Virtual Symposium: Update on AMX0035 CENTAUR and open-label extension trials

Reading Time: 5 minutes

This blog is part of the ‘Virtual Highlights’ collection of articles, where you can read about the content of some of the talks and posters presented at the Virtual 31^st International Symposium on ALS/MND. This presentation was part of Session 3: Clinical Trials.

Presented by Dr Sabrina Paganoni, from the Sean M. Healey & AMG Center for ALS at Massachusetts General Hospital and Harvard Medical School, USA, this presentation discussed results spanning the CENTAUR and Open Label Extension trials of AMX0035.

Statement from the MND Association April 14, 2021:
News that Amylyx is preparing to conduct a Phase 3 clinical trial of AMX0035 in Europe with the intention to submit an application to the European Medicines Agency (EMA) to license the investigational drug is encouraging for the MND community.
The Phase 3 trial will last 48 weeks and will examine whether AMX0035, which is made up of two compounds, is safe and effective in extending the life of people with MND and what effect it has on the rate of progression – as measured by the Amyotrophic Lateral Sclerosis Functional Rating Scale (ALSFRS, the scale which measures various physical functions, for instance speech and movement, of a person with MND.
The trial is expected to begin enrolment in Europe in the summer of 2021. We are in regular contact with the team involved with this trial and look forward to sharing more information with the MND community as it becomes available.

The CENTAUR trial

We have written about AMX0035 and the CENTAUR trial in a previous blog, but here’s a quick recap. AMX0035 is a combination of sodium phenylbutyrate (PB) and taurursodiol (TURSO or TUDCA) which was designed to reduce neuronal death by modifying endoplasmic reticulum stress (involved in protein synthesis) and mitochondrial dysfunction. AMX0035 can be taken dissolved in water by mouth or via a feeding tube. It is important to note that AMX0035 is an investigational product, and its safety and efficacy have not been established by the FDA or any other regulatory authority.

The CENTAUR trial met its primary endpoint which was a slowing of functional decline measured by ALSFRS-R. The group that was given AMX0035 declined at a rate of 1.24 points per month compared to a decline of 1.66 points per month in the placebo group. This equated to a difference of 2.32 points at the end of the six-month study.

Review of a pre-specified overall survival analysis showed that those originally randomised to the AMX0035 group had a 44% lower risk of death, or 6.5 months longer average survival, than those originally randomised to the placebo group. It is worth noting that 77% of participants were taking riluzole or edaravone or both at study entry, but analysis showed that AMX0035 effect on survival was consistent regardless of use of these drugs. These results were published in 2020.

In terms of safety, nearly all participants (97% AMX0035, 96% placebo) reported one or more adverse events during the 24-week CENTAUR trial. Events that occurred with greater frequency in the AMX0035 group were primarily gastrointestinal adverse events.

All participants who completed the CENTAUR trial, whether in the treatment group or placebo group, were given the option to take part in an open-label extension (OLE), in which they were all offered AMX0035. 92% of eligible participants opted to enrol in the OLE, which is still ongoing.

Key questions raised

The publication of the overall survival analysis results raised some key questions by many people in the MND community.

Were the groups originally randomised to AMX0035 and placebo well matched?
CENTAUR participants were from a fast-progressing population. What is it the expected survival for this type of population and how does the observed survival compare?
Did the original placebo group, who received AMX0035 in the OLE (with a 24-week delayed start), show longer observed survival than predicted?

To answer these questions, the team partnered with leading TRICALS ALS researchers at UMC Utrecht in the Netherlands, who produced survival predictions for every participant in CENTAUR using the ENCALS Survival Prediction Model. This model predicts survival of patients with ALS based on baseline characteristics from over 11,000 patient records collected from across Europe. Each participant has a projected curve predicting the probability of survival at several time points after diagnosis, and these predictions were used to compare actual survival in CENTAUR to predicted survival. It is important to note that this model is based on a European population of people living with ALS and, while the model is expected to perform similarly in a US population, it has not yet been modelled for those in the US living with ALS and is considered an exploratory analysis.

So, to go back to the questions, these are answered in relation to the exploratory analysis of the CENTAUR overall survival data.

Were the groups originally randomised to AMX0035 and placebo well matched?

A TRICALS risk profile, based on the ENCALS Survival Prediction Model, showed that the original AMX0035 and the original placebo groups were well matched at the beginning of the CENTAUR trial.

Participants were from a fast-progressing population. What is the expected survival for this type of population and how does the observed survival compare?

Participants who received AMX0035 at baseline lived significantly longer than predicted by the ENCALS Survival Prediction Model.

Did the original placebo group, who received AMX0035 in the OLE (with a 24-week delayed start), show longer observed survival than predicted?

Yes. Participants who received placebo at baseline (the majority of whom crossed over to the OLE and received AMX0035) showed longer survival than predicted by the model.

The full results from this analysis are planned to be published this year.

Going forward

In pre-specified analyses, administration of AMX0035 is associated with both a statistically significant functional and survival benefit in people with ALS. An exploratory analysis using the ENCALS Survival Prediction Model shows a potential longer survival benefit for the groups originally randomised to AMX0035 and originally randomized to placebo that received AMX0035 in the OLE than seen in the pre-specified overall survival analysis. This prediction model analysis must still be viewed with caution as the ENCALS Survival Prediction Model is currently designed for a European, not American, population. However, the results from the randomised trial would seem promising, with the exploratory analysis further supporting these results.

We look forward to receiving more information about AMX0035 as a potential treatment for ALS in due course.

“We were thrilled to share the results from CENTAUR at the 31^st International Symposium on ALS/MND. The statistically significant results for the prespecified overall survival and functional analyses provide strong evidence for the benefit of AMX0035 in ALS. The exploratory natural history analysis using the ENCALS survival prediction model adds further information and may also be an invaluable tool for ALS clinicians and future trial designs.”
Dr Sabrina Paganoni

With thanks to Dr Sabrina Paganoni for allowing us to share her work with the wider MND community, and for her input into this blog article. Slides were shared with permission from the presenter.