BY CHARLES PILLER
These MRI images show brain swelling in the cerebral cortex—the outer large section of the brain—of a Florida woman who died after receiving the experimental Alzheimer’s drug lecanemab. Before treatment with the antibody, a scan of her cerebral cortex (left) reveals characteristic folds of the temporal lobes. Afterward (right), extreme swelling made those folds impossible to distinguish. PROVIDED ANONYMOUSLY TO SCIENCE
As enthusiasm mounts for a new experimental antibody that appears to slow cognitive decline in some Alzheimer’s patients, a third death linked to the drug during its clinical testing may amplify concerns about its safety. Science has obtained medical records showing a 79-year-old Florida woman participating in an ongoing trial of the antibody died in mid-September after experiencing extensive brain swelling and bleeding, as well as seizures. Multiple neuroscientists who reviewed the records at Science’s request believe her death was likely caused by the antibody, lecanemab.
“The brain swelling and the microhemorrhages … could be a serious side effect of the study medication,” and should be evaluated by trial investigators, says Ellis van Etten, a neuroscientist and neurologist at Leiden University.
The clinical trial’s sponsor, Japanese biotech Eisai Co., did not divulge the fatality at a major Alzheimer’s meeting last month where it detailed data from lecanemab’s phase 3 trial. The death came in an extension of that trial, but some scientists say it should have still been noted at the conference. “The failure of Eisai and [lecanemab codeveloper] Biogen to disclose this case … is concerning and undermines my confidence that the reported safety data is complete,” says Vanderbilt University neurologist and neuroscientist Matthew Schrag, who also reviewed the woman’s records.
The newly revealed death comes on top on other reports of serious brain bleeding and swelling in the core clinical trial and two other deaths in the extension phase—the first reported by STAT and the second by Science—that some scientists have linked to lecanemab.
Eisai, which attributed the prior fatalities and brain injuries to factors unrelated to lecanemab, declined to comment on the Florida woman’s death, citing patient privacy concerns. “All serious events, including fatalities, are reported to Eisai and considered in our evaluation of the study,” a company spokesperson said in a written statement to Science. “This information is provided to the FDA [Food and Drug Administration] and other regulatory authorities,” as well as independent review boards for the study.
The spokesperson added that the age and medical condition of any trial participants should be considered when evaluating a death. The Florida woman, however, had no obvious health problems, other than her signs of early Alzheimer’s disease, according to her medical records.
Eisai has reported 13 deaths in the core clinical trial, which involved about 1800 people. Deaths are expected given the study population’s age and health, and the company says the numbers were similar in the groups receiving lecanemab and the placebo. But it has not made public the details of each death, so in most cases scientists have been unable to independently assess whether lecanemab contributed to the fatalities.
Lecanemab is one of several experimental Alzheimer’s drugs that target beta amyloid, the protein that builds up in the brains of people with the disease. Many in the field believe it is responsible for the brain cell death that robs people of memories and ultimately kills them, although deposits of the protein are also found in brains of healthy people.
Amyloid-seeking antibodies often cause brain swelling and bleeding, a condition known as amyloid-related imaging abnormalities (ARIA) because it is diagnosed through brain imaging. “We need a name change … because these are not just imaging abnormalities, as this case illustrates,” says Boston University neurologist and neuroscientist Andreas Charidimou, who examined the woman’s records for Science. “It’s a real clinical syndrome, which can be fatal.”
Although lecanemab targets a soluble version of beta amyloid, it also binds to the extracellular beta amyloid “plaques,” considered a hallmark of Alzheimer’s. About half of Alzheimer’s patients have a condition called cerebral amyloid angiopathy (CAA), in which beta amyloid plaques replace the smooth muscle of blood vessel walls. When antibodies such as lecanemab strip away those plaques, blood vessels can become inflamed and weakened, increasing a person’s susceptibility to ARIA.
In the two previous deaths tied to lecanemab, neurologists say the patients’ use of anticoagulants worsened brain swelling and bleeding. The Florida woman was given a minimal course of the anticoagulant heparin after being hospitalized, but several neurologists discounted it as a contributing factor to her sudden problems and ultimate death.
Whether the woman received infusions of the antibody or a placebo during the core 18-month trial is unclear. But she did get the drug over 6 weeks in the extension phase—in which any participant can opt for treatment. Before the extension trial started, a brain scan revealed signs of a few microhemorrhages, but they were not serious enough to rule her out of the trial.
One of the Florida woman’s daughters provided the medical documents to Science and authorized their review by others. To protect the family’s privacy, Science is withholding the names of the patient, the daughter, a friend who served as the woman’s helper during the study, and the clinical trial site where the patient received lecanemab.
A textbook case
The woman’s friend described a harrowing series of events that began with the patient’s first infusion of the antibody as part of the extension trial, in August. “She was so tired. She … didn’t get out of bed for 2 days other than to maybe eat a yogurt or go to the bathroom,” the friend says. A couple of weeks later, after the second infusion, the woman complained of severe headaches, “couldn’t complete sentences,” and increasingly felt confused about everyday matters, her friend recalls.
At a restaurant on 14 September, the woman experienced what seemed like a stroke. She was rushed to the hospital, where her friend informed doctors that the woman was taking the experimental drug. Seizures began, causing her to thrash her arms and legs, requiring restraints for her safety.
Before a Florida woman received lecanemab in an extension phase of a clinical trial, an MRI scan of her brain (left) had a few microhemorrhages—tiny bleeds (dark spots, examples marked by arrows). Afterward (right), dozens of microhemorrhages were obvious (examples marked by arrows). PROVIDED ANONYMOUSLY TO SCIENCE
Brain scans showed dozens of areas of bleeding and brain swelling so extensive that the characteristic folds of the cerebral cortex were “merged and squashed” in substantial parts of her brain, Charidimou says. He calls it “a textbook case of severe ARIA, both the clinical presentation and the imaging manifestations.” Given the absence of other potential causes for brain damage indicated in the medical records, he adds, lecanemab almost certainly was the culprit.
The hospital records show the clinical trial site investigator, contacted after the woman was hospitalized, suspected ARIA and urged treatment with steroids—which the physicians tried without significant benefit. She began to suffer from multiorgan failure and pneumonia, and died 5 days after being admitted.
“The patient had extensive swelling of her brain with some small areas of bleeding which caused her to have a seizure and ultimately to die,” says Schrag, who is a CAA specialist. “I am confident this was a side effect from lecanemab.”
Eric Smith, a neurologist at the University of Calgary who also reviewed the case materials, agrees the drug likely caused the death. He previously consulted for Eisai partner Biogen and was an investigator for the two companies’ other antiamyloid drug aducanumab (marketed as Aduhelm), which won FDA marketing approval last year.
The family has arranged for an autopsy, which could confirm that the woman had CAA and clarify the antibody’s role in her death, but it has not been completed, the daughter says. The Eisai spokesperson said the company “is thorough and proactive” in its efforts to obtain any safety information, including autopsy results for trial participants.
A lack of consensus
The deaths linked to the antibody cast a pall over recent trial results largely seen as hopeful. Eisai has reported that lecanemab slowed the rate of cognitive decline among early Alzheimer’s patients by an average of 27% over 18 months, a statistically significant effect. Neurologists differ on whether that benefit would be noticeable to many patients or caregivers, and some large subgroups in the trial, including women and people under age 65, did not benefit to a statistically significant threshold.
Still, the trial represented the most favorable results for any antiamyloid therapy so far and has prompted calls from some Alzheimer’s scientists and patient groups for FDA to quickly green-light the drug.
Earlier this month, a lecanemab “consensus statement”—whose initial signers have worked as consultants to Eisai or Biogen or have conducted research for the recent lecanemab trial—began to circulate online. Nearly half of the more than 200 scientists or medical practitioners who had signed the statement as of 20 December are recent consultants or grantees of one or both companies, Science has determined. (Some but not all disclosed a conflict of interest.)
The document describes lecanemab as a “foundational gamechanger” for the illness and calls for its approval and “no barrier” to widespread availability of the antibody. It notes possible safety concerns associated with ARIA but does not mention the deaths and serious brain injuries some have linked to the drug. FDA is expected to decide on lecanemab’s approval, and whether to require any warnings or cautions for prescribers, by 6 January 2023.
Smith, who has not signed the pro-lecanemab letter, acknowledges that people with early Alzheimer’s might deem the possibility of even very modest cognitive benefits worth the risk of debilitating cases of ARIA or even fatal outcomes. But he thinks any FDA approval should come with warnings.
An Alzheimer’s patient receiving lecanemab would require as many as five MRIs annually to properly monitor for ARIA, he says, and a vast education program would be needed to ensure that doctors outside of major medical centers can recognize problems found in the brain scans. Smith also calls for FDA to require a registry that records ARIA-related problems if it approves lecanemab.
Eisai’s spokesperson said that if lecanemab is approved, the company would work with FDA to ensure that doctors and patients “understand how to monitor the patient for side effects, such as ARIA,” adding that “for many patients the benefits will outweigh the risks.”This story was supported by the Science Fund for Investigative Reporting.
Leave a Reply