Elan and Wyeth's Alzheimer's hope fails to impress

Friday 1 August 2008
Tom Moberly - Reporter

Elan and Wyeth have presented further disappointing results from a closely watched and failed Phase II trial of bapineuzumab, an investigational Alzheimer's disease treatment. Observers had had high hopes for the drug as it is the most developmentally advanced monoclonal antibody directly targeting beta-amyloid, a class of drugs thought to have the potential to reverse patients' neurodegeneration.

Shares in Elan and Wyeth fell 42% and 12%, respectively, on the New York Stock Exchange on July 29th following presentation of the data at the International Conference on Alzheimer's Disease (ICAD) in Chicago. However, the companies said the results of the Phase II trial nevertheless support their four ongoing Phase III studies.

Top-line results presented last month showed that bapineuzumab had failed to meet the trial's primary endpoint (Scrip Online, June 25th, 2008). Further post-hoc analyses presented at ICAD showed mixed results and the treatment also seems to be associated with an increased risk of vasogenic oedema, a condition in which excess fluid accumulates in the brain. The affected patients were successfully treated, however, and Elan and Wyeth believe the condition is manageable.

Vasogenic oedema occurred in 12 patients in the trial, all of whom had been given bapineuzumab. Although the drug did not show a clear dose-response relationship, the incidence of vasogenic oedema appeared to be dose-related, with eight of the cases occurring in patients given the highest dose (2mg/kg) and three in those given the next highest dose (1mg/kg). 10 of the 12 cases of vasogenic oedema occurred in patients carrying the gene that produces apolipoprotein (ApoE4), a lipoprotein known to be associated with an increased risk of developing AD. Elan and Wyeth said that, in response to the results of this Phase II trial, the 2mg and 1mg doses will not be used in patients carrying the ApoE4 gene in Phase III trials.

The one bright spot of the trial was that bapineuzumab appeared to be more effective in patients who were not carriers of the ApoE4 gene, 50% of the trial's population.

Bapineuzumab is a humanised monoclonal antibody designed to bind directly to beta-amyloid and clear it from the brain, thereby preventing it from forming the beta-amyloid plaques thought to be a major cause of the symptoms of Alzheimer's disease. Patients in the trial were given either bapineuzumab (0.15, 0.5, 1.0 or 2.0mg/kg) or placebo, in addition to any other Alzheimer's medicines they were taking. Over 95% of the patients in the trial were taking acetylcholinesterase inhibitors, such as Eisai/Pfizer's Aricept (donepezil), or the NMDA receptor antagonist memantine (Forest Laboratories' Namenda).

trial data

The study's primary efficacy endpoints were change from baseline in scores on the cognitive subscale of the Alzheimer's Disease Assessment Scale (ADAS-cog) and on the disability assessment scale for dementia (DAD), which assesses ability to carry out functional activities. Besides not meeting the primary endpoint in the intent-to-treat population, a number of post-hoc analyses, examining subpopulations and using variations in statistical modelling, failed to show statistically significant improvements in favour of bapineuzumab.

However, an analysis looking only at patients who had completed treatment (63% of bapineuzumab patients and 71% of placebo patients) showed statistically significant improvements on ADAS-cog and DAD for all doses of bapineuzumab over placebo (p=0.003 and p=0.043, respectively).

A further post-hoc analysis, which was conducted in response to the vasogenic oedema findings and the known risk with the ApoE4 gene, compared ApoE4 carriers and non-carriers. In patients carrying the ApoE4 gene, treatment with bapineuzumab showed a trend towards improvements in ADAS-cog, but this was not statistically significant (0.9-point improvement, p>0.10). However, among non-carriers (50% of patients in the trial), all doses of bapineuzumab produced statistically significant improvements in ADAD-cog (5.0-points over placebo, p=0.026), though still not in DAD (6.9-point improvement, p>0.10).

Aside from vasogenic oedema, the adverse events seen in the trial were generally mild to moderate, transient and not dose related, Elan and Wyeth said. Adverse events which occurred at least twice as frequently in bapineuzumab-treated as in placebo-treated patients included anxiety, back pain, gait disturbance, hypertension, muscle spasms, paranoia, skin laceration, vomiting and weight loss. In addition, although there was no significant change in brain volume, bapineuzumab did lead to a small but statistically significant increase in ventricular volume compared with placebo (2.5cc increase, p=0.037). Wyeth and Elan said they did not know what had caused this increase.

beta-amyloid

Pfizer and Lilly also have monoclonal antibodies targeting beta-amyloid in development. Pfizer's PF-04360365 is in Phase I development and results from a Phase II safety trial showing that Lilly's LY2062430 was well tolerated were presented at the ICAD meeting. Bapineuzumab, PF-04360365 and LY2062430 bind to different sections of beta-amyloid. In conference calls following presentations at ICAD, Pfizer said it believed that PF-04360365 has an advantage of lower immunogenicity, while Elan and Wyeth said that the binding site of bapineuzumab allowed it to attach to a variety of types of beta-amyloid important in Alzheimer's disease (in addition to beta-amyloid in plaques).

However, the potential benefit of targeting beta-amyloid has been brought into question by a paper published by Clive Holmes from the University of Southampton, UK, and colleagues in The Lancet in July. This showed that, although Elan's antibody AN1792 was able to clear patients' amyloid plaques in a Phase I trial, it did not prevent the progression of neurodegeneration. Development of AN1792 was discontinued in 2002, following reports of CNS side-effects associated with the product. Bapineuzumab differs from AN1792 in that it delivers antibodies to beta amyloid directly to the patient, rather than stimulating the patient's own immune system to produce antibodies against the peptide, which it was hoped would reduce safety concerns associated with eliciting an immune response. An editorial commenting on the Lancet paper points out that removal of amyloid does not tackle other features of the disease, such as accumulation of tau. It suggests that although anti-amyloid therapies may not, on their own, be able to halt Alzheimer's, they might have a role in combination with other therapies or as a prophylactic treatment.

• Not a subscriber? Request a FREE trial
• Subscribe to Scrip
• Email to a friend

• Facebook
• digg
• del.icio.us
• furl