Conference Coverage

Alzheimer’s trial design problem throws a wrench in promising BAN2401 results



– BAN2401, a monoclonal antibody that targets soluble amyloid-beta oligomers, slowed cognitive decline by up to 47% while clearing brain amyloid in 81% of patients with mild cognitive impairment and very mild Alzheimer’s disease, a phase 2 study has shown.

At the Alzheimer’s Association International Conference, Eisai, which is developing the antibody along with Biogen, touted it the study results as an important step forward in a field desperate for success. But questions remain, especially about the unequal distribution of apolipoprotein E4 (APOE4)-positive patients across the six treatment groups: BAN2401 2.5 mg/kg biweekly, 5 mg/kg monthly, 5 mg/kg biweekly, 10 mg/kg monthly, and 10 mg/kg biweekly, or placebo. This troubled some researchers, who said the mix could have biased cognitive results in the antibody’s favor. APOE4 carriers comprised about 70%-80% of every unsuccessful treatment arm in the trial, and 29% of the arm that conferred significant cognitive benefits.

Dr. Keith Fargo, director of scientific programs for the Alzheimer's Association

Dr. Keith Fargo

“This is a big confound,” Keith Fargo, PhD, director of scientific programs and outreach at the Alzheimer’s Association, said when he saw the numbers. “An important point is that, according to the trial sponsors, for a period of time a regulatory body requested that people with the APOE4 Alzheimer’s risk gene not be included in the highest dose group, for safety reasons. Because of this, the people in the highest-dose group are different from people in the other groups on an important dimension; they were much less likely to have the APOE4 gene, which is known to be a major risk factor for cognitive decline. So it is plausible that the people on the highest dose declined differently due to genetic differences rather than due to being on the highest dose. A planned subgroup analysis will shed more light on this, and whether it reduces confidence in the overall findings. We eagerly await the results of this analysis.”

The decision to restructure the randomization was not Eisai’s, according to David Knopman, MD, chair of the Alzheimer’s Association Medical and Scientific Advisory Council.

Dr. David Knopman

Dr. David Knopman

“European regulators did not allow randomization of [some] APOE4 carriers to the highest dose,” he said in an interview. “I ultimately don’t know what it would do to the results, except make them even more difficult to justify as sufficient for registration. In general, in symptomatic Alzheimer’s dementia patients, APOE4 carriage has no substantial impact on rate of decline, but whether [APOE4] status interacted with the treatment is of course completely unknown. Bottom line: Just another feature that makes this a phase 2 study that needs to be followed by a phase 3 study with a simple design using the high dose.”

Alzheimer’s researcher Michael S. Wolfe, PhD, was more candid.

“Regardless of who made the decision and why, the data is what it is, and the question remains,” said Dr. Wolfe, the Mathias P. Mertes Professor of Medicinal Chemistry at the University of Kansas, Lawrence. “Given the numbers of E4-positive vs. E4-negative for each treatment group, the interpretation of the results is now seriously thrown into question. The one group that showed a clear slowing of cognitive decline versus placebo (10 mg/kg biweekly) also has far fewer E4-positive vs. E4-negative. This difference in proportion of E4-positive could be a major factor in the apparent reduced rate of cognitive decline and confounds the ability to tell if the 10-mg/kg biweekly dose is effective. In contrast, the 10-mg/kg monthly dose group has an E4-positive/-negative ratio more comparable to that of placebo, and the effect of the drug on cognitive decline under this dosing regimen is not clearly distinguishable from that seen with placebo.”

Dr. Michael S. Wolfe, the Mathias P. Mertes Professor of Medicinal Chemistry at the University of Kansas, Lawrence

Dr. Michael S. Wolfe

Additionally, BAN2401 failed to meet its 12-month prespecified primary cognitive endpoints, a conclusion determined by a complex Bayesian analysis that strove to predict an 80% probability of reaching at least a 25% cognitive benefit. In December, the company announced that the study hit just a 64% probability. But because Eisai felt the numbers were moving in the right direction, it continued with the additional 6 months of treatment, as allowed for in the study design, and then reanalyzed results with a simpler and more straightforward method. This analysis concluded that 10 mg/kg infused biweekly conferred a 30% slowing of decline on the Alzheimer’s Disease Composite Score (ADCOMS), a new tool developed and promoted by Eisai, and a 47% slowing of decline on the Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-Cog). It also cleared brain amyloid in 81% of subjects and sent cerebrospinal fluid (CSF) biomarkers in the right direction. CSF amyloid levels went up – to be expected if the antibody was clearing it from the brain – and CSF total tau went down, an indication of decreased neuronal injury.

Inquiries to Eisai to clarify these issues had not been returned at press time.

This is a developing story.

[email protected]

Next Article: