CONFERENCE COVERAGE SERIES
European Federation of Neurological Societies
Stockholm, Sweden
11 September 2012
Bapineuzumab Phase 3: Target Engagement, But No Benefit
Bapineuzumab prevents accumulation of Aβ in the brain of patients with mild to moderate Alzheimer’s disease and lowers phospho-tau (p-tau) in their cerebrospinal fluid (CSF), according to analysis presented September 11 at the European Federation of Neurological Societies (EFNS) annual meeting in Stockholm, Sweden. Reisa Sperling, Brigham and Women’s Hospital, Boston, and Stephen Salloway, Brown University, Providence, Rhode Island, reviewed much-anticipated biomarker data from Phase 3 clinical trials of the passive immunotherapy in 1,121 ApoE4 carriers and 1,331 non-carriers, respectively. Despite the positive biomarker results reported yesterday, clinical data released August 6 showed the drug failed to protect patients in these trials from cognitive and functional decline (see ARF related news story). Clinical development of intravenous bapineuzumab in mild to moderate AD patients will end with these trials.
“While we are disappointed with the clinical data, the biomarker evidence indicates target engagement and downstream effects on secondary markers of neurodegeneration without serious safety issues,” Salloway told Alzforum after the meeting. Niels Prins, VU University Medical Center, Amsterdam, The Netherlands, who chaired the session, considers this a positive sign. “It says there might be role for immunotherapy [in Alzheimer’s treatment], but that in order to be effective we may have to give it in the prodromal phase, and maybe even earlier,” he said in an interview with Alzforum. Sperling, Salloway, and other researchers in the field echoed this sentiment. Other researchers had a different interpretation, expressing caution about the safety data. Yet others thought the positive Aβ effects spurious.
Sperling reported that over the 71 weeks of the trial, bapineuzumab, given only once every 13 weeks at 0.5 mg/kg, reduced CSF p-tau. It also kept brain amyloid burden steady in ApoE4 carriers as it gradually increased in the placebo group. The researchers measured amyloid in a subset of patients using PIB PET imaging. In the ApoE non-carrier trial, 1.0 mg/kg of the antibody lowered CSF p-tau and showed a tendency to reduce brain Aβ compared to placebo. Some researchers contacted by Alzforum cautioned that the effect seen in the ApoE4 carrier trial might be due to an uncharacteristic jump in amyloid burden in the last six months of the trial in the untreated group. Most data suggest that amyloid accumulates gradually and has neared its peak by the time patients are symptomatic (see ARF Webinar). The drug sponsors made webcasts of Sperling’s and Salloway’s presentations slides of the two talks freely available on the EFNS website.
With the exception of vasogenic edema (also called ARIA-E or amyloid-related imaging abnormalities due to edema or effusion) and stroke, treated ApoE non-carriers had no more adverse events than untreated controls. However, safety data for the ApoE4 trial showed that 1.8 percent of the treated patients died during the course of the trial, compared to 0.7 percent of untreated patients. That one percent difference would equate to 10,000 additional deaths if one million patients were to be treated. “That is extremely concerning,” said Murali Doraiswamy, Duke University Medical Center, Durham, North Carolina, who is an investigator on the trials. Most of the deaths in the treated ApoE4-positive group resulted from cancer. No more cancer deaths than in the placebo group were seen in the ApoE4-negative trial, however. In her talk, Sperling said that the data monitoring safety board deemed the cancer deaths unrelated to the therapy. “Nonetheless, we would have to be very careful to consider the reasons for that difference before bapineuzumab could be administered to a normal cohort of ApoE4 carriers,” Doraiswamy told Alzforum. Researchers have suggested that bapineuzumab might prove effective if given earlier.
Salloway noted that analysis of PIB PET data in the ApoE4 non-carriers may be subject to strong outlier effects since the number of patients included in the analysis was smaller than expected. More than one third of patients fell below the threshold for amyloid positivity at baseline, and another 25 patients did not come for their planned follow-up scans. Patients not meeting threshold levels for amyloid were a major talking point after the presentations. “Some of these patients are likely misdiagnosed, but there may be other factors involved,” suggested Sperling. Salloway thought there might be technical issues. “We want to look more carefully at that data. Some had CSF taken, so we want to look at that to see if it is non-AD like, as well,” he said. Prins considered misdiagnosis the most likely explanation. Sperling said the figures were actually not that surprising given that, on autopsy, 15-20 percent of AD patients are found to not have had AD, after all. “That’s in the general population. In ApoE4 non-carriers you might expect that number to be higher,” she said. Among the ApoE4 carriers, 95 percent of trial participants tested reached the threshold on PIB PET scans for brain Aβ.
Another key point of discussion at EFNS centered on what to do going forward. “Why do we see evidence of an effect on downstream markers of neurodegeneration but still do not see a clinical effect?” asked Sperling. She suggested that perhaps the treatments are being given too late or the effects are too weak. On the latter, vasogenic edema forced the study sponsors, Janssen Alzheimer Immunotherapy and Pfizer, to drop the highest doses from the trials. Some immunotherapy trials deliver antibody monthly, not every 13 weeks. Other researchers suggested that combination therapy, perhaps with a gamma-secretase inhibitor, might be a better approach. Prins said it will be interesting to see subgroup analysis of the bapineuzumab data. Salloway told Alzforum that additional data will be presented at the American Neurological Association meeting in Boston in October. “We’d like to see an analysis specifically of the mild patients and also see behavioral data,” he told Alzforum. “That is yet to come.”—Tom Fagan.