Two new studies show how new Alzheimer’s treatment Leqembi is faring in the real world: How safe is it, who’s taking it, and who has access?
New drugs designed to slow the progression of Alzheimer’s in the early stages, like Leqembi and Kisunla, might perform well in a carefully controlled clinical trial, but the results don’t always translate to the real world.
Since Leqembi’s approval by the U.S. Food and Drug Administration in 2023, patients and the medical community at large have been keeping a close eye on this next-generation, disease-modifying drug to see whether it is as safe and effective in the real world, where patients are less healthy, on average, than trial participants.
There’s also, as with any new drug on the market, the question of access. Many people living in regions with the highest prevalence of Alzheimer’s don’t have a dementia specialist nearby, which is necessary to get a diagnosis, and may need to travel hours to an infusion site to get the treatment.
Two new studies provide a glimpse into Leqembi’s usage in the United States during its first year on the market. Eli Lilly’s anti-amyloid drug Kisunla has also been approved by the FDA but hasn’t been on the market long enough for doctors to conduct a similar analysis.
Leqembi Vs. Kisunla: An Alzheimer’s Drug Point-by-Point Comparison
How Leqembi fared at a major Alzheimer’s clinic
The neurology clinic at Washington University in St. Louis, one of the largest in the country, has been prescribing Leqembi since shortly after it was approved. The neurologists there recently published a study in the medical journal JAMA Neurology, tracking 234 of their patients who had been prescribed the drug between August 2023 and October 2024.
They found that only 16 percent of patients eligible for the drug ultimately decided to take it.
“We really try to set expectations with the patients,” Dr. Barbara Joy Snider, a dementia specialist at the university and a senior author on the study, told Being Patient. Anti-amyloid drugs, like Leqembi and Kisunla, which work to clear toxic forms of beta-amyloid proteins from the brain, “are not going to improve your memory and thinking; the goal is to [stop] you from getting worse as fast.”
The results, she said, were “reassuring”: Despite patients being older and less healthy on average than the participants in Leqembi’s Phase 3 trial, the drug was still relatively safe and well tolerated.
A major downside of Leqembi and other similar drugs is the potential side effect of amyloid-related imaging abnormalities (ARIA) — brain swelling, or small brain bleeds that appear on brain scans. These events are, for the most part, asymptomatic. Snider and team found that ARIA rates were comparable to the rates in clinical trials, that they were most often mild or asymptomatic, and they typically occurred within the first six months.
“Most people would not know they had [ARIA] if we weren’t doing all these MRI scans,” Snider said of the drug-label-recommended MRIs to monitor patients during the course of treatment. If ARIA is spotted on a brain scan, and there are symptoms, neurologists slam the brakes on Leqembi, providing time for it to subside. Doctors carefully monitor these patients because, in rare cases, ARIA can be serious or fatal. A lifelong prevention plan
ARIA is more likely for a certain group of people: those who carry ApoE4, also known as the “Alzheimer’s gene.”
Everyone carries two copies of the ApoE gene. There are four different variants of it. The E4 variant, in particular, is linked to a notably higher risk of one day developing Alzheimer’s. And this risk is all the higher for people who are “ApoE4 homozygous” — who inherited a copy of E4 from each parent: While only one out of 50 people in the general population carry two copies, about one in seven people with an Alzheimer’s diagnosis carry two copies. This group is also roughly twice as likely to experience ARIA while taking Leqembi.
Would carrying two copies be more wary of taking Leqembi? That appears to be the case: Snider’s study found that about 8.5 percent of people on Leqembi are ApoE4 homozygous.
While the number of people taking Leqembi in the clinic is still small, Snider’s team noticed something surprising: People taking the drug who had mild cognitive impairment were less likely to develop ARIA than those in the earliest stages of Alzheimer’s.
The drug is intravenously infused every two weeks. Based on the safety data from the clinical trial, more people than the research team expected, 37 percent, had infusion-related reactions — fever, chills, and a headache — after the first or second infusion. In line with the drug’s prescribing information, Snider said it might be possible to reduce the risk by having patients take Tylenol and other antihistamines half an hour before the infusion.
All in all, only 23 out of the 234 people decided to stop treatment with Leqembi: 10 because of ARIA, and eight because of infusion-related reactions. Others stopped taking the drug for various reasons: unrelated health issues, anxiety over potential ARIA, inability to undergo MRI scans, the burden of biweekly treatment and monitoring, and one because their insurance stopped covering the drug.
But is the drug actually doing its job — slowing cognitive decline? It’s too early to tell, according to Snider.
The researchers only have six months of data for all the participants, since many started treatment at different points in time. Right now, the rate of decline, as expected, is on par with people who haven’t been on any medications for that time period. In the Leqembi trial, the researchers only saw cognitive decline slow after 1.5 years of treatment.
Who is taking Leqembi?
Now doctors have a rough idea of Leqembi’s safety in the real world, but who is taking the drug, and are people most at risk of Alzheimer’s able to access it?
A short study, funded by multiple grants from the National Institutes of Health, published in the medical journal JAMA Network Open in May 2025 looked at Leqembi uptake through Medicare data. In total, the study tracked usage among 1,725 people who started taking Leqembi on July 1, 2023 through March 31, 2024.
The average age across this group was 76 years old. More than 90 percent of patients were white, 88 percent were from urban areas. And, the researchers found, 1.3 percent were socioeconomically disadvantaged. This stood out because people who are most at risk of developing Alzheimer’s, Black and Hispanic Americans, those who are socioeconomically disadvantaged were underrepresented among the patients taking Leqembi.
“Unfortunately, with some of these more complex therapies [mABs] require infusion centers, lots of eligibility testing, monitoring, MRIs and so on, that it makes it much harder for certain people to access therapy,” Frank Zhou, a medical student at UCLA and author of the study, which was funded by multiple grants from the National Institutes of Health, told Being Patient.
While it is impossible to know why these disparities exist based on Medicare data alone, Zhou pointed to some potential factors. For example, patients may not live close enough to a doctor who can diagnose them. And doctors may be less likely to recommend treatment to Black and Hispanic patients, since they were underrepresented in the trials.
Another disparity was gender: While women are twice as likely as men to develop Alzheimer’s, only about 52 percent of people taking the drug were women. Because the study is based on Medicare data, the researchers are unable to say why.
Patients who are prescribed Leqembi are supposed to stay on the drug until they progress to the moderate stage of Alzheimer’s where it is no longer thought to be effective. But about one in four people in the Medicare database were not taking Leqembi by the end of March 2024. It isn’t clear whether they discontinued the treatment altogether or paused treatment temporarily for side effects, like brain swelling, to subside.
Snider, the neurologist at Washington University in St. Louis, thinks that blood tests and injectable forms of Leqembi, which may hit the market later this year, will make it easier for people to get a diagnosis and receive treatment at home.
She added that the healthcare system will need to train more dementia specialists to meet the challenge, help people navigate their cognitive impairment, and get the right diagnosis and treatment.
“We can use this drug in a specialty neurology clinic, and we can use it safely,” Snider said of Leqembi. “The results were very similar to the clinical trial, so it’s reassuring.”
