The Aduhelm Approval
The recent FDA ruling on a new drug for Alzheimer Disease has implications for all of us.
Unfortunately, many of us know someone with dementia. Maybe this is your parent, your partner’s parent, a distant relative, or a neighbor. Maybe you have memories of intimate conversations and outings together. But now, their ‘conversations’ with you are non-sensical; they need help preparing their meals and navigating their own home; sometimes they’re suddenly upset with you and ask you to leave the room because they’re not even sure they know who you are; and sometimes they’re incontinent and need help from others to change their soiled clothing. This is what it means to have dementia.
Alzheimer Disease (AD) is the most common type of dementia, accounting for 60-75% of dementia diagnoses worldwide. Today, we don’t have any drugs that can stop the disease, much less reverse or prevent it from happening altogether. To treat AD means managing the symptoms with a combination of ineffective drugs, nurses and home health aides, and asking children and spouses of AD patients to sacrifice their livelihood to become caretakers. According to a recent study in the American Journal of Managed Care, Medicare spent an estimated $155 billion on AD in 2020, while patients and their families spent $66 billion out of their own personal bank accounts. To take another perspective on cost of care - in 2019, 16.3 million spouses, children, and other relatives provided an estimated 18.6 billion hours of unpaid care. If you value the cost of that labor at $13 per hour, that’s $244 billion of informal care provided to AD patients.
Preventing and/or stopping the progression of AD is one of the true “holy grails” in Medicine. Over two dozen pharmaceutical companies have a drug in phase 1, 2, or 3 trials currently. The Alzheimer’s Association, the largest patient advocacy group for the disease, has almost $300M in total assets for education and research programs (for comparison, the largest breast cancer education and research program, Susan Komen, has $200M). To say the least, many patients, their families, their doctors, and society at large have been waiting for something - anything - that could reduce the impact of AD on all of us.
And, so, it was no small affair recently when the FDA approved the first treatment for AD since 2003. In fact, aducanumab (Aduhelm), is the first approved drug to target what many believe to be the cause of the disease. Many are hopeful that we’ve now found the grail.
The amyloid hypothesis
For the last 30 years, most of the AD world has been studying the “amyloid hypothesis”, believing this explained the disease and would generate treatments. The hypothesis goes something like this: all of us have amyloid proteins in our brain. They exist as by-products of amyloid-precursor proteins (APP), which facilitate basic cellular activities that allow different cells in the brain to communicate with one another. These APPs often need to be recycled for housekeeping purposes. Most of us do this housekeeping without any problem, but some of us don’t recycle these APPs appropriately and end up generating beta-amyloid proteins that accumulate in the brain as “plaque”. The build up of amyloid plaque is believed to slow down and/or kill brain cells, leading to what we call Alzheimer Disease. So, if too much amyloid is bad, let’s find a way to target the amyloid proteins and get rid of them. And that’s exactly what Biogen and many others have been working on for the past several years.
A tale of 2 clinical trials
After a long and complicated process, Biogen and Aduhelm crossed the FDA’s finish line on June 6th. But not everyone celebrated the achievement. In fact, three people from the FDA’s own advisory committee that reviewed the Biogen data resigned in protest of the approval. Why would they do that? Because the advisory committee effectively recommended AGAINST the FDA approving the drug.
To understand this controversy and the conflicting views between the FDA and its own advisory committee, let’s talk a bit more about AD and look briefly at EMERGE and ENGAGE - the two phase 3 double-blind, randomized, placebo-controlled clinical trials (DBRPCT - a.k.a “the gold standard”). These trials, with 1600 patients each, generated the Aduhelm data submitted to the FDA for review.
When neurologists talk about patients with AD, they don’t start with amyloid or any other protein (testing for amyloid isn’t a great way to track AD, as we’ll discuss below. And it involves doing spinal taps / lumbar punctures!) Instead, they usually talk about “scores” from clinical surveys like “MMSE” or Mini Mental State Exam, the CDR-SB, or the ADAS-Cog13. With the MMSE, the most commonly used survey, a neurologist asks a patient a set of standardized questions ranging from “what is today’s date?” to “where are we right now?” and tells them to do tasks like “count backwards from 100 by 7” and “copy the shapes you see on this paper”. A score of 24-30 is considered “normal”, 19-23 is mild dementia, 10-18 - moderate, and less than 10 is severe dementia. A similar scoring system is used with the CDR-SB and other surveys. To be sure, these surveys aren’t always the only tool used to make an AD diagnosis. Sometimes, if the diagnosis is unclear, neurologists will use specialized MRI scans of the brain to help support or refute what they see on the MMSE or other surveys.
EMERGE and ENGAGE both asked the question: “can Aduhelm prevent people with mild dementia from progressing to moderate or severe dementia?” They answered that question with all of the surveys listed above, but relied most heavily on the CDR-SB.
So what’d they find? In short, ENGAGE said “no” and EMERGE said “yes”. The full story on this is long and winding one with a lot of biostatistics. (Note: no peer-reviewed publication of the data is available yet; all data used by the FDA and others commenting on it are from the Biogen slides referenced in the previous link).
On the surface, the results paint a 50-50-coin-flip like scenario for Aduhelm helping AD patients. However, when the FDA’s advisory committee met to review Aduhelm, they considered not only whether it improved CDR-SB scores (EMERGE showed a 22% improvement), but also the severity of side effects seen in EMERGE and ENGAGE (roughly 1 in 10 people treated with Aduhelm had symptoms from brain swelling and/or bleeding!).
They also considered how these two trials compared to the previous 25 clinical trials of amyloid-targeted drugs that failed to show any improvement in or slower progression of AD. Taken altogether, the advisory committee didn’t feel confident that the one positive result from EMERGE showed that Aduhelm is a safe and effective drug.
Surrogate outcomes & ‘Accelerated Approval’
So how did the FDA justify its Aduhelm approval? While Biogen’s EMERGE and ENGAGE looked at CDR-SB scores clinical outcomes, they also indirectly measured amyloid protein in the brain (via specialized PET scans) as a “surrogate outcome”. The role of brain amyloid as a surrogate outcome is the key to understanding FDA’s thinking that led to the approval - and the controversy surrounding it. In situations where measuring a clinical outcome would take an unacceptably long time or be logistically or financially limited, surrogate outcomes enable an “Accelerated Approval” process. In the FDA’s own words,
“A surrogate endpoint is a marker, such as a laboratory measurement, radiographic image, physical sign or other measure that is thought to predict clinical benefit, but is not itself a measure of clinical benefit. The use of a surrogate endpoint can considerably shorten the time required prior to receiving FDA approval.”
While the clinical outcomes (i.e. CDR-SB scores) are highly debatable, the reduction in brain amyloid on PET scans was less debatable. The FDA agreed with Biogen that “treatment with Aduhelm was clearly shown in all trials to substantially reduce amyloid beta plaques. This reduction in plaques is reasonably likely to result in clinical benefit.”
So, in amyloid plaque reduction we trust. But should we?
“The amyloid hypothesis is dead”
“The amyloid hypothesis has been the mainstream concept underlying AD research for over 20 years….all attempts to develop [amyloid]-targeting drugs to treat AD have ended in failure and recent findings indicating that the main factor underlying the development and progression of AD is…not amyloid”
That was the opinion of two prominent Japanese AD researchers in 2018. Interestingly, a few month later, in early 2019, Biogen halted the 2 clinical trials discussed above were halted early because their own data showed Aduhelm wasn’t working
While there is evidence supporting the amyloid hypothesis, there’s also a lot of evidence refuting it. The level of amyloid in the brain does not always correlate with the severity of symptoms in AD patients. Sometimes AD patients have little to no evidence of amyloid plaques, and sometimes PET MRI scans show a lot of amyloid plaque in patients with very mild symptoms. (If you’re interested in other hypotheses, check out the “tau hypothesis”).
The neuroscience community hasn’t completely buried the amyloid hypothesis yet. Some believe that amyloid drugs haven’t worked yet because we’ve used them too late in the game. They think we should try to find people at high risk of developing AD and give them anti-amyloid drugs before they even show signs or symptoms of AD. But after seeing the Aduhelm data, many more believe Biogen and everyone else working on amyloid drugs are wasting their money, and, most importantly, wasting the time of AD patients and their families and loved ones. They should abandon amyloid and explore other drug targets.
What happens next?
Biogen and Aduhelm aren’t off the hook yet. Yes, Aduhelm is now available on the market like any other FDA-approved drug. But, per the Accelerated Approval process, Biogen owes the FDA more ‘real world data’. The FDA has publicly said that Biogen has until 2029 to collect more data and present more results supporting the drug’s effectiveness. But, this time the data isn’t coming from a clinical trial, but rather from neurologists and patients who choose to use the drug per their own discretion and shared decision-making.
So, sometime in 2030, we’ll hopefully know better whether or not Aduhelm is worth it or not. What happens between now and then? And where does that leave all of us?
Hot Takes
Alzheimer patients and their significant others
I’m not sure what to tell you. AD is devastating and I’m glad there’s now a treatment on the market that *might* work. But, please keep in mind:
Biogen is asking for $56K per year of treatment. It’s unclear how Medicare and private payors will choose to cover Aduhelm. I doubt they’re going to pay Biogen’s list price given the drug’s questionable track record so far. But the social and political pressure on CMS to cover at least some of the cost is going to be hard to ignore. Ultimately, my worry for the hopeful if not desperate people in search of a neurologist who will prescribe Aduhelm is that they might be ignoring where that money could go if it weren’t spent on the lottery ticket that is this drug.
Taking Aduhelm means getting special brain MRIs every few months to monitor for brain swelling and/or bleeding. As with Aduhelm itself, it’s unclear how CMS and other payors will think about covering the MRIs and other monitoring. And, more importantly, it’s unclear if more, the same, or fewer patients will experience brain swelling than seen in the clinical trials.
Your money might be better spent paying for an assisted living facility or home health aide to help you deal with your symptoms. If insurance won’t cover the drug and Biogen won’t provide financial assistance, a better option might be to pay for skilled labor.
There are other drugs in development right now. If you believe the amyloid hypothesis still, it might be worth waiting to see what happens with Eli Lilly’s donanemab. And if you think amyloid is a red herring, then wait for Roche’s anti-tau protein drug. If the FDA approves either or both, that could create competition and force Biogen to lower Aduhelm’s asking price. No guarantees here, but if you can wait, it would make sense to do so for a bit.
And if you are determined to get your hands on Aduhelm, I would highly recommend you work with a neurologist specializing in AD. I didn’t mention this earlier, but EMERGE and ENGAGE found that: 1) higher doses of the drug led to better outcomes, 2) higher doses were associated with more brain swelling and/or bleeding, 3) genetic testing for certain APOE4 mutations could predict if certain people are at higher risk for bleeding with the higher doses. This drug is complicated, so you might want a neurologist how understands the complexity and can walk you through it all every step of the way. I would recommend going to an academic medical center for Aduhelm, not your nearest neighborhood neurologist.
Neurologists
Good luck managing the expectations of patients and their families. As I mentioned above, not all dementia is AD. Other causes and forms of dementia exist, and Aduhelm won’t work for them.
Good luck selecting patients who might benefit from Aduhelm. Even if you do have legitimate AD, the clinical trials tested patients with the mildest forms of AD. Are you going to recommend Aduhelm to anyone who shows even a single symptom of AD? Are you going to genetically test all these people? How are you going to tell people with severe AD that you don’t think the drug will work for them? These conversations with patients will not be easy nor straight-forward.
Medicare and Private Insurance Companies
You now have the unenviable task of choosing to either pay for some or all of this still-unproven drug, OR denying coverage to some patients who actually may benefit from it.
I think insurers will ultimately do what they always do: pass the costs down to the rest of us, either through higher premiums, co-pays, or taxes.
Biogen and Biogen’s Investors
Congratulations: your drug is FDA-approved for at least the next 8 years!
Through the Accelerated Approval process, you’ve now shifted some of the cost of drug development onto CMS, private payors, and patients. As neurologists prescribe Aduhelm over the next few years, you’ll be collecting sales revenue from some combination of those people. Even if this thing doesn’t work out, it wasn’t a complete loss!
But you’re still on the hook for collecting more data and preparing another report for the FDA. How are you going to do so? How are you going to find which neurologists are prescribing Aduhelm? How are you going to collect CDR-SB or MMSE survey scores from them?
Other Pharmaceutical Companies
Biogen just gave Eli Lilly, Roche, and several others a playbook on how to use surrogate endpoints and the Accelerated Approval process.
The FDA
You might have just put your credibility on the line with this approval. On one hand, if the 2029 results show that Aduhelm works, you’ll be celebrated for having given more patients access to an amazing drug. If it doesn’t work or there are reports of more brain swelling and bleeding than previously reported in EMERGE and ENGAGE, then you’ll be vilified for allowing a dangerous, unproven drug into the public market. And against the recommendations of your very own advisory committee no less.
It’ll be interesting to see if the FDA allows Eli Lilly, Roche and others to follow the Biogen precedent. There’s going to be a lot of media attention and probably some Congressional hearings on all of this. Let’s see how the dust settles.