Alnylam’s heart beats stronger
A win in Apollo-B should open up the multi-billion cardiomyopathy market, but questions remain.
The outcome of the keenly awaited Apollo-B trial of Alnylam’s Onpattro in amyloidosis with cardiomyopathy had looked tough to call. So it is no wonder that investors reacted with relief to news of a win, with shares in the company opening up 43% this morning.
While meeting the tricky primary endpoint, change in six-minute walk test, is undoubtedly a feat, details are thin on the ground. There will be plenty of questions for Alnylam to answer when it shares full data at the International Symposium on Amyloidosis on September 8 – not least how the placebo arm fared.
This is an important consideration because of the wildly different performances seen in the placebo groups of previous cardiomyopathy trials. Notably, Bridgebio’s Attribute-CM trial of acoramidis, which failed last year, showed minimal decline in 6MWT in both the treatment and placebo cohorts; the earlier successful Attr-act trial of Pfizer’s Vyndaqel found a 25m decline in 6MWT with therapy, but an even bigger drop with placebo.
If the placebo cohort in Apollo-B turns out to be more “Attribute-CM-like”, this would make Alnylam’s victory all the more impressive.
All Alnylam is saying for now is that Apollo-B hit the 12-month 6MWT endpoint with a p value of 0.016 – which even bullish Stifel analysts acknowledged was “close”.
The study also met its initial secondary endpoint, change in quality of life measured by the Kansas City cardiomyopathy questionnaire, with a p value of 0.04. However, it failed on the next secondary endpoint in the hierarchy to be tested, a composite of all-cause death, cardiovascular events and change from baseline in 6MWT.
The final secondary endpoints, which looked at a composite of death, hospitalisations and heart failure visits, were nominal failures, but Alnylam pointed out that these were not powered for statistical significance given the study's short duration.
As for safety, rates of serious adverse events were similar between the Onpattro and placebo arms. Side effects seen more often with Onpattro included infusion reactions, arthralgia and muscle spasms.
Alnylam plans to file a supplemental NDA for Onpattro in cardiomyopathy by the end of the year.
If the FDA gives Onpattro the nod here, it will open up a huge new pool of patients. The therapy is currently approved for the polyneuropathy subtype of ATTR amyloidosis, which affects around 50,000 worldwide. The cardiomyopathy subtype, meanwhile, is thought to affect more than 300,000 patients globally.
The only currently approved cardiomyopathy therapy is the aforementioned Vyndaqel. This is an oral transthyretin stabiliser, while Onpattro is an RNAi therapy designed to reduce production of transthyretin.
Vyndaqel brought in $2bn last year, and Alnylam will now want a piece of this massive market as it seeks to become profitable by 2025. However, the Pfizer drug has shown a benefit on deaths and hospitalisations, not just on 6MWT.
When asked during a conference call today how Alnylam might compete, its chief commercial officer, Tolga Tanguler, noted that Onpattro had positive data “across the full spectrum of ATTR amyloidosis” – perhaps a nod to the fact that Vyndaqel has never been approved in the polyneuropathy subtype in the US.
Mr Tanguler added that the amyloidosis market was growing, with increasing rates of diagnosis and treatment, and concluded: “This is not going to be a zero-sum game.” He was adamant that Alnylam would not promote Onpattro off-label before any decision in cardiomyopathy.
The drug might not have too long before its next rival comes along, but this looks likely to be an internal one: Alnylam expects data from the Helios-B study of its follow-on therapy Amvuttra (vutrisiran) in early 2024.
That study had always looked less risky, given that it is larger, longer, and is primarily evaluating a hard endpoint: mortality and cardiovascular events at 30 months. Given the result in Apollo-B, confidence in Helios-B will now only have increased.