Tony Huge calls himself a walking cardiovascular time bomb: one copy of APOE4, homozygous risk at all three standard 9p21 markers, none of the protective variants, and 14 years of testosterone and anabolics on top. Then he pulls the blood work. LDL 82.6. Lipoprotein(a) at the floor of the range. This video is his audit of both columns, what the genetics predicted, what the labs actually show, and the 30-year-old testosterone study he thinks got buried.
Video: My Heart Genetics Are Terrible. So I Tested Everything. Published September 8, 2026. Watch on YouTube.
What the video covers
Timestamps below are approximate, placed from the transcript against the 18-minute runtime.
- 00:00 The starting position: APOE4, 9p21, 14 years of anabolics, and the claim that biohacking can overcome genetics.
- ~01:45 The genetic file: No protective variant at PCSK9, the LDL receptor, SORT1, CETP, or LPL.
- ~03:00 What “worst genetics” means: Common variants, modest shifts, a loaded start, not a guaranteed heart attack.
- ~04:15 The panels: Total cholesterol 156, LDL 82.6, triglycerides 102, HbA1c 5.4, hs-CRP 0.087, normal EKGs.
- ~05:45 What he used: GW501516, citrus bergamot, rosuvastatin, bempedoic acid, ezetimibe, and the PCSK9 inhibitor he has never actually taken.
- ~08:00 HDL at 24: It was 24 before anabolics, 13 on a full cycle, and in the 40s on a supplement protocol.
- ~09:30 Hematocrit, hepatic lipase, liver enzymes: 59% hematocrit, the 1999 transdermal testosterone study, AST 60 and ALT 66.
- ~11:30 Lipoprotein(a) at 3.43: Why it matters more than people realize, and the 1996 and 2004 testosterone studies.
- ~13:00 The 93.9% injection: How small interfering RNA stops the protein from being built at all.
- ~14:30 Checking the LPA gene: Zero copies of both common risk variants, and the alternative explanations.
- ~16:00 Both columns: Credits, debits, and the outcome trial that does not exist.
- ~17:00 What he is doing next: Another phlebotomy, more TUDCA, and more cardio.
What a loaded starting position means
Tony is precise about what his genetics do and do not say. APOE has three versions and you get two; his E4 copy is the one linked to higher LDL and higher cardiovascular risk. Chromosome 9p21 is the strongest common coronary artery disease signal found in the human genome, replicated repeatedly, with three standard markers, and he is homozygous risk at all three. He then lists the protective variants he lacks: PCSK9, the LDL receptor gene, SORT1, CETP, LPL. Every common variant that would give him a break comes up empty.
What that does not mean is a guaranteed heart attack. These are common variants and each shifts risk by a modest amount. What it does mean is that standard reassurance was built on an average person, and he is not one. His broader argument is that asking about family history is the old-fashioned version of this question. Parents’ outcomes carry too many variables to be useful; the genetics themselves are the more accurate instrument.
The labs that should not exist
Tony reads from two full panels drawn ten days apart in late 2023, chosen because one includes lipoprotein(a). Total cholesterol 156, LDL 82.6, triglycerides 102, HbA1c 5.4, hs-CRP 0.087, and EKGs that always come back normal. For a man with his genetics, he says, an LDL of 82.6 is not the genetic baseline. Before he ever took a supplement, his LDL was extremely high and his HDL extremely low, which is what started his interest in the cardiovascular stack in the first place.
“When anybody tells me to be natural, I say they don’t understand that sometimes natural means your genetics programmed you for an early death.”
Tony Huge, approx. 05:20
The debit column is just as specific. HDL came back at 24 on one panel, against an optimal above 60. He is emphatic that it was 24 as a natural, before anabolics, so when a doctor blames the anabolics he corrects them: on a full cycle it dropped as low as 13, and on a supplement protocol he has pushed it into the 40s, higher than it ever was naturally. Hematocrit sat at 59 against a male ceiling of 55, hemoglobin 18.6 against 18. Liver enzymes were AST 60 and ALT 66, elevated on every test he has ever had, partly from training, which is why he suggests a two-day break from exercise before a liver panel. Testosterone was 194 with LH at 0.06 and FSH at 0.15, meaning natural production was shut down; he notes he has since used HMG and clomiphene and has five children conceived during those 14 years.
The hepatic lipase mechanism
Tony gives the HDL story a mechanism rather than leaving it as a lab value. An enzyme called hepatic lipase clears HDL out of the blood. Estrogen holds it back; androgens push it forward. He cites a 1999 study in which ten men on transdermal testosterone had hepatic lipase activity measured directly: it went up, and HDL subfractions went down. In plain terms, muscle-building hormones are also telling the liver to clear out the good cholesterol.
Lipoprotein(a), and the study he thinks got buried
This is the center of the video. Lipoprotein(a) came back at 3.43 on a reference range of 2 to 53. It is an independent heart risk factor, roughly one in five people carry a high level, almost nobody tests for it, and neither diet nor exercise lowers it. It is 70 to 90% genetic, about as heritable as height. With his genetics, Tony says, he should not be sitting at the bottom of that range, which means something he is doing has acutely overridden the genetic default.
His candidate is testosterone. Tony cites a 1996 cardiology paper titled “Testosterone decreases lipoprotein(a) in men,” with a 37% reduction. He then cites a 2004 study of bodybuilders on anabolic steroids in two groups: 35 men observed with Lp(a) falling from 189 to 32, and 16 men in a randomized, double-blind, placebo-controlled arm with Lp(a) falling from 103 to 65. That is 83% and 37%, and the second group had a placebo comparison. Statins are ruled out as his explanation, because a 2018 analysis found they cut LDL 39% while leaving Lp(a) untouched, and a network meta-analysis across 20 trials in more than 23,000 people agrees.
He contrasts this with the new injectable drugs, small interfering RNAs that find the messenger RNA for apolipoprotein(a) and destroy it before the cell reads it, so the protein never gets built. One shot, 93.9% reduction, still down six months later. That is why diet cannot work on Lp(a): diet acts on circulating particles, and the injection acts on whether they get made. Tony is clear he is not using that shot. He is pointing at it as the future, while asking why a cheap generic that may do part of the same job never got a follow-up trial. His answer is the usual one: nobody can patent testosterone.
Checking his own explanation
Then he tests himself and it gets complicated. He went into his raw genetic data for the two common variants that drive high Lp(a) and found zero copies of both. So his genetics are bad for LDL and HDL, but not necessarily for Lp(a). He offers the alternatives honestly. The biggest driver of Lp(a) is the number of repeating copies inside the LPA gene, which consumer DNA tests do not measure, so he could simply carry a low-producing version. Bempedoic acid, cardarine, SR9011, and citrus bergamot have no Lp(a) data he can find. And it was one test on one day, though he remembers a previous result also being low.
His hypothesis stays the same: most likely testosterone, secondarily the cardiovascular stack. His regret is not testing Lp(a) in his twenties, before any compound, so that the comparison could be made.
Cheat Sheet Pivot
What Tony reported using and testing. He states repeatedly that this is not advice, that he cannot give doses on the platform, and that the full protocol is in his community.
- The compound he credits most for fixing his cholesterol is GW501516 (cardarine), with citrus bergamot as a weaker second.
- He later added rosuvastatin at a dosing pattern that gave him most of the benefit without side effects, and ezetimibe, which he describes as low-risk and moderately effective.
- Bempedoic acid did less than expected in him, which he attributes to his genetics.
- He has never actually used a PCSK9 inhibitor, considers it the strongest option, and says he would if it were accessible and affordable.
- Fish oil, vitamin D, magnesium, and the basics are constant; TUDCA is his current liver support.
- For hematocrit he uses periodic therapeutic phlebotomy, about 500 mL, and reports feeling lighter afterward.
- His action item for viewers is testing: get a baseline, get Lp(a) specifically, and if testosterone is low and Lp(a) is high, raise the 1996 finding with a doctor who probably has not seen it.
Every compound above is indexed in the Miracle Molecules Cheat Sheet. For the full written version of the stack, read My Cholesterol Protocol: What I Take With the Worst Heart Genetics. For why Lp(a) is a test you only need once, see Lipoprotein(a): The Inherited Risk You Test Once.
Where the evidence stops
Tony lays both columns side by side and admits he does not know whether they cancel. Credits: LDL 82.6 with genetics that should have wrecked it, low inflammation, a clean EKG, Lp(a) at the floor with 30 years of studies suggesting testosterone as a reason. Debits: HDL 24, hematocrit 59, elevated liver enzymes, shut-down natural testosterone. There is no outcome trial. Nobody has shown that lowering Lp(a) with androgens prevents a single heart attack, and he notes the new siRNA injections face exactly the same gap until their outcome data arrives.
He also concedes the obvious: he does very little cardio, cardio is excellent for the heart, and taking handfuls of supplements instead of training is, in his word, ridiculous. He should be doing both.
Keep going
The earlier written breakdown of these same labs is at 14 Years on Gear: Heart Bloodwork and Lipoprotein(a). Companion articles for each new video go out through the tonyhuge.is email list, and the cardiovascular archive is at tonyhuge.is.