Beyond the Single Number: Toward a Functional Classification of Lipoprotein(a) Subfractions

Mykola Iabluchanskyi, Vladimir Shlyakhover, Pavlo Garkaviy, Andriy Yabluchanskiy

Abstract

Lipoprotein(a) [Lp(a)] is currently measured and interpreted as a single quantitative entity, with plasma concentration used as a population-level indicator of cardiovascular risk above defined thresholds. Yet established structural evidence already shows that Lp(a) is not one particle but a heterogeneous family of molecularly distinct subfractions, differing in apo(a) isoform size, oxidized phospholipid content, and fibrin-binding properties. This article advances the hypothesis that structural heterogeneity corresponds to functional heterogeneity: that distinct Lp(a) subfractions may operate in different biological modes, ranging from physiological and reparative to pathological and atherogenic, and that the balance among these modes, rather than total concentration alone, may help determine individual clinical outcome. If correct, this view challenges the adequacy of single-number Lp(a) measurement as a clinical standard and suggests that concentration-based assessment may conflate biologically different states. It also raises a therapeutic question: non-selective lowering of all Lp(a) subfractions may not necessarily produce optimal clinical benefit if physiological and pathological forms are reduced simultaneously. By analogy, earlier HDL-directed strategies that increased total HDL concentration without resolving its internal functional heterogeneity failed to demonstrate convincing clinical efficacy. The article therefore argues for a new research agenda centered on the functional classification of Lp(a) subfractions and on the identification of conditions that shift the system from adaptive to maladaptive operation.

Introduction

For most of its clinical history, lipoprotein(a) [Lp(a)] has been treated as a single measurable entity. A blood test returns one number, and that number is compared with a clinical threshold above which cardiovascular risk is considered elevated. This logic is epidemiologically supported: elevated Lp(a) is an independent, causal, and largely genetically determined risk factor for atherosclerotic cardiovascular disease and calcific aortic valve stenosis, and its predictive value at the population level is well established. This article is intended as a conceptual and hypothesis-generating contribution rather than as a systematic review.

Yet a single number is also a radical simplification. It collapses whatever internal structure and functional diversity the measured population of molecules may possess into one scalar. The question this article raises is whether that simplification is adequate — or whether it hides distinctions that may matter more than total concentration alone.

The history of natural science offers a sobering precedent. For much of the 19th century, the atom was defined by its indivisibility: it was the final, irreducible unit of matter. That assumption was productive and not entirely wrong — atomic theory organized chemistry and correctly predicted observable phenomena. But it was also incomplete. The electron was discovered in 1897. The proton and neutron followed. As experimental resolution improved, the atom revealed internal structure of extraordinary complexity: quarks, gluons, leptons, bosons. The number of known subatomic and composite particles now exceeds the number of elements in the periodic table by an order of magnitude. Each new level of resolution did not invalidate the previous one — it contextualized it, explaining why atomic-level observations were correct at their own scale while remaining incomplete at the next.

The parallel with Lp(a) is not metaphorical. It is structural.

Lp(a) is already known to be structurally heterogeneous

The structural heterogeneity of Lp(a) is not a new finding. It is already part of the established science, though its clinical implications have not been fully drawn.

Lp(a) consists of an LDL-like lipid core containing apolipoprotein B100, to which apolipoprotein(a) [apo(a)] is covalently linked. Apo(a) is a highly polymorphic protein structurally homologous to plasminogen. Its key source of diversity is the number of kringle IV type 2 (KIV-2) repeat copies, which vary from 2 to more than 40 copies across individuals, generating over 40 structurally distinct apo(a) isoforms. Smaller isoforms — with fewer KIV-2 repeats — are associated with higher plasma Lp(a) concentrations, while larger isoforms produce lower concentrations.

Beyond isoform size, individual Lp(a) particles differ in their content of oxidized phospholipids (OxPL), which are preferentially carried by Lp(a) and are increasingly recognized as a biologically active component that may be a stronger predictor of cardiovascular risk than total Lp(a) concentration. Particles also differ in their lysine- and fibrin-binding affinity, which determines their interaction with fibrin clots and their capacity to inhibit fibrinolysis. These are not trivial differences. They describe particles that, while sharing a common structural class, may behave differently in the biological environments they encounter.

What current clinical measurement does — reporting a single Lp(a) concentration — is equivalent to reporting the total atomic weight of a sample without specifying which isotopes or elements are present. It captures something real. But it compresses functionally relevant information into a number that cannot carry it.

The physiological and pathological modes of Lp(a)

A further and less appreciated dimension of Lp(a) biology is its apparent capacity to operate in distinct functional modes depending on biological context.

The physiological mode is now reasonably well documented, though it remains mechanistically incomplete. Lp(a) accumulates at sites of vascular and tissue injury. It participates in wound healing and vascular repair, in part by delivering cholesterol for membrane regeneration and in part by modulating fibrinolysis at sites of damage — a role that follows directly from the structural homology between apo(a) and plasminogen. Acute-phase responses to myocardial injury include an initial fall in Lp(a), followed by a rebound increase of up to threefold, consistent with a reparative deployment of the particle. These observations support the interpretation that Lp(a) evolved, at least in part, as a component of the organism's response to tissue damage: a mechanism for delivering lipid material and regulating clot dissolution at sites where both are acutely needed.

The pathological mode emerges when this system operates chronically and systemically at high concentrations. The same molecular properties that serve a reparative function in acute contexts become atherogenic and prothrombotic when present continuously in the circulation at high levels. Lp(a) deposits in the arterial wall and contributes to plaque formation. It inhibits fibrinolysis in a diffuse, non-localizing manner that increases thrombotic risk. Its OxPL cargo drives chronic vascular inflammation and may promote calcific remodeling of the aortic valve.

The critical observation is that these are not two different particles. They are the same particle family operating under different conditions. What distinguishes the physiological from the pathological mode is not primarily the molecular structure of Lp(a) itself, but the context: the concentration, the duration of exposure, the state of the vascular wall, and the metabolic condition of the organism. This is a distinction that the single-number clinical measurement cannot make.

The anomaly that demands explanation

The strongest empirical argument for this hypothesis is a well-recognized anomaly that current Lp(a) models handle poorly: a meaningful proportion of individuals with very high Lp(a) concentrations have no detectable arterial disease and no cardiovascular events over extended follow-up.

At the population level, the Lp(a)–cardiovascular risk relationship is statistically robust. At the individual level, it is substantially imprecise. A patient with Lp(a) of 200 nmol/L may have severe premature atherosclerosis. Another patient with the same measured level may have clean coronary arteries at the age of 65.

Several explanations have been proposed: differences in isoform size, differences in OxPL content, differences in co-existing risk factors, differences in arterial wall biology. All of these are plausible and probably partially true. But none of them has been organized into a coherent framework that explains the variance. The hypothesis presented here offers such a framework: the clinically relevant variable may not be how much Lp(a) is present, but in what functional mode it is operating — and the single concentration number conflates individuals whose Lp(a) is predominantly in a reparative mode with individuals whose Lp(a) is predominantly in an atherogenic mode.

Implications for measurement, research, and treatment

If the functional heterogeneity hypothesis is correct, several implications follow.

For measurement, the single Lp(a) concentration should eventually be supplemented by a functional profile: isoform characterization, OxPL quantification, and potentially fibrinolytic activity of the Lp(a) fraction. This would constitute a genuine advance in clinical precision, analogous to the transition from reporting "total white blood cell count" to reporting a differential with functional subclassification.

For research, the key question shifts. Rather than asking "how much Lp(a) is present?", the productive question becomes: "what kind of Lp(a) is present, in what functional state, and under what conditions does it shift from reparative to atherogenic?" This reframing opens a research agenda that current trial designs, which focus almost entirely on concentration reduction, are not equipped to address.

For treatment, current RNA-targeted therapies achieve profound reductions in total Lp(a) — by 80 to 95 percent in phase 2 and phase 3 trials. If the functional heterogeneity hypothesis is correct, these therapies reduce physiological and pathological Lp(a) subfractions simultaneously. Non-selective lowering of all Lp(a) subfractions may therefore prove insufficient for achieving clinically meaningful benefit with newly developing Lp(a)-targeted therapies. A similar lesson may be drawn from previous HDL-directed strategies: increasing total HDL levels without adequate consideration of the particle’s internal structural and functional heterogeneity did not result in convincing clinical efficacy. This may ultimately prove to be an appropriate strategy, but it also raises the question of whether future, more refined interventions might selectively neutralize the atherogenic mode while preserving the reparative one — a question that cannot even be formulated within the current single-number framework.

For conceptual framing, this hypothesis is consistent with and extends the view of atherosclerosis as a disorder of disrupted lipid homeokinesis. On that view, the pathological state arises not simply when Lp(a) concentration is high, but when the Lp(a) system loses its homeokinetic regulation — when what was an adaptive, context-sensitive reparative mechanism becomes a chronic, context-insensitive atherogenic process. The quantity rises, but more fundamentally, the functional character of the circulating population shifts.

Conclusion

The periodic table of Lp(a) — its full functional taxonomy — does not yet exist. What exists is a structurally heterogeneous particle family, a well-documented physiological role that coexists with an equally well-documented pathological one, and a clinical anomaly that the current single-number measurement cannot explain. Together, these observations are sufficient to formulate a testable hypothesis: that Lp(a) concentration, while epidemiologically valid as a population-level risk indicator, conflates functionally distinct subfractions whose ratio, rather than their sum, determines individual vascular outcome.

The atom, too, was once considered indivisible. Its indivisibility was not an error — it was a productive first approximation that correctly organized observable chemistry. The discovery of internal structure did not invalidate atomic theory. It completed it.

Lp(a) measurement may be at a similar stage. What is needed now is not a rejection of concentration-based risk assessment, but its extension toward a functional classification that the current number cannot provide. Building that classification is among the most important open tasks in lipidology.

Note on Intellectual Background

The views presented in this article were shaped by the work of many scientists who contributed to the understanding of atherosclerosis, lipoprotein metabolism, vascular inflammation, and thrombosis. Because this text is conceived as a synthetic conceptual reflection rather than a formal literature review, we did not attach individual citations to each idea. We acknowledge, with respect, that many investigators working in this field may find a part of their intellectual contribution reflected in the argument presented here.