Hormone Research Market

Atosiban

An oxytocin analog blocking oxytocin and vasopressin V1a receptors, authorized in the European Union since 2000 as Tractocile to delay preterm birth, never approved in the United States, and no better than placebo for neonatal outcomes in APOSTEL 8.

Oxytocin AntagonistTocolyticPreterm LaborVasopressinNot FDA Approved

Also referenced as: Tractocile, Antocin, Atosiban SUN, atosiban acetate, RWJ-22164, ORF 22164, RW-22164

Public product evidence No certificate records are bound to a tracked Atosiban product page A record counts here only when a vendor publishes a certificate against one of its own tracked Atosiban product pages. Other certificates naming this compound may still exist — search the wider certificate database.
Status
Research Market

This name primarily lives in the research market and should not be read like an approved pharmaceutical product.

Research area
Hormone signaling research

This profile is grouped by its dominant research area, not by vendor shelf placement.

Body systems
Reproductive · Hormones

Where the compound acts, as a browse axis. It is an editorial grouping for discovery, not a claim about labeled indications.

Aliases
7

Tractocile, Antocin, Atosiban SUN, atosiban acetate, RWJ-22164, ORF 22164, RW-22164

Signal depth
Low

No FDA label signal · 26 trials · 579 PubMed results

Promising

Atosiban has name-matched human trials with published or reported controlled evidence, but is not FDA-approved. The research is real and ongoing — treat findings as developing rather than settled.

Atosiban has 13 name-matched clinical trials (highest phase: Phase 4) and 579 PubMed-indexed publications and is not FDA-approved. 1 trial has posted results.

Human data
Phase 3
Trial quality
Large RCT
Outcomes
Clinical outcomes
Replication
Meta-analysis
Literature
Top-tier journals

Re-checked nightly against the registries — tracked since 2026-11-17. No band changes yet.

Grades evidence strength, not efficacy or safety. Research-use context; not medical advice. Graded 2026-11-18 from PubMed, ClinicalTrials.gov, ISRCTN, openFDA, Health Canada, and OpenAlex — computed deterministically and refreshed nightly, with a retraction check. How we grade →


What is Atosiban?

Atosiban is a synthetic nonapeptide analog of oxytocin that acts as a competitive antagonist at the oxytocin receptor. Its structure is written [Mpa1,D-Tyr(Et)2,Thr4,Orn8]-oxytocin: the N-terminal cysteine is replaced by 3-mercaptopropionic acid, tyrosine 2 by O-ethyl D-tyrosine, glutamine 4 by threonine and leucine 8 by ornithine, while the disulfide ring of oxytocin is kept. The molecular formula is C43H67N11O12S2 and the molecular weight 994.2 g/mol; the marketed form is atosiban acetate. The compound was developed by Ferring, which first tested it in women in preterm labor in Lund, Sweden, and it carried the development codes RW-22164, ORF 22164 and RWJ-22164. The US new drug application was filed by the R. W. Johnson Pharmaceutical Research Institute under the trade name Antocin. In Europe it is sold as Tractocile and as the generic Atosiban SUN.

Atosiban was the first oxytocin receptor antagonist to reach the clinic. It is a hospital tocolytic given intravenously to delay imminent preterm birth between 24 and 33 completed weeks of gestation, with essentially no footprint in the research-peptide market. It is a distinct molecule from carbetocin, an oxytocin agonist used after delivery, and from the later non-peptide antagonists retosiban, nolasiban and epelsiban.

How it works

  • Competitive oxytocin-receptor antagonist. Atosiban occupies the myometrial oxytocin receptor and blocks the Gq-phospholipase C pathway through which oxytocin raises intracellular calcium and drives contraction. Its affinity for the human oxytocin receptor is modest (Ki 76.4 nM) and its affinity for the human vasopressin V1a receptor higher (Ki 5.1 nM), so it is a mixed oxytocin/V1a antagonist, and later peptide antagonists were designed for far greater oxytocin-receptor affinity and selectivity (Manning et al., Journal of Peptide Science, 2005;11(10):593–608). The European product information notes the same dual binding and that atosiban inhibits the effect of vasopressin.
  • Biased agonism at Gi. Although it antagonizes oxytocin-receptor coupling to Gq, atosiban acts as an agonist on Gi coupling, stimulating GTPgammaS binding, persistent ERK1/2 activation and p21 induction and inhibiting growth of receptor-expressing cells, without internalizing or desensitizing the receptor (Reversi et al., Journal of Biological Chemistry, 2005;280(16):16311–16318). In human amnion cells it fails to block the proinflammatory effects of oxytocin and on its own activates NF-kappaB and MAP kinases, upregulating prolabor genes, whereas nolasiban suppresses oxytocin-driven NF-kappaB and p38 activation in amnion and myometrial cells (Kim et al., Molecular Pharmacology, 2017;91(4):403–415).
  • Rapid uterine quiescence and a short half-life. In eight pregnant women with preterm contractions, an infusion reached a steady-state plasma concentration of 442 ng/mL within an hour, contraction frequency fell by 75% during the third hour, and after the infusion stopped plasma levels fell with initial and terminal half-lives of 13 and 102 minutes and a clearance of about 42 L/h (Goodwin et al., American Journal of Obstetrics and Gynecology, 1995;173(3 Pt 1):913–917). The European product information adds that contractions are significantly reduced within 10 minutes and that the fetal-to-maternal plasma ratio at term is 0.12.

Research status

Pilot study. The analog was first infused in 13 women with uncomplicated premature labor; uterine activity was inhibited in every woman without maternal or fetal side effects (Akerlund et al., British Journal of Obstetrics and Gynaecology, 1987;94(11):1040–1044).

US placebo-controlled program. The pivotal US trial randomized 531 women with preterm labor (501 treated) to intravenous atosiban or placebo followed by subcutaneous maintenance, with standard tocolytics allowed as rescue after one hour. The primary endpoint, time to delivery or therapeutic failure, did not differ (median 25.6 versus 21.0 days; P = .6), although more women on atosiban remained undelivered without rescue at 24 hours, 48 hours and 7 days, a benefit confined to pregnancies of 28 weeks or more. Fourteen atosiban and five placebo patients were randomized before 24 weeks, and fetal and infant deaths were more frequent with atosiban in that subgroup (Romero et al., American Journal of Obstetrics and Gynecology, 2000;182(5):1173–1183). The European product information gives the pooled figures as 15 of 288 (5.2%) fetal or infant deaths on atosiban versus 5 of 295 (1.7%) on placebo, 11 of the 15 in pregnancies of 20 to 24 weeks, with no mortality difference above 24 weeks (1.5% versus 1.7%). A companion trial of subcutaneous maintenance in 513 responders lengthened the median time to recurrence of labor from 27.6 to 32.6 days (P = .02) (Valenzuela et al., American Journal of Obstetrics and Gynecology, 2000;182(5):1184–1190).

Beta-agonist comparator trials. In 247 women at 23 to 33 weeks, 84.9% on atosiban and 86.8% on ritodrine remained undelivered at 48 hours, while maternal cardiovascular side effects occurred in 4.0% versus 84.3% (Moutquin et al., American Journal of Obstetrics and Gynecology, 2000;182(5):1191–1199). The three pooled trials of the Worldwide Atosiban versus Beta-agonists Study Group randomized 742 women to atosiban or intravenous ritodrine, salbutamol or terbutaline; delivery was delayed 48 hours in 88.1% versus 88.9% (P = 0.99) and 7 days in 79.7% versus 77.6% (P = 0.28); cardiovascular adverse events occurred in 8.3% versus 81.2% (P < 0.001), with no difference in neonatal outcomes (Worldwide Atosiban versus Beta-agonists Study Group, BJOG, 2001;108(2):133–142). On this basis the CHMP concluded that Tractocile delayed birth as well as beta-agonists and that its better outcome was due to tolerability.

Regulatory history. The FDA’s Advisory Committee for Reproductive Health Drugs was convened on April 20, 1998 to discuss NDA 20-797, Antocin (atosiban injection, R. W. Johnson), for the management of premature labor. No approval followed; Drugs@FDA lists no atosiban application and no atosiban label exists on DailyMed. The European Commission granted Tractocile a marketing authorization valid throughout the European Union on January 20, 2000, with Ferring Pharmaceuticals A/S as holder, and the generic Atosiban SUN followed on July 31, 2013; the product is also licensed in the United Kingdom and in other countries, but not in the United States.

Cochrane review. The 2014 update included 14 trials and 2,485 women. Against placebo (four trials, 854 women) there was no difference in birth within 48 hours (RR 1.05, 95% CI 0.15–7.43) or perinatal mortality (RR 2.25, 95% CI 0.79–6.38), and the US trial showed more births before 28 weeks (RR 3.11) and more infant deaths to 12 months (RR 6.13), a finding the authors noted may be confounded by the randomization of more women below 26 weeks to atosiban. Against betamimetics (eight trials, 1,402 women) birth within 48 hours did not differ (RR 0.89, 95% CI 0.66–1.22) and far fewer women stopped treatment for adverse effects (RR 0.05). The authors concluded that the review did not demonstrate superiority of atosiban over placebo, betamimetics or calcium-channel blockers for pregnancy prolongation or neonatal outcomes (Flenady et al., Cochrane Database of Systematic Reviews, 2014;(6):CD004452).

APOSTEL III (NTR2947). This open-label Dutch and Belgian trial randomized 510 women at 25 to 34 weeks to oral nifedipine or intravenous atosiban for 48 hours. The composite adverse perinatal outcome occurred in 14% of babies on nifedipine and 15% on atosiban (RR 0.91, 95% CI 0.61–1.37); 16 (5%) babies died in the nifedipine group and 7 (2%) in the atosiban group (RR 2.20, 95% CI 0.91–5.33) (van Vliet et al., Lancet, 2016;387(10033):2117–2124). At 2.5 to 5.5 years, abnormal development was reported in 30% of nifedipine-exposed and 38% of atosiban-exposed children (OR 0.74, 95% CI 0.41–1.34) (van Winden et al., BJOG, 2020;127(9):1129–1137).

APOSTEL 8 (NCT05693688). The first placebo-controlled atosiban trial powered for neonatal outcomes randomized 755 women with threatened preterm birth at 30 weeks 0 days to 33 weeks 6 days in the Netherlands, England and Ireland between December 4, 2017 and July 24, 2023; 752 were analyzed. The composite of perinatal mortality and six severe neonatal morbidities occurred in 37 of 449 infants (8%) on atosiban and 40 of 435 (9%) on placebo (RR 0.90, 95% CI 0.58–1.40), with 3 and 4 deaths respectively; the authors concluded that superiority was not demonstrated and questioned standardized use at these gestations (van der Windt et al., Lancet, 2025;405(10483):1004–1013). The accompanying economic evaluation did not demonstrate cost-effectiveness (van der Windt et al., PLoS One, 2026;21(11):e0347475).

Comparator role. Atosiban was the active control in GlaxoSmithKline’s phase 3 trial of retosiban (NCT02292771), which enrolled 97 women, 29% of target, before termination for slow recruitment (Saade et al., American Journal of Perinatology, 2021;38(S 01):e309–e317).

Assisted reproduction. Atosiban has been given around embryo transfer to suppress uterine contractions. A double-blind trial of 800 women (NCT01501214) found live birth rates of 39.8% with atosiban and 38.0% with saline (P = 0.612) (Ng et al., Human Reproduction, 2014;29(12):2687–2694), and the 2021 Cochrane review rated the effect on live birth as uncertain (RR 1.05, one trial, low certainty) (Craciunas et al., Cochrane Database of Systematic Reviews, 2021;9(9):CD012375). As of November 2026, ClinicalTrials.gov lists three recruiting investigator-led atosiban studies, all in China, and none in the United States.

Common dosage forms

  • Tractocile 6.75 mg/0.9 mL solution for injection. A single-dose 0.9 mL glass vial of atosiban acetate for intravenous bolus injection, used in hospital obstetric units.
  • Tractocile 37.5 mg/5 mL concentrate for solution for infusion. A 5 mL vial (7.5 mg/mL) of atosiban acetate that is diluted for intravenous infusion in hospital.
  • Generic atosiban. Atosiban SUN and national generics in the same 6.75 mg/0.9 mL and 37.5 mg/5 mL vial or ampoule strengths.
  • Investigational subcutaneous formulation. A pump-delivered subcutaneous presentation used in the US trial program of the late 1990s; it was never marketed.

Key considerations

  • Regulatory status. The Research Market status reflects the absence of any FDA-approved atosiban product in Drugs@FDA. The FDA’s Advisory Committee for Reproductive Health Drugs met on April 20, 1998 to discuss NDA 20-797 for Antocin and no approval followed, while the European Commission authorized Tractocile on January 20, 2000, and no US-directed program is active as of November 2026.
  • Delays birth without improving neonatal outcomes. Atosiban prolongs pregnancy about as well as beta-agonists or nifedipine, but neither the Cochrane review nor APOSTEL 8 found a neonatal benefit over placebo. Its established value is tolerability, not efficacy.
  • Early-gestation mortality signal. The US placebo-controlled trial recorded 5.2% fetal or infant deaths with atosiban versus 1.7% with placebo, concentrated in unbalanced 20-to-24-week enrollment. The European label excludes use before 24 completed weeks, and the Cochrane authors singled out the finding as a reason for caution.
  • Mixed receptor pharmacology. Atosiban’s affinity for the vasopressin V1a receptor exceeds its affinity for the oxytocin receptor, and it acts as a biased agonist on Gi signaling without blocking oxytocin’s proinflammatory effects in amnion, properties the selective antagonists retosiban, nolasiban and barusiban were designed to avoid.
  • Naming. Atosiban, atosiban acetate, RW-22164, ORF 22164, RWJ-22164, Antocin, Tractocile and Atosiban SUN all denote the same peptide, an oxytocin antagonist not to be confused with the agonists oxytocin and carbetocin.