Finally, a Winner for Alzheimer's? Anti-amyloid Agent Shows Promise

Pauline Anderson

July 26, 2018

CHICAGO — Positive results for an anti-amyloid agent in patients with early-stage Alzheimer's disease (AD) is drawing praise, but experts are calling for caution.

Results from the new phase 2 study showed a statistically significant reduction in brain amyloid with a high dose of BAN2401 (Eisai Co. Ltd/Biogen Inc) at 18 months.

In addition, the study showed a dose-dependent, statistically significant, and clinically meaningful slower decline in cognition and function with the highest dose compared to placebo.

"This is the first large clinical trial to support the amyloid hypothesis," Lynn D. Kramer, MD, chief clinical officer and chief medical officer, Neurology Business Group, Eisai Co, Inc, told a press briefing.

The study was released here at the Alzheimer's Association International Conference (AAIC) 2018.

Unique Trial Design

The multicenter study included 856 patients with early AD (mild cognitive impairment due to AD or mild AD dementia) and amyloid pathology confirmed by positron-emission tomography (PET) or cerebral spinal fluid (CSF) tracer.

At baseline, the mean Mini‒Mental State Examination score was 25.6; the mean Clinical Dementia Rating Scale‒Sum of Boxes (CDR-SB) score was 2.9, and the mean Alzheimer's Disease Assessment Scale–Cognitive Subscale (ADAS-Cog) score was 22.2.

Patients were randomly assigned to six groups. One group received placebo, and each of the other groups received one of five doses of the experimental drug BAN2401. Doses ranged from 2.5 mg/kg biweekly to 10 mg/kg biweekly.

BAN2401 is a humanized monoclonal antibody that selectively binds to large, soluble aggregated amyloid beta (Aβ) protofibrils. The antibody was developed following the discovery of a mutation in amyloid precursor protein that leads to a form of AD that is marked by particularly high levels of Aβ protofibrils.

The study's design included two features not typically used in trials of AD dementia.

One was use of a complicated Bayesian statistical analysis. That analysis involves use of a computer model to assess data and probabilities of the most meaningful drug doses and allocates patients — in this case, starting with patient 197 — accordingly.

The approach can expedite and enhance a clinical trial and minimize the number of patients who receive ineffective doses, which, from a statistical point of view, is "a waste," said Kramer.

"Once you can decide those are ineffective doses, your trial efficiency is improved by having patients on doses that you might give to patients and that might show benefit," he said.

This approach predicted that the 10-mg/kg biweekly dose was the best, followed by the 10-mg/kg monthly dose.

Of the 856 patients in the trial, 414 received the top two doses — 161 received the highest dose, and 253 received the second highest (10-mg/kg monthly). There were 247 patients in the placebo group; the remaining patients received other doses of the active drug.

Primary Endpoint a High Bar

Kramer emphasized that all participants received the dose to which they had been initially randomized to receive for the duration of the study. They all received biweekly infusions, so, for example, if patients were taking a once-monthly dose of 5 mg, they would receive a placebo for the biweekly dose. This, he noted, helped ensure blinding.

In addition, for the first time, the study used ADCOMS (Alzheimer's Disease Composite Score), a composite of scores on existing outcome scales that is particularly sensitive to changes in function and cognition in patients with early-stage AD.

Results of the 12-month primary interim analysis using ADCOMS, which were released in December 2017, were negative.

"We prespecified that it had to have an 80% likelihood of being superior to placebo by 25% or more," said Kramer. "It was a very high hurdle."

The analysis found a 64% probability of being superior after 12 months. So although it was "relatively close," said Kramer, it failed to meet the prespecified outcome.

The company has now reported secondary outcomes, including 18-month changes on a number of cognitive scales.

One of these outcomes was amyloid burden. Here, the study found that the drug significantly reduced amyloid PET values (for adjusted mean change, P < .0001), "with a nice dose response," commented Kramer.

There were similar results with the Centiloid scale (a standardized measure of amyloid imaging plaque load). On this measure, the mean was 74.5 at baseline and 55 at 18 months, representing about a 93% reduction, said Kramer.

"There is insoluble plaque and soluble fibrils around that plaque," he explained. "The thought is that as you pull the soluble portions out of plaque, you are reducing that overall brain amyloid load."

There were also improvement of cognition and function at 18 months, as measured by various scales.

One of these measurements was ADCOMS. On this scale, at 6 months, the highest dose began to separate from placebo. This continued through the trial, reaching a 30% difference from placebo at 18 months (P = .034).

Robust Results

With regard to more traditional endpoints, there was a 47% difference in decline with the highest dose compared to placebo on the ADAS-Cog (P = .17).

An illustrative graph showed that the line for the highest dose was superior to that for the projected 25% less decline vs placebo, starting at 6 months.

There was a 5-point decline on ADAS-Cog in the placebo group compared to about a 2-point decline in the highest-dose treatment group, for a difference of about 3 points.

On the CDR-SB, there was a 26% reduction vs placebo, but this was not statistically significant.

Kramer noted that 65% of patients who received the highest dose were found to have converted to amyloid-negative status at 12 months (P < .0001); at 18 months, 81% were found to have converted (P < .0001).

The incidence of serious adverse events (AEs) and treatment-emergent AEs (TEAEs), including death, were similar across all groups. The most common TEAEs were infusion reactions and amyloid-related imaging abnormalities (ARIAs).

ARIA-edema occurred in 9.9% of patients in the highest-dose group and in an even higher percentage (14.6%) in those patients with APOE4 who were taking this dose. Most ARIA-edema cases occurred within the first 3 months of treatment and were asymptomatic.

Early in the study, regulatory authorities advised researchers not to administer the highest dose of the drug to patients who were APOE positive. This helps explain why the second-highest dose was larger.

There were no changes in ECG findings, laboratory results, and vital signs, said Kramer.

Still to come are detailed results for CSF biomarkers as well as subgroup analyses, including one comparing APOE4 genetic status.

Kramer described the results as "fairly conclusive."

"The results are certainly robust enough to approach regulatory authorities to discuss next steps in terms of what we might do to try to register this product, in terms of additional trials, and discussions of things like breakthrough status," he said.

Finally, a Winner?

A number of AD experts commented on the new research. Maria Carrillo, PhD, chief science officer, Alzheimer's Association, said it's "intriguing to think about what these results might mean."

She noted that this is the second agent that has been shown through PET imaging and other markers to clear amyloid in the brain. The other agent is aducanumab (under development by Biogen).

"It also has hints of some efficacy in cognitive tests," she said.

As with research to date on aducanumab, the BAN2401 trial was not large enough to definitely demonstrate cognitive efficacy.

Carrillo added that it's important that the study was released at the AAIC, where almost 6000 researchers and scientists had the opportunity to discuss the results.

"What we have been using for the last 15 years has not generated positive results, so it's important for people in the Alzheimer's and related dementia field to think more broadly about what we can do to get those more efficacious medicines to our loved ones as soon as possible," she said.

David Knopman, MD, professor of neurology, Mayo Clinic, Rochester, Minnesota, who is a member of the Alzheimer's Association Medical and Scientific Advisory Council, also found the results promising.

"After a number of failures of similar drugs and other anti-amyloid therapies over the last several years, to see a movement in the biomarker and a clinical outcome, however modest, for a number of subjects is encouraging," he said.

Julie A. Schneider, MD, Alzheimer's Association Medical and Scientific Advisory Council and associate director of Rush Alzheimer's Disease Center, Chicago, Illinois, said that although the study results hold "a lot of intrigue," many questions remain.

"I don't think removing amyloid is the end all and be all," said Schneider, noting that in some original studies, patients developed encephalitis after such plaque removal.

"So taking these plaques out doesn't mean you're going to do better. Having said that, this is showing that this drug is getting to its target," she said.

Schneider also noted that the patients in the study "are still declining" and may go on to develop AD, even if the cognitive decline is slowed.

She and other experts agreed that more and larger studies, possibly investigating additional agents, are needed.

"The scientific community is in agreement that combination therapy may be the future; it won't be a silver bullet," said Carrillo.

That may include approaches using anti-amyloid and the development of other treatments that address multiple aspects of the disease. These approaches should include both medications and lifestyle interventions.

"To be able to delay the disease even for a year can have a huge impact on lives," said Carrillo.

Dr Kramer is an employee of Eisai. The other investigators have disclosed no relevant financial relationships.

Alzheimer's Association International Conference (AAIC) 2018. Abstract 27531, presented July 25, 2018.

For more Medscape Psychiatry news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.