Key points are not available for this paper at this time.
We carried out a genome-wide association study (GWAS) of LDL-c response to statin using data from participants in the Collaborative Atorvastatin Diabetes Study (CARDS; n = 1,156), the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT; n = 895), and the observational phase of ASCOT (n = 651), all of whom were prescribed atorvastatin 10 mg. Following genome-wide imputation, we combined data from the three studies in a meta-analysis. We found associations of LDL-c response to atorvastatin that reached genome-wide significance at rs10455872 (P= 6.13 × 10−9) within the LPA gene and at two single nucleotide polymorphisms (SNP) within the APOE region (rs445925; P= 2.22 × 10−16 and rs4420638; P= 1.01 × 10−11) that are proxies for the ∊2 and ∊4 variants, respectively, in APOE. The novel association with the LPA SNP was replicated in the PROspective Study of Pravastatin in the Elderly at Risk (PROSPER) trial (P= 0.009). Using CARDS data, we further showed that atorvastatin therapy did not alter lipoprotein(a) Lp(a) and that Lp(a) levels accounted for all of the associations of SNPs in the LPA gene and the apparent LDL-c response levels. However, statin therapy had a similar effect in reducing cardiovascular disease (CVD) in patients in the top quartile for serum Lp(a) levels (HR = 0.60) compared with those in the lower three quartiles (HR = 0.66; P= 0.8 for interaction). The data emphasize that high Lp(a) levels affect the measurement of LDL-c and the clinical estimation of LDL-c response. Therefore, an apparently lower LDL-c response to statin therapy may indicate a need for measurement of Lp(a). However, statin therapy seems beneficial even in those with high Lp(a). We carried out a genome-wide association study (GWAS) of LDL-c response to statin using data from participants in the Collaborative Atorvastatin Diabetes Study (CARDS; n = 1,156), the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT; n = 895), and the observational phase of ASCOT (n = 651), all of whom were prescribed atorvastatin 10 mg. Following genome-wide imputation, we combined data from the three studies in a meta-analysis. We found associations of LDL-c response to atorvastatin that reached genome-wide significance at rs10455872 (P= 6.13 × 10−9) within the LPA gene and at two single nucleotide polymorphisms (SNP) within the APOE region (rs445925; P= 2.22 × 10−16 and rs4420638; P= 1.01 × 10−11) that are proxies for the ∊2 and ∊4 variants, respectively, in APOE. The novel association with the LPA SNP was replicated in the PROspective Study of Pravastatin in the Elderly at Risk (PROSPER) trial (P= 0.009). Using CARDS data, we further showed that atorvastatin therapy did not alter lipoprotein(a) Lp(a) and that Lp(a) levels accounted for all of the associations of SNPs in the LPA gene and the apparent LDL-c response levels. However, statin therapy had a similar effect in reducing cardiovascular disease (CVD) in patients in the top quartile for serum Lp(a) levels (HR = 0.60) compared with those in the lower three quartiles (HR = 0.66; P= 0.8 for interaction). The data emphasize that high Lp(a) levels affect the measurement of LDL-c and the clinical estimation of LDL-c response. Therefore, an apparently lower LDL-c response to statin therapy may indicate a need for measurement of Lp(a). However, statin therapy seems beneficial even in those with high Lp(a). Statin therapy is now widely accepted for the primary and secondary prevention of cardiovascular disease (CVD) in certain patient groups. However, there is considerable variation in response to statin therapy that remains poorly understood. For example, in the Collaborative Atorvastatin Diabetes Study (CARDS) trial (1Colhoun H.M. Betteridge D.J. Durrington P.N. Hitman G.A. Neil H.A. Livingstone S.J. Thomason M.J. Mackness M.I. Charlton-Menys V. Fuller J.H. Primary prevention of cardiovascular disease with atorvastatin in type 2 diabetes in the Collaborative Atorvastatin Diabetes Study (CARDS): multicentre randomised placebo-controlled trial.Lancet. 2004; 364: 685-696Abstract Full Text Full Text PDF PubMed Scopus (3261) Google Scholar), among self-reported and pill count-validated compliant recipients of atorvastatin 10 mg daily, the absolute change in LDL-c at one month post-randomization varied from −2 to −0.6 mmol/l, (5th and 95th centiles of the range), and the percentage lowering from baseline varied from 67% to 22%. Understanding the pathways and determinants involved in this variation in response to therapy could lead to improved treatments. Even without understanding the pathways, identifying predictors of poorer response could identify those most in need of additional or alternative therapeutic strategies. Two genome-wide association studies (GWAS) of statin response and several candidate gene association studies have been reported (2Thompson J.F. Hyde C.L. Wood L.S. Paciga S.A. Hinds D.A. Cox D.R. Hovingh G.K. Kastelein J.J. Comprehensive whole-genome and candidate gene analysis for response to statin therapy in the Treating to New Targets (TNT) cohort.Circ. Cardiovasc. Genet. 2009; 2: 173-181Crossref PubMed Scopus (164) Google Scholar–3Barber M.J. Mangravite L.M. Hyde C.L. Chasman D.I. Smith J.D. McCarty C.A. Li X. Wilke R.A. Rieder M.J. Williams P.T. et al.Genome-wide association of lipid-lowering response to statins in combined study populations.PLoS ONE. 2010; 5: e9763Crossref PubMed Scopus (186) Google Scholar, 4Thompson J.F. Man M. Johnson K.J. Wood L.S. Lira M.E. Lloyd D.B. Banerjee P. Milos P.M. Myrand S.P. Paulauskis J. et al.An association study of 43 SNPs in 16 candidate genes with atorvastatin response.Pharmacogenomics J. 2005; 5: 352-358Crossref PubMed Scopus (184) Google Scholar, 5Donnelly L.A. Palmer C.N. Whitley A.L. Lang C.C. Doney A.S. Morris A.D. Donnan P.T. Apolipoprotein E genotypes are associated with lipid-lowering responses to statin treatment in diabetes: a Go-DARTS study.Pharmacogenet. Genomics. 2008; 18: 279-287Crossref PubMed Scopus Google the is that in the APOE gene region are associated with variation in response. we a genome-wide analysis of LDL-c response from two clinical of CARDS and the Anglo-Scandinavian Outcomes Trial M. of and with atorvastatin in patients have or in the Anglo-Scandinavian Cardiac Outcomes a multicentre randomised trial.Lancet. Full Text Full Text PDF PubMed Scopus Google Scholar), to LDL-c response to We to determinants of LDL-c response to atorvastatin among those to atorvastatin in alternative to the of the effect of atorvastatin LDL-c using data from and treatment groups. However, we did not this for is of association and in we did not change LDL-c in the to were with clinical and in with the of for in CARDS have been patients with type 2 diabetes and were to or atorvastatin 10 mg and for a of was serum LDL-c baseline to had to and serum and were at and three and an LDL-c was with the of the of in without of the 18: PubMed Scopus Google Scholar), or serum mmol/l, and the change in was of For this genome-wide the were to those to and the of two LDL-c was the baseline LDL-c and a of post-randomization within the post-randomization was the or treatment with for month and for at and were an with from is the patients of with at three cardiovascular were to one of two in with of or at the had been additional atorvastatin 10 mg or patients the lipid-lowering of the For this genome-wide two from ASCOT were The to 10 mg atorvastatin in whom LDL-c was at the and LDL-c was the of at the and LDL-c was using the in Following the of the there was an observational The all not to 10 mg atorvastatin those to and those not for the were prescribed atorvastatin 10 mg. For LDL-c was the measurement the or to of and LDL-c was the measurement from the of data were from the PROspective Study of Pravastatin in the Elderly at Risk (PROSPER) J. M. et of a study of Pravastatin in the Elderly at Risk Study PROspective Study of Pravastatin in the Elderly at J. Full Text Full Text PDF PubMed Scopus Google was a placebo-controlled trial to treatment with the of in and we and in and the and of were had disease or of disease of or of were to or of to the Pravastatin of the trial were in the and were an at and at and was the The measurement was at baseline and the was the of the to associations and to the for single nucleotide polymorphisms we LDL-c levels to to the association analysis using the for and to the the We the of the in a of the to a We a of the to baseline and LDL-c The and were the of a with of the and response the and with the of the from the two The were in the were in the and in the observational the the in V. et al.Genome-wide association three and Genet. 2009; 5: PubMed Scopus Google Scholar, Li A.L. et associated with a Genet. 2009; PubMed Scopus Google Scholar), the of a the and the the for to a response was in ASCOT or the for in CARDS are The effect in and and the (PROSPER) were that the had studies using to was from was from 10 of using the from was with an in the of a that the of were at for 2 was treatment with was at for was with and at for the was in and in 10 was at were at using a SNP SNPs that had and were We a SNP of for SNPs in the and we that the for a of from was not SNPs for The was with SNPs with a of SNP was of the SNPs were in the analysis of the was was at of We from those had been atorvastatin 10 mg daily, had for and had a SNP the of we with and those with and were (n = LDL-c from at the was compliant with atorvastatin pill were was carried out at in two were had data SNPs were the had been to or in the or were or in or in the combined or had in or in the combined or had in or in the combined the were had and using a using a of SNPs and SNPs were showed in the at were in the combined were not in or had to or showed using in the combined and all the were (n = using analysis A.L. and Genet. 2: PubMed Scopus Google a of SNPs and were for the genome-wide was in the was for was with the and SNPs were for analysis M. D.J. J.J. et of in for Genet. PubMed Scopus Google The CARDS data were combined with from phase to of at all using the J. P. for genome-wide association studies of Genet. PubMed Scopus Google For ASCOT and genotypes at SNPs were using Genet. 2009; PubMed Scopus Google and from For a of was to and to SNPs and using for in Using and genotypes were for the of using for in The A.L. M.E. analysis for in genome-wide association Genet. PubMed Scopus Google was to for Using J. P. M.J. et a for whole-genome association and J. Genet. Full Text Full Text PDF PubMed Scopus Google Scholar), we a of SNPs that are in with in the CARDS analysis was using this of in the analysis were from the of for statin response. The three were and in all of for LDL-c for at and post-randomization were We the post-randomization LDL-c and that the reported APOE at was the association in a genome-wide analysis of response at P= × the to further we the of post-randomization LDL-c to the of the association of response with APOE at the for were combined in a with the a of and the of for with = × J. P. for genome-wide association studies of Genet. PubMed Scopus Google was to for association of response with in a with the post-randomization and with and the three of The was to for of genotypes at the of several post-randomization LDL-c a single to the We the analysis in J. P. M.J. et a for whole-genome association and J. Genet. Full Text Full Text PDF PubMed Scopus Google to for of SNP associations within the a the of that at the SNP was compared with a in the of were The is that the SNP for all associations of with response. those carried out to a of LPA the effect of atorvastatin was carried out the associated with to atorvastatin in a Cox of to and using a a with this treatment effect and one a for of × treatment We the response in Genet. 2009; PubMed Scopus Google Scholar, for genome-wide association analysis of 2010; PubMed Scopus Google is to for in genotypes in with improved Johnson M. V. P. for association genotypes and PubMed Scopus Google and were 10 The response was of of in and top three were The and for the effect were studies to a is the of the to a of the with to the with the that the of to and the three is a of the of For we the of the study and of effect at the most significance levels. the data those at the of was the studies have been We have the of × the for a genome-wide of variation treatment response LDL-c could the LDL-c effect is to for the LDL-c levels and to the of the effect of for the in is measurement and variation in the of we the effect of an is the of for for is the of and is the of for For we 0.8 a for the the in LDL-c two in baseline of participants in the three a of the for association of SNP with LDL-c response to effect the CARDS and ASCOT that the of the most of that there is a of 2 a of the 2 all at the for association a The of effect are for the response and the response was LDL-c with LDL-c a in the is to change in LDL-c with LDL-c a the in and for was in for an that the is associated with a in LDL-c and a response to of patients and studies in the = = = and and and was was at baseline = with LDL-c at the is are with baseline high for LDL-c for for LDL-c at baseline and for response to statin 10 mg 10 mg 10 mg for of SNPs not in for observational of of = with LDL-c at the is are with baseline high for for LDL-c at baseline and for response to statin for of SNPs not in in a of from of all SNPs that The association the of SNPs the genome-wide significance × the analysis ASCOT and ASCOT SNPs that in the the of the genes is of and that the is associated with a LDL-c a lower response to that the is associated with lower LDL-c a response to associated with LDL-c response to statins with of and ASCOT ASCOT of and that the is associated with a LDL-c a lower response to that the is associated with lower LDL-c a response to For SNPs that in the the of the genes is in a observational of of SNPs associated with LDL-c response to statins with of and ASCOT ASCOT The associations were with rs10455872 in the LPA gene and with SNPs in the gene region genome-wide associations were The SNPs in the LPA and APOE region of the in LDL-c response in The most was that for the region this did not genome-wide was of for all the top SNPs reported in the The effect for all top SNPs were similar in all the participants had type 2 and in of the participants had type 2 diabetes that diabetes was not a of the effect of the top the LPA gene rs10455872 showed a genome-wide association with LDL-c response The effect at rs10455872 was the in 2 is not the in for example, the percentage change in LDL-c with statin therapy at one month was in those with at one at rs10455872 = compared with in for the was effect of this SNP change in LDL-c post-randomization in those in the (P= the association with LPA we that LPA genotypes serum lipoprotein(a) Lp(a) had been in CARDS not in the of the for serum Lp(a) all associated were in the LPA with M. M. J. Li J. et analysis of candidate gene the LPA with Lp(a) levels in participants of Full Text Full Text PDF PubMed Scopus Google a that and SNPs in the LPA region had serum Lp(a) SNPs variation in the serum Lp(a) levels in most of this was to the rs10455872 SNP with levels and in those with and We the association of at SNPs in the LPA gene with LDL-c response to statin for serum Lp(a) levels in the CARDS data to the are to the effect of LPA serum Lp(a) levels The of the at the associated SNP in LPA in CARDS was from to with the effect of apparent response to statin Lp(a) levels. We that Lp(a) levels had an association with apparent LDL-c response to statin in (P= analysis in CARDS that there was effect of statin Lp(a) was to for in Lp(a) levels with atorvastatin at one for baseline and of for serum Lp(a) levels in Lp(a) Lp(a) that in the the of the genes is APOE of for serum Lp(a) levels in CARDS for SNPs associated with an LDL-c response to statins with a of For that in the the of the genes is in a of for serum Lp(a) levels in CARDS for SNPs associated with an LDL-c response to statins with a of serum Lp(a) levels alter of statin therapy we there was of from a of a high serum Lp(a) levels and atorvastatin in The for associated with statin was among those in the top quartile for serum Lp(a) compared with among those with serum Lp(a) this for P= was there of rs10455872 at LPA and atorvastatin for (P= for the of at rs10455872 the associated with atorvastatin in those for the was and the in those with at one was However, the to an was there were 16 among the trial participants with at one of the We the effect of LPA SNP in participants in the trial to of this of rs10455872 was associated with lower response to statins with a of P= The combined for the three studies was SNPs in the gene region reached genome-wide significance for statin response The effect LDL-c response to statin therapy associated with SNPs at in this region was in CARDS for the change in LDL-c with statin therapy at one month was in those with at one at compared with in and was in those with at one at compared with in were of the effect of at rs10455872 in LPA and of Lp(a) levels. the CARDS we that there was effect of SNPs change in LDL-c in those in the (P= We the in this region could accounted for the of to APOE SNP and respectively, at and The of the at the ∊2 with variants, of the at the ∊4 with two were not and could not are not in the The at we found to associated with a statin response is in with the at with a reported of is a for the ∊2 M. J. M. et genome-wide association of cardiovascular disease in the Genet. 2010; PubMed Scopus Google The at was associated with lower response to is in with the at with reported of with a of for gene and levels in disease among 2010; PubMed Google is a for the ∊4 two SNPs are in the and could in this analysis with percentage of and we for of SNP a for or a for the ∊4 the additional percentage is with and = P= the for the additional percentage is with and = P= that ∊2 for most of the in response at this associations of LDL-c response with APOE and genome-wide associations were The most were those in the gene region several SNPs had for 10 most to of the gene with one SNP within a for association with statin response of × that are to baseline and of the effect of at the SNPs for the APOE region LPA and in the CARDS the SNP without for baseline the apparent LDL-c response to statins that in of the effect of for baseline However, effect further measurement at baseline the apparent effect to there is effect of baseline to measurement the SNP and at the LPA the effect of baseline LDL-c is and the for measurement the association of LDL-c for baseline for baseline for measurement for baseline and for measurement and without for measurement in baseline in a and without for measurement in baseline reported associated with statin response in the associated with response to and associated with response to genes were not associated at with LDL-c response to statin in this study at an accepted genome-wide association or even at to However, and were at a of We have from the and of in the study of the and the studies in of the this genome-wide association study of LDL-c response to atorvastatin we that those with genotypes in the LPA gene that lead to Lp(a) levels have an apparently lower LDL-c response to and we the reported association of a response to statin in those with the at the The top three SNPs in the rs10455872 in LPA and the and variants, in the LDL-c response to statin is that that studies SNPs with effect or that there are at not is a of a with one of and an additional a levels of Lp(a) have a for example, in the serum Lp(a) was and with high of in Lp(a) levels been reported the type 2 in is to in the of in Lp(a) to genes is to the of the to is the with the we J.J. of the and and and PubMed Scopus Google measurement in data from the study the measurement is to of J.J. C.C. L.A. V. et and disease in the Full Text Full Text PDF PubMed Scopus Google The associated with the and high Lp(a) levels have been reported to associated with cardiovascular in several studies J. and the of and 2009; PubMed Scopus Google and of cardiovascular and analysis of Google Scholar, M.J. J. P. et a cardiovascular J. 2010; PubMed Scopus Google Scholar, V. for PubMed Scopus Google Scholar, J.F. et associated with Lp(a) and J. 2009; PubMed Scopus Google emphasize the of high Lp(a) and with M.J. J. P. et a cardiovascular J. 2010; PubMed Scopus Google The rs10455872 SNP that we found associated with LDL-c response is in with the in Lp(a) R.A. Comprehensive analysis of variation in the LPA and to lipoprotein(a) in and Cardiovasc. Genet. 2010; PubMed Scopus Google with variation at rs10455872 accounted for of in Lp(a) in the CARDS However, the for the apparently lower LDL-c response in those with genotypes associated with high Lp(a) in understanding LDL-c estimation The LDL-c levels from and and the that in Lp(a). For most this is of of is is to in Lp(a). However, is that of apparent LDL-c in Lp(a) Lp(a) levels are in the and Lp(a) is of serum lipoprotein(a) estimation of the PubMed Scopus Google we in the CARDS statin therapy did not lower Lp(a) levels. had an of carried Lp(a) had in and in Lp(a) For LDL-c response apparent LDL-c LDL-c Lp(a) levels. been in the of P. J. M. and in lipid-lowering 2004; Full Text Full Text PDF PubMed Scopus Google and been et J. the of statin therapy in with high lipoprotein(a) PubMed Scopus Google that those with at one of the at rs10455872 have a percentage lower apparent statin response in and in and that this association for Lp(a) levels is with the effect of the statin response is this for of in Lp(a) levels. The data a clinical that with Lp(a) levels for have a lower apparent response to statin therapy that an apparently lower LDL-c response to statin may an for Lp(a) levels. However, we that similar from with atorvastatin therapy was found in those with and without Lp(a). is accepted that Lp(a) J. and the of and 2009; PubMed Scopus Google Scholar, and of cardiovascular and analysis of Google Therefore, is Lp(a) levels are not and statin LDL-c statin therapy in with high Lp(a). We the association of LPA SNP that the association of with a lower response to statins in an of to mg of is the of a of response to statin treatment the APOE region in a genome-wide association We replicated the that at the is associated with variation in statin response. at one at is in with the the ∊2 was associated with baseline LDL-c and response to the for ∊4 was associated with lower LDL-c response to The analysis most of the variation is to the of ∊2 to the of ∊4 this is not that there is in the and that the SNPs are in whom a of is ∊2 are to of (2Thompson J.F. Hyde C.L. Wood L.S. Paciga S.A. Hinds D.A. Cox D.R. Hovingh G.K. Kastelein J.J. Comprehensive whole-genome and candidate gene analysis for response to statin therapy in the Treating to New Targets (TNT) cohort.Circ. Cardiovasc. Genet. 2009; 2: 173-181Crossref PubMed Scopus (164) Google Scholar), and in there may and to apparent LDL-c in those patients with an are at We found of an association statin response and in the gene this did not the genome-wide significance of in studies a for 10 the to the for to statin response remains to of is to of of and the of in the of a Full Text PDF PubMed Google Two of LDL-c response to statin therapy have been reported (2Thompson J.F. Hyde C.L. Wood L.S. Paciga S.A. Hinds D.A. Cox D.R. Hovingh G.K. Kastelein J.J. Comprehensive whole-genome and candidate gene analysis for response to statin therapy in the Treating to New Targets (TNT) cohort.Circ. Cardiovasc. Genet. 2009; 2: 173-181Crossref PubMed Scopus (164) Google Scholar, M.J. Mangravite L.M. Hyde C.L. Chasman D.I. Smith J.D. McCarty C.A. Li X. Wilke R.A. Rieder M.J. Williams P.T. et al.Genome-wide association of lipid-lowering response to statins in combined study populations.PLoS ONE. 2010; 5: e9763Crossref PubMed Scopus (186) Google the Treating to New Targets (TNT) trial with with a genome-wide there were at for association were in a of for candidate gene associations three SNPs in APOE and one SNP in reached genome-wide significance (2Thompson J.F. Hyde C.L. Wood L.S. Paciga S.A. Hinds D.A. Cox D.R. Hovingh G.K. Kastelein J.J. Comprehensive whole-genome and candidate gene analysis for response to statin therapy in the Treating to New Targets (TNT) cohort.Circ. Cardiovasc. Genet. 2009; 2: 173-181Crossref PubMed Scopus (164) Google a of three that M.J. Mangravite L.M. Hyde C.L. Chasman D.I. Smith J.D. McCarty C.A. Li X. Wilke R.A. Rieder M.J. Williams P.T. et al.Genome-wide association of lipid-lowering response to statins in combined study populations.PLoS ONE. 2010; 5: e9763Crossref PubMed Scopus (186) Google a SNP in the gene was at for association with and there was a association with SNP in APOE. the in the gene was to associated with change in gene have that a is associated with lipid-lowering response to treatment L.M. J. Smith J.D. Rieder M.J. X. D.A. of and variation lipid-lowering response to 2010; PubMed Scopus Google the gene polymorphisms were associated with LDL-c and with LDL-c response to The association of gene with statin response was reported in a of patients with diabetes L.A. Doney A.S. J. Whitley A.L. Lang C.C. Morris A.D. Donnan P.T. Palmer C.N. in the gene is associated with lipid-lowering response to statin treatment in diabetes: a study.Pharmacogenet. Genomics. 2008; 18: PubMed Scopus Google were in in a carried out in of a in have been reported to have from with to not with to or response et of the in with and disease in 2 the and 2008; PubMed Scopus Google association of the gene with statin response been reported J.F. Man M. Johnson K.J. Wood L.S. Lira M.E. Lloyd D.B. Banerjee P. Milos P.M. Myrand S.P. Paulauskis J. et al.An association study of 43 SNPs in 16 candidate genes with atorvastatin response.Pharmacogenomics J. 2005; 5: 352-358Crossref PubMed Scopus (184) Google Scholar, variation in the gene is associated with the effect of in 2009; PubMed Scopus Google from the APOE of associations were replicated we that the in this study are of the of further the of response to statin therapy may not in from statins in identifying therapeutic The the the and the participants of the CARDS and ASCOT studies for data for the M. Neil and J. M. M. and J. M. M. and and for a of and with Anglo-Scandinavian Cardiac Outcomes Trial Collaborative Atorvastatin Diabetes Study cardiovascular disease genome-wide association study lipoprotein(a) PROspective Study of Pravastatin in the Elderly at Risk single nucleotide
Deshmukh et al. (Sun,) studied this question.