Berobenatide: The Monthly GLP-1 Candidate

Where the description comes from

It is worth being explicit about the source of the phrase used to describe this molecule, because it is not a journalist's characterisation. The official title of VESPER-4, as registered, reads: 'Evaluating The Efficacy and Safety of MET097, a Fully-Biased, Ultra Long-Acting GLP-1RA, In People With Overweight or Obesity.' That is the sponsor's own description in a registered protocol title. It tells you what the molecule was designed to be. It does not, on its own, tell you that the design succeeded, and a registered title is not peer-reviewed evidence of receptor pharmacology.

What 'fully biased' is claiming

A GLP-1 receptor does two broadly separable things after an agonist binds: it couples to G proteins, raising cyclic AMP, and it recruits beta-arrestin, which contributes to receptor internalisation and desensitisation. An agonist that engages the first pathway strongly while recruiting little beta-arrestin is called G-protein biased. The rationale is that reduced internalisation leaves more receptor at the surface for longer, so signalling is sustained rather than blunted by desensitisation. Applied to a molecule intended for monthly dosing, that matters more than usual: a drug that has to hold an effect for four weeks between injections cannot afford to be desensitising its own target.

What the clinical data actually shows

VESPER-3 (NCT06973720) randomised 268 participants to one of four regimens of twelve weekly doses, with or without titration, followed by monthly doses, or to placebo on the same schedule. The primary endpoint was percent change in body weight at week 28. Reported topline gave placebo-adjusted weight loss of up to 12.3% on the efficacy estimand with a 4.8 mg monthly dose, and 10.5% for that arm on the treatment-policy estimand; another arm reported 10% and 8.4% on the same pair. Separately, VESPER-1's Part B extension reported 15.9% mean weight loss at 32 weeks on the top weekly dose, which is not placebo-adjusted and is not comparable with the figures above.

The tolerability question monthly dosing raises

Weekly incretin dosing lets clinicians titrate slowly, because the interval between dose increases is short. A monthly schedule cannot titrate the same way, and a monthly injection cannot be un-given if it is poorly tolerated. That is why the VESPER-3 design puts twelve weekly doses in front of the monthly phase: the titration happens on the weekly schedule, and the monthly dose is only reached once tolerance is established. Reported gastrointestinal events were predominantly mild to moderate, with no more than one instance of severe nausea or vomiting and no severe diarrhoea reported across the two arms described in the February 2026 topline. Discontinuation for adverse events was five participants during the weekly phase and five during the monthly phase across those arms combined, against zero in placebo.

How it compares with what is licensed

It does not, yet, in any meaningful sense. Semaglutide and tirzepatide have pivotal Phase 3 programmes, cardiovascular outcome data in the case of semaglutide, and years of post-marketing exposure. Berobenatide has Phase 2b data in a few hundred people and Phase 3 trials whose first primary completion is September 2027. A placebo-adjusted 12.3% at week 28 in 268 people is an encouraging Phase 2b result; it is not evidence that can be lined up against a licensed medicine's label.

What this means for anyone reading about it here

Berobenatide is an investigational compound. It is not licensed, not approved, and not supplied on this site in any form. It is covered because the engineering question behind it - whether a GLP-1 effect can be held for a month - is the most interesting unsolved problem in the class, and because the compound has had three names in two years, which makes it unusually easy to misattribute data to.

Quick reference