ARA-290 Dosage, half-life & reference guide

Week by week.

What to expect in the first 8 weeks

Onset

1–2 weeks (user reports; the trials' main readouts were at day 28)

Peak effect

4 weeks, the point the trials measured

Duration

4 weeks, often repeated

1. Day 1–14Community

   The first two weeks: for some, the burning eases

   Commonly reported: burning, tingling or pins-and-needles settling inside the first one to two weeks, and just as commonly, nothing to feel yet. ARA-290 switches on a repair receptor that calms the inflammation around damaged nerve endings rather than blocking pain signals. How your symptoms sit today is the baseline the fourth week compares with.

   The science

   The trials' main readouts came at day 28 or 35; one also tracked symptoms weekly and saw pain interference separate from placebo by week 3, so the first-two-weeks timing comes from user reports. The mechanism rests on Brines 2004 and Brines 2008, which identified the innate repair receptor — the EPO receptor joined to CD131 — and showed an 11-amino-acid peptide built from helix B's water-facing side carrying EPO's tissue protection without raising red cells. In rats with a spared nerve injury, ARA-290 reduced mechanical and cold allodynia for up to 15 to 20 weeks after a handful of doses, and the effect was absent in mice lacking CD131 (Swartjes 2011; Swartjes 2014).

   Sources

   • Brines et al. (2004)
   • Brines et al. (2008)
   • Swartjes et al. (2011)
   • Swartjes et al. (2014)
   • Dahan et al. (2013)

   Updated Oct 2026

2. Day 14–28RCT

   Four weeks is where the trials looked

   The randomized trials in people with nerve symptoms dosed for four weeks (daily SubQ in three, IV three times a week in the first) and read their main results at, or a week after, the end of dosing. In people with sarcoidosis and documented small-nerve-fiber loss, symptom scores improved clearly more than on placebo, and the nerve fibers counted in the cornea went up. Pain intensity alone improved just as much on placebo, which is worth knowing before reading your own week.

   The science

   Dahan 2013 (38 patients, 4 mg SubQ daily for 28 days): the nerve-symptom score fell 12.2 points against 3.8 on placebo (p = 0.005); median corneal nerve-fiber area rose 14.5% from baseline in the treated group against a 5.3% fall on placebo; the six-minute walk gained 18.7 m against a 15.1 m loss (p = 0.049); skin-biopsy fiber density did not change significantly. Culver 2017, a 64-subject phase 2b of 1, 4 and 8 mg a day, put the placebo-corrected corneal gain at 697 µm² in the 4 mg group (p = 0.012), with regenerating skin fibers up in that group too — and pain improving in every group, the 4 mg difference not reaching significance. Brines 2015, in 48 people with type 2 diabetes and painful neuropathy, found the PainDetect score improving 3.3 points against 1.1 (p = 0.037).

   Sources

   • Dahan et al. (2013)
   • Culver et al. (2017)
   • Brines et al. (2015)

   Updated Oct 2026

3. Day 28–56RCT

   What happened after the dosing stopped

   The unusual part of ARA-290's trial record is the tail: in the two trials that followed people after dosing stopped, the improvement outlasted the drug, which clears from blood in about 20 minutes. Four weeks after the last dose the gap over placebo was wider, not narrower, in the diabetes trial. If you have stopped, this is the window the trials looked at.

   The science

   Brines 2015: the PainDetect improvement grew from 3.3 points at day 28 to 4.2 at day 56, 28 days after the last dose, while placebo drifted back from 1.1 to 0.74; HbA1c stayed down (−0.21% at day 56 against +0.21% on placebo, p = 0.002 by repeated-measures ANOVA), with triglycerides and the cholesterol-to-HDL ratio also improved. Dahan 2013 followed its patients 12 weeks past dosing with the symptom gain largely held, and reached 19 of 21 treated patients again nine months later, still better than baseline against placebo (p = 0.036). Its developers read this as a switch being flipped rather than a drug level being maintained (Dahan 2016).

   Sources

   • Brines et al. (2015)
   • Dahan et al. (2013)
   • Dahan et al. (2016)

   Updated Oct 2026

4. Day 56–120Clinical

   Past the trials' window

   The longest published daily course is 12 weeks, in nine people with diabetic macular edema; it was tolerated, and the eye measures it was built to move did not move. Beyond that, open-ended use is uncharted — not known to be harmful, just unstudied. Community practice is a four-week run and a break rather than running it continuously.

   The science

   Lois 2020: 4 mg SubQ daily for 12 weeks, nine patients, eight completing. Visual acuity, retinal thickness, retinal sensitivity and tear production did not improve on average; the vision-function questionnaire score did, and individual participants improved on retinal thickness, tear production, diabetic control and albuminuria. All nine had at least one adverse event, none serious, none with antibodies to the drug. The 28-day trials (Heij 2012; Dahan 2013; Culver 2017; Brines 2015) are everything else there is.

   Sources

   • Lois et al. (2020)
   • Heij et al. (2012)
   • Dahan et al. (2013)
   • Culver et al. (2017)
   • Brines et al. (2015)

   Updated Oct 2026

5. Any dayRCT

   What ARA-290 is built on

   A clean mechanism, four small randomized trials, and one company. ARA-290 is the rare research peptide whose human studies were placebo-controlled and measured something objective — nerve fibers in the cornea — and its patient trials all come from its developer, in about 200 people, with no approval anywhere. Its trials say nothing about a nerve under mechanical pressure, and the one user report on that found no change.

   The science

   Brines 2008 delimited EPO's tissue-protective signal to helix B and then to an 11-amino-acid peptide, on the receptor Brines 2004 had identified. The randomized human record: Heij 2012 (22 patients, 2 mg IV three times a week), Dahan 2013 (38 patients, 4 mg SubQ daily), Culver 2017 (64 subjects, 1, 4 or 8 mg a day) and Brines 2015 (48 with type 2 diabetes), plus Lois 2020 in nine eyes and Cerit 2015, which gave 36 healthy volunteers a single 2 mg dose and found no clear antidepressant signal. Araim Pharmaceuticals, the developer, co-authored all five patient trials. Two animal papers in the wider helix-B peptide literature have since been retracted by their journals, neither of them a study in people.

   Sources

   • Brines et al. (2008)
   • Brines et al. (2004)
   • Heij et al. (2012)
   • Dahan et al. (2013)
   • Culver et al. (2017)
   • Brines et al. (2015)
   • Lois et al. (2020)
   • Cerit et al. (2015)

   Updated Oct 2026

6. With BPC-157 and TB-500Community

   Different jobs: the nerve and the tissue

   A pairing the community reaches for when an injury hurts in a nerve way as well as a structural way. The split is honest: ARA-290's human data is about nerve symptoms and nerve-fiber counts, while BPC-157 and TB-500 rest on animal work about blood supply and repair cells moving in. Nothing has tested them together.

   The science

   ARA-290 acts on the innate repair receptor and, in rats, suppressed the spinal microglia response that maintains neuropathic pain (Swartjes 2014). BPC-157's case is rat work on angiogenesis and tendon healing (Seiwerth 2018); TB-500's is lab and animal work on cell migration. The one controlled test of BPC-157 with TB-500, in 32 rats with repaired Achilles tendons, found the combination added nothing over either alone (Biçer 2026), and no study has combined either with ARA-290.

   Sources

   • Swartjes et al. (2014)
   • Seiwerth et al. (2018)
   • Biçer et al. (2026)

   Updated Oct 2026

Swipe for more ›

Mixing and a typical run.

Safety.

Goes with.

• BPC-157
• TB-500

Questions people ask.

Research.

Reference.

Cadence

Daily

Half-life

~20 minutes

~20 minutes (measured in healthy volunteers after a SubQ dose; about 2 minutes after IV, and its developers argue a brief exposure is enough because the receptor acts as a switch)

Research basis

●●● Moderate

Animal studies + some human data