The Lp(a) Conundrum

My paternal grandpa was a 6’ 8” gentle giant who had a monster fast pitch and made us kids laugh with his funny faces. I enjoyed riding on the tractor with him and searching for arrowheads in the creek beds. I also watched grandpa roll and smoke endless cigarettes. At 60, while driving to Florida with my grandmother, my beloved grandpa died of a massive myocardial infarction. As the family patriarch, it was a significant blow to us. Flash forward years, I’m in the medical education department at a laboratory that offers an advanced lipid panel. I discover that my father, sister and brother all have very high Lp(a). It’s not a big leap to consider that along with smoking, Lp(a) could’ve played a role in my grandpa’s untimely death.  And those of you attempting to lower Lp(a) in your patients know: It’s TOUGH. In my practice (and with my family), I’ve taken to addressing not just Lp(a) directly, but also the underlying Lp(a) pathogenic mechanisms and other CVD risk factors with the goal of reducing the impact of high circulating Lp(a) on multiple fronts. My father is 77 today and doing reasonably well, taking combination of supplements and medications (although if you are reading this dad, bump up the exercise!). My siblings are generally healthy and are versed on obtaining advanced lipid panels from their providers which is essential for those with elevated Lp(a).  ~DrKF

Despite widespread use of cardio-preventive therapies, around 40% of deaths are still down to cardiovascular disease. There’s great focus on what constitutes this residual risk, and lipoprotein(a), AKA Lp(a), may be attributable for much of it. Lp(a) is emerging as an important biomarker for atherosclerosis and coronary heart disease, but though it was discovered in 1963– and common, about 25% of us have higher levels– it’s taken more time for researchers to tease out what Lp(a) looks like and what its physiological ‘reason for being’ might be. 

Lp(a) has been linked to increased risk for all of the following: 

• Overall cardiovascular disease (CVD), especially atherosclerosis
• Cardiovascular events like myocardial infarction and stroke
• Coronary heart disease (CHD), including coronary artery disease (CAD)
• Cerebrovascular stroke
• Peripheral vascular events
• Endothelial dysfunction
• Calcific aortic valve stenosis (CAVS)

The studies

• An in-depth 2017 review of Lp(a) states that it is established as “an independent, genetic, and likely causal risk factor for CVD,” 
• Lp(a) level is also recognized as a possible independent risk factor for CAD. 
• A landmark 1981 study identified Lp(a) as an independent risk factor for myocardial infarction, with a risk threshold of 30 mg/dl in normolipidemic individuals and a 2.3-fold risk at levels >50 mg/dl. 
• A more recent study found a doubled overall risk for CVD at plasma levels >20 mg/dl,
• cerebrovascular stroke risk is about 1.8-2.7 times greater at plasma levels around 30 mg/dl. 
• The 2018 NHLBI working group tasked with developing Lp(a) recommendations has stated that well over 1 billion persons have elevated levels and that 20-30% of humans have Lp(a) levels within the >30-50 mg/dl range generally accepted as potentially atherothrombotic. 
• About one-third of those with familial hypercholesterolemia have Lp(a) levels >50 mg/dl. 
• Bottomline: These and other hard-hitting findings really point the finger at Lp(a) as being responsible for a meaningful portion of residual cardio risk. 

Lp(a)’s Uniquely Structure: An LDL-ish protein peppered apo(a) kringles

Structurally, Lp(a) consists of one LDL-like particle with two other proteins: apolipoprotein B100 (apoB100) complexed via disulfide bridge to apolipoprotein(a). (NOTE: Apolipoprotein(a) is also called apo(a), and is different from apoA, which is HDL-related.) Apo(a) contains multiple looping kringles, which are protein domains involved in coagulation. Think about that a minute: an LDL-ish lipoprotein plus two apolipoproteins with kringles. That constitutes a very busy molecule having several regions where binding and other biochemical changes can take place. See figure 1 below for the far-reaching pathogenic mechanisms of Lp(a). Scroll to the end of the blog for a rabbit hole discussion on the structural and genetic details of Lp(a)

Figure 1. Pathogenic Mechanisms of Lp(a)

The atherogenicity of Lp(a) can be broadly classified in 3 categories: proatherogenic, proinflammatory, and potentially antifibrinolytic. The major individual mechanisms within each category are listed. EC = endothelial cell; IL = interleukin; MCP = monocyte chemoattractant protein; PAI = plasminogen activator inhibitor; SMC = smooth muscle cell; TFPI = tissue factor pathway inhibitor; other abbreviation as in Figure 1.

The Lp(a) Conundrum: Important, yet impossible….?

Lp(a) is obviously important, so why hasn’t it had as much share of the limelight as other blood constituents? It’s really very simple: it’s hard to budge Lp(a) levels, and our understanding of Lp(a) metabolism is still quite limited.

Gender and age have little bearing on Lp(a) levels, even though associated health risks may differ by gender, age, or ancestry at comparable plasma levels. Most statins and fibrates don’t lower Lp(a) levels (some may even increase them), though niacin and flaxseed can, and atorvastatin may mitigate cardiac risk without necessarily altering levels. Lp(a) levels typically increase after menopause, and hormone replacement therapy is known to lower them. Insulin has been suggested as a modulator of Lp(a) levels, which decreased prior to diabetes onset in a 2017 study; this may reflect transitions in insulin resistance, immune activation, and redox balance in prediabetes. Lp(a) levels temporarily increase during inflammation yet don’t change much after a fatty meal. However, eating omega-3-enriched pork led to lower Lp(a) levels and LDL oxidation in men, which normal pork increased. Treatments for lowering Lp(a) levels are under research, and while a few meds can impact Lp(a), none yet do so selectively. Even moderate exercise doesn’t move Lp(a).  

So What’s Lp(a) Actually For, After All? 

We don’t yet fully understand why Lp(a) exists, and some researchers theorize that it isn’t entirely pathogenic. It might be involved in wound healing, controlling bleeding, or gearing up for or winding down inflammation—so why is it such a huge risk factor? Lp(a) gives us plenty of tricky clues: though it ignores many common therapies, it seems to respond to the use of things like N-acetylcysteine, estrogen, niacin, and aspirin. The first is a precursor for precious, versatile glutathione, the second a profound tissue protector when in balance, the third a nutrient that happens to help lower residual cardio risk, and the fourth a trigger for making pro-resolving lipid mediators. A pro-inflammatory environment stimulates Lp(a) synthesis, while a ‘recovering from inflammation’ environment slows it down. So redox balance, protection, and inflammatory processes seem central to Lp(a)’s mission.

Addressing Lp(a) in Clinical Practice: Considerations

Beyond biologics and plasma apheresis, addressing Lp(a) level is challenging. We all have our various protocols, likely most of us leaning on niacin as a foundational intervention. Here is a run-down of additional takeaways we gleaned from the research: 

• Given Lp(a) resistance to lowering, follow a full, foundational FxMed approach so general inflammation is dialed back, resilience is up, nutrients are optimized. 
• Reference ranges. In general, <10mg/dL is associated with a reduced risk of CVD and anything higher is associated with increased risk. Concern ranges are 30mg/dL or higher. (Some Lp(a) assays use nmol/L. Conversion calculators are readily available online.) 
• Optimize all lipid and inflammation markers using an advanced panel, even as Lp(a) is treatment resistant. 
  • Include fibrinogen in your work-up as a surrogate marker of clotting risk (recall Lp(a)’s clotting kringles). Since fibrinogen is fairly easy to address with omega 3’s and proteolytic enzymes, I try and keep it on the lower side of normal.
  • Risk from elevated Lp(a) levels would seem to be additive (maybe even synergistic) with high LDL or oxPL levels, very high HDL levels, or high mean HDL size. In the future, we’ll take a closer look at particular polymorphisms that may help clarify for whom they affect risk and if there are any known lifestyle considerations that modulate risk. 
• Consider obtaining a coronary artery calcium score as a part of a general work-up for Lp(a) folks
• In polycystic ovary syndrome (PCOS) patients with elevated Lp(a), supplementing with 1 g omega-3s from flax (including 400 mg ALA) along with 400 IU vitamin E daily downregulated mononuclear cell Lp(a) and oxLDL expression, improved total antioxidant capacity in the blood, and significantly lowered triglycerides, total cholesterol, LDL, and VLDL—an impressive combination of effects, especially since these women often show just the kind of oxidative and glycation stress that may increase Lp(a) modification. 
• A meta-analysis found that oral L-carnitine may significantly lower Lp(a) by an average of about 9 mg/dl; studies used mainly 1 or 2 grams daily, and results did not differ by dosage. 
• Another meta-analysis found that CoQ10 supplementation produced mild yet still significant Lp(a) reductions of about 3.5 mg/dl, especially at higher dosages (200-300 mg/day) and in those with baseline Lp(a) levels of ≥30 mg/dl. 
• In hypercholesterolemic men and postmenopausal women, consuming 40 g/day ground flaxseed in baked products lowered Lp(a) levels by 14% and HOMA-IR by 23.7%, and this intake level was well-tolerated.
• In clinical trials, aspirin dosages of 81 or 150 mg daily significantly lowered Lp(a) levels, especially in persons at higher baseline concentrations; all the same, risks inherent to aspirin and aspirin dosage should be considered for each individual. 
• A meta-analysis found that the use of extended-release niacin could significantly lower Lp(a) levels by almost 23%, and effects at dosages under 2000 mg/day were comparable to higher dosages in the trials studied; again, niacin-related risks should be considered for each individual. 
• One meta-analysis of studies in healthy women found hormone replacement therapy (HRT) use associated with around 19% lower Lp(a) levels and reduced CVD risk, while another in diabetic and prediabetic women found that it (oral forms) lowered fibrinogen and platelet activator inhibitor-1 levels as well as reducing Lp(a) levels by 25% in these women; again, though, HRT may be better for women with higher baseline Lp(a) levels, and postmenopausal use primarily to manage CVD risk may not be recommendable. 

Lingering Questions and the Lp(a) Rabbit Hole

It seems very likely that lifestyle interventions that don’t necessarily impact Lp(a) levels still influence vascular health and function in beneficial ways that limit the production and pathogenicity of Lp(a). Here are a few questions we have about Lp(a):

• Could Lp(a) be a milieu marker, a canary for the cardio coalmine? Might it communicate an increasingly pro-oxidant exposome, or reflect body burden of lipid damage?
• Is Lp(a) a chaperone protein to limit the activity of modified lipids that might otherwise induce further damage? 
• Could lifestyle therapies targeting redox biology and metabolic flexibility address production and modification of Lp(a)? Might fat-soluble vitamins, omega-3s, or antioxidants affect Lp(a) synthesis or enzymatic detoxification of damaged lipids? 

Lp(a)’s Fourth Dimension: Oxidation and Other Modifications

Lp(a) is quite prone to oxidation (more so than LDL), and can be glycated too—and these modifications clearly impact vascular health. Lp(a) is inherently pro-inflammatory, yet its atherogenic potential is increased by binding to oxidized phospholipids (oxPLs) generated by cell membrane stress. Lp(a) is the main lipoprotein carrier of these oxPLs, which have been called ”atherosclerosis-relevant antigens” for the strong immune response and altered genetic expression they generate within blood vessels. In fact, Lp(a)-related risk may follow directly from its oxPL contents, which are associated with small Lp(a) isoforms and predict cardiovascular events. OxPLs may mediate vascular remodeling, drive calcification in aortic valve stenosis, and they are found in unstable atherosclerotic plaques. OxPLs are additionally known to promote immune cell binding to endothelial cells, interact with pattern recognition receptors, and induce proinflammatory and proatherogenic gene expression. Let’s not forget about apolipoprotein B, though, because oxidation of apoB-related lipids correlates with cardiac events, and levels of these “oxPL-apoBs” predict coronary and peripheral artery disease. One study found that Lp(a) and oxPL-apoB levels showed similar associations with the severity and progress of atherosclerosis. 

Here are some brain-scratchers, though: according to human and animal studies, things like statins, garlic, and low-fat diets may actually increase oxPL-apoB or oxPL-Lp(a) levels even while lessening atherosclerosis, improving vascular function, or slowing coronary calcification. And though Lp(a) binds to tissues more tightly than LDL, this can be modified by N-acetylcysteine. So it isn’t just that particular Lp(a) component levels are “good” or “bad”—they reflect very dynamic processes. Some research points out that lipoprotein-associated phospholipase A2 may help degrade these deleterious oxPLs. 

But wait, there’s more: Lp(a) structure and genetic function

Lp(a) comes in 6 genetically-coded isoforms showing different functionality, and apo(a) contains looping “kringle” structures interlinked with glycosylated regions. Like Lp(a) itself, apo(a) and kringles also come in multiple forms, and smaller apo(a) isoforms may double the risk for CHD or ischemic stroke. These smaller apo(a) isoforms have fewer kringles, and are associated with higher Lp(a) levels and higher atherogenic and thrombotic potential, though as in other paradoxes of successful aging, some healthy octogenarians show just this small isoform/high concentration profile. Finally, Lp(a)’s overall size and structure can vary by apo(a) amount, the number of Lp(a) particles, and amount of cholesterol carried. So it seems worthwhile to speculate how much these differences (especially if oxidation is also involved—we’ll talk more about that) may contribute to Lp(a)’s pathogenic potential. 

Lp(a)’s Genetic Dimensions and Many Variants 

Lp(a) plasma levels are strongly heritable through multiple polymorphisms within the LPA gene, but there is also an APOE2 gene variant notably affecting them. Lp(a) level is further influenced by LPA variants that are more specific to kringle and/or apo(a) expression, encoding for varying numbers of kringles on apo(a) and thus different apo(a) size and mass. LPA genetics may explain over 90% of the variation in Lp(a) levels, with kringle number accounting for most of it. These differences in the molecular conformation of Lp(a) are reflected by the broad range of Lp(a) plasma levels seen across large populations, which can vary from under 0.1 mg/dl to over 200 mg/dl. It’s interesting to note that apo(a) shares around 94% molecular homology with plasminogen, which may provide some clues as to how both Lp(a) and apo(a) inhibit fibrinolysis and boost atherogenesis. 

One of the largest investigations into Lp(a) genetics, the INTERHEART study, found that the link between Lp(a) level and myocardial infarction risk was particularly high in Latin Americans and South Asians and intermediate for Europeans, Chinese (who showed the largest Lp(a) isoforms), and Southeast Asians, yet it was non-significant in Africans (even though they had the smallest Lp(a) isoforms) and Arabs. Risks for cerebral, aortic, and peripheral vascular incidents have also been associated with Lp(a) variants. 

Relevant Links

Lipoprotein(a) Foundation
www.lipoproteinafoundation.org