Learn Explainer
Familial hypercholesterolemia genes: LDLR, APOB, PCSK9 and Finland's founder variants
Familial hypercholesterolemia is one of the commonest inherited conditions, and three genes explain most of it. In Finland a handful of old variants explain most of those, and one of the two commonest is a deletion of 8,535 letters that a chip export or a plain VCF has no way to show.
Key takeaways
- Familial hypercholesterolemia affects about 1 in 311 people worldwide by one 2020 meta-analysis and 1 in 313 by another[3, 4]. In a UK Biobank exome study, 257 of 277 carriers had their variant in LDLR[8].
- Finland's FH is dominated by founder variants. FH-Helsinki, a deletion of 8,535 letters, was found in 38 percent of 199 Finnish patients, and FH-North Karelia, seven letters deleted from exon 6, in 34 percent of 201[14, 15, 16].
- A deletion that size leaves no record in a plain VCF unless the provider also ran a copy-number caller, and a chip reports it only if it carries a probe built for it. Aligned reads show it as a drop to about half the usual read depth[9], but our report has no caller for it.
- Many people with very high LDL carry no FH variant at all. In 2013, 52 percent of mutation-negative UK patients scored in the top three tenths of a 12-variant LDL score[24].
- The PCSK9 R46L variant works the other way: carriers in one US cohort had 15 percent lower LDL cholesterol[25], and the variant is nearly three times as common in Finnish genomes in gnomAD as in non-Finnish European ones[26].
Familial hypercholesterolemia (familial hypercholesterolaemia in British spelling, FH for short) is high LDL cholesterol from birth with a single inherited cause. Most cases come from a variant in LDLR, the gene for the LDL receptor; variants in APOB and PCSK9 cause it less often[1]. The receptor sits on liver cells and pulls LDL particles out of the blood[2], so anything that leaves a person with fewer working receptors, or with LDL the receptor cannot grip, leaves more LDL circulating. Two meta-analyses published in 2020 put the condition at about 1 in 311[3] and 1 in 313 people[4].
Three dominant genes and one recessive one
Each of the main genes breaks the same pathway at a different point. LDLR makes the receptor itself, and variants in it are the usual cause[1]. APOB makes apolipoprotein B-100, the protein on the surface of an LDL particle that the receptor binds; the classic APOB variant, described in 1990 as familial defective apolipoprotein B-100, changes a single amino acid and weakens that binding[5]. PCSK9 makes a protein that lowers the number of LDL receptors in the liver[6]. PCSK9 variants were first tied to dominant high cholesterol in 2003[7], and the ones that cause it make the protein more active: one, D374Y, cut receptor numbers in liver cells about ten times as strongly as the usual protein[6].
All three are dominant: one altered copy is enough to raise LDL[1, 7]. A fourth gene, LDLRAP1, works differently. A 2001 study mapped autosomal recessive hypercholesterolemia to it and found that the adaptor protein it makes is needed in the liver[2]; the receptor needs it to be pulled into the cell, and PCSK9 had no effect on liver cells that lacked it[6]. Recessive means a person needs two altered copies; one copy is carrier status.
1 in 311
people with familial hypercholesterolemia in the general population, pooled from 42 studies[3]
257 of 277
UK Biobank carriers of an FH variant whose variant was in LDLR; 13 were in PCSK9 and 7 in APOB[8]
about 10%
of FH cases attributed to deletions or duplications of whole LDLR exons[9]
The two meta-analyses agree closely: 1 in 311, with a 95 percent confidence interval from 1 in 250 to 1 in 397[3], and 0.32 percent, or 1 in 313, across almost 11 million people[4]. Genetic screening of unselected people lands in the same place. Exome sequencing of 50,726 people in one US health system found FH variants in 1 in 256[10], and in 48,741 UK Biobank participants the figure was 1 in 176[8]. Both meta-analyses left out studies of founder populations[3, 4], and Finland is one.
LDLR, APOB and PCSK9 are on the American College of Medical Genetics and Genomics list of genes whose findings it asks genome laboratories to return even when nobody asked for them, and LDLRAP1 is not[11]. Our guide to the ACMG secondary findings list covers why those 84 genes and how the report reads them.
Finland's founder variants
In Britain FH is spread thinly across many variants: the 257 LDLR carriers in the UK Biobank study had 110 different ones between them[8]. Finland's is concentrated in a few, each descended from a single ancestral copy, which is what a founder variant means and why the country also has its own set of rare recessive diseases. The first was characterised and named in 1989: a deletion of about 9.5 kb running from intron 15 into exon 18 of LDLR, which Helsinki researchers called FH-Helsinki and found in 23 of 46 unrelated Finnish patients[12]. A Norwegian group later measured it at 9.6 kb and showed that it removes the part of the receptor that crosses the cell membrane and the tail inside the cell[13].
The second was FH-North Karelia, reported in 1992. It deletes seven letters from exon 6, which shifts the reading frame and cuts the receptor short; carriers in that study averaged an LDL cholesterol of 10 mmol/l[14]. Its modern name is LDLR c.925_931delCCCATCA[15]. The two divide the country between them. FH-Helsinki was found in 26 to 58 percent of patients depending on the region, but in only 1 of 26 from North Karelia[16]; FH-North Karelia was found in 79 percent of patients from eastern Finland[14].
| Name | Change in LDLR | Kind of variant | Share of Finnish FH patients |
|---|---|---|---|
| FH-Helsinki | 8,535 letters deleted, intron 15 to exon 18 | Large deletion | 38% (75 of 199) |
| FH-North Karelia | c.925_931del, exon 6 | 7-letter deletion | 34% (69 of 201) |
| FH-Turku | c.2531G>A, Gly844Asp (Gly823Asp in 1995) | One letter changed | About 8%, with FH-Pori |
| FH-Pori | c.1202T>A, Leu401His (Leu380His in 1995) | One letter changed | About 8%, with FH-Turku |
Source: FH-Helsinki[16]; FH-North Karelia[14, 15]; FH-Turku and FH-Pori[15, 18, 19]. Each share comes from its own patient series, so they cannot be added into one total.
A 2015 study turned the question around and genotyped the founder variants in 28,465 Finns from population surveys, FINRISK and Health 2000 among them. By then five founder variants were known, together accounting for 78 percent of Finnish FH. Three turned up, in 0.12 percent of people combined, and the authors estimated that at least 0.17 percent of Finns, about 1 in 600, have FH[20]. That is a floor built on known variants, and it sits below the worldwide 1 in 311.
gnomAD, the largest public table of variant frequencies, shows how Finnish FH-North Karelia is. Among Finnish samples it appears 9 times in 52,942 exome alleles and 6 times in 10,580 genome alleles. Among non-Finnish Europeans it appears in none of 1,108,760 exome alleles[21]. FH-Helsinki is harder to count, for the same reason it is hard to see in a personal DNA file.
Why an 8.5 kb deletion is invisible in most DNA files
A chip export lists the genotypes at a fixed set of positions the chip was built to measure, and nothing in between. It can only report a deletion if the chip has a probe designed for it. Even for single-letter variants, chips do badly once a variant is rare: in a UK Biobank comparison, only 16 percent of chip calls for variants rarer than 1 in 100,000 were confirmed by sequencing[22]. Our guide to scary raw data results goes through what that means for a single file.
A VCF from a sequenced genome does better on small things. FH-North Karelia's seven missing letters are an ordinary small deletion that a variant caller writes as one record. A deletion of 8,535 letters is a different kind of event. The standard callers compare short reads with the reference letter by letter, and over the deleted stretch the person still has one intact copy that matches the reference perfectly. The caller sees reference letters, so it writes nothing at all into a plain VCF and writes reference blocks into a gVCF.
The reads hold the answer. When one copy is missing, about half as many reads land on the stretch, and a program that compares depth along the gene with depth in other samples can see it. A 2017 study did this on sequencing data from 388 FH patients and found every one of the 38 exon deletions and duplications that the older laboratory method, MLPA, had found, and no others. The heterozygous deletions came out at 0.43 to 0.65 times the expected depth[9]. That is a job for a copy-number caller run on a BAM or CRAM. A VCF carries its result only if the provider ran one.
High LDL from many small variants
Plenty of people with very high LDL carry no FH variant. Among 20,485 participants of control groups and prospective cohorts, 1,386 had an LDL cholesterol of 190 mg/dl or more, and sequencing found an FH variant in 24 of them, 1.7 percent[23]. Clinics see a filtered version of the same picture. A 2013 UK study noted that about 60 percent of patients with a clinical FH diagnosis had no variant found in the known genes, and asked whether common variants could add up to the same result[24].
They built a score from 12 common LDL-raising variants and compared 321 mutation-negative patients with 3,020 healthy controls from the Whitehall II study. 167 of the patients, 52 percent, scored in the top three tenths of the controls' distribution, and 35, 11 percent, in the bottom three[24]. For many of them, high LDL looked like the sum of many small pushes. How such a score is built and read is in polygenic scores in plain words.
| Group | People | Hazard ratio (95% CI) |
|---|---|---|
| Monogenic FH variant | 277 | 1.93 (1.34 to 2.77) |
| Polygenic: LDL score above the 95th percentile | 2,379 | 1.26 (1.03 to 1.55) |
| High LDL, no genetic cause identified | 2,232 | 1.00 (comparison) |
Source: Trinder, Francis and Brunham, JAMA Cardiology 2020[8]. Events were coronary or carotid revascularization, myocardial infarction, ischemic stroke and death from any cause.
The two kinds of high LDL are not interchangeable. In the UK Biobank study whose 223-variant score our report uses, people with an FH variant had about twice the rate of events of people with the same measured LDL and no genetic explanation, and people with a high polygenic score had about a quarter more[8]. The authors read this as lifelong exposure adding risk that a single measurement does not capture.
PCSK9 R46L, the variant that lowers LDL
PCSK9 gives FH when a variant makes it more active. A variant that makes it less active does the opposite, and the best known is R46L, written rs11591147, which swaps arginine for leucine at position 46. In 9,524 white participants of the US ARIC study, 3.2 percent carried it. Their LDL cholesterol was 15 percent lower, and over 15 years they had a 47 percent lower rate of coronary heart disease, with a hazard ratio of 0.50 and a confidence interval from 0.32 to 0.79[25].
15%
lower LDL cholesterol in white R46L carriers in the ARIC study[25]
4.4% vs 1.6%
R46L allele frequency in Finnish and in non-Finnish European genomes in gnomAD[26]
Finland carries more of it. In gnomAD's genomes the variant makes up 467 of 10,630 Finnish alleles and 1,104 of 68,046 non-Finnish European ones, 4.4 against 1.6 percent[26]. Our own arithmetic from the Finnish figure, assuming random pairing of alleles, puts about 1 Finn in 12 at one copy, an estimate. Nobody has measured what R46L does to the LDL of someone who also carries FH-Helsinki, and this article does not guess.
What our report shows, file by file
Four parts of the report touch this topic: the FH row of the clinical findings section, which covers LDLR, APOB and PCSK9; LDLRAP1 in the wider ClinVar screen, as a recessive gene; the 223-variant LDL polygenic score from the UK Biobank study above[8]; and an exploratory card for R46L. What each can say depends on the file.
| File | LDLR, APOB, PCSK9 | Large deletions such as FH-Helsinki | LDL polygenic score | R46L card |
|---|---|---|---|---|
| Chip export | Not screened | Not looked for | Yes, on the variants the chip typed | Only if the chip typed rs11591147; otherwise left unresolved |
| Plain VCF | ClinVar pathogenic and likely pathogenic variants; a clear result reads "No matching variant in your file" | Not looked for | Yes, with a caveat, because a missing variant can only be assumed | Yes; a missing record is read as the common version and labelled an inference |
| gVCF | The same check; "Screened, nothing found" when all three genes are at least 90% callable | Not looked for | Yes, with the positions the file proves | Yes, read from the file |
| BAM or CRAM (Deep Read) | Adds a coverage map from the reads to the variant file's result | Not looked for | From the variant file | From the variant file |
Source: The report's behaviour as of October 2026: the clinical findings panel and its FH condition content, the wider ClinVar screen, the LDL score (PGS Catalog PGS000115) and the R46L Frontier card.
On a VCF or gVCF, a variant in one of the three genes counts when ClinVar classifies it pathogenic or likely pathogenic with two or more review stars; single-submitter entries are listed apart as leads, and the FH explanation is attached when ClinVar names FH or a close term as the variant's condition. A gVCF earns the stronger "Screened, nothing found" only when each of the three genes is at least 90 percent callable at a depth of 10 reads or more, and Deep Read coverage from a BAM or CRAM counts as the same proof. The guide to genome files explains why a plain VCF cannot give it.
That screened row has a limit, and it falls on one of Finland's two commonest FH variants. The coverage test asks whether the positions were read, and a heterozygous FH-Helsinki deletion leaves them read: at half depth, about 15 reads in a 30x genome, above our floor of 10. The report has no copy-number caller for LDLR, so "Screened, nothing found" speaks for small variants only, and the report does not look for FH-Helsinki in read depth. The small founder variants depend on ClinVar. As read on 10 October 2026, FH-North Karelia is pathogenic with agreeing submitters and FH-Turku likely pathogenic after expert-panel review, so the report would flag either; FH-Pori has conflicting classifications, and conflicting entries are dropped[18].
A chip export gets no FH gene screen at all. Our chip report reads two single clotting variants as clinical findings and nothing in these genes, because a chip measures only fixed positions and its calls for rare variants are the least reliable it makes[22]. The LDL score and the R46L card run on chips, and both are statements about common variants; the score, like the report's other disease-related scores, is shown only with the reader's consent to sensitive findings. Neither measures cholesterol. A different inherited lipid, lipoprotein(a), has its own guide.
What Aimosti would (and wouldn't) show you
On a VCF or gVCF the clinical findings section checks LDLR, APOB and PCSK9 against ClinVar's pathogenic and likely pathogenic classifications, and a gVCF or Deep Read coverage can show that all three genes were read. The report's deletion caller reads two other genes from aligned reads and not LDLR, so FH-Helsinki is not looked for. The 223-variant LDL polygenic score and the PCSK9 R46L card run on every file type, within the limits each file sets; a chip export gets no FH gene screen.
What we won't claim
We won't present a clear FH row as proof that someone does not have familial hypercholesterolemia, call a polygenic score a diagnosis of anything, or turn an R46L genotype into a cholesterol reading. A gene result says what a file contains; a blood test measures cholesterol.
Bottom line. Three genes and a handful of Finnish founder variants explain most familial hypercholesterolemia, but one of the two commonest Finnish variants is a large deletion, and a raw DNA file that lists only small variants cannot show it. A clear FH row covers what the file could carry.
Questions people ask
Can a 23andMe or other chip raw data file show familial hypercholesterolemia?
Rarely, and not reliably. A chip reads a fixed set of positions, so it can only report the few FH variants it was built to measure, and chip calls for rare variants were confirmed by sequencing only 16 percent of the time in a UK Biobank comparison[22]. Our report does not run an FH gene screen on a chip export.
Does a clear FH result on a genome file rule out familial hypercholesterolemia?
No. It means no ClinVar pathogenic or likely pathogenic variant in LDLR, APOB or PCSK9 was found in the file. Large deletions such as FH-Helsinki are not looked for, and many people with a clinical FH diagnosis have no variant found in the known genes[24]. Cholesterol itself is measured in blood, not in a DNA file.
What is FH-Helsinki?
A deletion of 8,535 letters in the LDLR gene, from intron 15 into exon 18[15], first described in Finnish patients in 1989[12]. It was found in 38 percent of 199 unrelated Finnish FH patients in a 1992 survey[16]. Because one intact copy remains, it shows up only as reduced read depth in aligned reads.
Is FH-North Karelia only found in Finland?
In gnomAD it appears only in Finnish samples: 15 times among 63,522 Finnish exome and genome alleles and in none of more than 1.1 million non-Finnish European exome alleles[21]. That fits a founder variant that spread within one population.
What is the difference between FH and polygenic high cholesterol?
FH has one variant with a large effect; polygenic high LDL is the sum of many common variants with small effects. In a UK Biobank study, at comparable LDL cholesterol, people with an FH variant had a hazard ratio of 1.93 for cardiovascular events and people with a high polygenic score 1.26, both against people with no genetic cause found[8].
Does Aimosti check LDLRAP1?
Yes, in the wider ClinVar screen on a VCF or gVCF, as a recessive gene: one pathogenic variant is reported as carrier status, two as two copies. LDLRAP1 causes autosomal recessive hypercholesterolemia[2] and is not on the ACMG secondary findings list[11].
References
- Defesche JC, Gidding SS, Harada-Shiba M, Hegele RA, Santos RD, Wierzbicki AS. Familial hypercholesterolaemia. Nature Reviews Disease Primers, 2017. doi:10.1038/nrdp.2017.93
- Garcia CK, Wilund K, Arca M, et al. Autosomal recessive hypercholesterolemia caused by mutations in a putative LDL receptor adaptor protein. Science, 2001. doi:10.1126/science.1060458
- Hu P, Dharmayat KI, Stevens CAT, et al. Prevalence of familial hypercholesterolemia among the general population and patients with atherosclerotic cardiovascular disease: a systematic review and meta-analysis. Circulation, 2020. doi:10.1161/CIRCULATIONAHA.119.044795 General population: 1 in 311 (95% CI 1 in 250 to 1 in 397), 42 studies, 7,297,363 people. Founder-effect studies excluded.
- Beheshti SO, Madsen CM, Varbo A, Nordestgaard BG. Worldwide prevalence of familial hypercholesterolemia: meta-analyses of 11 million subjects. Journal of the American College of Cardiology, 2020. doi:10.1016/j.jacc.2020.03.057 General population: 0.32% (95% CI 0.26 to 0.39%), 1 in 313, 44 studies, 10,921,310 people; non-founder populations only.
- Innerarity TL, Mahley RW, Weisgraber KH, et al. Familial defective apolipoprotein B-100: a mutation of apolipoprotein B that causes hypercholesterolemia. Journal of Lipid Research, 1990. doi:10.1016/S0022-2275(20)42605-7
- Lagace TA, Curtis DE, Garuti R, et al. Secreted PCSK9 decreases the number of LDL receptors in hepatocytes and in livers of parabiotic mice. Journal of Clinical Investigation, 2006. doi:10.1172/JCI29383 Activity about 10-fold greater for the gain-of-function mutant PCSK9(D374Y); no effect in hepatocytes lacking ARH.
- Abifadel M, Varret M, Rabès JP, et al. Mutations in PCSK9 cause autosomal dominant hypercholesterolemia. Nature Genetics, 2003. doi:10.1038/ng1161
- Trinder M, Francis GA, Brunham LR. Association of monogenic vs polygenic hypercholesterolemia with risk of atherosclerotic cardiovascular disease. JAMA Cardiology, 2020. doi:10.1001/jamacardio.2019.5954 48,741 UK Biobank participants with exomes; 277 (1 in 176) carried a monogenic FH variant: LDLR 257, PCSK9 13, APOB 7. Polygenic: above the 95th percentile of a 223-variant LDL score.
- Iacocca MA, Wang J, Dron JS, et al. Use of next-generation sequencing to detect LDLR gene copy number variation in familial hypercholesterolemia. Journal of Lipid Research, 2017. doi:10.1194/jlr.D079301 About 10% of FH cases attributed to LDLR copy number variants; 38 of 38 MLPA-confirmed CNVs found from sequencing depth in 388 samples; heterozygous deletions at normalised depth ratios of 0.43 to 0.65 (table 1).
- Abul-Husn NS, Manickam K, Jones LK, et al. Genetic identification of familial hypercholesterolemia within a single U.S. health care system. Science, 2016. doi:10.1126/science.aaf7000 50,726 exomes; FH variants in 1 in 256 unselected participants; 24% of carriers met record-based criteria for probable or definite FH.
- Lee K, Abul-Husn NS, Amendola LM, et al. ACMG SF v3.3 list for reporting of secondary findings in clinical exome and genome sequencing: a policy statement of the American College of Medical Genetics and Genomics (ACMG). Genetics in Medicine, 2025. doi:10.1016/j.gim.2025.101454
- Aalto-Setälä K, Helve E, Kovanen PT, Kontula K. Finnish type of low density lipoprotein receptor gene mutation (FH-Helsinki) deletes exons encoding the carboxy-terminal part of the receptor and creates an internalization-defective phenotype. Journal of Clinical Investigation, 1989. doi:10.1172/JCI114192 9.5-kb deletion from intron 15 to exon 18; found in 23 of 46 unrelated FH patients.
- Rødningen OK, Tonstad S, Ose L, Berg K, Leren TP. Effects of a 9.6-kb deletion of the LDL receptor gene (FH Helsinki) on structure and levels of mRNA. Human Mutation, 1998. doi:10.1002/(SICI)1098-1004(1998)12:2<95::AID-HUMU4>3.0.CO;2-E
- Koivisto UM, Turtola H, Aalto-Setälä K, et al. The familial hypercholesterolemia (FH)-North Karelia mutation of the low density lipoprotein receptor gene deletes seven nucleotides of exon 6 and is a common cause of FH in Finland. Journal of Clinical Investigation, 1992. doi:10.1172/JCI115839 69 of 201 unrelated Finnish FH patients (34%); 79% of patients from eastern Finland.
- Jokiniitty A, Eskola M, Metso S, et al. Genetic testing for familial hypercholesterolemia in a Finnish cohort of patients with premature coronary artery disease and elevated LDL-C levels. Frontiers in Cardiovascular Medicine, 2024. doi:10.3389/fcvm.2024.1433042 Names FH-North Karelia as LDLR c.925_931delCCCATCA, p.(Pro309Lysfs), and FH-Pori as c.1202T>A, p.(Leu401His).
- Aalto-Setälä K, Koivisto UM, Miettinen TA, et al. Prevalence and geographical distribution of major LDL receptor gene rearrangements in Finland. Journal of Internal Medicine, 1992. doi:10.1111/j.1365-2796.1992.tb00528.x FH-Helsinki in 75 of 199 unrelated Finnish FH patients (38%); 26 to 58% by region; 1 of 26 in North Karelia.
- LDLR gene, canonical transcript ENST00000558518 (LDLR-208), GRCh38. Ensembl, 2026. Read through the Ensembl REST API on 10 October 2026: chr19:11,089,418-11,133,830, 18 exons.
- ClinVar variations 3729, 3734 and 3735: LDLR c.925_931del (FH North Karelia), c.2531G>A (FH Turku) and c.1202T>A (FH Pori). ClinVar, National Center for Biotechnology Information, 2026. Read through NCBI E-utilities on 10 October 2026. 3729: Pathogenic, multiple submitters, no conflicts. 3734: Likely pathogenic, reviewed by expert panel. 3735: Conflicting classifications of pathogenicity. Aliases as ClinVar lists them.
- Koivisto UM, Viikari JS, Kontula K. Molecular characterization of minor gene rearrangements in Finnish patients with heterozygous familial hypercholesterolemia: identification of two common missense mutations (Gly823→Asp and Leu380→His) and eight rare mutations of the LDL receptor gene. American Journal of Human Genetics, 1995. PMID 7573037. FH-Turku and FH-Pori together about 8% of FH genes in Finland, 30% in the southwest.
- Lahtinen AM, Havulinna AS, Jula A, Salomaa V, Kontula K. Prevalence and clinical correlates of familial hypercholesterolemia founder mutations in the general population. Atherosclerosis, 2015. doi:10.1016/j.atherosclerosis.2014.11.015 28,465 Finns genotyped for five founder mutations, which account for 78% of FH in Finland; three found, combined 0.12%; FH in Finland estimated at least 0.17%, at least 1 in 600.
- gnomAD v4: variant 19-11107497-AACCCATC-A (LDLR c.925_931del, rs387906304). Genome Aggregation Database, 2026. Read through the gnomAD API on 10 October 2026. Finnish: 9 of 52,942 exome alleles, 6 of 10,580 genome alleles. Non-Finnish European: 0 of 1,108,760 exome alleles, 0 of 67,970 genome alleles.
- Weedon MN, Jackson L, Harrison JW, et al. Use of SNP chips to detect rare pathogenic variants: retrospective, population based diagnostic evaluation. BMJ, 2021. doi:10.1136/bmj.n214
- Khera AV, Won HH, Peloso GM, et al. Diagnostic yield and clinical utility of sequencing familial hypercholesterolemia genes in patients with severe hypercholesterolemia. Journal of the American College of Cardiology, 2016. doi:10.1016/j.jacc.2016.03.520 Of 1,386 people with LDL cholesterol of 190 mg/dl or more, 24 (1.7%) carried an FH variant.
- Talmud PJ, Shah S, Whittall R, et al. Use of low-density lipoprotein cholesterol gene score to distinguish patients with polygenic and monogenic familial hypercholesterolaemia: a case-control study. The Lancet, 2013. doi:10.1016/S0140-6736(12)62127-8 12-variant weighted score; 167 of 321 mutation-negative UK patients (52%) in the top three deciles of the Whitehall II distribution, 35 (11%) in the lowest three.
- Cohen JC, Boerwinkle E, Mosley TH, Hobbs HH. Sequence variations in PCSK9, low LDL, and protection against coronary heart disease. New England Journal of Medicine, 2006. doi:10.1056/NEJMoa054013 9,524 white ARIC participants: 3.2% carried the variant; 15% lower LDL cholesterol; hazard ratio for coronary heart disease 0.50 (95% CI 0.32 to 0.79).
- gnomAD v4: variant 1-55039974-G-T (PCSK9 R46L, rs11591147). Genome Aggregation Database, 2026. Read through the gnomAD API on 10 October 2026. Genomes: Finnish 467 of 10,630 alleles (4.4%), non-Finnish European 1,104 of 68,046 (1.6%).
Last reviewed . Every number on this page links to the source it comes from; if one of them has moved, tell us.