Rusfertide: The Hepcidin Peptide, Explained

Rusfertide in brief

Three numbers frame the drug: how many patients it took off phlebotomy in the pivotal trial, how few phlebotomies were left, and the date the FDA turned an investigational peptide into a prescription drug.

Every figure on this page is taken from a PubMed abstract, from the MIMRYLO prescribing information, or from the FDA approval announcement, all listed at the end. Rusfertide doses are quoted only where the label or a trial states them; nothing here is a protocol for anyone outside the approved indication.

What this rusfertide profile covers

What is rusfertide? A synthetic hepcidin, and what hepcidin does

Rusfertide is an injectable peptide engineered to mimic hepcidin, the liver hormone that is the central negative regulator of iron trafficking (PMID 41100735). Hepcidin binds ferroportin, the only known iron exporter on gut cells and macrophages, and pulls it out of the membrane; with ferroportin gone, iron stays inside the cells that store it, serum iron falls, and the bone marrow, which needs a steady supply of iron to build hemoglobin, slows its output of red cells. The peptide binds ferroportin the same way and decreases iron delivery to the marrow, reducing red blood cell production (PMID 41175501). Natural hepcidin is cleared within hours; rusfertide was built to last long enough for a once-weekly subcutaneous injection.

The drug was discovered by Protagonist Therapeutics as PTG-300, later carried the Takeda code TAK-121, and reached the market on 28 August 2026 as MIMRYLO, approved for the treatment of erythrocytosis in adults with polycythemia vera. Polycythemia vera is a myeloproliferative neoplasm in which a JAK2 mutation drives red cell production regardless of iron or oxygen; the hematocrit rises, the blood thickens, and the standard defence is therapeutic phlebotomy to keep the hematocrit below 45 percent, a target that a randomised trial of 365 patients showed cut cardiovascular death and major thrombosis from 9.8 to 2.7 percent (PMID 23216616).

The point that matters most for readers of this site: polycythemia vera patients make too many red cells despite being iron deficient, because the JAK2 clone ignores the normal brakes, and hepcidin levels in the disease are relatively low (PMID 41100735). The mimetic restores the brake from outside. It is not a steroid, not an anabolic, not a growth hormone secretagogue and not a performance peptide of any kind; it reduces a physiological output rather than raising one, which is why it sits among the peptides on this site as an explanation and not as a product. How peptides are classified is on types of peptides.

What is measured on rusfertide, and in whom

The drug has an unusually complete record for a new peptide: two randomised placebo-controlled trials in the disease, an open-label study in poorly controlled patients, three pharmacology studies in healthy volunteers, a cardiac safety study, a trial in hemochromatosis, and a prescribing label with adverse event rates against placebo. What it does not have is a single study in anyone taking testosterone. The table keeps the two apart.

Read the weight column first. Every green row on rusfertide itself is in polycythemia vera patients or healthy volunteers, and the two green testosterone rows measured hepcidin, not rusfertide. The rows meet only at the mechanism. That is the evidence base, and the rest of this page stays inside it.

Rusfertide FDA approval: what the MIMRYLO label says

The rusfertide FDA approval came on 28 August 2026, when the agency approved MIMRYLO for the treatment of erythrocytosis in adults with polycythemia vera. The path was quick by the standards of a first-in-class drug: breakthrough therapy designation on 25 August 2025, a new drug application submitted on 5 January 2026, acceptance with priority review on 2 March 2026 and approval inside the third-quarter goal date. Protagonist Therapeutics discovered the peptide and Takeda, which licensed it worldwide in January 2024, commercialises it.

The label is where the drug stops being a research story and becomes a set of numbers. The recommended starting dose is 19 mg injected subcutaneously once a week, with the dose adjusted between 9.5 mg and 108 mg weekly to bring the hematocrit below 45 percent and keep it there; an increase is allowed only after at least two weeks at the current dose, with at least two weeks between increases, and the dose is reduced for anemia of grade 2 or worse or for any grade 3 drug-related toxicity (MIMRYLO prescribing information). It is supplied as single-dose vials of 9.5, 19, 28, 41 and 54 mg of lyophilised powder with a prefilled diluent syringe (MIMRYLO prescribing information).

Two things the rusfertide FDA approval label does not contain deserve saying. There is no boxed warning, and there is no malignancy warning. The programme carries a history on that point: on 17 September 2021 the FDA placed a clinical hold after a 26-week study in rasH2 mice, a strain bred to be sensitive to carcinogens, produced benign and malignant tumours under the skin; the hold was lifted on 11 October 2021 after review of the human safety database found no matching signal, and the trials resumed with closer monitoring. Those dates come from company announcements, not from a PubMed abstract.

The approved label five years later lists thrombocytosis, injection site reactions and fetal harm as the warnings, nothing about cancer.

The rusfertide trials: VERIFY, REVIVE and the open-label study

VERIFY is the trial behind the approval. It randomised 293 adults with polycythemia vera, 145 to rusfertide and 146 to placebo in the analysed population, each on top of their existing standard care, for a 32-week controlled period followed by open-label treatment to week 52 and an extension to week 156. The primary endpoint was response, defined as absence of phlebotomy eligibility during weeks 20 to 32: 113 of 147 patients on rusfertide reached it against 48 of 146 on placebo, 76.9 against 32.9 percent, a difference of 43.8 points with a p value below 0.0001 (MIMRYLO prescribing information).

The secondary endpoints moved the same way. The mean number of phlebotomies from baseline to week 32 was 0.53 on rusfertide and 1.82 on placebo; 92 of 147 patients on rusfertide, 62.6 percent, kept their hematocrit below 45 percent throughout, against 21 of 146, or 14.4 percent, on placebo; and the PROMIS fatigue score fell by 1.79 points on rusfertide while rising 0.19 on placebo, p equal to 0.0252 (MIMRYLO prescribing information). The topline data were presented at ASCO and ASH in 2025 and as of 30 September 2026 the full trial has not yet appeared as a PubMed-indexed paper, which is why this page cites the label for its numbers.

REVIVE, the phase 2 trial published in the New England Journal of Medicine, is the one with a peer-reviewed abstract. Seventy patients entered a 28-week dose-finding period; the estimated phlebotomy rate fell from 8.7 a year in the 28 weeks before rusfertide to 0.6 during treatment, and the mean maximum hematocrit fell from 50.0 to 44.5 percent.

Fifty-nine patients were then randomised to continue or switch to placebo for 12 weeks: 60 percent of those who stayed on the drug kept their response against 17 percent on placebo, p equal to 0.002. Grade 3 adverse events occurred in 13 percent of patients and there were no grade 4 or 5 events; injection site reactions of grade 1 or 2 were common (PMID 38381675).

The open-label study is worth a paragraph because it answers the question a reader with a high hematocrit actually has: how fast does it move. Patients with suboptimally controlled disease received 40 mg subcutaneously twice weekly until their hematocrit reached below 45 percent and weekly thereafter (PMID 41175501). The mean hematocrit was below 45 percent by week 4 and stayed there for the whole 52 weeks; red cell count fell from 5.98 to 4.79 times ten to the twelfth per litre, ferritin and mean corpuscular volume improved, 85 percent of patients had a treatment-emergent adverse event and most were grade 1 or 2 (PMID 41175501). Seven of the 20 discontinued, which is the number to hold beside the 85 percent.

What the trials measured is hematocrit, phlebotomy count and symptoms. What they have not measured is thrombosis or disease course: a 2026 review in Blood Reviews puts it plainly that the evidence for iron-directed therapy rests on surrogate hematologic endpoints and that whether it reduces thrombotic risk remains undefined (PMID 42660705). The hematocrit target itself is borrowed from the CYTO-PV trial, where keeping polycythemia vera patients below 45 percent rather than at 45 to 50 cut cardiovascular death and major thrombosis from 9.8 to 2.7 percent, a hazard ratio of 3.91 for the higher target (PMID 23216616).

Rusfertide in people without polycythemia vera: the healthy-volunteer studies

The most relevant rusfertide data for a testosterone user are not in the disease trials at all; they are in the three pharmacology studies in healthy adults, because healthy adults have a normal marrow and a normal JAK2 gene, which is closer to a man on testosterone than a polycythemia vera patient is.

In the multiple-dose study, 18 healthy subjects received 60 mg once weekly for five weeks, randomised to injection in the abdomen, upper arm or thigh (PMID 39888264). Serum iron, transferrin saturation, hemoglobin and hematocrit all fell as a mechanism-based effect, with no difference between injection sites, no serious adverse events, and injection site reactions, fatigue and headache as the events seen in two or more subjects (PMID 39888264).

The dose-ranging study of the lyophilised formulation, the one now sold, gave the peptide at 10 to 60 mg in a crossover design (PMID 39546273). Time to peak was 24 hours at 10 to 30 mg and 2 to 4 hours at 45 and 60 mg, the terminal half-life ranged from 19.6 to 57.1 hours, exposure rose less than proportionally with dose, serum iron and transferrin saturation fell in a dose-related way, and the treatment-related events in 10 percent or more of subjects were injection site erythema and pruritus (PMID 39546273).

The first-in-human study, single doses of 1 to 80 mg of an aqueous formulation in healthy men, found no serious or severe events and dose-dependent falls in iron and transferrin saturation (PMID 38785334).

The label consolidates the kinetics: a mean half-life of 28.6 hours with a standard deviation of 11.3, a median time to peak of 24 hours with a range of 4 to 48, protein binding above 99 percent, and steady state after about three weekly doses (MIMRYLO prescribing information). The cardiac question was closed separately: in a thorough QT study of 60 healthy adults given 90 mg, the placebo-corrected change in QTc varied between minus 2.0 and plus 1.8 milliseconds, the concentration-QTc slope was not significant, and an effect above 10 milliseconds could be excluded up to plasma concentrations of about 2,130 nanograms per millilitre (PMID 41033871).

One more disease has been tested, and it is the mirror image of polycythemia vera. In HFE-related hemochromatosis, where hepcidin is genetically insufficient and iron overloads the organs, 16 patients on maintenance phlebotomy received rusfertide starting at 10 mg once weekly for 24 weeks; the mean number of phlebotomies fell from 2.31 in the 24 weeks before to 0.06 during treatment, 15 of 16 were phlebotomy-free, and liver iron concentration moved from 1.4 to 1.1 milligrams per gram without reaching significance (PMID 37863080). A 2026 review still lists rusfertide as investigational in that disease (PMID 42764193). The approval is for polycythemia vera only.

Rusfertide side effects by the evidence

Rusfertide side effects are the side effects of switching hepcidin on: less iron reaches the marrow, so hemoglobin can fall too far, and platelets, which the iron restriction does not throttle, tend to rise. On top of that sits the one problem every subcutaneous peptide has, the injection site. The table gives each with the strongest source that exists and the placebo rate where the label provides one.

The honest summary of rusfertide side effects is that they are predictable from the mechanism and mostly mild, that the label carries three formal warnings and no boxed warning, and that the two events a reader should remember are the platelet rise, because it is monitored with blood counts every two to four weeks, and anemia, because it is what happens when a drug built to lower hematocrit is not adjusted. The general page on peptide side effects explains why injection site reactions dominate this whole drug class.

Testosterone, hepcidin and hematocrit: why rusfertide keeps coming up on TRT forums

Here is why a polycythemia vera drug is on a testosterone site. One of the best documented reasons testosterone raises hematocrit is that testosterone suppresses hepcidin. In 109 healthy younger and older men whose own production was switched off and who received weekly testosterone enanthate at 25, 50, 125, 300 or 600 mg for 20 weeks, the high doses markedly suppressed serum hepcidin within one week; the suppression was dose-dependent in the older men and more pronounced than in the young, which matched their greater rise in hemoglobin, and hepcidin at weeks 4 and 8 predicted the change in hematocrit from baseline to peak (PMID 20660052).

The follow-up trial showed testosterone pulls from both ends. In older men randomised to testosterone gel or placebo for six months, hemoglobin rose 7 percent and hematocrit 10 percent, erythropoietin rose while ferritin and hepcidin fell at months 1 and 3, and by month 6 erythropoietin and hepcidin had drifted back toward baseline while the hematocrit stayed high, which the authors read as a new erythropoietin-to-hemoglobin set point (PMID 24158761). In mice the same group showed testosterone inhibits hepcidin transcription in the liver and raises renal erythropoietin, and that mice engineered to over-express hepcidin failed to raise hemoglobin on testosterone (PMID 23399021).

That knockout result is the asterisk on every forum post that says rusfertide is the perfect answer to testosterone-driven erythrocytosis. A hepcidin mimetic would put the brake back, and the healthy-volunteer study shows a hepcidin mimetic does lower hemoglobin and hematocrit in a normal marrow (PMID 39888264). But testosterone also raises erythropoietin (PMID 24158761), stimulates erythroid progenitors directly, as a review of the hypoxia, testosterone and erythropoietin axis describes (PMID 41118472), and drove erythropoiesis in mice with the hepcidin route deleted (PMID 31001931). How much of a testosterone-driven hematocrit a hepcidin mimetic could hold, at what dose, and at what cost in platelets and iron, has not been measured in a single human being.

A third mouse study adds a warning of its own: in iron-deficient mice, testosterone reduced the red cell count and produced ineffective erythropoiesis, so restricting iron under androgen pressure is not a neutral act (PMID 32667087).

The scale of the problem such a drug would be aimed at is real. Reviews of testosterone therapy list erythrocytosis as its most frequent adverse event, highest with short-acting injectables, with hepcidin suppression, iron turnover, erythropoietin and marrow stimulation as the proposed mechanisms (PMID 28526632, PMID 34987178).

A blood centre that reviewed three years of requests found therapeutic phlebotomy orders for testosterone patients rose 71 percent between 2014 and 2016, and after it stopped the service, 32 percent of those patients kept coming as volunteer donors (PMID 32134468). Among 519 transgender adults on testosterone with paired counts, 20 percent developed a hematocrit above 50 percent and 4.8 percent underwent phlebotomy, while thromboembolic events occurred in 0.9 percent (PMID 34592628). The testosterone cypionate profile has the dose-response and route data behind that hematocrit rise.

What rusfertide means if your hematocrit is climbing on testosterone

Nothing that you can act on, and the reasons are worth stating precisely rather than as a warning. MIMRYLO is approved for erythrocytosis in adults with polycythemia vera, a JAK2-driven blood cancer; secondary erythrocytosis on testosterone is a different diagnosis with a different management, in which the 2026 update on JAK2-unmutated erythrocytosis lists testosterone among the drug causes, advises against cytoreductive therapy, and reserves phlebotomy for symptom control (PMID 41123216). No trial of the peptide has enrolled anyone on testosterone. The label warnings, thrombocytosis above all, have been characterised only in patients whose platelets were already being watched by a hematologist.

The second reason is that even the standard tool is less settled than the forums assume. A 2024 review asked directly whether phlebotomy can be justified for testosterone-induced erythrocytosis and found the evidence for its efficacy or safety lacking, noting that phlebotomy lowers tissue oxygen tension and depletes iron, which triggers pathways that might raise thrombotic risk; it recommended shared decision-making rather than a reflex (PMID 39212549). A review of testosterone therapy and secondary erythrocytosis reached a similar place: the hematocrit cut-offs in guidelines appear arbitrarily chosen and the evidence that secondary polycythemia causes harm during testosterone therapy is scarce (PMID 34987178). Adding an unstudied hepcidin mimetic on top of an unsettled intervention is not a shortcut.

The third reason is practical. MIMRYLO is a lyophilised peptide in single-dose vials priced for a rare disease, titrated in steps no closer than two weeks apart against blood counts every two to four weeks. A vial of something labelled rusfertide from a research-chemical vendor carries none of that: no assay, no dose, no platelet count, and a drug whose entire safety record was built under hematology supervision. What does have evidence for a rising hematocrit on testosterone is on high hematocrit on TRT: dose and injection frequency, route, hydration and draw timing, and phlebotomy when a physician judges it needed. Where your own number sits is on the hematocrit calculator.

Where the drug is interesting for this audience is as a proof of concept. It is the first drug to show, in a randomised trial, that red cell overproduction can be controlled by iron restriction alone, without removing blood, and that the patients felt less fatigued rather than more. A trial of a hepcidin mimetic in testosterone-induced erythrocytosis is the study that would turn the forum idea into evidence. As of 30 September 2026 that study does not exist, and this page will be updated when it does.

Rusfertide bloodwork: what the label monitors, and what the trials moved

The label monitoring is short because the mechanism is single: iron. The markers below are the ones the prescribing information asks for or the trials moved, in the order a hematologist would read them; for a testosterone user they are also the markers that describe the problem the drug is not approved to treat.

The trial numbers say the effect is measurable within four weeks (PMID 41175501) and the label asks for blood counts on the same schedule. For anyone on testosterone the relevant panel is the standard one: hematocrit, hemoglobin, ferritin and a full count at baseline and at the trough, on TRT bloodwork and blood tests before steroids.

Five mistakes people make about rusfertide

Each of these comes from reading the mechanism as if it were a trial, or the approval as if it were broader than it is.

1. Saying rusfertide is still investigational. It was, until 28 August 2026. It is now MIMRYLO, approved for erythrocytosis in adults with polycythemia vera, with a label, five vial strengths and three warnings. The rusfertide FDA approval covers that disease and nothing else.

2. Treating testosterone erythrocytosis as a hepcidin problem only. Testosterone suppresses hepcidin (PMID 20660052), but mice without hepcidin still raised their hematocrit on testosterone (PMID 31001931) and erythropoietin rises as well (PMID 24158761). A hepcidin mimetic addresses one lever.

3. Expecting it to be gentle because it is a peptide. Platelets rose 31 percent within four weeks and exceeded 600 in 36 percent of patients on the label; anemia hit 16 percent against 4.1 on placebo; injection site reactions 56 percent. The label asks for blood counts every two to four weeks for a reason.

4. Quoting the 2021 mouse tumours as a human cancer finding. The hold came from a carcinogen-sensitive mouse strain and was lifted within a month; the approved label carries no malignancy warning and no PubMed abstract reports a human signal. Equally, nobody should quote the absence as proof of long-term safety in a drug approved weeks ago.

5. Buying a vial of rusfertide from a research vendor. MIMRYLO is a lyophilised peptide dosed by titration against blood counts, priced for a rare cancer, and never tested in anyone on testosterone. A grey-market vial has none of the assay, monitoring or evidence that made the approval possible.

Verdict: a real drug for a specific disease, and a mechanism for everything else

Rusfertide, after the label and the papers rather than the forums, is a hepcidin mimetic that controls polycythemia vera erythrocytosis by restricting iron: 76.9 percent of patients needed no phlebotomy against 32.9 on placebo, phlebotomies fell to 0.53 from 1.82, and 62.6 percent held a hematocrit below 45 percent (MIMRYLO prescribing information), on top of a phase 2 trial that took the phlebotomy rate from 8.7 a year to 0.6 (PMID 38381675). Its costs are anemia, a platelet rise and a sore injection site, and its unknowns are thrombosis and disease course (PMID 42660705).

For readers here, the drug is the first proof that iron restriction alone can hold a hematocrit down without removing blood, and a signpost to a trial that has not been run. Until it is, the evidence for a rising hematocrit on testosterone stays where it was: on high hematocrit on TRT, with the mechanism on do steroids cause high hematocrit and the wider drug class on peptide research vs human use.