EPO: The Drug That Works — and the Trial That Showed Why That Is the Problem

EPO (Erythropoietin) is the one compound on this site whose problem is too much evidence rather than too little: a 165-amino-acid glycoprotein hormone of about 30,400 daltons, made by the kidneys, approved in the United States since 1989, and carrying a boxed warning that names six harms, death first among them. The trial that showed most clearly what it does to blood was halted in 1,233 dialysis patients with heart disease, with more deaths and more heart attacks in the group given a normal haematocrit than in the group left anaemic.

That trial was not a doping study. It tested a target most people would assume is obviously good, in patients under medical supervision, and the assumption failed. Everything downstream — the dose ceiling, the warning label, the reason a performance user is running an experiment nobody has run safely — follows from it. EPO is grouped with the other immune support & longevity peptides on this site, and almost nothing about verifying a short synthetic peptide transfers to it.

The short version

• EPO is a glycoprotein hormone made by the kidneys, not a peptide in the sense the rest of this site uses the word, and the sugar chains attached to it are why standard purity testing cannot establish what is in a vial.

• It works for what it was approved for: in the trial behind its approval, 12 of 18 dialysis patients who had needed transfusions stopped needing them.

• Three randomised trials in kidney patients found more deaths, heart attacks or strokes in the groups assigned the higher target, and the label now caps the target instead of normalising it.

• The only double-blind randomised trial of EPO in well-trained cyclists found a laboratory improvement and no difference in their actual race time.

• Peptigrity holds 1 published lab certificate for EPO across 83 shops that list it (verified September 2026), which is not enough for the Purity Index to report a figure at all.

What is EPO, and is it a peptide?

EPO is a glycoprotein hormone rather than a peptide. Erythropoietin is a 165-amino-acid hormone produced mainly by the kidneys, weighing about 30,400 daltons once its sugar chains are attached, and it works by binding the erythropoietin receptor on the immature red cells in bone marrow, which UniProt records as activating JAK2 and then STAT signalling. The elimination half-life of the approved product is 4 to 13 hours given into a vein.

Those figures come from the label, EPOGEN® (epoetin alfa) injection, for intravenous or subcutaneous use (revised 12/2024), which describes epoetin alfa as "a 165-amino acid erythropoiesis-stimulating glycoprotein manufactured by recombinant DNA technology" containing "the identical amino acid sequence of isolated natural erythropoietin". A glycoprotein is a protein with sugar chains bonded to it. Glycosylation is the attachment of those chains, and in EPO's case it accounts for a large share of the finished molecule's mass. The receptor and signalling detail comes from UniProt rather than the label.

The identity attributes come from the UniProt entry P01588 · EPO_HUMAN, which numbers every position on the 193-residue precursor; the label and most of the literature count from the mature hormone, 27 residues later. A source that gives a glycosylation position without saying which numbering it uses is ambiguous by 27, and the two sources differ by one residue at the other end: UniProt's chain runs to 166 residues, the label's product is 165.

A CAS registry number exists for erythropoietin, and it settles less here than it would elsewhere. Two vials could share it and differ in the one attribute that decides whether either is the drug.

The sugar chains are not decoration. They extend how long the hormone survives in circulation, and because a manufacturing process produces a slightly different pattern of chains than a human kidney does, they are also what the anti-doping test reads. That same pattern is the thing no ordinary certificate measures, which is why the single certificate on file for this compound, listed among Peptigrity's Peptide Lab Tests, answers less about an EPO vial than a certificate answers about a 9-amino-acid peptide.

What has EPO been shown to do in humans?

EPO corrects the anaemia of kidney failure, and the evidence is 40 years old and undisputed. In the trial behind its approval, 25 dialysis patients received recombinant human erythropoietin intravenously; of the 18 given effective doses, 12 who had required transfusions no longer needed them and 11 reached a haematocrit of 35% or more. US approval followed in 1989. The approved indications now cover anaemia in chronic kidney disease, anaemia from chemotherapy, anaemia in HIV infection treated with zidovudine, and reducing transfusions around elective surgery.

The trial is Correction of the anemia of end-stage renal disease with recombinant human erythropoietin. Results of a combined phase I and II clinical trial (Eschbach et al., 1987), which reports dose-dependent increases in effective erythropoiesis and concludes that the hormone "is effective, can eliminate the need for transfusions with their risks of immunologic sensitization, infection, and iron overload, and can restore the hematocrit to normal in many patients". Haematocrit is the share of blood volume made up of red cells; haemoglobin is the oxygen-carrying protein inside them, and the label works in haemoglobin.

Two things in that table sit oddly together, and the rest of this article is about the gap between them. The drug does what it was approved to do. The three large trials that tried to give patients more of it found harm.

What did the Normal Hematocrit Study find, and what did it not?

The Normal Hematocrit Study found that dialysis patients with heart disease did worse when treated to a normal haematocrit, and it was halted before it could prove why. The effects of normal as compared with low hematocrit values in patients with cardiac disease who are receiving hemodialysis and epoetin (Besarab et al., 1998) randomised 1,233 patients to a haematocrit of 42% or 30%. After 29 months there were 183 deaths and 19 non-fatal heart attacks in the normal-haematocrit group against 150 and 14 in the low group.

The numbers need stating carefully, because the page's own argument does not need overstating. The risk ratio on the trial's primary endpoint was 1.3, with a 95% confidence interval of 0.9 to 1.9 — a result that did not reach the study's prespecified stopping boundary, and the study was halted anyway. The label reports the same trial at 1,265 patients and gives all-cause mortality as 35% against 29%, a hazard ratio of 1.27 with a confidence interval of 1.04 to 1.54 and p = 0.018, and records that non-fatal myocardial infarction, vascular access thrombosis and other thrombotic events were also more common in the higher group. The authors' conclusion is the operative sentence: administering epoetin to raise haematocrit to 42% in these patients "is not recommended".

Two later trials point the same way from different populations. CHOIR randomised 1,432 kidney patients not on dialysis to 13.5 or 11.3 g/dL of haemoglobin and was also halted early, with a major cardiovascular event in 125 of 715 patients against 97 of 717 — 18% against 14%, hazard ratio 1.34. TREAT was double-blind and placebo-controlled in 4,038 patients, and found the stroke rate nearly doubled, 2.1% against 1.1%.

The usual explanation is viscosity: more red cells make blood thicker, thicker blood moves less easily through vessels already narrowed by disease, and past some point that costs more than the extra oxygen delivery gains. It is a plausible mechanism rather than a demonstrated one, and the 1998 paper contains a finding that complicates it — within both groups, mortality fell as achieved haematocrit rose. What the trials establish is that patients assigned to be driven to a higher target did worse, not that a high haematocrit is itself the injury. The distinction matters, because the intervention and the number it produces are not the same thing, and neither trial arm tells you where either sits in a healthy person.

Why does EPO carry a boxed warning?

EPO carries a boxed warning because erythropoiesis-stimulating agents increase the risk of death, myocardial infarction, stroke, venous thromboembolism, thrombosis of vascular access, and tumour progression or recurrence — the warning's own list, in the warning's own order. For kidney disease it states that patients "experienced greater risks for death, serious adverse cardiovascular reactions, and stroke when administered erythropoiesis-stimulating agents (ESAs) to target a hemoglobin level of greater than 11 g/dL", and that no trial has identified a target, dose or strategy that avoids those risks.

The instruction that follows is the shape of the whole drug: use the lowest dose sufficient to reduce the need for transfusions. Not the dose that normalises a number, and not the dose that makes someone feel best.

The cancer half of the warning is the part that surprises people. The label's Limitations of Use states plainly that the drug "has not been shown to improve quality of life, fatigue, or patient well-being", and it is not indicated when the anticipated outcome of chemotherapy is cure, or when anaemia can be managed by transfusion. A reader who assumed erythropoiesis-stimulating agents are given to cancer patients because they help them feel better has the reason wrong. How risk profiles across this market are reported by class and route is set out in Peptide Side Effects, which puts EPO's in context against compounds with no human data at all.

What harms are documented, and at what frequency?

Four harms in the label are documented with a frequency or a mechanism rather than a general caution. Blood pressure: after starting and adjusting the drug, about 25% of dialysis patients required antihypertensive therapy to be started or increased, and hypertensive encephalopathy and seizures have been reported. Seizure risk rises in kidney disease, with the first months the period of concern. Blistering and skin-exfoliation reactions including Stevens-Johnson syndrome have been reported after marketing. And the drug can cause the body to stop making red cells altogether.

That last one is EPO's signature signal, and it is the one a grey-market buyer should understand first. The label records cases of pure red cell aplasia, and of severe anaemia with or without other cytopenias, "that arise following the development of neutralizing antibodies to erythropoietin" — the immune system learns to attack the injected hormone, and the antibodies also neutralise the body's own. It has been reported predominantly with subcutaneous administration in kidney patients. A product that is less well purified, or aggregated from bad handling, is exactly the kind of product that provokes an antibody response.

The label's contraindications are short and absolute: uncontrolled hypertension; pure red cell aplasia that began after treatment with an erythropoietin product; serious allergic reactions to the drug. Multiple-dose vials contain benzyl alcohol and are additionally contraindicated in neonates, infants, pregnant women and lactating women. Overdose, in the label's own framing, "can cause hemoglobin levels above the desired level", managed by stopping or reducing the drug and sometimes by removing blood; severe hypertension has been seen after overdose with this class.

The label also addresses performance use directly, which is unusual. Under drug abuse it states that abuse "may be seen in athletes for the effects on erythropoiesis", and that abuse of drugs that increase erythropoiesis "by healthy persons may lead to life-threatening cardiovascular complications (e.g., stroke, myocardial infarction, or thromboembolism)".

Stop and seek care: swelling of the face or throat, difficulty breathing or hives after an injection are signs of a serious allergic reaction; spreading redness, warmth or pus at an injection site with fever is an infection; severe or persistent abdominal pain with vomiting on a GLP-1 drug can be pancreatitis; confusion, sweating or shaking on an IGF-1 or insulin-adjacent compound can be hypoglycaemia. Call your local emergency number (911 in the US, 112 in the EU, 999 in the UK, 000 in Australia) or go to an emergency department, and bring the vial and its label. Research-use vials are not approved products in any jurisdiction, even where a prescription product of the same molecule exists; no article on this site replaces a clinician who knows your history.

Chest pain, breathlessness, one-sided limb swelling and sudden vision changes are the thrombosis symptoms this drug's warning is built around, and they are reasons to be seen rather than to adjust anything.

Does EPO improve endurance performance?

EPO improves laboratory measures of maximal exercise and has not been shown to improve an actual race. The one double-blind, randomised, placebo-controlled trial in well-trained cyclists, Effects of erythropoietin on cycling performance of well trained cyclists: a double-blind, randomised, placebo-controlled trial (Heuberger et al., 2017), enrolled 48 men and found maximal power output and VO₂ max higher than on placebo, while the submaximal test and the Mont Ventoux race time did not differ.

The race times are the figure to hold on to: 1 h 40 min 32 s on the drug against 1 h 40 min 15 s on placebo, a difference of 0.3% with a confidence interval running from 8.3% faster to 9.6% slower. The authors' own summary is the sentence to keep: although the treatment "improved a laboratory test of maximal exercise, the more clinically relevant submaximal exercise test performance and road race performance were not affected". All adverse events in that trial were grade 1–2 and similar in both groups, which matters for a different reason further down.

Smaller earlier studies found what the laboratory half of that result predicts. In Effect of rhEPO administration on serum levels of sTfR and cycling performance (Birkeland et al., 2000), a double-blind, placebo-controlled study in 20 male athletes, VO₂ max rose from 63.6 to 68.1 mL/kg/min in the treated group — a 7% increase — with no change on placebo, and haematocrit went from 42.7% to 50.8%. The largest reported gains come from untrained subjects. In Prolonged administration of recombinant human erythropoietin increases submaximal performance more than maximal aerobic capacity (Thomsen et al., 2007), eight subjects, described in the paper as healthy non-athletes, gained 12.6% and 11.6% in VO₂ max at weeks 4 and 11, and time to exhaustion at a fixed workload rose by about 54% — the study's headline result, and the source of the double-digit figures in circulation. The same abstract carries the caveat almost never quoted with it: once the workload was re-set to the same relative intensity, time to exhaustion was 26.8% lower than before treatment. The gain, in other words, was largely the gain of working at a lower fraction of a raised maximum.

The systematic review Erythropoietin doping in cycling: lack of evidence for efficacy and a negative risk-benefit (Heuberger et al., 2013) reached the conclusion the 2017 trial then tested: "there is no scientific basis from which to conclude that rHuEPO has performance-enhancing properties in elite cyclists", the studies having been done in the wrong population against a surrogate endpoint.

None of that makes EPO ineffective. It makes the popular figure — a decisive double-digit gain in competition — an extrapolation from untrained subjects in a laboratory, and it means the one trial that took the question to a road race found no effect on the finishing time. That trial's confidence interval is wide enough to contain a real improvement, so "not affected" is what it established, not "makes no difference". The administered schedules from these studies are deliberately not reproduced on this page.

Which EPO claims survive the evidence?

Two of eight common EPO claims survive intact. Erythropoietin does correct anaemia and remove transfusion dependence, which is what it was approved for, and it does raise maximal oxygen uptake by about 7% in a laboratory test. Of the rest, two are contradicted outright by the trials or by the label's own wording, one is unproven rather than disproven, one is false, one misdescribes which law applies, and the last understates a risk the boxed warning states in full.

The last row is the one that reframes the rest. This is not a compound whose risks are hypothesised from animal work. They were measured in thousands of monitored patients, at doses a physician chose, and they were serious enough to change the label.

How is EPO used medically, and what does the label actually state?

The label sets a dose for each approved indication and bounds all of them with haemoglobin thresholds rather than a target to hit. For adults on dialysis it states a starting dose of 50 to 100 Units/kg three times weekly, begun below 10 g/dL and reduced or interrupted as haemoglobin approaches 11 g/dL. Increases are capped at one every four weeks, with a reduction of 25% or more if haemoglobin climbs more than 1 g/dL in two weeks. No dose is established for performance use, in any evidence tier.

Two supporting requirements travel with every one of those rows. Iron status is checked before and during treatment, with supplemental iron when ferritin is below 100 mcg/L or transferrin saturation below 20%, because rapid red cell production consumes iron and the drug cannot work without it — the label notes that most kidney patients will need it. Haemoglobin is measured weekly until stable and at least monthly after that.

The label's storage instructions are equally specific: 2 °C to 8 °C, protected from light, never frozen and never shaken, with a multiple-dose vial discarded 21 days after the first needle enters it. Shaking is not a cosmetic concern; the label states a shaken vial is not to be used. Why a purity figure survives storage abuse that the molecule does not is the subject of Why a Badly Stored Peptide Still Passes an HPLC Purity Test, which carries this compound's temperature figure.

No dose calculator is offered for EPO on this site, and the arithmetic is not the difficulty here. A figure in Units per kilogram, a haemoglobin threshold measured in a laboratory and a clinician reading the trend are one instrument, and separating the arithmetic from the other two produces a number with nothing behind it.

Can EPO use be detected in sport, and where does it sit on the WADA List?

Recombinant EPO has been detectable in urine for a quarter of a century, and the second layer of testing does not look for the drug at all. Recombinant erythropoietin in urine (Lasne & de Ceaurriz, 2000) described a method that separates injected from natural erythropoietin by their isoelectric patterns — the different arrangement of sugar chains a manufacturing process leaves behind — and it has been in anti-doping use since. WADA's Athlete Biological Passport then "monitor[s] selected biological variables over time that indirectly reveal the effects of doping".

On the World Anti-Doping Code International Standard Prohibited List 2026, in force since 1 January 2026, EPO sits in class S2, prohibited at all times, in and out of competition, and every substance in that class is non-Specified — the category that carries the standard sanction rather than a reduced one.

Section S2.1 covers "Erythropoietins (EPO) and agents affecting erythropoiesis" and is wider than the hormone, in five parts: erythropoietin receptor agonists, giving darbepoetins, erythropoietins, EPO-based constructs and EPO-mimetic agents as examples; hypoxia-inducible factor activating agents, naming cobalt, daprodustat, molidustat, roxadustat, vadadustat, IOX2 and xenon; GATA inhibitors; transforming growth factor beta signalling inhibitors, naming luspatercept and sotatercept; and innate repair receptor agonists, naming asialo EPO and carbamylated EPO.

Which settles a question the subsection is often used to dodge. An oral drug that raises the body's own erythropoietin is in the same prohibited subsection as an injection of the hormone, one item below it. How this compares with the detection picture for other compound classes, most of which are not tested for at all, is set out in Do Peptides Show Up on Drug Tests?.

Is EPO legal to buy, and could a pharmacy make it?

EPO is a prescription biologic in the United States, it is not a scheduled controlled substance, and no pharmacy can lawfully compound it. The label states directly that epoetin alfa "is not a controlled substance", so the penalties, the enforcement agency and the risk model people import by analogy from anabolic steroids are the wrong ones. What applies instead is licensure: this is a biological product licensed under section 351 of the Public Health Service Act, and selling one lawfully takes a licence, not a prescription pad.

That has a consequence with no parallel anywhere else on this site. Every unapproved peptide here has a theoretical pharmacy route — the 503A bulks list, the compounding pathway, the advisory committee votes. FDA's final guidance of January 2018, Mixing, Diluting, or Repackaging Biological Products Outside the Scope of an Approved Biologics License Application, closes it for biologics: sections 503A and 503B "do not provide an exemption from the requirement to obtain premarket approval under section 351 of the PHS Act", so licensed biological products "are not eligible for the exemptions for compounded drugs", and "the FD&C Act does not provide a legal pathway for marketing biological products that have been prepared outside the scope of an approved BLA".

For a buyer, then, there are two routes and not three: a licensed product on prescription, or research-labelled material from the grey market for peptides, where nobody upstream answers for a batch. The pharmacy middle ground that exists for BPC-157 or ipamorelin does not exist here. The statutory mechanics of those two routes are compared in Compounding Pharmacy vs Research Peptide, and how jurisdictions treat prescription medicines differently is set out in Are Peptides Legal?. This page states what published documents say; it is not legal advice, and anything touching a particular situation belongs with a professional.

Why is grey-market EPO a different risk category?

Grey-market EPO carries the market's usual risks and then inverts the usual arithmetic. An underdosed research peptide generally does nothing. An erythropoiesis stimulator that works, used without a haemoglobin measurement, moves the reader toward the region three randomised trials associated with death, heart attacks and strokes — and the label's own no-trial-has-found-a-safe-strategy sentence was written about patients who were being monitored weekly.

There is a second failure mode specific to this molecule. Because pure red cell aplasia arises from neutralising antibodies the body makes against the injected hormone, the quality of the protein is not only a question of whether the dose is right — and the consequence, when it happens, is not a wasted vial but a body that has stopped responding to erythropoietin, including its own. Whether unlicensed material provokes that response more often than a licensed product is not something this platform can show, and no study reached in writing this page measures it. What can be said is narrower and still worth saying: the outcome exists, the label ties it to an immune response against the injected protein, and nothing on a purity certificate speaks to it. There is no equivalent failure mode for a 15-amino-acid synthetic peptide.

The cold chain compounds both. A product held outside 2 °C to 8 °C, frozen in transit, or shaken may have lost activity, and a buyer who titrates by how they feel escalates — with the escalation landing whenever an intact vial finally arrives.

The standard peptide-testing toolkit was not built for a glycoprotein. An HPLC purity figure on a 30,000-dalton glycosylated protein establishes homogeneity against a declared analyte; it does not establish the glycosylation pattern, comparability to a licensed product, or biological potency. Why a purity percentage and an identity are different questions is the subject of Why a Peptide Can Be 99 Percent Pure and Still Be Entirely the Wrong Molecule, and the gap is wider here than for anything else in the library. What each available test method measures, and what it costs, is set out in How to test your peptides; which laboratories publish verifiable certificates is in Where peptides get verified.

Peptigrity's own record for this compound is close to empty, and that is the finding. One published certificate is on file — a single vial tested by one laboratory on 23 May 2025, reporting 99.52% HPLC purity and a measured content of 12.97 mg against a 10 mg label, roughly 130% of the labelled amount, with no endotoxin result — while 83 shops list EPO (verified September 2026). The Peptigrity Purity Index publishes no figure for it at all, because a compound needs at least five recency-weighted tests to enter the index; across the compounds that do qualify, 7.24% of 12,629 net-content tests came in under 95% of their label. Certificates on this platform are the ones shops chose to publish, verified for authenticity before listing, a selection bias stated openly in How Peptigrity actually works. For EPO the sample is one vial, no price comparison is possible because fewer than three shops publish comparable prices, and a single HPLC figure would not answer the glycosylation question even if there were a hundred of them. Shops listing it, with their reviews and lab records, are compared in the peptide shops directory.

What are the alternatives, and do any of them avoid the problem?

Three approaches exist alongside EPO and none of them is a clean substitute. Oral agents that stabilise hypoxia-inducible factor raise the body's own erythropoietin and are prohibited in sport under the same subsection as EPO itself. Iron repletion corrects the actual deficiency where iron deficiency is the cause and is first-line there. Altitude exposure raises endogenous erythropoietin transiently and is not named anywhere on the 2026 List. The compound engineered specifically to keep EPO's tissue protection without the red cells remains investigational.

The second row is the instructive one. Reaching the same physiological endpoint by a different molecular route did not persuade anti-doping authorities that the endpoint was the safe part — the List treats the oral agents and the injection as one problem. ARA-290 goes the other way, keeping the tissue-protective signalling and dropping the erythropoiesis, which is an attempt to separate the benefit from the risk this article describes. It has not been approved for anything, and the sugar-chain and antibody problems above apply to grey-market ARA-290 far less than to grey-market EPO, because it is a short synthetic peptide.

What has the performance question actually settled, and what is still open?

The performance trial that was widely called impossible has been run, and what remains open is the long term rather than the effect. Heuberger and colleagues ran a double-blind, randomised, placebo-controlled trial in 48 well-trained cyclists in 2016, published it in 2017, recorded only grade 1–2 adverse events, and presented it as "a model clinical study for doping research on other substances". It settled the immediate question in the direction nobody selling EPO wants: the laboratory number moved and the race did not.

The second and fourth rows together are the honest position. A drug whose measurable benefit did not reach the finish line in the one trial designed to check, and whose harm in patients took more than two years of follow-up to show up, is not a compound where an absence of published catastrophe among users means much.

Where does this leave EPO?

EPO is the clearest case on this site of "it works" and "it is safe to use" being different questions, and of a third question — "does it do the thing people buy it for?" — turning out to be separate from both. It is an effective, approved, well-characterised medicine with four decades of clinical data behind it, and its most instructive trial is one where giving people more of what the drug does was associated with more of them dying.

Used medically, inside a capped haemoglobin target with weekly bloodwork and iron monitoring, it is a valuable therapy that removed transfusion dependence for a generation of dialysis patients. Used to raise haematocrit for performance, without bloodwork, from an unverified source, outside a cold chain, it removes every safeguard that made the medical use defensible and buys a laboratory improvement that did not show up in a race.

The gap between those two sentences is not a matter of dose discipline. It is the difference between a drug bounded from above by measurement, and the same drug bounded by how someone feels. A reader who already holds a vial can start by asking whether any certificate exists for that lot, and whether it says anything at all about glycosylation.

Frequently Asked Questions

Does EPO improve endurance performance?

EPO raises maximal oxygen uptake, by about 7% in a double-blind placebo-controlled study of 20 male athletes, but the only randomised trial to measure an actual race found no difference: 48 well-trained cyclists rode Mont Ventoux in 1 h 40 min 32 s on the drug against 1 h 40 min 15 s on placebo. The commonly quoted double-digit figures come from untrained subjects in laboratory tests.

Is EPO a peptide?

Not in the sense the rest of this site uses the word. EPO is a 165-amino-acid glycoprotein hormone of about 30,400 daltons, much of that mass being the sugar chains attached to it, and those chains are why standard peptide purity testing cannot establish what is in a vial.

Is EPO banned in sport?

Yes, at all times, in and out of competition, under section S2.1.1 of WADA's 2026 Prohibited List, and as a non-Specified substance. It has been detectable in urine since 2000 by isoelectric focusing, and the Athlete Biological Passport monitors an athlete's own blood values over time rather than looking for the drug.

Is possessing EPO a criminal offence?

The Epogen label states that epoetin alfa is not a controlled substance, so the Controlled Substances Act framing people import from anabolic steroids does not apply. EPO is a biologic licensed under the Public Health Service Act, and unlicensed commercial supply is a licensure and marketing matter. That is a description of the statutes rather than advice about any individual's position.

Could a compounding pharmacy legally make EPO?

No. FDA's guidance states that sections 503A and 503B do not exempt licensed biological products from premarket approval under section 351 of the Public Health Service Act, so biologics "are not eligible for the exemptions for compounded drugs". The pharmacy route that exists on paper for unapproved peptides does not exist for EPO.

Why is there a boxed warning on a drug that treats anaemia?

Because three randomised trials found more deaths, cardiac events or strokes in the patients assigned a higher target — a haematocrit of 42% in the 1998 trial, haemoglobin of 13.5 and 13 g/dL in the two later ones — with two of the three halted early. The usual explanation is that more red cells make blood thicker and clot more readily. The label's own position is blunter: no trial has identified a target, dose or strategy that avoids the risks.

What is a safe haematocrit on EPO?

There is no self-managed answer, and the label does not work in haematocrit. It sets haemoglobin thresholds, and they differ by indication: on dialysis, start below 10 g/dL and reduce or interrupt as haemoglobin approaches or exceeds 11 g/dL; not on dialysis, reduce or interrupt above 10 g/dL. In every case the instruction is the lowest dose that avoids a transfusion, with haemoglobin measured weekly until stable. Talk to a clinician before using EPO at all, and particularly with any history of clots, stroke, uncontrolled high blood pressure, cancer or a blood disorder: the boxed warning names all of those, and pure red cell aplasia from anti-erythropoietin antibodies is a reason that any unexplained loss of response needs investigating rather than a larger dose.

This article is for educational and informational purposes only and does not constitute medical advice. Peptides discussed may be investigational compounds not approved by the FDA (or equivalent regulators in your jurisdiction) for human use; research peptides are sold for laboratory use and are not manufactured, tested or labelled to pharmaceutical standards. Peptigrity does not recommend doses: every dose figure above is an approved-label value, a published-trial value or a community report, and is labelled as such — a community report is not a recommendation. Self-injection carries risks of infection, allergic reaction and dosing error; signs of a serious allergic reaction, an injection-site infection with fever, severe abdominal pain, or confusion and sweating on an IGF-1 or insulin-adjacent compound are medical emergencies — call your local emergency number. Always consult a qualified healthcare provider before using any peptide or research compound. Peptigrity is an independent review platform and does not sell, endorse, or recommend specific products or vendors.