PHILADELPHIA — Ever since Eisai first revealed data from its pivotal Alzheimer’s disease study two years ago, the company has argued that treating patients with Leqembi earlier and longer would likely lead to more dramatic effects.
Now, the company claims it has new data to back that hypothesis. But the results unveiled Tuesday also raise questions about how the data in the study were analyzed, and if the results are really as strong as Eisai appears to be saying they are.
In a long-term follow-up of patients who got Leqembi for three years, Eisai said the drug kept working and maintained a similar slowing of cognitive decline as in its original 18-month study that was the basis of the drug’s approval by the FDA last year. And in a small sub-study of 41 patients in the earliest stages of the disease, Eisai said that half of them actually improved, rather than declined.
Eisai said the results bolster a core part of its scientific and commercial strategy: that treating patients earlier and for longer will generate the best results. It’s a view that’s in partial opposition to its biggest competition, Eli Lilly, whose own Alzheimer’s drug Kisunla was approved earlier this month. Lilly believes that patients should stop receiving amyloid-targeting drugs once plaques are cleared from the brain.
“Alzheimer’s is a chronic, progressive disease that needs to be continuously interrupted, just like cancer or arthritis,” chief clinical officer Lynn Kramer told Endpoints News in an interview. “Earlier treatment is better. It gives you a longer runway. And the benefit continues to increase.”
Making the case for the drug with doctors is crucial for Eisai and its partner Biogen. Sales of the drug have gotten off to a slow start, in part because of physicians’ doubts about the real-world clinical impact of the treatment, as well as barriers to access and concerns about safety, Endpoints has reported. Biogen and Eisai will report quarterly results in the coming days, giving a glimpse at how their rollout of the drug is proceeding.
There are several important limitations to Eisai’s study, which was presented at the Alzheimer’s Association International Conference in Philadelphia.
When Eisai’s 18-month clinical trial wrapped up, people who got a placebo were then offered the real drug. And so for the 36-month data, Eisai used historical data on how patients typically progress in the disease to make conclusions about the second half of the study.
“Making inferences from open-label observations is limited,” Lon Schneider, a professor of psychiatry, neurology, and gerontology at USC’s Keck School of Medicine, told Endpoints in an email. He also noted that 600 of 1,800 patients dropped out by the end of the three-year study. “They are not accounted for, but could be assumed as doing worse,” Schneider said.
There are also questions about how to interpret patients’ decline. Neither of the approved drugs stop or reverse the progression of Alzheimer’s; they merely slow it. In Eisai’s original study, patients on Leqembi progressed 0.45 points more slowly than those who got placebo, when measured on an 18-point scale of cognitive decline called the CDR-SB.
In the open-label extension, patients who got the drug for another 18 months had their disease slowed by 0.95 points when compared to a historical control group.
On its face, that seems like a consistent effect. But the study also shows that, during the second half of the study, patients got worse at a much faster rate than in the first half. In the first 18 months, Leqembi-treated patients progressed about 1.2 points. But after three years, they had progressed a total of 3.09 points.
Eisai estimated that without the drug, patients would have progressed 4.04 points, but without a placebo group to compare the patients to, it’s hard to be certain — especially when talking about the relatively small effect of the drugs on the disease.
And by another calculation Eisai previously conducted, the results are even less dramatic. A 2023 study co-authored by Eisai that modeled the potential effects of the drug beyond 18 months showed that the company thought Leqembi recipients would only worsen by about two points on the CDR-SB over three years, rather than the three points seen in real life.
At the conference, Eisai scientists presented other models that aligned with the company’s results. Kramer told Endpoints that the 2023 study modeled different kinds of patients, and said that the results for the three-year study were exciting and consistent with the company’s expectations.