Summary
- AMLX got relyvrio approved through an interesting set of events.
- There are questions surrounding high price, low benefit, trial design, and so on.
- However, simply the high demand from a vocal and very sick patient population may well save the day for the company.
I covered Amylyx ( AMLX ) after relyvrio was approved, and I said that while the approval itself derisks the future to an extent, there are certain uncertainties that make me cautious. Some of these are the nature of the drug combo itself, which may cause generic and reimbursement problems, trial data, and so on.
I discussed the various flip flops that happened during the approval process of relyvrio. While interesting, most of those are now moot, except for one thing:
…the FDA had AMLX's CEO publicly pledge to withdraw the drug if the ongoing confirmatory phase 3 trial failed - which will top line sometime in 2024.
It is standard practice to withdraw drugs that fail confirmatory trials, but given the emotions surrounding this particular drug, I can see why the FDA would like such a pledge from the company. Of course, pledge or not, it is completely within the FDA's rights to withdraw approval. While the following relates to oncology drugs, it also applies to neuro drugs:
Q: What happens if these confirmatory trials don't verify clinical benefit?
A: Products approved under the accelerated approval program may be withdrawn voluntarily by the company or may be withdrawn by FDA after a public hearing.
But note that the CEO has modified that pledge, saying later, in essence, that the trial results should not be read in isolation to real life effects of the drug.
My previous article led to a very interesting conversation with someone who seems to be either an aggrieved shareholder (?) or an interested and knowledgeable party . They raised a few objections which I think are worth discussing:
First, they said that relyvrio does not have a lot of efficacy, and it neither helps with survival nor stops disease progression
Secondly, relyvrio consists of two well-known, widely available compounds - Sodium phenylbutyrate and ursodoxicoltaurine - and they said that these compounds are available on amazon at a low cost
Third, that relyvrio requires one to mix these two drugs in a fixed proportion with water, and one may as well simply buy the two components and do so at home
Fourth, insurers will not cover the drug, or, if they cannot stop coverage due to political reasons, they will at least make it difficult for patients to access the drug, and make the two OTC drugs easily available
Fifth, they mentioned the high drop off rates of the CENTAUR trial, and suggested that similar drop offs will be seen in real life as well, making it difficult for the company to stick to its revenue guidance
Lastly, they said the PHOENIX trial will not be able to show statistical significance just like CENTAUR did not except for the modified intent to treat population or mITT analysis.
These are logical points so let me try analyzing them one by one. Now, going back to the data from the CENTAUR trial, here's the key efficacy data :
There was a statistically significant difference in the rate of reduction in the ALSFRS-R total score from baseline to Week 24 in RELYVRIO-treated patients compared to placebo-treated patients (p = 0.034)
This is the mITT population of 87 patients in the drug arm and 48 patients in the placebo arm. Now, here's the ITT population:
A total of 137 patients were randomized 2:1 to receive either RELYVRIO (n=89) or placebo (n=48) for 24 weeks (Intent-to-Treat [ITT] population).
So where did the two additional patients go from the drug arm?
These two patients died, and therefore were not considered for analysis. The p-value anyway was pretty close to baseline, which in this case is 0.05. If those two patients were considered, then the p-value would be greater than 0.05, thus making the data statistically non-significant. One can argue that there's a solid reason to modify the ITT - the two patients died due to unrelated causes. However, even with mITT, the p-value is borderline. Thus, this is a problem. Relyvrio also does not improve survival considerably (5 months longer survival), it simply slows down disease progression to a small extent. As the data says:
In a post hoc, long-term survival analysis, vital status was ascertained in 136 of 137 patients who were enrolled in Study 1. Longer median overall survival was observed in the patients originally randomized to RELYVRIO compared to those originally randomized to placebo. This exploratory analysis should be interpreted cautiously given the limitations of data collected outside of a controlled study, which may be subject to confounding.
Now, as to the second point, it is true that relyvrio consists of pharmacological variations of those two well-known and easily available compounds. However, unlike that the argument claimed, relyvrio comes in a single sachet, not two separate ones. So, it is compounded. I have not seen any research that shows that the two versions are equivalent. I have read that while the approval was getting delayed, patients had been trying to compound the drug at home - with what success, I have not yet been able to figure out. Here's an interesting bit of anecdotal information though:
Oliff previously took what he calls "do-it-yourself" Relyvrio, by buying TUDCA online and having the drug's other chemical compounded at a local pharmacy. He said he believes the combination has helped stabilize his disease.
TUDCA is the colloquial name for one of the two compounds, and it is easily available on Amazon and elsewhere, and people already use it, along with radicava and generic riluzole. However, with Amylyx offering to help with the co-pay - which according to some estimates comes to around $7400 in certain plans - the high price tag is not a problem - for patients. It is, however, a problem for insurers like Cigna, who have broadly refused to cover relyvrio due to lack of clear efficacy, or, as they said in a statement , due to the drug being "experimental, investigational or unproven for any use."
Thus, not everything that the comments on my article said were wrong - indeed, quite a few of them were quite correct. However, there's something that the comments do not cover - patient desperation. As this article says:
"These are people who have a fatal disease," [Sabrina] Paganoni [a physician scientist at the Healey Center who also led that key clinical trial which supported Relyvrio's approval], said. "So anything that's new and can provide even a little bit more time with their families, even a little bit more time with good function, that's certainly something. It's really a matter of life or death."
For further details see:
Amylyx: Patient Advocacy, If Not Drug Merit, Will See It Through