In my opinion, people with two copies of the APOE-4 allele should not be given anti-amyloid monoclonal antibodies

So far, three anti-amyloid monoclonal antibodies have been approved in the US by the FDA for treatment of dementia or mild cognitive impairment (MCI) caused by Alzheimer’s disease: aducanumab, lecanemab, and most recently donanemab. Aducanumab has been withdrawn from the market probably because it is less effective than the other two. Both lecanumab and donanemab are very effective at removing amyloid plaques from the brain, and both have modest benefit is slowing cognitive decline, especially when used by those with early disease. Side effects are usually but not always mild. The most problematic side effects are caused by small areas of swelling and/or bleeding in the brain called amyloid-related imaging abnormalities (ARIA). ARIA are likely caused by blood-brain barrier disruption and damage to the small blood vessels in the brain induced by the rapid clearance of amyloid. Most of the time, ARIA are associated with mild symptoms or no symptoms at all, and they usually resolve after stopping the anti-amyloid medication. However, in 1.5% of cases, ARIA are severe, sometimes resulting in hospitalization or even death. Seven cases of death associated with severe ARIA have now been reported. The most recent case was reported this week by J.N. Briard and colleagues:  Refractory status epilepticus in a patient with aducanumab-induced amyloid-related imaging abnormalities Neurology2024;103:e209582. doi:10.1212/WNL.0000000000209582. (Unfortunately, this paper is behind a paywall, but I can try to send a PDF to those who don’t have access.) All of the patients had two copies of the APOE-4 allele. The APOE-4 allele, especially when two copies are present, seems to be associated with a form of cerebral amyloid angiopathy in which amyloid protein is present in the smooth muscle of small blood vessels, making the blood vessels susceptible to leak or rupture when the amyloid is removed by treatment. 

It has been recommended that Alzheimer’s patients who like me have two copies of the APOE-4 allele receive extra monitoring to detect ARIA before symptoms occur. I don’t think that is adequate. I had severe ARIA resulting in two days of ICU care and six months of recovery (see figure below). My symptoms began after just three doses of aducanumab, well before a screening MRI would have been scheduled. In my opinion, the risk-benefit ratios do not support the use of anti-amyloid antibodies in the treatment of Alzheimer’s patients who have two copies of the APOE-4 allele.