Human Genetics
Contributions
Type: Oral Presentation
Presentation during EHA20: From 14.06.2015 08:30 to 14.06.2015 08:45
Location: Room Strauss 2
Background
Polycythemia vera (PV) and essential thrombocythemia (ET) are characterized primarily by an excess production of erythrocytes and platelets, respectively. Whilst it is accepted that somatic mutations in JAK2 and other genes drive clonal proliferation, less is known about the factors that determine the precise disease phenotype. Compared to ET, PV is characterized by higher levels of the JAK2 V617F mutation and much more frequent outgrowth of homozygous mutant clones, however mutation burden alone cannot distinguish the two disorders and it is clear that other factors must also be important. The finding that leukocyte counts in murine JAK2 V617F retroviral-mediated transfection/transplantation models differ in a strain-dependent manner was the first suggestion that host genetic factors may influence the disease phenotype.
Aims
Our aim is to identify inherited genetic factors that influence whether JAK2 V617F positive MPN patients develop PV or ET.
Methods
We undertook a two stage genome-wide association study. At stage 1,556 ET patients and 556 PV patients that were positive for JAK2 V617F and from the UK were genotyped. Following standard quality control, allelic chi-square tests were used to compare SNPs between ET or PV cases against controls (WTCCC2 n=5200). SNPs with subtype specific effects were identified by using Breslow-Day tests to assess the homogeneity of effect sizes. SNPs with significantly different effects in ET and PV cases were followed up at stage 2 by additional genotyping in two independent cohorts of JAK2 V617F negative cases with either ET or primary myelofibrosis from the UK (n=524), Germany (n=187) and Austria (n=99). At stage 2, controls from the UK (WTCCC2 NBS n=2706) and Bavaria (KORA n=1805) were used for comparison. The final effect size and significance of SNPs was determined by a fixed effects meta-analysis of the 3 cohorts. To investigate SNP function, we used RT-qPCR to analyse gene expression in myeloid progenitors cultured in vitro from a series of healthy controls and MPN cases.
Results
After quality control, 490,755 SNPs were tested in 499 ET and 505 PV cases, and 5200 WTCCC2 controls at stage 1. Excluding the JAK2 region due to recurrent aUPD in PV and to a lesser extent in ET, rs9376092 in the intergenic region between HBS1L and MYB was identified as having one of the strongest subtype specific effects (Breslow-Day test p=4.4x10-5) which increases the risk of ET (allelic chi-square p=6.1x10-7, OR=1.42) and may reduce the risk of developing PV (allelic chi-square p=0.068, OR=0.87). After testing stage 2 and using a fixed effects meta-analysis to combine evidence across stages, rs9376092 achieved genome-wide significance (meta-analysis p=1.8x10-11, OR=1.35) without heterogeneity between cohorts (Cochran’s Q test p=0.50). Using RT-qPCR we associated the rs9376092 risk allele with reduced expression of MYB (5.8-9.7 fold) and, less prominently, HBS1L (1.9-2.4 fold) in healthy individuals. We also found that MYB expression in ET was significantly lower in JAK2 V617F-heterozygous BFU-E colonies compared to JAK2-wild type colonies from the same patient (Mann-Whitney test P=0.0009; q=0.01) but this difference was not seen in colonies from PV cases.
Summary
We conclude that constitutional genetic variation in the intergenic region between HBS1L and MYB influences whether JAK2 V617F mutant MPN develop ET or PV. In JAK2 V617F-negative MPN, the same variation predisposes to the development of ET. Our findings thus link constitutional differences in MYB expression to both predisposition to MPN and to MPN phenotype.
Keyword(s): Essential Thrombocytemia, Genetic polymorphism, Polycythemia vera
Session topic: Novel insights into the mechanisms involved in MPNs
Type: Oral Presentation
Presentation during EHA20: From 14.06.2015 08:30 to 14.06.2015 08:45
Location: Room Strauss 2
Background
Polycythemia vera (PV) and essential thrombocythemia (ET) are characterized primarily by an excess production of erythrocytes and platelets, respectively. Whilst it is accepted that somatic mutations in JAK2 and other genes drive clonal proliferation, less is known about the factors that determine the precise disease phenotype. Compared to ET, PV is characterized by higher levels of the JAK2 V617F mutation and much more frequent outgrowth of homozygous mutant clones, however mutation burden alone cannot distinguish the two disorders and it is clear that other factors must also be important. The finding that leukocyte counts in murine JAK2 V617F retroviral-mediated transfection/transplantation models differ in a strain-dependent manner was the first suggestion that host genetic factors may influence the disease phenotype.
Aims
Our aim is to identify inherited genetic factors that influence whether JAK2 V617F positive MPN patients develop PV or ET.
Methods
We undertook a two stage genome-wide association study. At stage 1,556 ET patients and 556 PV patients that were positive for JAK2 V617F and from the UK were genotyped. Following standard quality control, allelic chi-square tests were used to compare SNPs between ET or PV cases against controls (WTCCC2 n=5200). SNPs with subtype specific effects were identified by using Breslow-Day tests to assess the homogeneity of effect sizes. SNPs with significantly different effects in ET and PV cases were followed up at stage 2 by additional genotyping in two independent cohorts of JAK2 V617F negative cases with either ET or primary myelofibrosis from the UK (n=524), Germany (n=187) and Austria (n=99). At stage 2, controls from the UK (WTCCC2 NBS n=2706) and Bavaria (KORA n=1805) were used for comparison. The final effect size and significance of SNPs was determined by a fixed effects meta-analysis of the 3 cohorts. To investigate SNP function, we used RT-qPCR to analyse gene expression in myeloid progenitors cultured in vitro from a series of healthy controls and MPN cases.
Results
After quality control, 490,755 SNPs were tested in 499 ET and 505 PV cases, and 5200 WTCCC2 controls at stage 1. Excluding the JAK2 region due to recurrent aUPD in PV and to a lesser extent in ET, rs9376092 in the intergenic region between HBS1L and MYB was identified as having one of the strongest subtype specific effects (Breslow-Day test p=4.4x10-5) which increases the risk of ET (allelic chi-square p=6.1x10-7, OR=1.42) and may reduce the risk of developing PV (allelic chi-square p=0.068, OR=0.87). After testing stage 2 and using a fixed effects meta-analysis to combine evidence across stages, rs9376092 achieved genome-wide significance (meta-analysis p=1.8x10-11, OR=1.35) without heterogeneity between cohorts (Cochran’s Q test p=0.50). Using RT-qPCR we associated the rs9376092 risk allele with reduced expression of MYB (5.8-9.7 fold) and, less prominently, HBS1L (1.9-2.4 fold) in healthy individuals. We also found that MYB expression in ET was significantly lower in JAK2 V617F-heterozygous BFU-E colonies compared to JAK2-wild type colonies from the same patient (Mann-Whitney test P=0.0009; q=0.01) but this difference was not seen in colonies from PV cases.
Summary
We conclude that constitutional genetic variation in the intergenic region between HBS1L and MYB influences whether JAK2 V617F mutant MPN develop ET or PV. In JAK2 V617F-negative MPN, the same variation predisposes to the development of ET. Our findings thus link constitutional differences in MYB expression to both predisposition to MPN and to MPN phenotype.
Keyword(s): Essential Thrombocytemia, Genetic polymorphism, Polycythemia vera
Session topic: Novel insights into the mechanisms involved in MPNs