ApoB, LDL and Lp(a): What Is Actually Inside What

Three words get used as if they were three items on the same list: ApoB, LDL and Lp(a). People ask whether they should get ApoB "instead of" LDL, or whether Lp(a) is a type of cholesterol, and the answers are confusing because the question has a category error inside it. These are not three things floating side by side in your blood.

Here is the whole relationship in one sentence. ApoB is a protein — a single enormous molecule wrapped around the outside of a particle, exactly one per particle, for life. LDL is a particle that carries one. And Lp(a) is that same particle with a second protein, apo(a), covalently attached to it. Not a separate substance, not a separate family — a member of the same family wearing an extra part.

Once that is clear, the things that sound like trivia on a lab report stop being trivia: why two people with identical LDL cholesterol can be at very different risk, why Lp(a) hides inside both of your other numbers, and why "high-density lipoprotein" never meant "good". The atlas below builds the picture one step at a time, starting from a single particle.

ApoB is a physical object, not a category

The first step of the atlas magnifies a single LDL particle. The red band around the outside is apoB-100, one protein molecule of 4,536 amino acids, long enough to wind right around the particle like a strap. It is put on when the particle is assembled and it never leaves, never swaps and is never joined by a second copy.

Drag the cargo slider and watch what happens: the amount of cholesterol inside changes continuously, and the count of apoB stays at one. That is the entire basis of the distinction. "How much cholesterol is in circulation" and "how many particles are carrying it" are genuinely different quantities, and one strap per particle is what makes the second one countable at all.

So ApoB is not an abstract risk category or a summary score. It is a measurement of how much of one specific protein is in a millilitre of your plasma — which, because of the one-per-particle rule, is a census of the particles that can enter an artery wall.

What a lipoprotein particle is actually made of

Step 2 cuts a particle open. There are only five ingredients in the entire family, and every particle is built from the same five in different ratios: protein, phospholipid, free cholesterol, cholesteryl ester and triglyceride. They are arranged in two zones — an oily core that avoids water, and a surface exactly one molecule thick that touches it.

Notice where cholesterol appears: twice. Esterified cholesterol sits in the core as cargo. Free cholesterol stands in the surface as structure, because that form needs to be in contact with water. When a lab reports "LDL-C" it adds both together, which is a reasonable thing to do and also a reminder that the number is a mass of a molecule, not a description of a particle.

That single fact also explains the names. These particles were sorted by spinning blood in a centrifuge long before anyone knew their function, and since fat floats and protein sinks, density is really the protein-to-fat ratio. "High-density lipoprotein" means enough protein to be dense. It has never meant good.

LDL-C weighs the cargo. ApoB counts the trucks.

Step 3 makes the consequence unavoidable. It shows a bloodstream in which LDL cholesterol is pinned at 100 mg/dL and cannot move, and lets you change only the size of the particles carrying it. Small particles hold less cholesterol each, so it takes more of them to add up to the same total — and the particle count on screen swings enormously while the cholesterol number does not move at all.

Every one of those particles carries one apoB, and every one is an independent opportunity to be retained in an artery wall. This is why the two tests can point in different directions in the same person: LDL-C answers "how much cholesterol is being carried", ApoB answers "how many carriers are there", and the artery responds to the second question. Where the two disagree — most often in people with high triglycerides, insulin resistance or metabolic syndrome, who tend to carry many small cholesterol-poor particles — risk follows the particle count.

What to do with a specific ApoB number, including what targets are used and how it is lowered, is the subject of our ApoB guide. This page is about why the measurement exists in the first place.

Lp(a) is an LDL particle, plus one extra protein

Step 4 answers the question people find hardest: how is Lp(a) "part of" anything? Take an ordinary apoB-100 particle — the one from step 1, cholesterol core and all — and covalently attach a second protein called apo(a). The result is called Lp(a). Nothing was removed. The apoB is still there, the cholesterol is still there.

Which produces a genuinely awkward situation on a lab report. Your ApoB result counted that particle, because it wears an apoB — but as one ordinary particle, with no indication of the apo(a). Your LDL-C result weighed its cholesterol, because standard assays cannot separate Lp(a) cholesterol from LDL cholesterol. Lp(a) is inside both numbers, and invisible in both.

The apo(a) is not decorative. It makes the particle stickier in the artery wall, and it closely resembles plasminogen, the protein the body uses to dissolve clots — which is the mechanistic reason Lp(a) is associated both with atherosclerosis and with clotting. Levels are largely genetic and stable through life, which is why it is generally measured once rather than tracked, and why it is reported in nmol/L, a count of particles, wherever modern assays are available.

Which blood test can see which particles

Step 5 lays the whole family out as boxes inside boxes, then lets you click a blood test and watch the particles it cannot see go dark. There are nine particle types on screen; here is what each test covers.

• •ApoB — every apoB-carrying particle: chylomicrons and their remnants, VLDL, IDL, both LDL sizes, and Lp(a). One count for the entire atherogenic fleet.
• •Non-HDL-C — the cholesterol in exactly the same set of particles. Same coverage as ApoB, but by weight rather than headcount, and it is already derivable from any standard panel.
• •LDL-C — only the LDL-sized particles, plus Lp(a)'s cholesterol folded in silently. Blind to every remnant, and blind to how many particles the cholesterol is spread across.
• •Remnant cholesterol — non-HDL-C minus LDL-C: the cholesterol in chylomicron remnants, VLDL and IDL. These enter the artery wall as readily as LDL does, and almost no panel reports them as a line item.
• •Lp(a) — exactly one box, the one that both ApoB and LDL-C were already including without telling you. It is the only test that can detect the apo(a).
• •HDL-C — the cholesterol in the apoA-I particles, a different family entirely. No apoB anywhere in that box.

Seen side by side, the standard panel's weakness is obvious: it reports in detail on one box, ignores the remnants next to it, and hides Lp(a) inside both. That is not an argument for ordering every test — it is the reason ApoB and non-HDL-C keep moving up the guidelines while LDL-C stays the number everyone quotes.

So is the plaque made of apoB?

No — and step 6 is worth watching for this alone, because it resolves a confusion that follows naturally from everything above. If apoB is what matters, why is the plaque made of cholesterol?

Because apoB is the anchor, not the payload. A particle drifts into the artery wall, and its apoB grips the proteoglycan chains of the matrix. Particles that grip stay; particles that do not drift back out and are harmless. The retained particle then unloads its cholesterol, which stays essentially forever, while the apoB protein itself is broken down and disappears. ApoB counts the deliveries. Cholesterol is what gets dumped.

That gives two independent levers, and the step lets you move both: send fewer particles, or make each one grip less well. Binding avidity is raised by small dense LDL, by glycated apoB in poorly controlled diabetes, by apo(a) on Lp(a), and by changes in the wall matrix itself. It is why the same ApoB level is not equally dangerous in everyone, and why the two levers multiply rather than add.

What happens to that cholesterol once it is retained — how it becomes a necrotic core under a fibrous cap, and what makes that cap fail — is the subject of the plaque rupture atlas. This page ends where that one begins.

The number that decides everything: ApoB × years

The final step plots a lifetime. Plaque does not respond to your ApoB today; it responds to every year you have spent above the level you were born with, which is roughly 30 mg/dL. What matters is the shaded area under the curve — concentration multiplied by time.

Drag the treatment start age and watch the crossing point move. Starting at 35 pushes the threshold much further than starting at 55 does, and by far more than the difference in ApoB alone would suggest, because early years are multiplied by more remaining time. This is the same asymmetry that genetic studies keep finding: people who inherit lifelong lower ApoB get substantially more protection per unit of reduction than treatment started in middle age delivers.

It also reframes a common sentence. "My cholesterol is fine now" describes a point. The arteries are integrating an area.

Frequently asked questions

The honest limits of this atlas

The structural facts are standard lipoprotein physiology: the composition ratios, the diameters, the one-apoB-per-particle rule, the apoA-I stoichiometry. The numbers that move as you drag sliders are not measurements. Particle counts are derived from cholesterol-per-particle scaling with core volume, calibrated so that an average 22 nm particle diameter gives roughly ApoB 85 mg/dL at LDL-C 100 — chosen to illustrate published discordance, not to reproduce anyone's panel.

The retention step in the atlas is a cartoon of a process that takes decades, run at a speed you can watch. The lifetime exposure curve uses a single event threshold applied to everyone, when in reality that threshold varies with blood pressure, smoking, inflammation, diabetes and genetics — none of which appear here.

Use it to understand what these tests are measuring and how they relate. Do not use it to interpret your own results, and do not use it to decide what to test — both of those need a clinician who knows the rest of your picture.