FGF21- Her Cholesterol Fell 53 Points. That Wasn’t the Interesting Part.

Abstract

A patient came to us in March with a total cholesterol of 298 and an LDL of 222. She made one change — she cut most of the dairy out of her diet — and by May her total cholesterol was 217 and her LDL was 143.

That is a large response to a dietary change, and it tells you something about her before we go any further: this is a woman whose lipid metabolism moves. Keep that in mind, because it matters for how much weight the rest of this deserves. She then held her diet steady. On May 25 we drew a fresh baseline panel — deliberately, the day before treatment, so that the starting point was current rather than two months stale. On May 26 she received FGF21 and Follistatin gene therapy. We drew the follow-up panel 45 days later, on July 9.

Here is what her panel looked like before and after.

Almost everyone reading a panel like this looks at the first line and the third. Total cholesterol down 53 points. LDL down 38. Good news, move on. The line that actually matters is the one in bold.

Counting particles instead of weighing cargo

Cholesterol does not travel loose in the blood. It is packaged into lipoproteins — LDL, VLDL, and their remnants — and every one of those particles carries exactly one molecule of apolipoprotein B on its surface. One particle, one ApoB. Always.

So LDL cholesterol and ApoB measure two different things. LDL-C tells you how much cholesterol is being carried. ApoB tells you how many vehicles are carrying it.

This distinction is not academic. What damages an artery wall is a particle small enough to lodge in it. A person with many small particles and a person with few large ones can have identical LDL cholesterol and quite different risk, because risk tracks the number of particles far more closely than the cholesterol inside them.

Now look at what happened to this patient. Her cholesterol load fell by a quarter. Her particle count fell by half.

Divide the two and you get the cholesterol carried per particle: 1.34 before, 1.93 after. Her remaining particles are 44% more loaded than the ones she started with. She has far fewer of them, and the ones left are large and buoyant rather than small and dense.

That is not a smaller version of the same lipid profile. It is a different profile.

Why a metabolic hormone does this

FGF21 is a hormone your body makes, principally in the liver, and it governs how you draw energy from fat and glucose. It is not a lipid drug. Its effect on cholesterol is downstream of what it does to metabolism — and it works at several points at once rather than blocking a single step.

It reduces what the liver ships out. FGF21 increases AMPK signaling in the liver, which drives fatty acid oxidation and suppresses new fat synthesis. Less fat available means fewer VLDL particles assembled and secreted. Since LDL is what VLDL becomes after it delivers its triglyceride cargo, fewer VLDL particles leaving the liver means fewer LDL particles arriving downstream. This is the most direct explanation for a 49% fall in ApoB: the particles were never made.

It increases what the liver takes back. FGF21 raises hepatic LDL receptor expression, accelerating clearance of VLDL remnants through the ApoE-LDLR pathway. The LDL receptor also degrades ApoB after translation, which reduces secretion further — the same effector working from both ends.

It puts fat tissue to work. Acting through the FGFR1c/β-Klotho complex on adipose tissue, FGF21 activates brown fat and induces browning of white fat via UCP1. Brown and beige fat are metabolically expensive tissue, and they accelerate the breakdown of triglyceride-rich lipoproteins. Much of the systemic benefit appears to run through increased adiponectin.

It moves cholesterol out of cells. A separate line of work describes FGF21 inducing autophagy-mediated cholesterol efflux through RACK1.

The triglyceride fall of 41% here is squarely in line with what FGF21 does in controlled trials — the analogs pegozafermin, LLF580 and efluxifermin have reduced triglycerides by 29% to 63%. The LDL and total cholesterol changes are also within the published range; efluxifermin at high dose has produced total cholesterol reductions up to 20% and LDL reductions up to 40%.

In other words, nothing about the atherogenic side of this panel is anomalous. It is what this hormone is known to do, showing up in one person.

The gene therapy question

There is one reason to suspect that delivering the gene may not be equivalent to injecting the protein, and it concerns how the body disposes of cholesterol permanently.

The only route that removes cholesterol from the body rather than moving it around is conversion to bile acids and excretion. The rate-limiting enzyme is CYP7A1.

Given as a protein, FGF21 acutely suppresses CYP7A1 — it induces ERK phosphorylation and inhibits the gene, though less potently than its close relative FGF19.

Under sustained expression, the direction reverses. When FGF21 was overexpressed continuously in mice using a viral vector, CYP7A1 expression rose and the bile acid pool grew significantly. The mechanism appears to be antagonism of FGF15/19 at the liver β-Klotho/FGFR4 receptor, lifting the normal feedback brake on the enzyme.

A weekly injection and continuous endogenous production are not the same exposure, and in the one model that examined it, that difference flipped this pathway from off to on.

We want to be careful here. This is a mouse finding, using a viral vector rather than our non-viral construct, and it has not been demonstrated in humans. We raise it because it is the most scientifically interesting open question about our modality, not because we are claiming it happened in this patient. We have not measured her bile acids.

What we should add

Two things in this panel do not fit, and we would rather say so.

Her LDL was calculated, not measured. The VLDL figure in both panels is exactly one fifth of the triglycerides, which means LDL was derived by the Friedewald equation rather than measured directly. Friedewald becomes less reliable as triglycerides change, and her triglycerides changed a great deal. This is one reason to trust the ApoB number more than the LDL number — ApoB is measured directly, and it does not care what her triglycerides were doing.

And this is one person. No control, no washout, one follow-up timepoint at 45 days — which may not represent steady state, since expression from a construct takes time to plateau. She received two therapies on the same day, so nothing here separates the contribution of FGF21 from that of Follistatin. She had already demonstrated, two months earlier, that her lipids respond strongly to what she eats. A single case tells you what is possible. It cannot tell you what is typical, and it is not evidence that anyone else will respond this way.

The reason we are writing this up anyway

Because the shape of the result is mechanistically coherent, and because the marker that moved most is the one most people never order.

If you have had a lipid panel recently, there is a reasonable chance it reported total cholesterol, HDL, triglycerides, and a calculated LDL — and no ApoB. That panel would have shown this patient a 24% improvement. It would have missed that she halved her circulating particle count.

Ask for ApoB. Whatever you are doing about your metabolic health, it is the number that will tell you whether it is working.

Blast Longevity is a biotechnology company. We develop and manufacture gene therapy products; we do not diagnose, prescribe, or treat. The case described here reflects one patient’s laboratory results and is not a claim of typical or expected outcome. Our products have not been evaluated or approved by the FDA or any comparable regulatory authority and are not intended to diagnose, treat, cure, or prevent any disease. Decisions about lipid management — including any decision regarding prescribed medication — belong with your physician.