A bad day for Molecular Partners

Amgen’s move to drop MP0310, one of four Molecular Partners darpin projects currently in clinical trials, completed a double whammy of bad news for the Swiss biotech yesterday.

Earlier in the day Novartis revealed that another of Molecular Partners’ clinical-stage darpins, the Covid antiviral ensovibep, might require a phase 3 trial to be carried out before being given US emergency use authorisation. Not only does this cast doubt over the viability of this asset, questions might again be raised over the entire darpin approach.

That had already taken a knock in 2020 when the FDA hit the most advanced darpin, abicipar, with a complete response letter for wet age-related macular degeneration. Last August Abbvie gave up on abicipar, having inherited rights to it via Allergan’s 2011 deal; however, the setback was put down to formulation problems, and Molecular Partners continues to list abicipar in its pipeline as a phase 3 asset.

Most promising darpin

After abicipar it was ensovibep that took on the mantle of the most promising darpin in development. Darpins are antibody mimetic proteins that incorporate several sequences, which can hit several binding sites, and while still injectable they are significantly smaller than MAbs.

For a while things looked highly optimistic for ensovibep: in January a hit in the phase 2 Empathy trial in Covid outpatients prompted Novartis to exercise an option on a full licensing deal. This was despite this space being largely controlled by Pfizer’s oral drug Paxlovid, and though ensovibep had failed in hospitalised Covid patients its darpin modality gave it pan-variant activity, the partners claimed.

An EUA request came shortly afterwards, based on the Empathy data, but it is this that has now been knocked back. According to Novartis’s first-quarter call yesterday the FDA might need more data before issuing ensovibep with an EUA; analysts say this means a phase 3 study with a hospitalisation or death endpoint, something that would be extremely difficult to carry out in the current Covid climate.

As such, one option might be to wait until autumn or winter, when Covid in the west might again rise in severity – while in the meantime the likes of Paxlovid become even more established. It is possible that the small number of events in Empathy, lack of dose response and lingering questions over toxicity caused the FDA to take its tough stance.

Novartis said it would now “make a sober evaluation” of ensovibep, and Molecular Partners’ Nasdaq-listed stock sank 18% yesterday. There was worse to come: after market close Amgen canned development of MP0310/AMG 506, a darpin bispecific hitting FAP and 4-1BB, and this morning the shares lost a further 37%.

No data

It is not clear what caused Amgen’s decision. MP0310 is in a phase 1 solid tumour trial, and as recently as last month Molecular Partners was promising to report data from this in the second half of this year; so far no results have been revealed.

The logic behind MP0310 is to localise it in the tumour, as FAP is a cell surface protein present on some cancers, before the co-stimulatory 4-1BB receptor prompts immune system activation. Thus a key advantage is to reduce off-tumour toxicity, and Amgen giving up on the molecule suggests that this has not played out in the clinic.

Molecular Partners’ fourth clinical-stage darpin, MP0317, also targets FAP, combining this with CD40 co-stimulation. Notably, Roche boasts no fewer than three clinical-stage anti-FAP assets, including traditional bispecific MAbs hitting FAP/4-1BB and FAP/CD40.

Perhaps this competitor situation played a part in Amgen’s move; if so it would serve as another black mark against the darpin approach.

This story has been amended to correct the definition of a darpin.