Background

Bapineuzumab is a humanized form of murine monoclonal antibody 3D6, which targets the N-terminal region of Aβ. The rationale of this passive immunotherapy approach is that antibody binding will clear excess Aβ. Bapineuzumab is an IgG1 antibody that binds fibrillar and soluble Aβ and activates microglial phagocytosis and cytokine production. A large portion of the extensive preclinical work supporting passive anti-Aβ immunotherapy for Alzheimer's disease was obtained with 3D6. A small fraction of peripherally administered antibody enters the CNS of PDAPP and other mouse models of Aβ amyloidosis. The antibody was shown to bind to amyloid plaques, lower plaque burden, improve measures of synaptotoxicity, and improve performance on mouse behavioral assays.

Findings

In a 12-month Phase 1 study, 0.5, 1.5, or 5 mg/kg of bapineuzumab appeared overall safe and well-tolerated in patients with mild to moderate AD. MRI monitoring showed abnormalities indicative of potential new microhemorrhage in three of 10 patients on the highest dose. Two patients were asymptomatic, and one experienced mild, transient confusion. This finding prompted the Alzheimer’s Association Research Roundtable Workgroup to coin the term amyloid-related imaging abnormalities (ARIA) to encompass imaging findings associated with amyloid-lowering therapies, specifically ARIA-edema/effusions (ARIA-E) to capture vasogenic edema.

In Phase 2, 124 patients with mild to moderate AD received either 0.15, 0.5, 1, or 2 mg/kg of bapineuzumab and 110 received placebo every 13 weeks for an 18-month trial. No significant difference was seen in any of the dose cohorts on either ADAS-Cog or DAD, the two prespecified primary outcomes. Even so, a Phase 3 program was initiated based on prespecified exploratory analyses on pooled treatment groups versus placebo, and on the subpopulation of patients who completed the trial. Also taken into consideration was a post-hoc exploratory efficacy analysis by apolipoprotein ApoE4 carrier status, which hinted at a treatment response in noncarriers. CSF biomarkers measured in a substudy of 35 study participants showed no treatment differences in CSF Aβ or total tau levels, but did show a trend toward P-tau reduction with treatment. MRI volumetry showed no overall differences between bapineuzumab and placebo groups. A substudy of serial amyloid PET scans in 28 patients demonstrated that bapineuzumab cleared some fibrillar cerebral Aβ by week 78. ARIA-E, back pain, anxiety, and paranoia occurred more commonly with treatment than placebo, in at least 5 percent of bapineuzumab patients. Deep-vein thrombosis, syncope, seizures, vomiting, hypertension, weight loss, skin laceration, gait disturbance, muscle spasm, and pulmonary embolism also occurred more frequently in the bapineuzumab group. ARIA-E was detected in 12 patients treated with apineuzumab  but none who had been treated with placebo, and more frequently at higher doses. Six patients with ARIA-E had no clinical symptoms, and six experienced headache, confusion, vomiting, or gait disturbance. Symptoms were transient; one patient required steroid treatment. ApoE4 carriers made up 10 of the 12 ARIA-E cases. For ARIA-E, rates were 33.3 percent in ApoE4 homozygotes, 7.1 percent in ApoE4 heterozygotes, and 4.3 percent in noncarriers. These findings have been ascribed in part to the increased load of Aβ, including vascular amyloid, in ApoE4 carriers. ARIA-E is thought to result from transient leakiness of cerebral vessels following vascular amyloid clearance. ARIA similar to the findings linked to bapineuzumab treatment have been reported to occur spontaneously in AD, as well. Other Phase 2 trials tested bapineuzumab in Japanese patients and as a subcutaneous formulation.

Two 18-month Phase 3 trials testing intravenous bapineuzumab in a total of 2,452 patients with mild to moderate Alzheimer’s disease were completed, one in people who carry at least one copy of ApoE4, and one in noncarriers. They were part of a four-trial Phase 3 program slated to test bapineuzumab in 4,000 patients across North America and Europe, plus an extension trial and a trial of a subcutaneous formulation of bapineuzumab. All subsequent trials were discontinued after the first two completed trials showed no treatment effect on either cognitive or functional outcomes. Biomarker analyses indicated that bapineuzumab engaged its target but had no benefit. Commonly cited explanations focus on the low dose necessitated by bapineuzumab's side effect profile and late-stage treatment initiated years after brain amyloid deposition has begun. The bapineuzumab clinical trials program established procedures that were instituted subsequently in other Alzheimer's immunotherapy programs, for example, the practice of using a central reader to assess ARIA at baseline and throughout the trial. For all bapineuzumab trials, see clincialtrials.gov. 

Clinical Trial Timeline

All phrase 3 trials were terminated on August 6, 2012 because two large Phase 3 studies showed no clinical benefit. This decision was not based on any new safety concerns.