Testing for Early Diagnosis of Alzheimer’s Disease - Episode 5
Dr Atri explains how aducanumab plays a role in diagnostic evaluations for patients with AD.
Alireza Atri, MD, PhD: If you’re sitting with a patient and considering aducanumab for them, there are multiple things to keep in mind. At this point, I’m not going to go into insurance coverage and costs. That’s another issue. CMS [Centers for Medicare & Medicaid Services] is considering a national coverage determination. That’s a separate issue. But if you think about who this would be appropriate for, there are a number of things to consider. I refer you to the appropriate use recommendations, by [Jeffrey] Cummings, [MD,] et al, I happen to be one of the authors, published in JPAD [The Journal of Prevention of Alzheimer's Disease]. It came out in the summer of 2021. It lists who can be considered in some ways and what would be appropriate use in clinical practice that reasonably follows with the data in the phase 3 ENGAGE and EMERGE trials.
There’s the FDA label, and we absolutely should refer to that. Nothing in the appropriate use is contrary; it falls within the FDA label indications. But compared with the appropriate use criteria, many colleagues and I feel that the FDA label is maybe a bit too liberal. The appropriate use has narrowed it and made it a little more restrictive. First, the patient has to be in the appropriate stage. This isn’t something that’s appropriate when there’s evidence for efficacy or even safety in individuals who are in the severe or moderate stages of Alzheimer dementia. It isn’t appropriate in clinical practice for individuals who are preclinical and don’t have symptoms. This is for individuals who should be in the mild cognitive impairment or mild dementia stages of Alzheimer disease. They must have the right clinical diagnosis and right stage.
Second, there should be evidence for amyloid positivity. In the clinical trials, this was done by amyloid PET [positron emission tomography]. In the appropriate use criteria, our recommendations are that it can be done by spinal fluid or amyloid PET. This is a drug based on a mechanism of removing amyloid plaques, so our clinical diagnosis isn’t enough. We have to show that the patient has the plaques in the brain.
A number of other things are important. One of them is for safety reasons. In the studies, individuals were excluded if at baseline they had moderate to severe amounts of white matter burden in the brain, multiple infarcts, and importantly if they had microhemorrhages above a certain amount. In the studies, this required individuals to have 4 or fewer microhemorrhages at baseline. This is important because individuals with Alzheimer disease at baseline can get microhemorrhages. It’s not uncommon: 10%, 15%, or 20% can have it, depending on the stage. This is often seen in a gradient-echo T2-star or SWI [susceptibility weighted imaging] sequence. Those are some of the important inclusion criteria.
If individuals can’t undergo MRI at this stage, I don’t think it’s appropriate use for them to get on the drug because we can’t monitor them with MRIs, and the adverse effect of ARIA [amyloid-related imaging abnormalities] as well. If they’re on blood thinners, if there are microhemorrhages, they may bleed more. At this point, the thought is to be a bit more restrictive. In the FDA label, there were no contraindications listed. This is a contrast. This is within the framework, but a bit more restrictive than that.
One of the things that the appropriate use recommendations talk about is that individuals should be presented with the expectations for this drug in a very patient-centered way. For example, it’s a monthly infusion and requires 1 hour of infusion. You would start on low doses at 1 mg/kg for 2 doses. After that, we give 2 more doses at 3 mg/kg. If everything continues to be fine, we go to 6 mg/kg. Ultimately, the dose achieved would be 10 mg/kg. But in between, there should be monitoring for symptoms of ARIA—we’ll talk a little more about ARIA—Including long-term MRI. It will require monitoring and infusions. It may come with potential costs. The expectations are important to understand that based on the data we have, on average individuals seems to be able to have a lot of plaque removed from their brain, and there may be a likelihood of benefit down the line of slowing decline. But this isn’t completely proven, and we don’t know how much we’re slowing decline.
Then the part that’s important is the safety, the amyloid-related abnormalities that can occur. It’s thought that these are due to the antibodies removing amyloid around the perivascular spaces. As they’re being removed, they accumulate. When they accumulate, they can cause an inflammatory response. When they do, you already have in Alzheimer disease leaky blood vessels, so they become more leaky. You could have areas of micro edema or frank swelling and effusions. From that, you get the ARIA-E, which is edema, or the ARIA-H, which are microhemorrhages.
In the studies for aducanumab, these were basically at the high dose of 10 mg/kg, for about 35% of individuals at some point, usually in the first 8 months as we’re escalating the doses, you could see ARIA radiographically. I liken this to having a little bruise on your hand. You get a bit of swelling. You may get some pain there. For most individuals, based on those monthly MRIs, they had no symptoms. About three-quarters of individuals who had ARIA anytime had no symptoms. For those who had symptoms, they were mild or moderate for the most part. It could have been feeling dizzy, having a headache, or having some confusion. In about 10% to 13% of individuals, they were severe. But serious adverse events during those trials only occurred in about 1 in 200 to 300, where it required a visit to the hospital, etc. This is on the background rate of about 2% to 3% of individuals accumulated over the 18-month portion of the study. Even on placebo, you have about 2% to 3% ARIA background. It’s mostly because of the microhemorrhages.
This transcript has been edited for clarity.