LDL cholesterol

LDL is the 'bad' cholesterol that can build up in artery walls. This score estimates your inherited tendency toward higher or lower LDL.

What this LDL score reads

This is PGS000115, used by Trinder and colleagues in JAMA Cardiology in 2020 for a study that set monogenic against polygenic causes of high cholesterol. It sums 223 established LDL-associated common variants, from GWAS data covering 297,626 people, about 72% of them of European ancestry.

The study behind it matters for how you read your result: its purpose was to show that two different genetic routes to high LDL carry different risks, and that this score identifies only one of them.

How well it predicts LDL

Measured in the UK Biobank, the score explained about 9% of the variation in LDL cholesterol in European-ancestry participants (n = 439,871), 6% in East Asian and 4% in African-ancestry participants.

It also tracked outcomes, not just the lab value. People in the top fifth of the score had 2.78 times the odds of uncontrolled high cholesterol compared with the bottom fifth, and the top versus bottom decile carried a hazard ratio of 1.35 for cardiovascular events. That is a real signal, and it is still a modest one next to what a cholesterol test plus blood pressure and smoking status tell a clinician.

How to read a high or low score

A high percentile means many small LDL-raising variants have added up in your genome. It does not tell you your LDL number, and it does not set your cardiovascular risk on its own, since LDL is only one input into that.

Unlike most traits on this site, the thing being predicted is measured routinely and treated effectively. If your LDL is high, the measurement and the treatment conversation are what matter; the score explains part of why, not what to do.

What this score cannot see: familial hypercholesterolemia

Familial hypercholesterolemia is caused by a single damaging variant in LDLR, APOB or PCSK9, affects roughly 1 in 250 people, and raises LDL from birth, carrying a 3 to 10 fold risk of early coronary disease when untreated. A polygenic score does not detect it, and neither does a consumer genotyping chip, which reads only a fixed set of common positions rather than sequencing these genes.

This is the general rule the American College of Medical Genetics states plainly: isolated polygenic testing is not the appropriate test when a single-gene cause is suspected. Very high LDL, especially with a family history of early heart attacks, is a reason for clinical evaluation and possibly gene sequencing, whatever this score says.

Ancestry and accuracy

The score explained 9% of LDL variation in European-ancestry participants but 4% in African-ancestry participants in the same biobank, using the same measurement. That gap is not biology, it is the composition of the studies the score was built from, and it is the reason a percentile computed for someone outside the training population carries less information.

Variants in this score you can read on their own

This score includes variants that this site also explains individually, where you can see what each genotype means on its own: Alpha-1 antitrypsin deficiency (rs28929474) · Drug acetylation speed (NAT2) (rs1495741) · Hereditary hemochromatosis (HFE gene) (rs1800562).

How much of this score your file covers

All 223 variants in the published score are present in a standard consumer DNA file, so this reconstructs the published score in full.

Common questions

Does a high LDL polygenic score mean I have familial hypercholesterolemia? No, and it cannot tell you either way. Familial hypercholesterolemia is a single damaging variant in LDLR, APOB or PCSK9, and this score is a sum of common small-effect variants that has nothing to say about those genes. The two routes to high cholesterol are separate, which is exactly what the study behind this score examined.

Can diet and exercise offset a high LDL score? Diet, weight and exercise move LDL for most people, though usually less than statin therapy does. The score does not set a floor you cannot get under: it shifts your starting point, and treatment decisions are made on your measured LDL and overall cardiovascular risk, not on genetics.

How much of my LDL does genetics explain? For this score, about 9% in European-ancestry people, less in others. Common-variant heritability of LDL is higher than that, so a larger score would explain more, but the honest summary is that most of the difference between two people's LDL is not read by this score.

Related

Adult lactose digestion · Alcohol response (flushing & dependence risk) · Beta-carotene to vitamin A conversion (BCO1) · Folate metabolism (MTHFR gene) · Omega-3/6 conversion efficiency · Sweet tooth / carbohydrate preference (FGF21) · Total cholesterol