ARA-290

What ARA-290 is and how it works

ARA-290 is eleven amino acids long. Its sequence is a copy of a stretch of the outer surface of helix B of erythropoietin, the hormone that drives red blood cell production, and it was built specifically so that it would not do what erythropoietin does.^[4]

Three names, one molecule. The same peptide appears as ARA-290, as cibinetide — its international nonproprietary name — and as pyroglutamate helix B surface peptide, abbreviated pHBSP. A review of the compound's pharmacology names two of the three together in one sentence, introducing the eleven-residue peptide modelled on the helix B region of erythropoietin as "pyroglutamate helix B surface peptide (pHBSP; ARA-290)"; that review's own text does not use "cibinetide" at all, and the name appears in its PubMed record only as a chemical indexing term.^[4] What establishes the three-name equivalence is the public chemical registry rather than that review: one compound record, checked on 1 September 2026, carries ARA290, ARA-290, pHBSP, cibinetide and "pyroglutamate helix B surface peptide" as synonyms of a single substance. That is not a pedantic point: on 1 September 2026 a PubMed search for ARA290 returned 80 records, ARA-290 61, cibinetide 53 and pHBSP 73, and the four sets do not overlap cleanly, so a reader searching only the trade-style name misses a substantial part of the corpus — including two papers in it that have since been retracted and are indexed under pHBSP.

It is not erythropoietin, and no erythropoietin finding transfers to it. Erythropoietin is a glycoprotein of roughly 30 kilodaltons; this peptide is about 1.26 kilodaltons and eleven residues. Preclinical papers in this corpus routinely run erythropoietin as a separate condition alongside ARA-290 — the independent receptor study described below is one of them — which makes accidental attribution easy. The arm matters. Carbamylated and asialo erythropoietin, which appear in the same discussions, are full-length modified erythropoietins sharing a design goal with this peptide and nothing else.

The proposed mechanism, and why it is contested. The account offered by the group developing the compound is that tissue protection is mediated not by the erythropoietin receptor homodimer that stimulates red-cell progenitors, but by a heterodimeric complex of that receptor with the beta-common receptor — an "innate repair receptor" said to be induced by injury or inflammation rather than normally present — and that an eleven-residue peptide can engage it selectively.^[4] An independent structural group tested that association directly, computationally and in vitro, with and without erythropoietin and with ARA290 itself, opened by noting that the complex is controversial and that no biophysical evidence confirmed a direct association, and concluded that the extracellular regions of the two receptors do not specifically associate.^[9] The mechanism is therefore a live hypothesis with a published counterweight, not an established fact.

Why anyone built a peptide like this. Erythropoietin itself showed protective effects in preclinical injury models, but four published clinical trials of high-dose erythropoietin after renal transplantation found no protective effect on short-term renal function and reported an increased risk of thrombosis.^[3] Stripping the erythropoietic activity out was the point of the exercise.

It does not last long. In blood specimens in vitro the peptide underwent extensive exopeptidase-driven metabolism, with one main metabolite, PyrEQLERALN.^[8] The proponents' own review gives a plasma half-life of about two minutes and treats the mismatch between that and the duration of the reported effects as the compound's central pharmacological puzzle.^[4]

What the research actually shows

Three trials in sarcoidosis, and they are the core of the record. A double-blind, placebo-controlled pilot in 22 patients with sarcoidosis and symptoms of small fiber neuropathy administered 2 mg intravenously three times weekly for four weeks. The small fiber neuropathy screening list score improved significantly against placebo (P < 0.05), and no safety concerns were raised by clinical or laboratory assessments — but mean Brief Pain Inventory and Fatigue Assessment Scale scores improved significantly and equivalently in both groups, which is to say the trial found no separation from placebo on pain or fatigue, and the authors describe the study as exploratory.^[1] A second blinded, placebo-controlled trial gave 28 days of daily subcutaneous administration and reported improved neuropathic symptoms together with increased corneal small nerve fiber density, changed cutaneous temperature sensitivity and increased six-minute walk distance; its abstract states neither the number of people randomized nor the amount administered, and the record carries a published erratum.^[2]

The phase 2b is the best human evidence that exists, and it is a split decision. Sixty-four adults with sarcoidosis-associated small nerve fiber loss and neuropathic pain received 1, 4 or 8 mg per day, or placebo, for 28 days. The primary endpoint was change in corneal nerve fiber area. It was MET in the 4 mg arm — a placebo-corrected 697 µm², 95% confidence interval 159 to 1236, P = 0.012 — and MISSED in the 1 mg arm (109; −429 to 647) and in the 8 mg arm (431; −130 to 992). The highest dose tested did not work. Pain improved significantly in every group including placebo, and the decrease the authors describe as clinically meaningful in the moderate-to-severe 4 mg subgroup carried P = 0.157, which is not statistically significant: the pain endpoint MISSED.^[7] The results posted to that trial's registry entry add detail the abstract does not, and they cut in both directions. The mean changes in corneal nerve fiber area — the primary measure, in µm² — were −64.3 at 1 mg, +533.8 at 4 mg, +203.8 at 8 mg and −170.0 on placebo, but every one of those figures carries a standard deviation larger than itself (1110.02 on the +533.8, for instance), and the registry entry posts no p-values and no between-group analysis at all, so the posted primary numbers are dispersed enough that quoting the 4 mg figure without its spread would overstate what is there. Intraepidermal nerve fiber density moved the wrong way in the 8 mg arm (−0.3, SD 1.81) while placebo showed the largest increase (+0.8, SD 1.13). The 8 mg arm lost three of sixteen participants — one to an adverse event, two whose treatment assignment could not be confirmed — and it carried the trial's only serious adverse event, one case of suicidal ideation, with no such event in any other arm. On the neuropathic pain symptom inventory the 4 mg arm actually did better than placebo (−14.50 against −12.88); the only sub-scale where placebo edged the 4 mg arm was Brief Pain Inventory pain interference, and the margin there was 0.075 against standard deviations of roughly 1.6 to 2.2, which is noise rather than a finding in either direction.

Type 2 diabetes, and the one place this page grades down. A placebo-controlled phase 2 study administered 4 mg subcutaneously daily for 28 days with a further month of observation, and reported improvement in hemoglobin A1c and lipid profiles over the 56 days, significant improvement in PainDetect scores, and a significant increase in corneal nerve fiber density confined to the subgroup already more than one standard deviation below normal. The abstract states no sample size, and the authors' conclusion is a hedge: the peptide "may benefit" both and "deserves continued clinical evaluation".^[5] This is the one claim on the page graded below "Human RCT evidence", and the reason is narrow: the word "randomized" does not appear in that abstract, and PubMed, checked on 1 September 2026, types the record as a Controlled Clinical Trial and a Phase II Clinical Trial rather than a Randomized Controlled Trial. The study was double-blind and placebo-controlled; randomization is simply not reported, and the rubric grades down rather than up when that is the case. The registered trial that matches it — a randomized, triple-masked phase 2 at a Swedish academic centre — records an estimated enrolment of 24, has sat at "unknown" status since its last update in September 2015, and has never posted results, so it cannot settle the question either.

Healthy adults, once. Thirty-six healthy participants received a single 2 mg administration or placebo in a randomized parallel-group design and were assessed a week later, with neural processing of fearful versus happy faces and behavioural recognition of emotional expressions as the primary measures. Some shifts in emotional processing appeared — a lower neural response to happy faces in the fusiform gyrus, faster categorisation of positive versus negative words — but no effects were observed on mood and affective symptoms, and the authors concluded that the direction and strength of the results do not unequivocally support an antidepressant-like profile.^[6]

Diabetic macular edema. A single-arm, open-label trial with no control group recruited nine adults, eight of whom completed 12 weeks of 4 mg per day subcutaneously. There was no improvement in best corrected visual acuity, central retinal thickness, central retinal sensitivity or tear production — the primary and secondary measures. A vision-related quality-of-life score improved, some individual participants improved on individual measures, and there were no serious adverse events and no antibodies to the peptide.^[10] Its registry entry records the trial as terminated, and the reason it gives reads in full: "Expiry of study drug - no replacement available."

Animals. In a mouse islet-allograft model, cibinetide reduced local inflammatory responses and delayed the onset of allograft loss, while durable long-term graft survival required combination with low-dose tacrolimus; cell work in the same paper found reduced maturation of bone-marrow-derived immature dendritic cells and a reduced allogeneic T-cell response.^[11] In mice bred to develop Alzheimer's-like pathology, an independent group reported that early treatment decelerated amyloid-beta accumulation and improved cognition in young animals, increased circulating monocytes and bone-marrow monocyte progenitors, and failed to reverse pathology or raise monocytes at all in aged animals.^[13]

Where the evidence is weak

Almost the entire human record belongs to one network. A PubMed search on 1 September 2026 for (ARA290 OR "ARA-290" OR cibinetide OR pHBSP) restricted to clinical-trial publication types returned five records, and adding the single-arm macular edema trial, which carries no clinical-trial indexing, gives six published human studies in total. The author list of every one of them includes staff of the company developing the peptide, the anesthesiology department of one Dutch university medical centre, or both. Exactly one — the 36-person healthy-volunteer study — has no author from the developing company, and it is the negative one.^[6] The word "independent" therefore does not belong anywhere near the human evidence on this page. The independent primary research that does exist is preclinical, and it cuts against the compound's premises rather than for them.^[9][13][3]

Why the badge on this page reads "Human RCT evidence." The evidence grade describes the quality of the evidence testing a claim, not the direction it points. Randomized, placebo-controlled human trials of this peptide were conducted and published, and the alternative grades — "animal studies only" and "anecdotal reports only" — would both misdescribe that record. What the badge does not mean is that the compound works: the phase 2b met its surrogate endpoint in one of three dose arms and missed its pain endpoint, the healthy-adult study found no effect on mood and its authors declined to read their own results as supporting an antidepressant-like profile, and the macular edema study missed both its primary and its secondary measures.^[7][6][10] A high grade means the question was asked properly.

The dose response is not monotonic, and the positive endpoints are surrogates. A middle dose separating from placebo while the doses either side of it fail, in arms of sixteen, is the shape a false positive takes as often as the shape a narrow therapeutic window takes. Corneal nerve fiber area and density are imaging biomarkers, presented by the phase 2b authors themselves as surrogate endpoints for disease modification rather than as symptoms.^[7] The endpoints closest to what a patient feels — pain, fatigue — are the ones that did not separate.^[1]

Nothing here is a monotherapy result. The sarcoidosis trials enrolled people whose current therapy for this complication, in the words of one of those trials, "consists primarily of immune suppression and symptomatic treatment".^[2] No published abstract read for this page describes a washout of that background therapy, so the reported changes are changes on top of whatever else participants were taking. In the animal transplantation work, the durable benefit explicitly required combination with low-dose tacrolimus.^[11]

"Non-erythropoietic" is well supported as a design claim and routinely overstated as a safety claim. A PubMed search for (ARA290 OR cibinetide OR pHBSP OR ARA-290) AND (hemoglobin OR haemoglobin OR hematocrit OR haematocrit OR reticulocyte), restricted to human records, on 1 September 2026 returned three records, and none of them reports a hemoglobin, hematocrit or reticulocyte dataset from people given the peptide — so the premise the whole compound rests on has never been measured as an endpoint in a person, in either direction. Meanwhile an independent group reports that it raises circulating monocytes and bone-marrow monocyte progenitors in mice.^[13] Non-erythropoietic, not hematologically inert, and never measured for it.

Nothing addresses the reason most readers arrive here. A PubMed search for (ARA290 OR cibinetide OR pHBSP OR ARA-290) AND (exercise OR athletes OR sport OR training OR muscle recovery) on 1 September 2026 returned five records, three of them analytical-chemistry method papers about detecting the peptide; the same search combined with (tendon OR ligament OR wound healing OR musculoskeletal injury) returned six, all mouse, cell-culture or biomaterials work; and the same search combined with (healthy volunteers OR healthy subjects OR healthy adults) returned two, one of them the mood study above and the other a corneal-imaging methodology paper. Across those three searches on that date, no record is a study of injury recovery, athletic performance or immune function in a healthy person.

The development programme stopped. A query of the ClinicalTrials.gov API on 1 September 2026 across ARA 290, ARA290, cibinetide and pHBSP returned four registered studies in total and no fifth: all of them phase 1 or phase 2, the largest with 64 actual participants, with no phase 3 registered there under any of those names. That same query on that date returns nothing at all for pHBSP, because the one registry record carrying the abbreviation spells it transposed as "pHSBP" — a spelling artefact rather than an absence, and a reason a reader searching the registry under that name alone would wrongly conclude there is nothing there. The latest study conduct in any of them ended in August 2017, which is not the same as the latest registry activity: the macular edema trial was not first posted until 4 October 2024. One is terminated, with the reason given as "Expiry of study drug - no replacement available"; and one has sat at unknown status since its last update in September 2015, more than a decade.

Legal and regulatory status

ARA-290 is not approved by the FDA for any indication, and there is no approved US labeling for it. Searches of Drugs@FDA and of the approved-labeling database on 1 September 2026 returned no record under cibinetide or ARA290 in any query shape, while control searches run in the same session for approved substances returned records normally. The National Drug Code directory does return four rows carrying these names, and every one of them is registered as a bulk ingredient rather than a finished drug product, with no application number and no route of administration — an ingredient registration is not an approval.

Orphan-drug designations are real here, and a designation is not an approval. FDA's orphan designation database, searched on 1 September 2026, holds four designations for this peptide, listed under its amino-acid sequence rather than under any trade name and held by Araim Pharmaceuticals: prevention of delayed graft function following renal transplant, designated 18 September 2009; treatment of neuropathic pain in patients with sarcoidosis, 13 September 2011; treatment of sarcoidosis, 13 June 2016; and treatment to increase survival and improve functioning of pancreatic islets following transplantation, also 13 June 2016. Each of the four records carries FDA's own status line, "Not FDA Approved for Orphan Indication", and a control search in the same session did return other products marked "Designated/Approved". Two European orphan designations sit on the same footing: EU/3/13/1191 for sarcoidosis, by Commission decision of 7 October 2013, and EU/3/16/1721 for prevention of graft loss in pancreatic islet transplantation, 29 August 2016, both held by the sponsor's European entity, both listed as active in the European Commission's Union Register on 1 September 2026, and neither attached to any marketing authorisation. An orphan designation is a statement about how rare a disease is, plus a package of development incentives. It is not evidence that a drug works and it is not permission to market it — seventeen years after the first of these, the searches this page ran on 1 September 2026 found no US approval, and neither EU designation is attached to a marketing authorisation in the Union Register. Those are the only two jurisdictions checked; no other national regulator was searched, and this page asserts nothing about the ones it did not look at. ARA-290 is also widely described as holding an FDA fast-track designation; FDA publishes no searchable register of fast-track designations and the orphan records carry no such field, so this page asserts neither that it holds one nor that it does not.

Compounding. The substance appears exactly once in FDA's 503A bulk drug substance nomination category lists, in the document dated 14 May 2026, as "Cibinetide (ARA-290)" under Category 3, "Bulk Drug Substances Nominated Without Adequate Support". That is a fact about a nomination someone filed and about the support submitted with it. It is not an approval, not a compounding authorisation and not a safety clearance. Checked on the same day, it does not appear anywhere on FDA's page listing bulk drug substances used in compounding that may present significant safety risks, whose content was stamped current as of 22 April 2026 and which does name several other research peptides.

Anti-doping. The 2026 World Anti-Doping Agency International Standard Prohibited List, in force from 1 January 2026, was retrieved on 1 September 2026 from the WADA-hosted document linked by the USADA prohibited-list page. Searching that document for ARA-290, ARA290, cibinetide, pHBSP and "helix B" returns no hit, and the peptide is likewise absent from the 2026 Monitoring Program checked the same day. That is not the finding. Section S2 covers peptide hormones, growth factors, related substances and mimetics, is prohibited at all times, contains only non-Specified Substances, and opens with a reach clause: "The following substances, and other substances with similar chemical structure or similar biological effect(s), are prohibited." Within it, S2.1 covers erythropoietins and agents affecting erythropoiesis "Including, but not limited to", and S2.1.5 reads in full: "Innate repair receptor agonists, e.g. asialo EPO; carbamylated EPO (CEPO)." That subsection is titled by mechanism, and the mechanism it names is the one this peptide is designed and described on; both of the examples it gives are themselves non-erythropoietic erythropoietin derivatives, so being non-erythropoietic is the defining feature of the subsection rather than an exemption from it. Carbamylated and asialo erythropoietin are the substances named, and they are different molecules from this one — the accurate statement is that ARA-290 is captured by the class, not that it is named on the List. Absence of a name from the List is never a statement that a substance is permitted. The List is reissued annually, so the position above is dated rather than standing.

The absence of an approved product is a scientific question as much as a regulatory one. An independent 2021 systematic review of drug development for diabetic neuropathy — sole-authored, and listing cibinetide among its search terms — recorded that over the preceding half century a number of agents had undergone unsuccessful trials for that condition, and argued for regulatory changes that would let agents with demonstrated safety and a suggestion of benefit reach patients while long-term trials continue. This compound sits inside that landscape rather than outside it.^[12]

PHL does not sell peptides and does not tell anyone where to obtain anything. What the regulatory position means for a reader is narrower and more useful: there is no approved product, no label, no assigned indication and no regulated quality standard behind material carrying any of this compound's three names, and the human evidence behind it is preliminary and small.

Questions to bring to a provider

Most people who reach this page have a specific problem — a neuropathy, a sarcoidosis diagnosis, a nerve-pain syndrome — and that problem, not the peptide, is where a useful conversation starts. Questions worth raising:

• What is the actual diagnosis, and what does the standard-of-care evidence base offer for it? Sarcoidosis-associated small fiber neuropathy in particular is usually managed with immune suppression and symptomatic treatment.^[2]
• Given that the largest randomized trial met a surrogate imaging endpoint in one of three dose arms and missed its pain endpoint, what would have to be true for this to be worth considering at all?^[7]
• The animal work reports this peptide blunting dendritic-cell maturation and the T-cell response that follows it. For someone already on immune-suppressing therapy for sarcoidosis or a transplant, is adding another agent that points the same direction a question worth asking at all?^[11]
• Does a personal or family history of thrombosis, or current erythropoietin therapy, change how a clinician would think about anything derived from that hormone?^[3]
• Which objective measures would be followed to tell whether anything actually changed? Corneal confocal microscopy and skin-biopsy nerve fiber density are the measures the trials used, and both are specialist procedures.^[7]
• For anyone competing under anti-doping rules, what is the process for checking a substance against the current Prohibited List before it is ever a question?
• What is the plan if a biomarker moves and symptoms do not, which is the pattern the published trials most often produced?^[1]

A clinician who answers that the human evidence here is six small published studies, most of them from one group, is describing the literature accurately rather than dismissing the question.