Peter H. Jones, MD, is associate professor, Center for Cardiovascular Disease Prevention, Baylor College of Medicine, Houston, TX.

The well-established causative role of low-density lipoprotein (LDL) particles in atherosclerotic plaque development has led to this biomarker being a primary target of treatment in the prevention of coronary artery disease (CAD). A related lipoprotein particle, lipoprotein(a), or Lp(a), has also been implicated in CAD risk for at least 2 decades.¹ Lp(a) is an LDL cholesterol particle that is bound to apolipoprotein(a) through a covalent bond to apolipoprotein(B).

Apolipoprotein(a) has multiple repeating kringles, the number of which is genetically determined. These kringles have extensive sequence homology to plasminogen; thus, an elevation in the number of Lp(a) particles has been implicated in both atherosclerosis and thrombosis.² It is not surprising that Kamstrup and Nordestgaard found that the hazard ratios for myocardial infarction in patients with very high Lp(a) values (≥85 to 119 mg/dL and ≥120 mg/dL) were 2.6- to 3.6-fold greater than for those with the lowest Lp(a) values (<5 mg/dL) over 10 years of follow-up in the Copenhagen City Heart Study. Generally, there is a continuous risk of CAD events across increasing Lp(a) levels, and this observational study suggests that the threshold for risk in a predominantly white population is ≥30 mg/dL, with the most significant risk at levels >85 mg/dL.

Although most observational studies confirm that an increased Lp(a) level is a risk factor for CAD, the National Cholesterol Education Program (NCEP) Adult Treatment Panel (ATP) has not recommended routine screening of this analyte. There are at least 2 reasons why Lp(a) has not become a mainstream biomarker. First, the population distribution for Lp(a) levels in whites is not Gaussian, but rather, is shifted to the lowest levels for the majority. This is very clear from the Copenhagen City Heart Study because more than half of the population had levels <30 mg/dL, and fewer than 10% had levels >85 mg/dL. But even more importantly, the distribution of Lp(a) levels varies in different ethnic groups, such that blacks and South Asians have a greater frequency of higher Lp(a) levels, yet these groups may not share the same CAD risk as whites with similar Lp(a) levels.^3,4 The variation in Lp(a) among different populations has made it difficult to establish cut points for risk. Second, the measurement of Lp(a) has not been standardized, and there is considerable controversy regarding which measurement technique is preferred. Some experts have suggested that Lp(a) levels should be reported as moles of Lp(a) particles per liter, rather than milligrams of cholesterol per deciliter, because the molecular weight of apolipoprotein(a) has considerable variability.

Both of the outlined factors were considerations that weighed heavily in the ATP's decision not to recommend routine Lp(a) screening, except for individuals with premature CAD or a strong history of a first-degree relative with premature CAD. A very important reason for reserving Lp(a) testing for families with premature CAD is that there have not been any randomized clinical trials showing a reduction in surrogate measurements of atherosclerosis or CAD events with reduction in Lp(a) levels. At doses of ≥1000 mg/day, niacin reduces Lp(a) by 20% to 25% and is the only known therapy that consistently lowers Lp(a). Clinical lipid experts have recommended that the therapeutic approach to individuals with high Lp(a) levels (>30 mg/dL) is to aggressively lower LDL cholesterol first and then consider adding niacin at a dosage of ≥1000 mg/day.⁵ Kamstrup and Nordestgaard found very high Lp(a) values in about 10% of a white population, which conferred a high risk of future CAD events in this group of patients. We await clinical trial evidence that will establish whether niacin can add incremental CAD event reductions to intensive LDL cholesterol-lowering treatment (usually with statins) in these individuals before routine Lp(a) screening can become a recommendation.

References