
- continuation of the face-to-face PHASE meetings;
- updating the PHASE website;
- performing pharmacogenetic analyses with low-density lipoprotein (LDL) lowering, clinical outcomes and adverse events;
- collaborative activities with other genome-wide association studies (GWAS);
- replication of the WOSCOPS study with the MetaboChip array;
- whole exome sequencing of high and non-responders;
- dissemination by publications and dissemination report;
- performing pharmacogenetic analyses with LDL lowering, clinical outcomes and adverse events.
The main objective for the second reporting period was to perform all proposed pharmacogenetic analyses, in which we succeeded.
We have performed first a GWAS on baseline LDL levels to validate the PROSPER / PHASE study, replication was performed in the WOSCOPS and CARE studies. With this proof-of-principle study we show that the PROSPER / PHASE study can be used to investigate genetic associations in a similar way to population-based studies. Moreover, we can also assume from these results that the PROSPER / PHASE study is likely to have sufficient power to detect genome-wide significant hits with large effects for other quantitative traits. Results are published in the following article: Trompet et al., BMC Med Gen, 2011, 12:131.
Secondly, we investigated the pharmacogenetic interaction between genetic variation, statin use, and LDL lowering in response to statin treatment. We performed this association within the PHASE study and collaborated and meta-analysed our results with the ASCOT and CARDS trials. Both studies have a large sample size and GWAS data that can be used in this analysis. We found that the APOE gene and the LPa gene are both involved in the pharmacogenetics of statins of LDL lowering. These findings are published in the following article: Deshmukh et al., J Lipid Res, 2012, 53:1000.
Thirdly, we investigated the pharmacogenetic interaction between genetic variation, statin use and the incidence of myocardial infarction in collaboration with the WOSCOPS and CARE trials. Within the WOSCOPS and CARE trials a GWAS has been performed in the cases of the studies only. To perform the interaction with the genetic variants a case only design was used for these studies. Results were meta-analysed with the PROSPER / PHASE results. We found independently in all three studies that genetic variation within the DNA JC5B gene influences the risk of a myocardial infarction after statin use. The results are published in the following article: Shiffman et al., PLoS One, 2012, 7: e38240
For these last two main analyses, the pharmacogenetic interaction between statin use and genotypes on LDL response and other clinical outcomes, we organised a new collaboration under PHASE leadership, named the genomic investigation of statin treatment (GIST) with other randomised controlled clinical trials, to assess which genetic variation is responsible for the variability in statin drug response. All randomised controlled trials, participating in this collaboration are similar to the PROSPER study, namely the ASCOT study, the CARDS study, the PRINCE study, the CAP study and the TNT trial. Positive associations will be replicated and confirmed in two comparable large clinical trials, the WOSCOPS and the CARE study. This gives us a unique opportunity to assess this research question in a large clinical setting, since this collaboration will achieve the largest possible study population. This research is still ongoing, the first analyses and meta-analyses are performed. We expect at least two articles from this GIST consortium to be submitted at the end of this year.
Collaborative activities with other GWAS studies
For our main analysis, the pharmacogenetic interaction between statin use and genotypes on LDL response and other clinical outcomes, we, as indicated, organised our own collaboration, the GIST consortium. The consortium consists of two parallel working groups, one with the focus on randomised controlled trials as explained above, the other with the focus on observational studies. The PHASE study has the lead and organisation of both working groups. The second working group consists mainly of cohorts also participating in the CHARGE consortium and is therefore closely working together with the pharmacogenetics group of the CHARGE consortium led by Prof. Bruce Psaty of the University of Seattle. Combining and meta-analysing these results will give us a unique opportunity to assess genetic variation responsible for the variability in statin drug response in a large clinical setting.
We have contacted the CHARGE consortium to collaborate with the PHASE study. This CHARGE consortium consists of five large cohorts with genome wide scan data, who agreed to work with each other and share the results for a meta-analysis resulting in a GWAS in a very large study population to achieve genome-wide statistical significance. The five cohorts are the 'Age, genes / environment, susceptibility' (AGES) Reykjavik study, the 'Atherosclerosis risk in communities' (ARIC) study, the 'Cardiovascular health study' (CHS), the 'Framingham heart study' (FHS), and the Rotterdam study. We have made collaborating agreements with the CHARGE consortium and we will combine our research efforts on several phenotypes, like cancer, haematological measures, and electrocardiography (ECG) parameters.
All other objectives, like continuation of the face-to-face meetings, updating the PHASE website, and dissemination activities were closely monitored by the management team and also successfully achieved.
Project results:
Summary of management activities
During the first six months of the PHASE project, we have made several steps to make sure that the project management office was in place. The management office comprises the project coordinator (Professor Dr J. W. Jukema), the project manager (Dr S. Trompet), the deputy coordinator (Professor Dr R.G.J. Westendorp), and the Head of the Biobank of the PROSPER study (Dr A.J.M. de Craen).
A first steering committee was arranged in the month before the start of the PHASE study to make sure all members knew about the project and knew their management duties. This has facilitated the collaboration between the three beneficiaries and has enhanced efficient communication, management and administration throughout the project. A consortium agreement was agreed upon between the three beneficiaries of the PHASE project. In the agreement, it was stated what was expected from the three beneficiaries during the project. Moreover, a project agreement was written and agreed upon for the members of the PROSPER study group. In this agreement between the beneficiaries, it was stated how to handle the intellectual property and ownership of the results that came out of the PHASE study.
To ensure that computer problems could not occur during the project analysis and that there was sufficient hard drive capacity to run and save all analyses with the genome wide scan data, we bought a new computer for the project management office with a large amount of working memory and hard drive capacity to handle these large computer files.
PHASE executive committee meetings
The start date of the PHASE project was 1 January 2009. Three weeks before that date we had our unofficial kick-off meeting in Cork, Ireland to start up the first necessary issues for the PHASE project. In April 2009, the grant agreement with the European Union (EU) was signed. In May 2009, we had therefore our official kick-off meeting in Leiden, the Netherlands. All members of the PHASE project have attended this meeting. During this meeting, we had fruitful discussions on the progression of the PHASE project.
In the first reporting period, we had the official kick-off meeting in May 2009, in Leiden, the Netherlands. After that meeting we organised three other meetings in the first period. In November 2009 and June 2010, we had two steering committee meetings in Glasgow, Scotland and Cork, Ireland respectively. In the second reporting period, we have organised four face-to-face meetings and one four-hour conference call, to ensure that all members could participate.
List of websites: http://www.lumc.nl/con/9099
Cardiovascular disease is the leading cause of death in industrialised countries. Advancing age is an important risk factor for cardiovascular disease. With the rising number of elderly people in our society, cardiovascular disease has a major impact on healthcare. The prevention of cardiovascular disease is critically dependent on lipid lowering therapy including the 3-hydroxymethyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors (statins). Statins are the most prescribed class of drugs worldwide and therapy is generally associated with a reduction of cardiovascular events by 20-30 %. However, clinical response is highly variable and adverse effects occur in a minority of patients. Recent research provides evidence that genetic variation contributes importantly to this variable drug response.
Pharmacogenomics focuses on unravelling the genetic determinants of such variable drug responses, both in intended, beneficial effects and unintended, adverse effects. The overall goal of the PHASE study is to perform a genome-wide scan in all 5804 participants of the 'Prospective study of pravastatin in the elderly at risk' (PROSPER), a multinational, randomised, placebo-controlled trial in order to develop personalised cholesterol lowering drug therapy based on an individual's genetic make-up to obtain optimal cardiovascular event reduction, minimal side effects, and major cost reduction for society.
The objectives for the first reporting period were the following:
- management office in place;
- kick-off meeting with all members of the PHASE study;
- completion of the PHASE website;
- execution of the PHASE genome wide scan in 5804 participants of the PROSPER study;
- contact other studies for collaborative activities.
The main objective for the first reporting period of the PHASE study was to make a good start with the project by setting up the management office to ensure efficient management and administration of the project. Furthermore, it is essential to ensure efficient communication between the participants of the project. Before the official start of the study, we organised a get-together with all participants of the PHASE project. After receiving the signed grant agreement of the European Commission (EC), we had an official kick-off meeting in Leiden, the Netherlands, all members of the PHASE study came to the Netherlands to attend this important meeting. Furthermore we have set-upped the official PHASE website (please see http://www.lumc.nl/con/9099 online).
Our main aim for this reporting period was to execute the genome wide scan in all 5804 participants of the PROSPER study and we succeeded in reaching that goal. In the first period of the PHASE study, we have performed the genome-wide scan in all 5804 participants of the PROSPER study according to the proposal. The genotyping of all these participants has been performed with the Sentrix Humanhap660K-Quad DNA analysis beadchip from the Illumina platform. With the Illumina 660Kb beadchip we have genotyped 557 192 genetic variants, which we have imputed up to 2.5 million single-nucleotide polymorphisms (SNPs) within the whole genome with MACH imputation software. Quality control has been implemented on all the 5763 successfully genotyped samples of the participants of the PHASE study. This resulted in 5244 samples that are investigated in the PHASE study.
The objectives for the second reporting period were the following:
- continuation of the face-to-face PHASE meetings;
- updating the PHASE website;
- performing pharmacogenetic analyses with low-density lipoprotein (LDL) lowering, clinical outcomes and adverse events;
- collaborative activities with other genome-wide association studies (GWAS);
- replication of the WOSCOPS study with the MetaboChip array;
- whole exome sequencing of high and non-responders;
- dissemination by publications and dissemination report;
- performing pharmacogenetic analyses with LDL lowering, clinical outcomes and adverse events.
The main objective for the second reporting period was to perform all proposed pharmacogenetic analyses, in which we succeeded.
We have performed first a GWAS on baseline LDL levels to validate the PROSPER / PHASE study, replication was performed in the WOSCOPS and CARE studies. With this proof-of-principle study we show that the PROSPER / PHASE study can be used to investigate genetic associations in a similar way to population-based studies. Moreover, we can also assume from these results that the PROSPER / PHASE study is likely to have sufficient power to detect genome-wide significant hits with large effects for other quantitative traits. Results are published in the following article: Trompet et al., BMC Med Gen, 2011, 12:131.
Secondly, we investigated the pharmacogenetic interaction between genetic variation, statin use, and LDL lowering in response to statin treatment. We performed this association within the PHASE study and collaborated and meta-analysed our results with the ASCOT and CARDS trials. Both studies have a large sample size and GWAS data that can be used in this analysis. We found that the APOE gene and the LPa gene are both involved in the pharmacogenetics of statins of LDL lowering. These findings are published in the following article: Deshmukh et al., J Lipid Res, 2012, 53:1000.
Thirdly, we investigated the pharmacogenetic interaction between genetic variation, statin use and the incidence of myocardial infarction in collaboration with the WOSCOPS and CARE trials. Within the WOSCOPS and CARE trials a GWAS has been performed in the cases of the studies only. To perform the interaction with the genetic variants a case only design was used for these studies. Results were meta-analysed with the PROSPER / PHASE results. We found independently in all three studies that genetic variation within the DNA JC5B gene influences the risk of a myocardial infarction after statin use. The results are published in the following article: Shiffman et al., PLoS One, 2012, 7: e38240
For these last two main analyses, the pharmacogenetic interaction between statin use and genotypes on LDL response and other clinical outcomes, we organised a new collaboration under PHASE leadership, named the genomic investigation of statin treatment (GIST) with other randomised controlled clinical trials, to assess which genetic variation is responsible for the variability in statin drug response. All randomised controlled trials, participating in this collaboration are similar to the PROSPER study, namely the ASCOT study, the CARDS study, the PRINCE study, the CAP study and the TNT trial. Positive associations will be replicated and confirmed in two comparable large clinical trials, the WOSCOPS and the CARE study. This gives us a unique opportunity to assess this research question in a large clinical setting, since this collaboration will achieve the largest possible study population. This research is still ongoing, the first analyses and meta-analyses are performed. We expect at least two articles from this GIST consortium to be submitted at the end of this year.
Collaborative activities with other GWAS studies
For our main analysis, the pharmacogenetic interaction between statin use and genotypes on LDL response and other clinical outcomes, we, as indicated, organised our own collaboration, the GIST consortium. The consortium consists of two parallel working groups, one with the focus on randomised controlled trials as explained above, the other with the focus on observational studies. The PHASE study has the lead and organisation of both working groups. The second working group consists mainly of cohorts also participating in the CHARGE consortium and is therefore closely working together with the pharmacogenetics group of the CHARGE consortium led by Prof. Bruce Psaty of the University of Seattle. Combining and meta-analysing these results will give us a unique opportunity to assess genetic variation responsible for the variability in statin drug response in a large clinical setting.
We have contacted the CHARGE consortium to collaborate with the PHASE study. This CHARGE consortium consists of five large cohorts with genome wide scan data, who agreed to work with each other and share the results for a meta-analysis resulting in a GWAS in a very large study population to achieve genome-wide statistical significance. The five cohorts are the 'Age, genes / environment, susceptibility' (AGES) Reykjavik study, the 'Atherosclerosis risk in communities' (ARIC) study, the 'Cardiovascular health study' (CHS), the 'Framingham heart study' (FHS), and the Rotterdam study. We have made collaborating agreements with the CHARGE consortium and we will combine our research efforts on several phenotypes, like cancer, haematological measures, and electrocardiography (ECG) parameters.
All other objectives, like continuation of the face-to-face meetings, updating the PHASE website, and dissemination activities were closely monitored by the management team and also successfully achieved.
Project results:
Summary of management activities
During the first six months of the PHASE project, we have made several steps to make sure that the project management office was in place. The management office comprises the project coordinator (Professor Dr J. W. Jukema), the project manager (Dr S. Trompet), the deputy coordinator (Professor Dr R.G.J. Westendorp), and the Head of the Biobank of the PROSPER study (Dr A.J.M. de Craen).
A first steering committee was arranged in the month before the start of the PHASE study to make sure all members knew about the project and knew their management duties. This has facilitated the collaboration between the three beneficiaries and has enhanced efficient communication, management and administration throughout the project. A consortium agreement was agreed upon between the three beneficiaries of the PHASE project. In the agreement, it was stated what was expected from the three beneficiaries during the project. Moreover, a project agreement was written and agreed upon for the members of the PROSPER study group. In this agreement between the beneficiaries, it was stated how to handle the intellectual property and ownership of the results that came out of the PHASE study.
To ensure that computer problems could not occur during the project analysis and that there was sufficient hard drive capacity to run and save all analyses with the genome wide scan data, we bought a new computer for the project management office with a large amount of working memory and hard drive capacity to handle these large computer files.
PHASE executive committee meetings
The start date of the PHASE project was 1 January 2009. Three weeks before that date we had our unofficial kick-off meeting in Cork, Ireland to start up the first necessary issues for the PHASE project. In April 2009, the grant agreement with the European Union (EU) was signed. In May 2009, we had therefore our official kick-off meeting in Leiden, the Netherlands. All members of the PHASE project have attended this meeting. During this meeting, we had fruitful discussions on the progression of the PHASE project.
In the first reporting period, we had the official kick-off meeting in May 2009, in Leiden, the Netherlands. After that meeting we organised three other meetings in the first period. In November 2009 and June 2010, we had two steering committee meetings in Glasgow, Scotland and Cork, Ireland respectively. In the second reporting period, we have organised four face-to-face meetings and one four-hour conference call, to ensure that all members could participate.
List of websites: http://www.lumc.nl/con/9099