WASHINGTON – An antiamyloid antibody that showed significant cognitive benefit at 10 mg/kg failed to deliver similar good news at a lower dose in the first randomized, antiamyloid antibody trial comprising only patients who screened positive for amyloid-beta plaques on PET imaging.

At 6 mg/kg, aducanumab, a fully human monoclonal antibody, cleared amyloid brain plaques almost as well as the higher dose. But although there were numerical improvements on cognitive measures, these were not statistically significant, Dr. Jeff Sevigny said at the Alzheimer’s Association International Conference 2015.

©roberthyrons/thinkstockphotos.com

The drug is being developed by Biogen. Dr. Sevigny is the company’s senior medical director of neurodegenerative disorders.

The PRIME study enrolled 166 patients with prodromal or mild Alzheimer’s who screened positive for amyloid-beta plaques on PET imaging. This ensured a pure cohort of patients with amyloid-beta plaques – something that has been lacking in antiamyloid antibody trials. Some researchers have estimated that up to 15% of those enrolled in previous trials actually had no amyloid-beta in the brain, a finding that could seriously dilute any efficacy signal of a drug designed to remove the protein.

The 52-week phase Ib study randomized patients to placebo or to 1-, 3-, 6-, or 10-mg/kg monthly infusions of aducanumab. In all, patients received 14 infusions. While the trial was designed to evaluate safety and tolerability, it did include preliminary cognitive data in the form of the Mini-Mental State Examination (MMSE) and the Clinical Dementia Rating Scale–Sum of Boxes (CDR-SB).

The initial study readout was reported in March amid global hoopla. The 3- and 10-mg/kg doses removed amyloid-beta from the brain; the 10-mg/kg dose delivered significant benefits on both cognitive measures. Patients taking the 10-mg/kg dose dipped about half a point below baseline on the MMSE, compared with nearly a 3-point drop for those on placebo. The 3-mg/kg dose also conferred significant benefits against cognitive decline.

Aducanumab was hailed as what could be the “Goldilocks drug” for Alzheimer’s – the first evidence of an antibody that both engaged the amyloid-beta target and provided significant cognitive improvements, although some researchers cautioned against overinterpretation of such early-phase results.

Patients in PRIME had a mean MMSE of 25, and about two-thirds were apolipoprotein E–epsilon 4 (APOE4) positive. Close to half were in the prodromal stage and the remainder had mild Alzheimer’s.

At 52 weeks, patients taking placebo worsened by 2.81 points on the MMSE. Patients taking 6 mg/kg lost 1.96 points – not a significant difference. The 10-mg/kg MMSE results were statistically significant, with a decline of 0.56 in the active group. Even the 3-mg/kg dose performed better than 6 mg/kg, with a decline of 0.70 points – significantly better than the placebo group.

The story was similar on the CDR-SB measure. The placebo group worsened by 1.87 points at week 52, while the 6-mg/kg group worsened by 1.11 points – again not statistically significant. Only the 10-mg/kg dose achieved a statistically significant difference relative to placebo (0.63 points). The 1-mg/kg dose was not significantly better than placebo on either the MMSE or CDR-SB.

All doses except for 1 mg/kg lowered amyloid-beta plaque burden. In the placebo arm, the level of amyloid remained unchanged at both imaging time points (26 and 52 weeks). Plaque moved out in a dose- and time-dependent manner, with 6 and 10 mg/kg removing increasing amounts of the protein.

Amyloid-related imaging abnormalities (ARIA), a brain inflammation related to the removal of amyloid plaques, occurred in 13% of the 3-mg/kg group, 33% of the 6-mg/kg group; and 47% of the 10-mg/kg group. Most of these (89%) occurred in the initial 5 weeks; most (65%) were asymptomatic. The remainder caused mild and transient symptoms, with headache, visual disturbances, and cognitive worsening that resolved by 12 weeks.

APOE4 carriers were particularly subject to ARIA. It occurred in 5% of patients taking the 1-mg/kg and 3-mg/kg doses, 43% of those taking 6 mg/kg, and 55% of those receiving 10 mg/kg. In noncarriers, there were no cases in the 1-mg/kg group. The incidence was 9%, 11%, and 17% in the 3-, 6-, and 10-mg/kg groups of noncarriers respectively. Patients who developed ARIA had the choice of dropping out, opting for a lower dose, or continuing the study at their original dose. Most who developed ARIA (56%) chose to continue treatment.

There were three deaths: two in the placebo group and one in the 10-mg/kg group. None were related to the study drug.

According to Biogen, aducanumab is a fully human IgG1 monoclonal antibody that binds aggregated forms of amyloid, not monomer, and preferentially attaches to parenchymal amyloid rather than vascular amyloid. It was “derived from a deidentified library of B cells collected from healthy elderly subjects with no signs of cognitive impairment and cognitively impaired elderly subjects with unusually slow cognitive decline.”

msullivan@frontlinemedcom.com

On Twitter @Alz_Gal

Biogen’s aducanumab stumbles on cognitive endpoints at 6 mg/kg dose