LONDON – "This is the best news we've had in my 25 years of doing Alzheimer's research."

So said Stephen Salloway, director of the Memory and Aging Program at Butler Hospital in Providence, R.I., and professor of clinical neurosciences and psychiatry at Brown Medical School, discussing the final results of Biogen Idec Inc.'s phase Ib study of the fully human antibody aducanumab (BIIB037), which are published in Nature Sept. 1.

Salloway, who has conducted approximately 100 Alzheimer's disease (AD) clinical trials, treated 40 patients in the phase Ib Biogen trial and was lead investigator for another AD antibody, bapineuzumab (Johnson & Johnson). He also was first to report the occurrence of the serious side effects caused by anti-amyloid antibodies that now are collectively referred to as amyloid-related imaging abnormalities (ARIA).

There are four features of the findings reported in Nature to underscore, Salloway said. They are:

  • • the "unique" isolation and development of aducanumab from humans who may be resistant to AD;
  • • the degree of amyloid lowering which is "unprecedented so far in our work";
  • • the evidence of slowing of cognitive decline at higher doses;
  • • establishing a route forward in managing ARIA side effects to balance the risk/benefit profile in phase III.

"I'm very encouraged that the amyloid lowering is quite pronounced and there is a hint of slowing of cognitive decline. If it could be reproduced in phase III, it would be terrific," Salloway said during a teleconference held to discuss the results.

As ever, a picture paints a thousand words, and the paper includes compelling before and after positron emission tomography (PET) images of one patient in the highest dose group (10 mg/kg), showing complete removal of amyloid plaque at 54 weeks.

That compares with no change from baseline in the PET image of a placebo group patient and dose-dependent reductions at 3 mg/kg and 6 mg/kg.

Symptoms of AD start when amyloid in the brain reaches its maximum extent, noted Roger Nitsch, co-founder and president of Neurimmune Therapeutics AG, of Zurich, Switzerland, the discoverer of aducanumab.

"The baseline images show 10-15 years of buildup [of amyloid]. One year later the placebo group is unchanged," Nitsch said. "In the three dose groups a clear reduction of amyloid is seen. There is no [amyloid] on the image in the 10 microgram group. Compared to other studies, the effect size of the drug is unprecedented."


Although the 165-patient PRIME phase Ib study of aducanumab was not powered to detect clinical changes, analysis of change from baseline on the clinical dementia rating (CDR) scale demonstrated dose-dependent slowing of clinical progression at one year, with the greatest slowing for 10 mg/kg (p<0.05 v. placebo).

A dose-dependent slowing of clinical progression on the mini mental state examination was also observed at one year, with the greatest effects at 3 mg and 10 mg/kg (p<0.05 vs. placebo).

A post-hoc analysis to see if there was a correlation between the degree of amyloid removal and cognitive change addressed the burning question of whether the PRIME study proves the amyloid cascade theory, which posits that amyloid beta peptide is the root cause of AD pathology.

Patients were divided into two groups showing low levels and high levels of amyloid removal. The group with small reductions in amyloid had the same level of cognitive decline as seen in an untreated population. The group with the greatest reduction in amyloid levels were stable.

"While not a prospective experiment, the data creates the hypothesis that removal of beta amyloid has a cognitive effect," Nitsch said.

Data showing a dose and time-dependent reduction in amyloid plaques in the brains of AD patients in PRIME – and a subsequent effect on cognitive function – were presented at the Alzheimer Association's International Conference in Washington in July 2015 and at the conference on Alzheimer's and Parkinson's diseases in Nice, France, in March 2015. (See BioWorld Today, March 23, 2015.)

However, the Nature paper is the first time all the data have been presented in a comprehensive form, in combination with preclinical research that Biogen conducted in murine models to unpick the precise mode of action of aducanumab.

"It's nice to see all the preclinical and clinical data together," said Alfred Sandrock, chief medical officer and senior vice present of Biogen. "We are starting to understand the biology of how we are getting the reaction in humans."

Following on from the presentation of the phase I data, in September last year Biogen began recruitment of two phase III studies, triggering a milestone payment of $60 million to Neurimmune.

The two trials will involve 2,700 patients with prodromal AD or mild cognitive impairment who have a positive amyloid PET scan.

Sandrock told BioWorld Today it is planned to test 3 mg/kg, 6 mg/kg and 10 mg/kg doses, with the CDR scale as the primary endpoint. "That will be acceptable to the regulators; we've had discussions, if [the trials] are positive aducanumab can be made available to patients," Sandrock said.

The two trials are scheduled to run until 2020, with full results expected in 2022.


In the PRIME study, the highest dose 10 mg/kg group in which the greatest effect on amyloid levels was seen also was most likely to experience ARIA.

ARIA also was genotype dependent, being more likely to occur in carriers of APOE e4, which is associated with a higher risk of AD. That finding is consistent with trials of other anti-amyloid beta antibodies.

It is theorized that ARIA is caused by increases in extracellular fluid resulting from changes in vascular integrity or local inflammatory processes associated with amyloid beta targeting therapies.

There was no apparent difference in the treatment effect of aducanumab in patients with and without ARIA.

ARIA was generally observed early in the course of treatment, typically resolving in four to 12 weeks, according to MRI scans. Of 27 patients who developed ARIA in the PRIME trial, 15 continued treatment.

Based on these findings, it is intended to mitigate ARIA in phase III by checking APOE e4 status and forward titrating the dose of aducanumab. APOE e4-positive patients will not be excluded, but dosing will be adjusted. All patients will be given frequent scans to detect ARIA in the earlier stages.

"If ARIA is mild, patients will stay on the current dose, but we will do more MRI [scans]. If moderate to severe, we suspend dosing. [ARIA] usually clears in one to three months and you then restart [dosing]," Salloway said.


As to why aducanumab displayed a greater effect than seen in other anti-amyloid beta antibodies, the Nature paper points to, "the inability of other drugs to adequately engage their target, or the proper target in the brain, or selecting the wrong patient population."

Sandrock pointed to differences in molecular structure of different anti-amyloid antibodies, noting preclinical data indicated aducanumab avoids soluble amyloid beta in the blood stream, crossing the blood-brain barrier and binding specifically to all morphological types of brain amyloid beta in vivo.

In addition, aducanumab has an immune-competent Fc region. Sandrock said experiments in which this was removed showed that it is an important component. "We found it was important to maintain [the Fc region]. Without it, [aducanumab] lost the ability to clear plaques as effectively," he said.