Serelaxin
Recombinant human relaxin-2, rhRlx, RLX030, Human relaxin H2
Serelaxin is recombinant human relaxin-2, a peptide hormone of pregnancy that produces systemic vasodilation and renal adaptation. It was developed for acute heart failure and looked highly promising after an early trial suggested a mortality benefit. A far larger definitive trial then found no benefit at all, and serelaxin was never approved anywhere. It is included here as an instructive example of how promising signals fail.
Mechanism
Relaxin-2 is a member of the insulin superfamily and shares its structural architecture: two peptide chains, A and B, cross-linked by two interchain disulphide bonds with a third bridge internal to the A chain. Despite this structural kinship with insulin it acts on entirely different receptors, principally RXFP1 (relaxin family peptide receptor 1), a leucine-rich repeat-containing G protein-coupled receptor. In normal human physiology relaxin-2 rises during pregnancy and mediates a set of striking maternal cardiovascular adaptations: systemic vasodilation, a fall in vascular resistance, increased arterial compliance, and a marked rise in cardiac output and renal blood flow with increased glomerular filtration.
RXFP1 activation raises intracellular cyclic AMP and triggers several downstream effects: upregulation of endothelial nitric oxide synthase, increased vascular endothelial growth factor expression, and activation of endothelin type B receptor signalling via matrix metalloproteinase-mediated processing of big endothelin. Relaxin also has documented antifibrotic activity, inhibiting transforming growth factor beta signalling and collagen deposition, and anti-inflammatory effects on the endothelium.
The therapeutic rationale for acute heart failure followed logically. Patients with acute decompensation have elevated filling pressures, high systemic vascular resistance and often worsening renal function. A vasodilator with concurrent renal and antifibrotic effects, which the body itself deploys to expand cardiac output during pregnancy, appeared an unusually well-matched intervention. That the logic was elegant and the mechanism genuine did not make the drug work.
What the research shows
The programme began encouragingly. Pre-RELAX-AHF, a phase 2b dose-finding trial, identified 30 micrograms per kilogram per day as the dose to carry forward and suggested improvement in dyspnoea. RELAX-AHF then randomised 1161 patients hospitalised with acute heart failure to a 48-hour infusion of serelaxin or placebo. It met one of its two primary dyspnoea endpoints, and produced a finding nobody had designed the trial to detect: a significant reduction in death from any cause at 180 days. A mortality benefit from a 48-hour infusion, appearing as a secondary endpoint, was biologically surprising and statistically fragile, but it generated enormous interest and drove regulatory submissions. Both the FDA and the EMA declined to approve serelaxin on this basis in 2014, correctly noting that a secondary endpoint in a single trial was not adequate evidence.
RELAX-AHF-2 was designed to settle the question and was very large: 6545 patients randomised to a 48-hour serelaxin infusion or placebo, with two co-primary endpoints, cardiovascular death at 180 days and worsening heart failure at 5 days. It found nothing. Cardiovascular death at 180 days occurred in 8.7 per cent of the serelaxin group and 8.9 per cent of the placebo group. Worsening heart failure at day 5 was not reduced either. A subsequent meta-analysis pooling the serelaxin trials was consistent with no meaningful treatment effect. Development was discontinued.
The episode is a textbook illustration of why regulators insist on prespecified primary endpoints and independent replication. The 180-day mortality signal in RELAX-AHF was almost certainly a chance finding in a secondary analysis, and a trial nearly six times larger showed it to be exactly that.
Evidence assessment
Mixed evidence
This tier requires explanation, because the evidence quality is high while the conclusion is negative. Serelaxin was tested in two large randomised placebo-controlled trials whose results directly conflict: RELAX-AHF in 1161 patients suggested improved dyspnoea and an unexpected reduction in 180-day mortality, whereas RELAX-AHF-2 in 6545 patients found no effect on either co-primary endpoint. Conflicting human trials is the defining feature of the mixed category. The correct practical conclusion, however, is not uncertainty but a negative one: the larger, definitive trial was null, a pooled meta-analysis was consistent with no effect, serelaxin holds no marketing authorisation anywhere, and its development was discontinued. Readers should not interpret "mixed" here as meaning the drug probably works. AUDIT NOTE: all four PMIDs verified against PubMed. The CAS number and molecular formula stated in the draft could not be independently confirmed and have been blanked; the two chain sequences were checked residue by residue against the known human relaxin-2 structure and are retained.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Relaxin for the treatment of patients with acute heart failure (Pre-RELAX-AHF): a multicentre, randomised, placebo-controlled, parallel-group, dose-finding phase IIb study Preclinical only
Identified 30 micrograms per kilogram per day as the dose to advance and suggested improvement in dyspnoea relief. Dose-finding rather than confirmatory, with endpoints not powered for clinical outcomes.
Serelaxin, recombinant human relaxin-2, for treatment of acute heart failure (RELAX-AHF): a randomised, placebo-controlled trial Preclinical only
Met one of two primary dyspnoea endpoints and reported an unexpected reduction in 180-day all-cause mortality as a secondary endpoint. This mortality signal was not a prespecified primary outcome and both the FDA and EMA declined approval on this basis in 2014.
Effects of Serelaxin in Patients with Acute Heart Failure Preclinical only
Definitively null. Cardiovascular death at 180 days occurred in 8.7 per cent with serelaxin versus 8.9 per cent with placebo, and worsening heart failure at day 5 was not reduced. Neither co-primary endpoint was met.
Safety
Serelaxin was generally well tolerated across a development programme encompassing well over seven thousand randomised patients, which is a substantial safety database even though the efficacy result was null. The pharmacologically predictable adverse effect is hypotension, reflecting the drug's vasodilator action; blood pressure reduction was consistently observed and required infusion rate adjustment or discontinuation in some patients. Because relaxin increases renal blood flow, effects on renal function were closely monitored and did not emerge as a safety concern. No unexpected serious toxicity signal was identified, and mortality was not increased. The practical safety point for readers is different in kind: serelaxin has never been approved by any regulator, was only ever given as a 48-hour intravenous infusion in a monitored hospital setting, and was never formulated for subcutaneous or self-administered use. Any product marketed as relaxin or serelaxin outside a clinical trial is unlicensed, has no established human dosing outside the hospital infusion context, and is of unverified identity and purity. Recombinant two-chain proteins of this size are also considerably harder to manufacture correctly than short synthetic peptides, making misidentification and incorrect disulphide folding realistic concerns in unregulated material.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Not approved. The European Medicines Agency adopted a negative opinion in 2014, and no United Kingdom marketing authorisation was ever granted by the MHRA. Serelaxin is not an available medicine in the United Kingdom. |
| United States | Not approved. The FDA declined approval for acute heart failure in 2014, and no approval followed the definitive RELAX-AHF-2 trial. Development has been discontinued and serelaxin has no United States marketing authorisation for any indication. |
| WADA (sport) | Not listed by name on the WADA Prohibited List. Relaxin-2 is not covered by the current prohibited hormone and growth factor classes, though athletes should note that section S0 (Non-Approved Substances) prohibits substances with no current regulatory approval for human therapeutic use, a category into which serelaxin now falls. |
Questions
It is at the boundary, and this entry flags that rather than glossing it. Serelaxin is recombinant human relaxin-2, a two-chain protein of 53 total residues held together by disulphide bonds, structurally in the same family as insulin. At around 5963 daltons it is usually described as a small protein rather than a peptide, and unlike the short synthetic peptides in this library it is produced by recombinant expression rather than chemical synthesis.
Because the promise rested on a secondary endpoint. RELAX-AHF found a reduction in 180-day mortality that the trial had not been designed to detect, from a 48-hour infusion, biologically surprising and statistically fragile. Both the FDA and EMA declined approval on that basis. RELAX-AHF-2, nearly six times larger, then found no effect on either co-primary endpoint. The original signal was almost certainly chance.
Yes, and this is worth separating from the drug's failure. Relaxin-2 is a genuine hormone that rises in pregnancy and drives real physiological adaptations (systemic vasodilation, increased cardiac output and increased renal blood flow) with documented antifibrotic effects. The biology is sound. What failed was the hypothesis that supplying it for 48 hours would improve outcomes in acute heart failure.
Not as a medicine. It holds no marketing authorisation in the United States, United Kingdom or European Union, and development was discontinued after RELAX-AHF-2. It was only ever administered as a hospital intravenous infusion under monitoring, and no subcutaneous or self-administered formulation exists. Anything sold as serelaxin is unlicensed material of unverified quality.