Final Report Summary - INDEFIC (Identification of new genetic determinants of plasma fibrinogen concentration)
The main goal of the INDEFIC project was to identify and characterise novel genes, genetic variants and biological pathways implicated in the regulation of plasma fibrinogen concentration, which is an independent risk factor for atherosclerosis, arterial thrombosis and occurrence of clinically manifest Cardiovascular disease (CVD). Specifically, INDEFIC aimed to achieve a better understanding of the pathways and biological mechanisms regulating fibrinogen concentration.
The initial part of the project focused on localising new genomic regions that determined plasma fibrinogen concentration based on the Genome-wide association study (GWA) technique. For this purpose, the applicant first aimed to use the databases and biobanks of PROCARDIS, a large European multi-centre study supported by the EU (co-PI: Prof. Anders Hamsten, Project LSHM-CT- 2007- 037273) and designed to identify susceptibility genes for Coronary artery disease (CAD).
However, after preliminary results were obtained using the PROCARDIS sample (genome-wide genotyping data on 6 000 CAD patients and 7 200 healthy control persons), it became obvious that the genetic architecture of fibrinogen regulation consists of many loci with small effect, and that a bigger sample size would be necessary in order to obtain the statistical power required for the detection of such small effects in fibrinogen regulation.
Thus, the next step was to establish an efficient collaboration with 28 other cohorts coming from Europe and the United States, which then performed individual GWAs analyses of plasma fibrinogen levels under the guidance of the applicant (project acronym: FIBRINOGEN) who take major responsibility of coordination and meta-analyses of individual GWAs to obtain the global association results.
After the main associated loci had been identified and since one of the main goals of INDEFIC was to focus on a better understanding of the biological mechanisms regulating the plasma fibrinogen concentration, the applicant used the meta-analysis results to infer the main biological pathways determining fibrinogen regulation in collaboration with the Womens Genome Health Study, using two different algorithms, implemented in the publicly accessible software GRAIL (see http://gettinggeneticsdone.blogspot.com/2010/02/grail-gene-relationships-across.html online) and MAGENTA (see http://www.broadinstitute.org/mpg/magenta/ online for further details).
In addition, we used the analysis of expression of quantitative trait loci (expression QTL, eQTL) methodology to dissect these newly discovered associated regions as a main tool to confirm the 'in silico' network analysis performed in GRAIL and MAGENTA and to prioritise candidate genes in previously described chromosomal regions associated with plasma fibrinogen concentration.
Further, a Genetic risk score (GRS) was computed using data from all European-ancestry individuals to model the increase in fibrinogen levels according to number of fibrinogen-raising alleles for each of the lead Single nucleotide polymorphism (SNP)s and we calculated which proportion of variance in fibrinogen concentration was explained by the variants found in the present project. Finally, associations of the newly identified fibrinogen-associated variants with prevalent CAD, stroke and VTE were also examined in large appropriate cohorts. Results were meta-analysed for the individual SNPs using METAL, and the association of the pooled-effect of all variants found in the present project was also calculated for different CVD outcomes.
Within the frame of this 2-years project, we have finalised the largest meta-analysis ever of fibrinogen GWAS, including a discovery analysis performed in 91 435 individuals of European ancestry plus further analysis in an additional 8 307 African Americans from 7 cohorts, using around 2.7 million SNPs. Overall, results from the meta-analysis in 91,435 individuals of European ancestry have identified 23 genome-wide significant loci for fibrinogen levels, among which 8 represent replications of previously described associations and 15 are newly identified fibrinogen-associated loci.
Gene-set enrichment analysis have identified that the most represented pathways in fibrinogen regulation are related to inflammation (acute-phase response, interleukin signaling), adipocytokine signaling and thyrotrophin-releasing hormone signaling. According to these results, several genes (LEPR, IL6R, IL1R, IL1F10/IL1F5/IL1F8/IL1RN, FGA/FGB, ACTN1 and CPT1B) have been prioritised as most plausible candidate genes within our 23 genomic regions.
Association analyses have shown that rs4129267 located in the IL6R locus, rs6734238 in the IL1F10/IL1RN locus and rs1154988 in the PCCB locus are significantly associated with CAD; however, the direction of effect is consistent only for rs4129267 in the IL6R locus. The pooled association for the 24 lead SNPs with CAD is not significant (OR(CI95%)= 1.00 (0.97,1.03)). None of the fibrinogen-associated lead SNPs are significantly associated with stroke or VTE after correction for multiple testing. The pooled results are suggestive for stroke (stroke OR(CI95%)= 1.03 (1.00,1.07); but not for VTE OR(CI95%)= 0.96 (0.92,1.01)).
Taken together, our findings reveal novel loci that regulate plasma fibrinogen concentration, and demonstrate that there is a substantial influence of inflammatory and adipocytokine-related proteins in fibrinogen regulation. With regards to important clinical implications, the main result of this project is that overall, our mendelian randomisation analyses argue against a functional role of fibrinogen in CVD. We believe that these results greatly improve our understanding of the implication of fibrinogen in CVD.
The initial part of the project focused on localising new genomic regions that determined plasma fibrinogen concentration based on the Genome-wide association study (GWA) technique. For this purpose, the applicant first aimed to use the databases and biobanks of PROCARDIS, a large European multi-centre study supported by the EU (co-PI: Prof. Anders Hamsten, Project LSHM-CT- 2007- 037273) and designed to identify susceptibility genes for Coronary artery disease (CAD).
However, after preliminary results were obtained using the PROCARDIS sample (genome-wide genotyping data on 6 000 CAD patients and 7 200 healthy control persons), it became obvious that the genetic architecture of fibrinogen regulation consists of many loci with small effect, and that a bigger sample size would be necessary in order to obtain the statistical power required for the detection of such small effects in fibrinogen regulation.
Thus, the next step was to establish an efficient collaboration with 28 other cohorts coming from Europe and the United States, which then performed individual GWAs analyses of plasma fibrinogen levels under the guidance of the applicant (project acronym: FIBRINOGEN) who take major responsibility of coordination and meta-analyses of individual GWAs to obtain the global association results.
After the main associated loci had been identified and since one of the main goals of INDEFIC was to focus on a better understanding of the biological mechanisms regulating the plasma fibrinogen concentration, the applicant used the meta-analysis results to infer the main biological pathways determining fibrinogen regulation in collaboration with the Womens Genome Health Study, using two different algorithms, implemented in the publicly accessible software GRAIL (see http://gettinggeneticsdone.blogspot.com/2010/02/grail-gene-relationships-across.html online) and MAGENTA (see http://www.broadinstitute.org/mpg/magenta/ online for further details).
In addition, we used the analysis of expression of quantitative trait loci (expression QTL, eQTL) methodology to dissect these newly discovered associated regions as a main tool to confirm the 'in silico' network analysis performed in GRAIL and MAGENTA and to prioritise candidate genes in previously described chromosomal regions associated with plasma fibrinogen concentration.
Further, a Genetic risk score (GRS) was computed using data from all European-ancestry individuals to model the increase in fibrinogen levels according to number of fibrinogen-raising alleles for each of the lead Single nucleotide polymorphism (SNP)s and we calculated which proportion of variance in fibrinogen concentration was explained by the variants found in the present project. Finally, associations of the newly identified fibrinogen-associated variants with prevalent CAD, stroke and VTE were also examined in large appropriate cohorts. Results were meta-analysed for the individual SNPs using METAL, and the association of the pooled-effect of all variants found in the present project was also calculated for different CVD outcomes.
Within the frame of this 2-years project, we have finalised the largest meta-analysis ever of fibrinogen GWAS, including a discovery analysis performed in 91 435 individuals of European ancestry plus further analysis in an additional 8 307 African Americans from 7 cohorts, using around 2.7 million SNPs. Overall, results from the meta-analysis in 91,435 individuals of European ancestry have identified 23 genome-wide significant loci for fibrinogen levels, among which 8 represent replications of previously described associations and 15 are newly identified fibrinogen-associated loci.
Gene-set enrichment analysis have identified that the most represented pathways in fibrinogen regulation are related to inflammation (acute-phase response, interleukin signaling), adipocytokine signaling and thyrotrophin-releasing hormone signaling. According to these results, several genes (LEPR, IL6R, IL1R, IL1F10/IL1F5/IL1F8/IL1RN, FGA/FGB, ACTN1 and CPT1B) have been prioritised as most plausible candidate genes within our 23 genomic regions.
Association analyses have shown that rs4129267 located in the IL6R locus, rs6734238 in the IL1F10/IL1RN locus and rs1154988 in the PCCB locus are significantly associated with CAD; however, the direction of effect is consistent only for rs4129267 in the IL6R locus. The pooled association for the 24 lead SNPs with CAD is not significant (OR(CI95%)= 1.00 (0.97,1.03)). None of the fibrinogen-associated lead SNPs are significantly associated with stroke or VTE after correction for multiple testing. The pooled results are suggestive for stroke (stroke OR(CI95%)= 1.03 (1.00,1.07); but not for VTE OR(CI95%)= 0.96 (0.92,1.01)).
Taken together, our findings reveal novel loci that regulate plasma fibrinogen concentration, and demonstrate that there is a substantial influence of inflammatory and adipocytokine-related proteins in fibrinogen regulation. With regards to important clinical implications, the main result of this project is that overall, our mendelian randomisation analyses argue against a functional role of fibrinogen in CVD. We believe that these results greatly improve our understanding of the implication of fibrinogen in CVD.