IGF-1 LR3: The Downstream Signal, Injected, and Never Studied in a Person

Skip the gland and inject the signal. That is the pitch.

What it is, and what it was built for

Francis and colleagues described it in 1992 at CSIRO in Adelaide. Native IGF-1 is 70 amino acids, this is 83: glutamic acid at position 3 swapped for arginine, the change aimed at cutting affinity for the IGF binding proteins, plus a 13 residue extension on the front, 11 amino acids of methionyl porcine growth hormone and a Val-Asn linker, which began as an E. coli expression construct and lowers that affinity further (PMID 1378742). Much reduced IGFBP affinity is the whole premise, and also the problem.

It sells as a cell culture supplement, on pages reading "WARNING This product is not for human or veterinary use", and a 2006 paper calls it "specifically engineered for use in biopharmaceutical protein production in mammalian cells" (PMID 17172665). Do not over-read that. It came out of a therapeutic analogue programme and went through rats, guinea pigs, pigs and sheep as a drug candidate. It failed to become a drug. It was not designed to avoid being one.

The drug it is not

Mecasermin, sold as Increlex, approved in the United States in 2005. Recombinant human IGF-1, "70 amino acids in a single chain with three intramolecular disulfide bridges", the molecule your own liver makes, for growth failure in children 2 years and older with severe primary IGF-1 deficiency, subcutaneously at 0.04 to 0.12 mg/kg twice daily.

None of that transfers to LR3, and most vendor copy behaves as though it does. Read the label anyway, because it is what the parent hormone looks like when somebody measures. Roughly 42 percent of subjects had hypoglycaemia at least once, with 5 cases of severe hypoglycaemia and 4 hypoglycaemic seizures. It is contraindicated in active or suspected malignancy, and section 5.7 records postmarketing malignant neoplasms in paediatric patients, more often at doses above those recommended.

Twenty to thirty hours, from nowhere

Most IGF-1 in blood sits in a ternary complex with IGFBP-3 and the acid labile subunit, and that complex is what extends its serum half-life (PMID 35907924). LR3 was engineered not to join it, then sold as long acting. That is taking the lid off the pot to make the water boil faster and quoting the time it took with the lid on.

The primary literature says so outright: LR3 "has very low affinity for the IGF-binding proteins in the rat and hence is cleared from the circulation more quickly than is IGF-I" (PMID 8708565). After an intravenous bolus in virgin rats, plasma metabolic clearance ran 9.84 ml/min/kg for LR3 against 0.90 for native IGF-I (PMID 7693845). I could not trace the 20 to 30 hour figure to a published measurement in any species. Native recombinant human IGF-1 has a subcutaneous half-life of about 20 hours in healthy volunteers, and the number looks inherited from the parent.

What the animals actually grew

In 150 g rats dosed subcutaneously for seven days at about 2,133 to 2,667 micrograms/kg/day, continuous infusion made LR3 1.5 to 2-fold more potent than IGF-I for body weight gain, visceral organ weights and feed use efficiency, an advantage much reduced under once daily injection, which is how the compound actually gets used. For reversal of carcass muscle loss in the steroid treated rats it was barely equipotent with IGF-I (PMID 8708565). Rats held at 100, 78, 56 or 33 percent of free feeding and infused at about 893 micrograms/kg/day kept body weight 3 to 8 percent higher and retained more nitrogen at every level of intake, both at P below 0.001, and muscle protein was not conserved (PMID 11472075). In guinea pigs, a seven day infusion at about 343 micrograms/kg/day raised the fractional weight of adrenals, gut, kidneys and spleen, and the authors state overall growth was not stimulated (PMID 7561636).

Then the pigs. Four days of infusion at 180 micrograms/kg/day in finisher pigs cut average daily gain and food intake, with mean growth hormone down 23 percent and the area under the growth hormone peaks down 60 percent. That paper states as settled background that analogues binding poorly to the IGFBPs stimulate growth in the rat and inhibit growth in the pig (PMID 9488001). Which of the two a human resembles is not a small question, and I found no resolution of it.

A week of intravenous LR3 into normal fetal sheep at 6.6 micrograms/kg/hour, about 158 micrograms/kg/day, left heart, adrenal and spleen heavier and lowered insulin, while fetal weight itself was not statistically different from saline, P = 0.15 (PMID 33427051). In growth restricted fetuses at about 28 micrograms/kg/day, a week changed nothing but a fall in circulating amino acids (PMID 39679943).

Gut, kidney, spleen, adrenals, heart. Skeletal muscle is the one thing it is bought for and the one thing that record is thinnest on.

The site injection claim is everywhere in community writing. I found no study testing intramuscular LR3 injection for localised muscle growth in any species. One site specific study does have a growth endpoint, and it was negative: long-R3-IGF-I infused into a skin artery in six conscious sheep for four hours raised local blood flow, oxygen consumption and amino acid uptake, and "No effects of IGF-I were found on replicating cell numbers in the bulbs of wool follicles" (PMID 8133213).

The harm that has been measured

Pigs, bolus injections at 20 and 50 micrograms/kg. The low IGFBP affinity variants, LR3 among them, were consistently 2 to 3-fold more potent than native IGF-I at driving plasma glucose to its nadir, and cumulative suppression across four hours showed roughly a 4 to 8-fold increase in the extent of hypoglycaemia. Maximum glucose lowering in the pig, 4.8 mmol/l (PMID 9415072). That paper's title names marmoset monkeys too, and the marmoset arm is real, but those animals received only IGF-I and des(1-3)IGF-I. LR3 is in the pig arm alone. If you see LR3 hypoglycaemia attributed to monkeys, that experiment was not run.

The same authors note that these variants clear faster than native IGF-I and yet suppress glucose harder and for longer. Short acting in blood, long acting on blood sugar. And there is no published measurement of blood glucose in a human given IGF-1 LR3, at any dose, by any route.

Where 20 to 100 mcg came from

The convention that circulates is roughly 20 to 100 mcg per day subcutaneously, sometimes split into two injections, in cycles of about 4 to 6 weeks. I found no derivation for it anywhere. No allometric scaling from a named animal study, no dose ranging work, no pharmacokinetic anchor.

Per kilogram, for an 80 kg adult, that is 0.25 to 1.25 micrograms/kg/day, and every published dose sits above it: growth restricted fetal sheep about 28, which is 22 to 112 times higher, normal fetal sheep about 158, pigs 180, guinea pigs about 343, food restricted rats about 893, steroid treated rats about 2,133 to 2,667. The pig bolus doses that produced measurable hypoglycaemia were 20 to 50 micrograms/kg in one injection, 16 to 200 times the convention. None of which means the convention dose is safe, because nobody has measured anything in a person at any dose.

What is in the vial, and what I could not check

Three anti-doping laboratories have opened black market material and found His-tagged Long-R3-IGF-I, read as likely a research by-product rather than something made for injection (PMID 20675162), unpurified long-R3-IGF-1 among confiscated products analysed in 2009 (PMID 21204286), and abundant oxidised forms (PMID 33587816). None was surveying vials against their labels, so none gives you a rate. I also found no carcinogenicity, repeat dose toxicology or genotoxicity study of LR3 in any species.

One of those papers states that these analogues including LongR3-IGF-I "were never approved for use in humans, they are readily available as black market products for bodybuilding", and I found no approval anywhere, in any jurisdiction. WADA prohibits it at all times under S2, Peptide Hormones, Growth Factors, Related Substances and Mimetics, though I could not open the 2026 Prohibited List itself, so I am not printing a sub-section number I did not read. If you go and check any of this yourself, use the spellings LR3IGF-I and LR3-IGF-I as well as LongR3-IGF-I, because adding them takes the corpus from roughly 78 to 114 records and the sheep skin study lives in that gap.

My read

What bothers me most is the borrowing. Increlex hands this category a real label, real trials, real hypoglycaemia numbers and a real oncology warning, and that document gets waved over a bacterial expression construct. None of it applies. What has been measured about LR3 itself is that in pigs it drops blood sugar harder and for longer than the native hormone while leaving the bloodstream faster.

If you have bought IGF-1 LR3, I want to know two things about the page you bought it from: what half-life figure it printed, and whether it said anything at all about blood sugar.

---

This content is for educational and informational purposes only and is not medical advice. Peptides discussed are research compounds and may not be approved for human use. Nothing here should be used to diagnose, treat, cure, or prevent any disease. Always consult a qualified healthcare professional before starting any peptide, supplement, or protocol. Individual responses vary. Do not self-administer compounds without proper medical supervision.