Lipoprotein(a) [Lp(a)]

Lipoprotein(a) is an LDL-like particle with apolipoprotein(a) attached. Its concentration is set almost entirely by the LPA gene, specifically by the number of kringle IV type 2 repeats, and it stays roughly constant across adult life. Diet, exercise and statins move it very little, which makes it different from every other lipid on a standard panel.

Pia Kamstrup, Anne Tybjaerg-Hansen, Rolf Steffensen and Borge Nordestgaard established the dose-response in JAMA using three Copenhagen populations: the Copenhagen City Heart Study with 8,637 people followed 16 years and 599 myocardial infarctions, the Copenhagen General Population Study with 29,388 people and 994 infarctions, and a case-control study with 2,461 participants and 1,231 infarctions [1]. Against the reference group below the 22nd percentile, multivariable-adjusted hazard ratios rose steadily: 1.2 (95% CI 0.9-1.6) at the 22nd to 66th percentile, 1.6 (95% CI 1.1-2.2) at the 67th to 89th, 1.9 (95% CI 1.2-3.0) at the 90th to 95th, and 2.6 (95% CI 1.6-4.1) above the 95th percentile, trend P < .001 [1]. Event rates went from 55 to 108 per 10,000 person-years across that span [1]. Kringle IV type 2 repeats ranged from 6 to 99 summed across both alleles [1].

Because the level is genetic, the therapeutic effort has gone into RNA-based silencing. Michelle O'Donoghue, Marc Sabatine and the OCEAN(a)-DOSE investigators ran a phase 2 dose-finding trial of olpasiran, a small interfering RNA, in 281 patients with established atherosclerotic cardiovascular disease and screening Lp(a) above 150 nmol/L, about 70 mg/dL. 227 received olpasiran, 54 placebo, across 34 sites in seven countries, with the primary endpoint the placebo-adjusted percent change in Lp(a) at week 36 [2]. The outcome trial that matters is Lp(a)HORIZON, testing pelacarsen (TQJ230) against placebo in 8,323 patients with established cardiovascular disease; Novartis lists actual study completion on 16 July 2026 [3].

How it is measured: immunoassay on serum, reported either in mg/dL of particle mass or nmol/L of particle number. The two are not interchangeable, because particle mass depends on isoform size; nmol/L is the preferred modern unit. One measurement in adult life is usually enough given the genetic determination.

How to move it: nothing on a supplement shelf. Niacin and lipoprotein apheresis lower Lp(a), and the siRNA and antisense drugs lower it far more, but no trial has yet reported that lowering Lp(a) lowers events [2][3]. Dose-ranging safety work in 29 healthy Japanese subjects, using single doses of 20, 40 and 80 mg subcutaneously plus four monthly 80 mg doses, reported no serious adverse events [4]. Until Lp(a)HORIZON reports, Lp(a) is a risk-stratification number that should intensify treatment of everything else that is modifiable.