A fatal case of multiple brain hemorrhages associated with lecanemab

Today the FDA granted accelerated approval of the anti-amyloid monoclonal antibody lecanemab, now going by the trade name Leqembi. I have mixed feelings about this. Lecanemab is the only disease-modifying drug so far to show statistically significant slowing of cognitive impairment in subjects with mild Alzheimer’s disease, and it is very effective in removing beta-amyloid from the brain. This is exciting. On the other hand, while the slowing of cognitive impairment was statistically significant, it was very slight and might not be recognizable by patients or their families. It is quite possible that this drug may be more effective if started earlier, and in fact a trial in subjects with pre-symptomatic Alzheimer’s, the AHEAD Trial, is already underway. I am keeping my fingers crossed that the AHEAD Trial will ultimately show that lecanemab can meaningfully slow the onset of Alzheimer’s symptoms in these subjects who already have evidence of Alzheimer’s pathology in their brains.

However, there is a darker side to anti-amyloid monoclonal antibodies including lecanemab. ARIA, an acronym for amyloid-related imaging abnormalities, have occurred in all trials of anti-amyloid MABs.  They occur in two forms, swelling of the brain and microhemorrhages (small areas of bleeding). Most of the time they are harmless and without symptoms, and they usually resolve within a month or two after the drug is stopped. Rarely, they can be severe.  As described in a case study, I was one of the 2% of subjects in the aducanumab trial who had severe ARIA of both types, swelling and bleeding.  I required ICU care for two days, but I eventually fully recovered over the next few months.So far, three research subjects have died during the lecanemab clinical trial. Biogen has stated that the deaths were not caused receiving lecanumab, but a case study of one of these deaths was published this week in the New England Journal of Medicine and suggests otherwise. The patient was a 65-year-old woman who had received three doses of lecanemab as part of the open label extension phase of the trial. It is not known if she received active lecanemab or placebo during the preceding double-blind portion of the trial. She was seen in an emergency room 30 minutes after the sudden onset of inability to speak (expressive aphasia) and a left gaze preference.  A CT scan showed evidence of an acute ischemic stroke in the left temporal-parietal region and an occlusion of a distal branch of the left middle cerebral artery. There were no hemorrhages.  She received the clot-buster t-PA (tissue plasminogen activator) for treatment of her acute, ischemic stroke.  During the t-PA infusion, her condition worsened. An MRI showed multiple areas of new bleeding on both sides of her brain. She subsequently died, and at autopsy, she had extensive hemorrhages with swelling throughout her brain. She also had beta-amyloid deposition in the walls of small blood vessels of her brain as well as the typical amyloid plaques and neurofibrillary tangles of Alzheimer’s disease. This amyloid within the walls of brain blood vessels is called cerebral amyloid angiopathy (CAA), and it is one of the major causes of brain hemorrhages in the elderly.

Based on autopsy evidence, CAA is found in 48% of subjects with Alzheimer’s pathology in the brain. Unfortunately, there are limited ways to assess CAA during life. An indirect measure is based on the presence of lobar microhemorrhages seen on blood-sensitive MRI scans, but this is much less sensitive as these are found in only 22% of people with Alzheimer’s disease. The patient who died after receiving lecanemab along with the clot-buster t-PA had autopsy proven CAA, but she had no microhemorrhages at the beginning of the lecanemab study. I had no sign of microhemorrhages before my ARIA episode, but I almost certainly have CAA.

In my opinion, patients who have CAA should not receive anti-amyloid monoclonal antibodies like lecanumab or aducanumab, and they should be given clot-busting drugs like t-PA only with great caution. We really need more sensitive biomarkers for CAA. The presence of microhemorrhages on MRI is not good enough.