8 August 2012. Janssen Alzheimer Immunotherapy will discontinue all studies of intravenous bapineuzumab, an amyloid-β antibody, in patients with mild to moderate Alzheimer’s disease, according to August 6 press releases by Pfizer and Johnson & Johnson, partners in development of the immunotherapy. Trials of subcutaneous bapineuzumab will continue, though some researchers contacted by Alzforum doubt that those will be successful. The decision to halt the Phase 3 trials—two in North America and two primarily outside North America—comes less than a month after the companies announced negative topline results of one of the U.S. trials. Bapineuzumab had failed to have any effect in patients with mild to moderate AD who carry an ApoE4 allele—the strongest genetic risk factor for AD (see ARF related news story). Following that setback, the field awaited news of the ApoE4 non-carriers. It turns out they fared no better.
“This news is disappointing,” said Steve Salloway, Brown University, Providence, Rhode Island. “We hoped [bapineuzumab] would be better in milder non-ApoE4 carriers, but it was not to be.” News of the failure has already made The New York Times and other general media.
The results came as no surprise to many, since earlier trials had failed to impress. Traditionally, in drug development, a Phase 3 program is a set of confirmatory trials of a solid Phase 2 efficacy finding. This was not the case here. In Phase 2, ApoE carriers showed no benefit from Bapineuzumab, and though non-carriers seemed to do better on cognitive tests (the ADAS-Cog) in a post-hoc subgroup analysis, that was only in comparison to a placebo group that declined much more rapidly than expected. “The problem with claiming there was a benefit in [ApoE4] non-carriers was that the placebo group dropped approximately 11 points on the ADAS-Cog over 18 months, while the treatment group dropped six points, but in other trials, placebos only drop around seven to eight points,” explained Ryan Watts, Genentech, San Francisco, California. Nevertheless, the companies are pioneers in this field, and with the uncertainty of cognitive measures in limited numbers of patients, they took a calculated risk that was worth taking, said Watts.
“I think the data are pretty clear from this trial that bapineuzumab is not effective for mild to moderate dementia, and the company was wise to stop the ongoing trials in that population,” said Salloway, who leads one of the trial sites. Researchers now hope that some good can come from these four 18-month studies by way of informative biomarker data. The companies will present data from the two North American trials (301 and 302) this fall at the European Federation of Neurological Societies Conference in Stockholm, Sweden, the American Neurological Association annual meeting in Boston, and the Clinical Trials on Alzheimer’s Disease meeting in Monte Carlo, Monaco. “That is going to be important,” said Salloway. “If there are signals that it modifies amyloid load or secondary markers of neurodegeneration, it may suggest that we can use this drug earlier in the disease.” A subgroup of trial participants underwent PET scans to quantify amyloid load in the brain, and spinal taps in a subset of patients will reveal levels of tau and phospho-tau in the cerebrospinal fluid. MRI imaging in some patients will also reveal any changes in brain volume.
Pundits have already used this latest failure to pronounce the amyloid hypothesis dead (see related article in Forbes). Such reports, ironically, come less than a month after the popular press claimed that the discovery of a protective mutation in amyloid precursor protein strongly supports the hypothesis (see Forbes, The New York Times, and ARF related news story). “This is the sad reality of binary thinking,” said Watts, who believes amyloid plays a crucial role in the disease. He sees this latest disappointment not as a failure of the hypothesis, but a failure of drug development. “The real question is how we will succeed as drug developers,” he told Alzforum.
Why did bapineuzumab fail? Like many researchers in the field, Watts thinks it may be a question of too little too late. He noted that the appearance of vasogenic edema (now commonly called ARIA, or amyloid-related imaging abnormalities) in some patients restricted dosing of bapineuzumab to a maximum of 1 mg/kg once a quarter, severely limiting the amount of antibody that gets into the brain. While a pilot study demonstrated the treatment lowered brain fibrillar Aβ by about 10 percent (see ARF related news story), that may be insufficient, particularly in later stages of disease. Genentech is addressing the “too little” and “too late” questions with their own trials of crenezumab, an antibody that binds all forms of Aβ, including oligomers (see ARF related news story). The ABBY and BLAZE trials of patients with mild to moderate AD will address the dosing question by looking at behavioral and biomarker outcomes, respectively, at up to 15 mg/kg once per month. The Alzheimer’s Prevention Initiative will test whether giving the antibody earlier will help (see ARF related news story).
For bapineuzumab, it remains to be seen if earlier intervention would succeed. “The skeptics might be right, but we wouldn’t know until we tested it,” said Salloway. The partners in the Alzheimer Immunotherapy Program (AIP) have not discounted that possibility. “We share the scientific belief that it is possible to reduce the burden of Alzheimer’s through earlier intervention in the illness, and the AIP is exploring research opportunities in this area,” a Janssen spokesperson told Alzforum.—Tom Fagan.