210 - Lp(a) and its impact on heart disease | Benoît Arsenault, Ph.D.& Peter Attia, M.D.

KEYPOINTS

• 01

  Genetics resurrected Lp(a) as a cardiovascular risk factor.

  Lp(a) was discovered in 1963 by Swedish scientist Kåre Berg. Interest faded when studies in the 1990s and early 2000s came back negative — partly because assays were unreliable — and PubMed activity dipped. Three genetic studies published in 2009, including Robert Clarke's PROCARDIS work with about 3,000 cases and 3,000 controls, identified variants in the LPA gene linked to heart disease, and Mendelian randomization established Lp(a) as causal.

• 02

  Lp(a) is an LDL-like particle with an extra protein attached.

  Every LDL carries exactly one apoB-100. Lp(a) adds apolipoprotein(a), which binds apoB — Arsenault says good evidence suggests this happens inside liver cells. Apo(a) contains kringle domains; kringle IV has 10 subtypes, and the type 2 subunit varies in copy number from one to as many as 40 repeats. Smaller isoforms are associated with higher Lp(a) levels.

• 03

  Measurement units and assays matter.

  Old assays overestimated large isoforms and underestimated small ones; newer antibodies bind other regions and count particles better. Lp(a) can be reported in mg/dL or nmol/L; 50 mg/dL corresponds to roughly 125 nmol/L, so a patient tested both ways may think their level 'doubled' when it didn't.

• 04

  Lp(a) is largely genetic and stable, so measure it at least once.

  Most guidelines suggest measuring once in a lifetime because levels are remarkably stable. More than 2,000 variants in the LPA region are associated with high levels, so a parent and child can both have high Lp(a) for different genetic reasons. Attia notes an apoB test costs about $14 — and the same logic applies to Lp(a).

• 05

  Risk rises with concentration.

  At highish levels — around 50 mg/dL or 125 nmol/L — risk may be increased 50% to 100%; at very high levels it climbs further. Roughly 20% of people (depending on ethnicity, at least 15%) have levels above 50 mg/dL. Attia admits he had been telling patients 10%, understating it.

• 06

  Oxidized phospholipids make Lp(a) especially harmful.

  Lp(a) carries far more oxidized phospholipids than LDL, affecting endothelial cells, smooth muscle cells, macrophages and aortic valve cells. Its sequence similarity to plasminogen may also impair clot breakdown, though Arsenault considers oxidized phospholipids the key.

• 07

  Aortic valve stenosis has the strongest link.

  Comparing top versus bottom quintile, aortic valve stenosis shows the biggest relative risk, followed by peripheral artery disease and heart attack, then ischemic stroke. Arsenault's group studied Lp(a) particles from valve-stenosis patients, about 20 per study arm, to explore mechanisms.

• 08

  Statins don't lower Lp(a), and why.

  Statins raise LDL receptor density, but Lp(a) levels are determined mainly by production rate, not clearance. Unexpectedly, PCSK9 increases apo(a) production in liver cells; PCSK9 inhibitors reduce Lp(a) production and lower it on average about 25–30%, with enormous variability.

• 09

  The PCSK9 story is a triumph of genetics.

  PCSK9 was discovered in 2003 by Nabil Seidah in Montreal and Catherine Boileau in France, studying a family with familial hypercholesterolemia without an LDL-receptor mutation — gain-of-function in PCSK9. In 2006 Helen Hobbs's group at UT Southwestern showed loss-of-function variants caused lifelong lower LDL and less heart disease. People with no functional PCSK9 have LDL around 10–20 mg/dL with no apparent harm to reproduction, hormones or other disease risk.

• 10

  Niacin lowers Lp(a) but hasn't proved outcome benefit.

  Niacin reduces Lp(a) on average about 20–30%. Large trials of HDL-modifying drugs, such as one of about 30,000 patients, illustrate why large outcome trials are needed in this field.

• 11

  Antisense drugs could transform treatment.

  Ionis's antisense oligonucleotide (pelacarsen) showed in phase 2 data published in 2020 that monthly dosing brought about 90% of patients below 50 mg/dL. An outcomes trial in about 8,000 patients with established cardiovascular disease was expected to read out around 2025.