Need to know:

  • Aduhelm is part of a class of antibodies that targets amyloid plaque build-up in the brain, which is thought to be implicated in the Alzheimer’s disease pathway.
  • Amyloid-targeting antibodies are known to cause brain swelling and bleeding, or ARIA, though symptoms typically resolve after lowering or stopping dosing.
  • While there is currently no evidence that symptomatically resolved ARIA is linked to long-term harm, some experts say there is insufficient public data on ARIA’s possible connection to cognitive decline and brain shrinkage to fully rule it out.
  • A less rigorous monitoring protocol in real-world settings compared to clinical trials could also make ARIA harder to detect in patients taking Aduhelm.

Seemingly incomplete public data on Biogen’s recently approved Aduhelm (aducanumab) has left some experts wary of the Alzheimer’s disease drug’s long-term safety.

Amyloid-related imaging abnormality (ARIA) is a known class side effect of antibodies like Aduhelm that target amyloid plaque in the brain. Its most common form, a type of cerebral edema called ARIA-E, can be associated with headache, confusion, nausea and gait disturbances, among other symptoms.

In the two Phase III trials backing Aduhelm’s approvalEMERGE and ENGAGE—35% of patients on the approved dose had ARIA-E, making it the most common adverse event. According to a Biogen presentation, however, 98% of ARIA-E cases resolved on MRI during the studies, and 76% of ARIA-E cases were asymptomatic.

But Biogen is yet to publicly release the more detailed data necessary to demonstrate that these ARIA cases don’t pose long-term threats, multiple experts said. Without published analyses linking ARIA cases to cognition and brain volume loss—or longer-term data beyond the two Phase III trials’ timeframe—concerns over ARIA may linger.

Biogen did not respond to a request for comment. 

How well do you really know your competitors?

Access the most comprehensive Company Profiles on the market, powered by GlobalData. Save hours of research. Gain competitive edge.

Company Profile – free sample

Thank you!

Your download email will arrive shortly

Not ready to buy yet? Download a free sample

We are confident about the unique quality of our Company Profiles. However, we want you to make the most beneficial decision for your business, so we offer a free sample that you can download by submitting the below form

By GlobalData
Visit our Privacy Policy for more information about our services, how we may use, process and share your personal data, including information of your rights in respect of your personal data and how you can unsubscribe from future marketing communications. Our services are intended for corporate subscribers and you warrant that the email address submitted is your corporate email address.

Though there is currently no evidence that resolved ARIA has long-term harmful consequences, further analysis of its impact on individual patients is warranted, said Yale University Alzheimer’s Disease Research Center director Dr Christopher van Dyck. As an Alzheimer’s disease clinical trial investigator, Van Dyck has treated around 300 patients with antiamyloid drugs—including Aduhelm—and has seen more than 50 cases of ARIA.

To gain more clarity, Biogen should perform posthoc analyses comparing the cognitive scores of patients with resolved ARIA-E cases to patients who never developed the side effect, Van Dyck said. If Biogen does not have the statistical power to perform the analysis with its own studies, a meta-analysis should be performed using additional data from other antiamyloid drug trials, he added.

Still, compounding factors, such as the presence of the APoE4 gene associated with a higher risk of ARIA, could complicate results of these analyses, van Dyck noted. In Aduhelm patients treated with the approved dose, the incidence of ARIA was 42% among APoE4 gene carriers compared to 20% in noncarriers.

Though Biogen released extensive data on the frequency and severity of ARIA-E in its two studies in a 26 July poster, there was no included data linking ARIA-E cases to changes in the measure of cognition used as the study primary endpoint.

But even with additional analyses of existing Aduhelm data, there is still uncertainty over the lasting implications of ARIA beyond the study’s timeframe, said University of Texas at San Antonio chair of neurobiology Dr George Perry. Without a placebo arm, ongoing extension studies are not sufficient to measure Aduhelm’s long-term safety and effect on cognition, Perry added.

The Aduhelm studies were placebo-controlled for 78 weeks, but patients can take the now-approved Aduhelm indefinitely. Biogen has nine years to complete a confirmatory trial as part of the accelerated approval process – which would include placebo-controlled data – but Biogen can market Aduhelm in the interim.

Multiple antiamyloid drugs are known to cause both brain volume loss and ARIA, so it is reasonable to expect they are related, said Dr Scott Ayton, head of the Translational Neurodegeneration laboratory at Melbourne, Australia’s Florey Institute of Neuroscience and Mental Health. But neither the FDA nor Biogen appear to have released any public data on brain volume loss during Aduhelm’s two Phase III trials, despite this being a prespecified endpoint, Ayton explained.

In the 343-page FDA Advisory Committee briefing document, brain volume change as measured by MRI at weeks 30 and 78 is listed as a prespecified pharmacodynamic endpoint. The document does not list any specific data for this outcome.

Without this data, it is difficult to tell if ARIA is tied to brain shrinkage or if this safety concern was ever extensively discussed during Aduhelm’s FDA approval process, Ayton said. It’s possible that antiamyloid drugs lead to some initial benefit but long-term neurodegeneration, Ayton and Perry said.

In addition to releasing brain volume data, Biogen should compare this data among those who developed ARIA and those who didn’t, determining whether brain shrinkage is accelerated in those with ARIA, Ayton said. However, brain shrinkage often presented after more than a year in the Phase II donanemab trial, so current placebo-controlled data, even if released, may not be sufficient, Ayton noted. As a result, future antiamyloid trials should prioritise measuring brain volume loss over an extended period, he said.

Even though Biogen may have trouble releasing brain volume loss data in peer-reviewed journals, this should not preclude the company from releasing the data in medRxiv, Ayton noted. MedRxiv, which publishes preprints of studies, had 12 articles referencing aducanumab as of 10 August but no data on brain volume loss.

Concerns over safety guidance, compliance

Though cases of ARIA were detected early on during Aduhelm’s Phase III trials, a less rigorous real-world monitoring protocol could make the side effect more difficult to catch, said Michigan State University bioethicist Dr Leonard Fleck. Still, the label’s dose titration guidance is a useful measure for lowering risk, van Dyck added.

During the two Phase III trials of Aduhelm, patients had seven MRIs during the studies’ 78-week placebo-controlled period to monitor for ARIA. Aduhelm’s FDA label calls for two MRIs to detect for ARIA – one prior to the seventh infusion at week 28 and one prior to the 12th infusion at week 48. Patients receive Aduhelm intravenously every four weeks, but do not receive the entirety of the approved 10mg/kg dose until their seventh infusion.

In addition to the two recommended MRIs in the label, van Dyck would advocate for a third MRI after four infusions, or 16 weeks. Almost all cases of ARIA occur during the first 8-12 months of treatment, making the timeframe for MRIs in the label reasonable, he noted.

But with a $56,000 price tag, and uncertainty over if and how Medicare and private insurers will cover the drug, any out-of-pocket costs associated with MRIs could hurt compliance, Fleck said. Getting patients to pay for additional MRIs, even if their prescribers recommend them, will be challenging if Medicare does not fully cover, he explained.