ARTICLE IN BRIEF
Two new reports find that most cases of the side effect associated with bapineuzumab for Alzheimer disease — formerly referred to as vascular edema and now called amyloid-related imaging abnormalities of edema/effusion, or ARIA-E — appear to be asymptomatic, with risk declining over time.
PARIS—A serious adverse event that had raised concerns about the safety of the experimental monoclonal antibody bapineuzumab is much more common than previously reported, according to two new reports presented here at the Alzheimer's Association International Conference.
However, most cases of the event — formerly referred to as vascular edema and now called amyloid-related imaging abnormalities of edema/effusion, or ARIA-E — appear to be asymptomatic, with risk declining over time, the new research suggests.
A biologic under development for the treatment of Alzheimer disease (AD), bapineuzumab binds to amyloid beta (Abeta). The hope is that the experimental compound can clear the brain of the sticky amyloid that clumps into plaques, causing AD.
CONFIRMS UNDETECTED CASES
In one new report, a re-examination of MRIs taken during phase 2 studies involving 210 patients on bapineuzumab not only confirmed 21 previously detected cases of ARIA-E, but also identified 15 previously undetected cases, for an incidence rate of about 17 percent.
However, all of the new cases were asymptomatic, as were all but eight of the originally detected cases — a finding that is “reassuring,” said presenter Reisa A. Sperling, MD, associate professor of neurology at Harvard Medical School and director of clinical research in the Memory Disorders Unit at Brigham and Women's Hospital in Boston.
Of the total 36 ARIA-E cases identified among patients with two or more MRIs, eight (22 percent) were symptomatic, marked by headache, confusion, and visual and gait abnormalities.
Incident micro-hemorrhages (ARIA-H) were visible on the image review in about half — 17 — of the cases.
Thirteen of 15 of the newly identified cases received multiple doses of bapineuzumab and remained asymptomatic, she said.
As for risk factors, apolipoprotein e4 (APOE4) carriers and patients treated with higher doses were at increased risk of ARIA-E, Dr. Sperling reported. Patients homozygous for the APOE4 allele had more than double the risk of developing ARIA-E, she noted. And risk approximately doubled for each 1mg/kg increase in dose.
Among the eight symptomatic patients, seven were carriers of the risk allele and six were receiving doses of 1 mg/kg or greater.
For the study, more than 2500 fluid attenuated inversion recovery (FLAIR) images from the 210 subjects were reviewed by two independent neuroradiologists blinded to demographics, medical history, and treatment assignment.
Interim analysis of a second, open-label extension study reported at the meeting found that the rate of ARIA-E declined after the third dose.
“Overall, bapineuzumab was well tolerated,” said principal investigator Stephen P. Salloway, MD, professor of neurology and psychiatry at the Warren Alpert Medical School of Brown University and director of neurology and the Memory and Aging Program at Butler Hospital in Providence, RI.
“The majority of cases of ARIA are asymptomatic and almost all resolve by holding the dose,” he said.
The study involved 194 patients who received one of three doses of bapineuzumab IV — 0.15 mg/kg, 0.5 mg/kg, or 1.0 mg/kg — every 13 weeks. (When the first reports of VE surfaced, investigators dropped the highest 2.0 mg/kg dose and patients were switched to 1.0 mg/kg.)
A total of 121 patients dropped out of the study, 24 due to adverse events. Another 72 were labeled “voluntary withdrawals.”
Asked to comment on the high discontinuation rate, Dr. Salloway said the reasons for the voluntary withdrawals are still being analyzed. A preliminary review suggests most are due to disease progression and caregiver and study fatigue, he said.
“Patients discontinued gradually over a period of years,” he added. Patients took bapineuzumab for an average of 2.6 years, with 94 of them on the drug for at least three years and 46 on it for at least four years.
About two-thirds were APOE4 carriers.
About 24 percent of patients (46/194) reported adverse events that were considered related or possibly related to bapineuzumab. Of these, 85 percent (39/46) were mild to moderate.
ARIA-E was the most common drug-related adverse event, affecting 9.3 percent of patients. “This is very similar to what we saw in [a previous] phase 2 trial, and the rate goes down with time,” Dr. Salloway said.
Specifically, the risk of developing ARIA-E dropped from 6.7 percent for infusions one through three to 2.7 percent for infusions four through 10, he said.
There also appeared to be a dose effect on ARIA-E incidence, with risk dropping from 12.5 percent for the 1.0 mg/kg dose to 9.5 percent for the 0.5 mg/kg dose and 2.7 percent for the 0.15 mg/kg dose.
“Overall, bapineuzumab was generally well tolerated and adverse effects tended to be mild to moderate,” Dr. Salloway said. “Importantly, there were no new safety signals with exposure over five years of treatment.”
IS THE DRUG WORKING?
Preclinical and early human findings suggest that ARIA-E may be a sign that bapineuzumab and other amyloid-clearing drugs are doing what they are supposed to be doing — removing amyloid from the brain and blood vessels, Dr. Salloway said.
APOE4 carriers have higher rates of amyloid in their brain and blood vessels, which is why they are at greater risk for ARIA, he said.
ARIA is not unique to bapineuzumab treatment. It has been reported in a number of studies of amyloid-modifying drugs and can occur naturally in the course of AD.
During her talk, Dr. Sperling showed the audience FLAIR/MRI images consistent with ARIA-E in placebo-treated patients to illustrate the potential for spontaneous cases.
Additionally, she presented a separate paper at the meeting reporting that three patients taking the gamma secretase inhibitor BMS-708163 appeared to develop ARIA-E.
While the mechanisms of ARIA remain unknown, Dr. Sperling noted that “risk analyses, animal data, and frequent co-occurrence of vasogenic edema with microhemorrhage suggest transient increases in vascular permeability — leaky vessels — due to mobilization of amyloid from plaque into vessels and clearance of vascular amyloid.”
The bottom line, she said, is that the current analyses “made me a little more comfortable about the risk side. I do think the leaky vessel problem will affect the whole class [of amyloid-modifying drugs], but it's manageable.”
Samuel E. Gandy, MD, PhD, the Mount Sinai Professor of Alzheimer's Disease Research, Professor of Neurology and Psychiatry and associate director of the Mount Sinai Alzheimer's Disease Research Center in New York City, agreed. “My sense is that ARIA-E seems to be a rather benign entity that should not impede development of the product.” Other researchers had a few reservations.
Jeffrey L. Cummings, MD, professor of neurotherapeutics and drug development at the Cleveland Clinic Lerner College of Medicine and director of the Cleveland Clinic Lou Ruvo Center for Brain Health in Las Vegas, said that an open-label study is prone to selection bias, as “many of the patients who were most symptomatic may have already dropped out. As a result, you're left studying the drug in patients selected for their resistance to the side effects or the effects of the side effects,” he said.
Whether ARIA-E will limit the usefulness of bapineuzumab depends on what other treatments are developed, Dr. Cummings said.
If the FDA requires regular MRI monitoring of patients, “it will be cumbersome and expensive. So if there is a drug without that requirement, it would have the advantage,” he said.
John M. Ringman, MD, associate clinical professor of neurology at the University of California, Los Angeles, said that the “high discontinuation rate due to voluntary withdrawal [in Dr. Salloway's study] makes me wonder about the tolerability of the drug. [Until the data are analyzed,] we don't know what voluntary withdrawal means: It might mean study fatigue or that the intervention wasn't working, or it could be that the subjects withdrew because of unspecified safety concerns,” he said.
On the other hand, “the Sperling paper shows that while ARIA-E is not terribly uncommon, it doesn't necessarily have devastating neurologic sequelae,” Dr. Ringman said.
Dr. Salloway stressed that no conclusions as to bapineuzumab's efficacy can be drawn from the open-label study. Four large, randomized, phase 3, double-blind trials that are underway in patients with mild to moderate disease should help answer that question, he said.
Dr. Ringman is an investigator on one of the studies, an 18-month trial pitting lower doses of bapineuzumab against placebo in patients with mild to moderate AD who are not APOE4 carriers. Infusions are given every 13 weeks, with safety MRIs midway between each dose. Enrollment is complete, but patients are still being treated in a blinded fashion, he said.
And it could be that the drug will work better in patients with less advanced disease, the experts said.
“If bapineuzumab moves the biomarkers in the right direction but does not produce a clinical benefit, the next step would be to use it in earlier stages of the disease when patients are more likely to respond,” Dr. Salloway said.
The studies were funded by Janssen and Pfizer Inc.©2011 American Academy of Neurology
Neurology Today Quick Links