Early Data Suggest Benefit of Aducanumab in Alzheimer’s Disease

BOSTON—The antiamyloid antibody aducanumab may slow cognitive decline and reduce amyloid burden in patients with Alzheimer’s disease, according to results presented at the 10th Edition of Clinical Trials on Alzheimer’s Disease (CTAD). The results are 36-month data from the phase Ib PRIME trial.

Patients who have been taking the highest dose of aducanumab, 10 mg/kg, for the duration of the study declined the least on two measures of cognition, the Mini-Mental State Exam (MMSE) and the Clinical Dementia Rating Scale–Sum of Boxes (CDR-SB). Some of the participants taking the 10-mg/kg dose became amyloid negative on PET by 24 months and stayed at a low level of amyloid until month 36, said Samantha Budd Haeberlein, PhD, Vice President of Clinical Development at Biogen in Cambridge, Massachusetts.

Samantha Budd Haeberlein, PhD

It is likely that the high-dose group continued to have amyloid, despite the imaging findings, said Dr. Haeberlein. “I would challenge the idea that [aducanumab] completely removed amyloid, since I think the instrument is not perfect,” she said, adding that the decreased level represents a drop below the threshold for positivity set by Eli Lilly, maker of the imaging agent florbetapir. “But we have to say that we are in a different realm here, where it can be difficult to determine whether an individual is positive or negative for amyloid pathology.”

The 36-month data support the continued development of aducanumab, said Dr. Haeberlein. The antibody is now being tested in two phase III studies, ENGAGE and EMERGE.

“The aducanumab data reported at CTAD are good news for safety and good news for the signals we need to see in the phase III trials,” said Maria Carillo, PhD, Chief Science Officer of the Alzheimer’s Association. “These are hopeful signs, but based on what we have learned from past Alzheimer’s studies, we need to wait for the phase III trial results.”

Study Examined Four Doses

Aducanumab is a monoclonal human antibody derived from B cells collected from a cohort of cognitively normal elderly subjects and cognitively impaired elderly subjects who exhibited unusually slow decline, according to Biogen. It binds to fibrillar and oligomeric amyloid aggregates, thus directly reducing amyloid plaques in the brain.

Investigators enrolled 165 patients with prodromal or mild Alzheimer’s disease into the PRIME study. All of the participants had brain amyloid on PET imaging. PRIME is the first randomized trial of an antiamyloid compound to rely solely on PET to establish participants’ amyloid positivity. These patients were randomized to placebo or 1 mg/kg, 3 mg/kg, 6 mg/kg, or 10 mg/kg of aducanumab for one year. The treatment period was followed by a two-year open-label extension. Patients who had been randomized to placebo or 1 mg/kg of aducanumab were switched to 3 mg/kg of aducanumab or to a 3-mg/kg to 6-mg/kg titration regimen in the long-term extension. Patients randomized to aducanumab at 3 mg/kg, 6 mg/kg, 10 mg/kg, or titration in the placebo-controlled period continued to receive the same dose.

The PRIME trial’s primary outcomes are safety and tolerability. The cognitive and functional outcomes, which are not usually assessed in a phase Ib study, are exploratory. The numbers in each dosing group are quite small, said Dr. Haeberlein. Of the original cohort, 117 entered the extension study, and 50 continued until 166 weeks, at which time 10 to 16 patients were in each of the dosage cohorts.

Amyloid Burden Decreased in Some Patients

At 36 months, the mean change in amyloid plaque level was greatest for the 10-mg/kg group, which, on average, no longer met the threshold of amyloid positivity on florbetapir PET. The amyloid level in the 6-mg/kg group declined to the threshold, but did not fall below it. The 1-mg/kg and 3-mg/kg groups declined at similar rates, but the decreases were not as large as in the higher-dose group.

All participants declined on the MMSE and CDR-SB. The decline, however, was clearly attenuated in some of the active groups, where the best results were seen in the 10 patients who received the 10-mg/kg dose. The average decline from baseline on the CDR-SB was 2.84 points among those patients. In the other groups, declines from baseline on the CDR-SB were 5.28 points in those who switched from placebo to 3 mg/kg, 6.11 points in those who switched from 1 mg/kg to 3 mg/kg, 3.86 points in the 3-mg/kg treatment group, and 4.49 points in the 6-mg/kg treatment group.

Patients taking 10 mg/kg also fared best on the MMSE, declining 4.10 points on average. Declines in the other groups were 7.98 points in those who switched from placebo to 3 mg/kg, 6.35 points in those who switched from 1 mg/kg to 3 mg/kg, 4.83 points in the 3-mg/kg treatment group, and 8.97 points in the 6-mg/kg treatment group. These differences were not statistically significant, said Dr. Haeberlein. “In this extension trial, we are not talking about statistical significance.”

Investigators Observed Cases of ARIA

The incidence of amyloid-related imaging abnormalities (ARIA), however, did not follow this dose-dependent pattern. All eight cases of edematous ARIA (ARIA-E) in the long-term extension phase occurred in the placebo group that switched to 1 mg/kg and in the 1-mg/kg group that was titrated to 3 mg/kg. All cases occurred early in the extension phase, no new cases occurred during the past year, and all but one case occurred in carriers of APOE4.

Hemorrhagic ARIA occurred in two controls who switched to 1 mg/kg of aducanumab, five participants taking 3 mg/kg, two participants taking 6 mg/kg, and one patient taking 10 mg/kg. These cases occurred early in the trial. All of the ARIA cases, regardless of etiology, were considered mild and resolved spontaneously. In all, 46 patients in the PRIME trial have experienced ARIA, and six have had more than one episode.

The most common adverse events in the long-term extension phase were falls, headache, and ARIA. Two patients in the extension phase died, one in the 6-mg/kg group and one in the 10-mg/kg group. Neither death was considered to be related to the study medication.

—Michele G. Sullivan