Author + information
- Received December 27, 2016
- Revision received February 3, 2017
- Accepted February 7, 2017
- Published online March 29, 2017.
- Nathan O. Stitziel, MD, PhDa,∗ (, )
- Amit V. Khera, MDb,c,d,
- Xiao Wang, PhDe,
- Andrew J. Bierhals, MD, MPHf,
- A. Christina Vourakis, BAg,
- Alexandra E. Sperry, BAg,
- Pradeep Natarajan, MDb,c,d,
- Derek Klarin, MDb,c,h,
- Connor A. Emdin, DPhilb,c,d,
- Seyedeh M. Zekavat, BScd,
- Akihiro Nomura, MDb,c,d,
- Jeanette Erdmann, PhDi,j,
- Heribert Schunkert, MDk,l,
- Nilesh J. Samani, MDm,n,
- William E. Kraus, MDo,
- Svati H. Shah, MD, MPHo,
- Bing Yu, PhDp,q,
- Eric Boerwinkle, PhDp,q,
- Daniel J. Rader, MDe,r,
- Namrata Gupta, PhDd,
- Philippe M. Frossard, PhDs,
- Asif Rasheed, MBBSs,
- John Danesh, DPhilt,u,v,
- Eric S. Lander, PhDd,
- Stacey Gabriel, PhDd,
- Danish Saleheen, MBBS, PhDs,w,
- Kiran Musunuru, MD, PhD, MPHe,∗∗ (, )
- Sekar Kathiresan, MDb,c,d,∗∗∗ (, )
- PROMIS and Myocardial Infarction Genetics Consortium Investigators
- aCardiovascular Division, Department of Medicine, Department of Genetics, and McDonnell Genome Institute, Washington University School of Medicine, St. Louis, Missouri
- bCenter for Human Genetic Research, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- cCardiovascular Research Center and Cardiology Division, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- dProgram in Medical and Population Genetics, Broad Institute, Cambridge, Massachusetts
- eCardiovascular Institute, Division of Cardiovascular Medicine, Department of Medicine, and Department of Genetics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
- fMallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, Missouri
- gHarvard College, Harvard University, Cambridge, Massachusetts
- hDepartment of Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- iInstitute for Integrative and Experimental Genomics, University of Lübeck, Lübeck, Germany
- jDZHK (German Centre for Cardiovascular Research), partner site Hamburg/Lübeck/Kiel, Lübeck, Germany
- kDeutsches Herzzentrum München, Technische Universität München, Munich, Germany
- lDZHK (German Centre for Cardiovascular Research), partner site Munich Heart Alliance, Munich, Germany
- mDepartment of Cardiovascular Sciences, University of Leicester, Leicester, United Kingdom
- nNIHR Leicester Cardiovascular Biomedical Research Unit, Glenfield Hospital, Leicester, United Kingdom
- oDuke Molecular Physiology Institute and the Division of Cardiology, Department of Medicine, Duke University, Durham, North Carolina
- pHuman Genetics Center, The University of Texas Health Science Center at Houston, Houston, Texas
- qHuman Genome Sequencing Center, Baylor College of Medicine, Houston, Texas
- rInstitute of Translational Medicine and Therapeutics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
- sCenter for Non-Communicable Diseases, Karachi, Pakistan
- tCardiovascular Epidemiology Unit, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom
- uWellcome Trust Sanger Institute, Hinxton, Cambridge, United Kingdom
- vNational Institute of Health Research Blood and Transplant Research Unit in Donor Health and Genomics, University of Cambridge, Cambridge, United Kingdom
- wDepartment of Biostatistics and Epidemiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
- ↵∗Address for correspondence:
Dr. Nathan Stitziel, Cardiovascular Division, Washington University, 660 South Euclid Avenue, Campus Box 8086, Saint Louis, Missouri 63110.
- ↵∗∗Dr. Kiran Musunuru, University of Pennsylvania, 3400 Civic Center Boulevard, Bldg 421, 11-104 Smilow Center for Translational Research, Philadelphia, Pennsylvania 19104.
- ↵∗∗∗Dr. Sekar Kathiresan, Center for Genomic Medicine, Massachusetts General Hospital, Simches Research Center, 185 Cambridge Street, CPZN 5.830, Boston, Massachusetts 02114.
Background Familial combined hypolipidemia, a Mendelian condition characterized by substantial reductions in all 3 major lipid fractions, is caused by mutations that inactivate the gene angiopoietin-like 3 (ANGPTL3). Whether ANGPTL3 deficiency reduces risk of coronary artery disease (CAD) is unknown.
Objectives The study goal was to leverage 3 distinct lines of evidence—a family that included individuals with complete (compound heterozygote) ANGPTL3 deficiency, a population based-study of humans with partial (heterozygote) ANGPTL3 deficiency, and biomarker levels in patients with myocardial infarction (MI)—to test whether ANGPTL3 deficiency is associated with lower risk for CAD.
Methods We assessed coronary atherosclerotic burden in 3 individuals with complete ANGPTL3 deficiency and 3 wild-type first-degree relatives using computed tomography angiography. In the population, ANGPTL3 loss-of-function (LOF) mutations were ascertained in up to 21,980 people with CAD and 158,200 control subjects. LOF mutations were defined as nonsense, frameshift, and splice-site variants, along with missense variants resulting in <25% of wild-type ANGPTL3 activity in a mouse model. In a biomarker study, circulating ANGPTL3 concentration was measured in 1,493 people who presented with MI and 3,232 control subjects.
Results The 3 individuals with complete ANGPTL3 deficiency showed no evidence of coronary atherosclerotic plaque. ANGPTL3 gene sequencing demonstrated that approximately 1 in 309 people was a heterozygous carrier for an LOF mutation. Compared with those without mutation, heterozygous carriers of ANGPTL3 LOF mutations demonstrated a 17% reduction in circulating triglycerides and a 12% reduction in low-density lipoprotein cholesterol. Carrier status was associated with a 34% reduction in odds of CAD (odds ratio: 0.66; 95% confidence interval: 0.44 to 0.98; p = 0.04). Individuals in the lowest tertile of circulating ANGPTL3 concentrations, compared with the highest, had reduced odds of MI (adjusted odds ratio: 0.65; 95% confidence interval: 0.55 to 0.77; p < 0.001).
Conclusions ANGPTL3 deficiency is associated with protection from CAD.
Loss-of-function (LOF) mutations leading to complete deficiency of angiopoietin-like 3 (ANGPTL3) cause familial combined hypolipidemia, a Mendelian disorder characterized by low circulating concentrations of low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides (TG) (1). ANGPTL3 is a hepatically secreted protein first identified via positional cloning of a hypolipidemic mouse strain (2). ANGPTL3 acts as a potent inhibitor of lipoprotein lipase (LPL), the primary mechanism by which triglyceride-rich lipoproteins are cleared from the circulation (3). In addition, ANGPTL3 is an endogenous inhibitor of endothelial lipase (EL) (4). Loss of ANGPTL3 function appears to decrease triglyceride-rich lipoprotein and HDL cholesterol concentrations through loss of LPL and EL inhibition, respectively. The mechanism by which ANGPTL3 regulates LDL cholesterol metabolism remains unclear (5). The seemingly favorable implications of ANGPTL3 deficiency in reducing TG concentrations and circulating LDL cholesterol catalyzed drug development programs aiming to inhibit ANGPTL3 with either a monoclonal antibody (5) or an antisense oligonucleotide (6).
Although decreased atherosclerotic burden was observed in Angptl3 knockout mice (7), the relationship of ANGPTL3 deficiency to coronary artery disease (CAD) in humans remains uncertain. Individuals who carry LOF mutations in ANGPTL3 have lifelong reductions of circulating ANGPTL3 (8); as such, the clinical phenotypes of these individuals may inform the potential therapeutic efficacy of pharmacological ANGPTL3 inhibition.
Here, we tested the hypothesis that ANGPTL3 deficiency reduces risk of CAD in humans. We compared coronary atherosclerotic plaque burden in individuals who had complete ANGPTL3 deficiency (caused by compound heterozygous LOF mutations in ANGPTL3) with wild-type first-degree relatives. Next, we examined the coding regions of ANGPTL3 in up to 180,180 individuals, identified those who carried LOF mutations in this gene with the aid of mouse models, and determined whether those mutations were associated with a lower risk of CAD. Finally, we measured circulating ANGPTL3 concentrations in individuals presenting with a first-ever myocardial infarction (MI) and compared them to concentrations in individuals without MI.
The original kindred used to map ANGPTL3 as a cause of familial combined hypolipidemia was recruited in the Lipid Research Clinic at the Washington University School of Medicine between 1994 and 1997. We recontacted all individuals from the kindred who inherited compound heterozygous LOF mutations in ANGPTL3 and invited them to participate in the current study. Three of the 4 compound heterozygous carriers (individuals II-1, II-2, and II-4 in Online Figure 1) were available to participate and were matched to 3 first-degree relatives who did not carry any ANGPTL3 LOF mutation (individuals II-8, II-7, and II-10 in Online Figure 1, respectively). Fasting laboratory values, including plasma lipids, were measured in all participants with standard clinical assays. Coronary computed tomography angiography (CCTA) was used to quantify coronary artery calcification and atherosclerotic plaque burden. Details of the image acquisition and post-acquisition processing are included in the Online Appendix. The Washington University School of Medicine Institutional Review Board approved all study protocols.
Ascertainment of LOF mutations in ANGPTL3
We identified carriers of an LOF mutation in ANGPTL3 using previously generated exome sequencing data from 9 case-control studies of the Myocardial Infarction Genetics Consortium. These included the ATVB (Italian Atherosclerosis Thrombosis and Vascular Biology) study (9), the ESP-EOMI (Exome Sequencing Project Early-Onset Myocardial Infarction) study (10), the South German Myocardial Infarction study (11), the OHS (Ottawa Heart Study) (12), PROCARDIS (Precocious Coronary Artery Disease Study) (13), PROMIS (Pakistan Risk of Myocardial Infarction Study) (14), the Registre Gironi del COR (Gerona Heart Registry or REGICOR) study (15), the BHF-FHS (British Heart Foundation Family Heart Study) (16), and the Lubeck Myocardial Infarction study (16). Furthermore, we extracted ANGPTL3 sequence data from exome sequencing performed in the Jackson Heart Study (17), the BioImage study (18), and the ARIC (Atherosclerosis Risk In Communities) population-based cohort study (19) in addition to targeted sequencing in the Duke CATHGEN case-control study (20). LOF mutations included those leading to truncation via a premature stop codon (nonsense), insertions or deletions that scramble protein translation beyond the variant site (frameshift), or point mutations at sites of pre-messenger ribonucleic acid splicing that alter the splicing process (splice site). Additional data for rs372257803, an intronic splice region variant in ANGPTL3 previously linked to significantly reduced circulating TG levels (21), was obtained by high-quality genotype imputation in the United Kingdom Biobank (22), the PennCath study (23), and the Wellcome Trust Case Control Consortium Coronary Artery Disease study (24). All variant positions were based on the ANGPTL3 canonical transcript (ENST00000371129). Additional details on gene sequencing, the imputation of rs372257803, and study cohorts are included in the Online Methods and Online Table 1.
Functional validation of missense variants in ANGPTL3 leading to LOF
Beyond the mutations that lead to LOF due to nonsense, frameshift, or splice-site disruption, studies based on evolutionary conservation have suggested that approximately 20% of all missense mutations lead to severe decrements in protein function (25,26). We sought to experimentally define such variants in ANGPTL3 using a mouse model. Rare (minor allele frequency <1%) missense variants were prioritized if they were 1) predicted to be damaging or possibly damaging by each of 5 in silico prediction algorithms (LRT [likelihood ratio test] score, MutationTaster, PolyPhen-2 HumDiv, PolyPhen-2 HumVar, and Sorting Intolerant From Tolerant) and 2) present in at least 2 sequenced individuals of the Myocardial Infarction Genetics Consortium cohorts.
For the set of ANGPTL3 missense variants identified previously, the functional significance of each variant was determined with adenoviral vectors developed to reconstitute the expression of the human ANGPTL3 ortholog in the livers of Angptl3 knockout mice. Vectors were engineered to contain either the wild-type ANGPTL3 gene or the missense variant of interest. Missense variants were annotated as LOF if they conferred <25% of wild-type activity as assessed by percent change in circulating TG levels and percent change in circulating cholesterol levels induced by expression. Additional details are described in the Online Appendix.
ANGPTL3 plasma concentration
Using a previously validated enzyme-linked immunosorbent assay (BioVendor, Prague, Czech Republic) (27), plasma ANGPTL3 concentrations were measured in individuals from PROMIS (14), a study that included cases presenting with a first-ever MI and control subjects free of MI.
For changes in circulating TG and total cholesterol levels in the mouse models, normality of data was assessed with the Shapiro-Wilk test and equality of variance was assessed with the F test; p values were calculated with 2-sided paired-samples Student t tests. The association of ANGPTL3 LOF mutations, analyzed in aggregate, with total cholesterol, LDL cholesterol, HDL cholesterol, and log-transformed TGs was assessed by linear regression with adjustment for the covariates of age, age squared, sex, study cohort, CAD status, and the first 5 principal components of ancestry. We accounted for the effect of lipid-lowering therapy in participants reporting such use at the time of lipid measurement by dividing the measured total cholesterol and LDL cholesterol by 0.8 and 0.7, respectively (28,29); HDL cholesterol and TG values were not adjusted. The association of ANGPTL3 mutations with risk of CAD was determined via meta-analysis with Cochran–Mantel–Haenszel statistics for stratified 2-by-2 tables as implemented previously (30). In calculating the study-specific odds ratio of disease, an adjustment of 0.5 was added to all counts in studies with zero mutation carriers in cases or controls. The association of circulating plasma ANGPTL3 concentration with MI was determined by multivariable logistic regression after stratification of the population into tertiles of ANGPTL3 concentration. Statistical analyses were performed with R version 3.2.2 software (The R Project for Statistical Computing, Vienna, Austria).
From the original kindred used to map this gene as a cause of familial combined hypolipidemia (1), we studied 3 individuals with complete ANGPTL3 deficiency due to compound heterozygous LOF mutations in ANGPTL3 and 3 matched first-degree relatives without an LOF ANGPTL3 mutation. As shown in Online Table 2, participants with complete ANGPTL3 deficiency continued to exhibit very low plasma lipid concentrations nearly 20 years after the initial report. An updated medical history was obtained. One participant with complete ANGPTL3 deficiency reported a history of type 2 diabetes mellitus, hypertension, and past tobacco use. Other characteristics and laboratory values are listed in Online Table 2. We performed CCTA in all 6 individuals. The coronary calcium score was 0 Agatston units (AU) for all participants with complete ANGPTL3 deficiency (Figure 1A). By contrast, 2 of the 3 matched control subjects had positive coronary calcium scores (6 AU for individual II-8 and 610 AU for individual II-7) (Figure 1B).
We next calculated total plaque burden (a combination of both calcified and noncalcified plaque) for each participant. The total plaque burden was lower in the participants with complete ANGPTL3 deficiency (mean = 0%) than in control subjects (mean = 39%) (Figure 1C, Online Table 2, Online Figure 2). The small number of phenotyped individuals precluded robust statistical comparisons between groups.
Ascertainment of ANGPTL3 LOF mutations
We next sought to characterize the clinical effects of ANGPTL3 LOF mutations in the population. Sequence data for ANGPTL3 were available in 13,914 individuals with CAD and 26,198 control subjects free of CAD. From these data, 21 LOF variants were identified, including 7 variants leading to premature stop codons, 2 variants predicted to disrupt splicing, and 12 frameshift indels (Online Table 3). Eleven rare missense variants underwent functional validation in a mouse model, of which 2 (p.Asp42Asn and p.Thr383Ser) were additionally included as validated LOF variants (Figure 2).
In aggregate across all sequencing studies, an ANGPTL3 LOF mutation was identified in 130 of 40,112 participants (0.32%; 95% confidence interval [CI]: 0.27% to 0.39%). One homozygote was identified with a Gln192ArgfsTer5 frameshift mutation, a 56-year-old woman of African ancestry free of clinical CAD with LDL cholesterol of 112 mg/dl, HDL cholesterol of 44 mg/dl, and TGs of 56 mg/dl.
Among sequenced individuals of European ancestry, the most frequently observed inactivating variant was the intronic splice region variant rs372257803 (minor allele frequency = 0.17%). This variant was imputed in an additional 8,066 CAD case subjects and 140,068 control subjects, identifying an additional 68 heterozygous carriers of an ANGPTL3 LOF mutation.
Association of ANGPTL3 LOF mutations with circulating lipid levels and CAD risk
Plasma lipid levels were available in up to 20,092 people in the Myocardial Infarction Genetics Consortium studies, including 60 heterozygous carriers of an ANGPTL3 LOF mutation. We found that individuals carrying an LOF ANGPTL3 mutation, compared with noncarriers, had 11% lower total cholesterol (p = 0.0008), 12% lower LDL cholesterol (p = 0.04), and 17% lower TGs (p = 0.01) (Table 1). HDL cholesterol was not significantly different between the groups (p = 0.17).
A cohort-based meta-analysis stratified by ancestry was performed to determine the relationship between LOF mutations in ANGPTL3 and risk of CAD (Figure 3, Online Table 3). We observed a 34% reduced risk of CAD among carriers of an ANGPTL3 LOF mutation compared with noncarriers (odds ratio [OR] of disease for carriers: 0.66; 95% CI: 0.44 to 0.98; p = 0.04). This effect estimate was similar in a sensitivity analysis restricted to individuals in whom complete gene sequencing (as opposed to rs372257803 imputation) was performed (OR: 0.70; 95% CI: 0.46 to 1.06; p = 0.09).
Circulating plasma ANGPTL3 and risk of MI
Protection from CAD among carriers of a rare LOF mutation in ANGPTL3 led to the hypothesis that individuals with lower levels of circulating ANGPTL3 protein might similarly have reduced coronary risk. Plasma ANGPTL3 concentrations were measured in 1,493 case subjects presenting with a first-ever MI and 3,231 control subjects free of CAD from the PROMIS study. Consistent with our expectations, individuals in the lowest tertile of ANGPTL3 concentrations had significantly reduced risk of MI compared with those in the highest tertile (adjusted OR: 0.65; p = 2.2 × 10−7) (Table 2). This finding was modestly attenuated after additional adjustment for observed plasma LDL cholesterol and TGs (adjusted OR: 0.71; p = 0.0001).
We have provided multiple lines of evidence suggesting that ANGPTL3 deficiency is associated with protection from CAD (Central Illustration). Detailed atherosclerotic phenotyping demonstrated an absence of coronary atherosclerotic plaque in individuals with complete ANGPTL3 deficiency. Genomic analysis of ANGPTL3 LOF variants, including functionally validated missense variants in up to 180,180 people, showed a 34% reduction in risk of CAD among heterozygous carriers. Finally, circulating ANGPTL3 concentrations were lower in healthy control subjects than in those presenting with MI.
These results permitted several conclusions. First, identifying families with extreme phenotypes of interest could facilitate both gene discovery and hypothesis-based phenotyping. Multiple independent groups have confirmed the impact of inactivating mutations in ANGPTL3 on decreasing lipid levels using family-based study designs (31–34). Here, we extended these observations by demonstrating that individuals with complete deficiency due to 2 inactivating mutations in the gene (effectively “human knockouts” for ANGPTL3) tended to have less coronary atherosclerosis as assessed by CCTA. This apparent protection from coronary atherosclerosis extended to a middle-aged participant (Online Table 2, individual II-1) with significant cardiovascular risk factors of type 2 diabetes mellitus, hypertension, and a history of cigarette smoking. Although suggestive, these results were based on a small number of family members. Large-scale gene sequencing in the population as presented here was used to confirm this observation.
Additionally, these findings lend support to ongoing drug development efforts focused on ANGPTL3 inhibition as a therapeutic strategy. Beyond a significant reduction in plasma LDL cholesterol and TG concentrations, heterozygous ANGPTL3 LOF mutation carriers had a 34% decreased risk of CAD. Similar results were noted in a preliminary report from Dewey and colleagues (35). We also found that individuals with circulating plasma ANGPTL3 concentrations in the lowest tertile of the population (in a sense, mimicking the effect of pharmacological inhibition of ANGPTL3) had a 35% reduced risk of MI. This study adds ANGPTL3 to the list of therapeutic targets for coronary disease, which includes ANGPTL4 (16,36), APOC3 (11,37), LPA (38), NPC1L1 (30), and PCSK9 (39), that have been validated by finding LOF mutations that associate with protection from disease, highlighting the promise and potential of human genetic studies in identifying such targets (40).
Furthermore, these data add to a growing body of human genetics evidence linking regulation of lipoprotein lipase activity, the major mechanism by which circulating TG-rich lipoproteins are hydrolyzed, with atherosclerosis. Studies in both mice and humans have demonstrated ANGPTL3 to be a potent inhibitor of LPL, particularly in the post-prandial state (41). Consistent with effects of rare mutations in 3 additional endogenous regulators of LPL, APOA5 (10), APOC3 (11,37), and ANGPTL4 (16,36), loss of ANGPTL3 function appears to result in gain of LPL activity, reduced TG-rich lipoproteins, and protection from coronary disease. Beyond effects on TGs and LDL cholesterol, ANGPTL3 might affect glucose homeostasis and remodeling of HDL cholesterol particles by EL (4,42). The relative contributions of each of these mechanisms, as they relate to risk of CAD, warrant additional investigation.
Finally, we have provided proof-of-concept for rare ANGPTL3 missense variant prioritization using a combination of bioinformatics tools and experimental characterization in vivo. Any given ANGPTL3 missense variant might perturb protein function via numerous potential mechanisms, including decreased expression, impaired hepatic secretion, or inability to bind and inhibit LPL. We developed an Angptl3 knockout mouse that exhibited a phenotype of very low TGs. We attempted to rescue this phenotype using adenoviral vectors producing either wild-type ANGPTL3 or a protein that included the missense variant of interest. This proved useful in determining that only 2 of 11 screened missense variants led to near complete loss of ANGPTL3 protein function (i.e., <25% of wild-type activity). The ability to confidently annotate the functional consequences of rare missense variants in a gene of interest remains one of the biggest hurdles to rare variant gene discovery efforts (43). We present one approach to address this challenge.
First, CCTA was performed in only a subset of the original family used to identify ANGPTL3 as the cause of familial combined hypolipidemia. Second, the functional characterization of missense variants was performed in only 11 variants, and alternative thresholds for defining LOF are possible. Third, genotype imputation was used to identify carriers of an ANGPTL3 splice-site imputation in some cohorts; however, a sensitivity analysis that was restricted to those studies in which complete sequencing of ANGPTL3 was available yielded similar results.
Deep phenotyping in a family, gene sequencing in the population, and biomarker analysis in case and control subjects showed ANGPTL3 deficiency to be associated with a reduced risk of CAD. Whether pharmacological inhibition of ANGPTL3 function will prove useful in the treatment or prevention of CAD remains to be determined.
COMPETENCY IN MEDICAL KNOWLEDGE: Hereditary combined hypolipidemia is caused by mutations that inactivate the gene ANGPTL3 and is characterized by low blood levels of all 3 major lipid fractions. Individuals with a loss-of-function ANGPTL3 mutation have reduced odds of CAD, which suggests that ANGPTL3 deficiency protects against CAD.
TRANSLATIONAL OUTLOOK: Interventions targeting ANGPTL3 should be considered for potential clinical application.
The authors are very grateful to the family presented here for participating in this study and for their ongoing contributions to furthering biomedical research. The authors also thank Ms. Teresa Roediger for coordinating the clinical imaging portion of the study.
For a supplemental Methods section as well as supplemental figures and tables, please see the online version of this article.
This study was supported by grants from the National Heart, Lung, and Blood Institute (NHLBI) (R01HL131961 and K08HL114642 to Dr. Stitziel; R01 HL118744 to Dr. Musunuru; and R01 HL127564 and R21 HL120781 to Dr. Kathiresan); the National Human Genome Research Institute (U54HG003067 to Dr. Gabriel and Dr. Lander, UM1HG008895 to Dr. Kathiresan Dr. Gabriel, and Dr. Lander, and UM1HG008853 to Dr. Stitziel); the Barnes-Jewish Hospital Foundation (Dr. Stitziel); the Fannie Cox Prize for Excellence in Science Teaching, Harvard University (Dr. Musunuru); and the MGH Research Scholar Award (Dr. Kathiresan). PROMIS fieldwork has been supported through grants awarded to Dr. Saleheen, Dr. Danesh, and Dr. Frossard. Biomarker assays in PROMIS have been funded through grants awarded by the NHLBI (RC2HL101834 and RC1TW008485) and Fogarty International (RC1TW008485). This work was funded by the National Institutes of Health (NIH), which had no involvement in the design and conduct of the study, in the collection, analysis, and interpretation of the data, or in the preparation, review, and approval of the manuscript. Dr. Stitziel has received a research grant from AstraZeneca; and consulting fees from Aegerion Pharmaceuticals. Dr. Khera is supported by an American College of Cardiology Foundation/Merck Cardiovascular Research Fellowship, a John S. Ladue Memorial Fellowship at Harvard Medical School, and a KL2/Catalyst Medical Research Investigator Training award from Harvard Catalyst funded by the NIH (TR001100); and has received consulting fees from Merck and Amarin. Dr. Klarin is supported by the NHLBI (T32 HL007734). Dr. Samani is supported by the British Heart Foundation and is a National Institute for Health Research Senior Investigator. Dr. Rader has received consulting fees from Aegerion Pharmaceuticals, Alnylam Pharmaceuticals, Eli Lilly and Company, Pfizer, Sanofi, and Novartis; is an inventor on a patent related to lomitapide that is owned by the University of Pennsylvania and licensed to Aegerion Pharmaceuticals; and is a cofounder of Vascular Strategies and Staten Biotechnology. Dr. Danesh has received funding from the UK Medical Research Council, British Heart Foundation, UK National Institute of Health Research (NIHR), NIHR Cambridge Comprehensive Biomedical Research, European Commission Framework Programme, European Research Council, GlaxoSmithKline, Merck, NHLBI, NHS Blood and Transplant, Novartis, Pfizer, Wellcome Trust, and UK Biobank. Dr. Saleheen has received grants from Pfizer, Regeneron, Eli Lilly, and Genentech. Dr. Kathiresan has received grants from Bayer Healthcare, Aegerion Pharmaceuticals, and Regeneron Pharmaceuticals; and consulting fees from Merck, Novartis, Sanofi, AstraZeneca, Alnylam Pharmaceuticals, Leerink Partners, Noble Insights, Quest Diagnostics, Genomics PLC, and Eli Lilly and Company; and holds equity in San Therapeutics and Catabasis Pharmaceuticals. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Stitziel, Khera, Wang, Saleheen, Musunuru, and Kathiresan contributed equally to this work.
- Abbreviations and Acronyms
- angiopoietin-like 3
- Agatston units
- coronary artery disease
- coronary computed tomography angiography
- confidence interval
- endothelial lipase
- high-density lipoprotein
- low-density lipoprotein
- loss of function
- lipoprotein lipase
- myocardial infarction
- odds ratio
- Received December 27, 2016.
- Revision received February 3, 2017.
- Accepted February 7, 2017.
- 2017 American College of Cardiology Foundation
- Shimizugawa T.,
- Ono M.,
- Shimamura M.,
- et al.
- Shimamura M.,
- Matsuda M.,
- Yasumo H.,
- et al.
- Gusarova V.,
- Alexa C.A.,
- Wang Y.,
- et al.
- Ando Y.,
- Shimizugawa T.,
- Takeshita S.,
- et al.
- Minicocci I.,
- Santini S.,
- Cantisani V.,
- et al.
- Atherosclerosis, Thrombosis, and Vascular Biology Italian Study Group
- McPherson R.,
- Pertsemlidis A.,
- Kavaslar N.,
- et al.
- Senti M.,
- Tomas M.,
- Marrugat J.,
- Elosua R.,
- REGICOR Investigators
- Peloso G.M.,
- Lange L.A.,
- Varga T.V.,
- et al.
- Baber U.,
- Mehran R.,
- Sartori S.,
- et al.
- Davies R.W.,
- Wells G.A.,
- Stewart A.F.,
- et al.
- Yampolsky L.Y.,
- Kondrashov F.A.,
- Kondrashov A.S.
- Mehta N.,
- Qamar A.,
- Qu L.,
- et al.
- Baigent C.,
- Keech A.,
- Kearney P.M.,
- et al.
- Pisciotta L.,
- Favari E.,
- Magnolo L.,
- et al.
- Noto D.,
- Cefalu A.B.,
- Valenti V.,
- et al.
- Dewey F.E.,
- Gusarova V.,
- O'Dushlaine C.,
- et al.
- Stitziel N.O.,
- Kathiresan S.
- Minicocci I.,
- Tikka A.,
- Poggiogalle E.,
- et al.
- Robciuc M.R.,
- Maranghi M.,
- Lahikainen A.,
- et al.
- Zuk O.,
- Schaffner S.F.,
- Samocha K.,
- et al.