Hype for Phase 1 Alzheimer Study More Impressive Than Results

Last week, the NYT, WSJ, and Wall Street reacted enthusiastically, if not irrationally, to results of a phase 1B (yes, phase 1B) study of another anti-amyloid MAb (yes, another anti-amyloid MAb) in Alzheimer disease–this time from Biogen Idec (or just Biogen now). The early-phase “interim-analysis” results for aducanumab (BIIB037) were presented at the AD/PD 2015 Conference in Nice, France, and Biogen’s stock (BIIB) got a hefty 10% 1-day boost on the news and the meeting attendees’ reactions to it.

Enthusiasm stems from the drug’s apparent dose-dependent effects on cognition (via the somewhat crude 30-point MMSE) and global function (via the CDR-SB) at up to 54 weeks in patients with prodromal or mild AD. This fervor was generated despite a very small number of enrollees (N = 166) in the early-phase trial. And consider these outcomes particularly when the study comprised 5 subgroups, each with only 30-40 patients. Nevertheless, the cognitive outcome was statistically significant for the low (3 mg/kg IV) and high (10 mg/kg IV) doses of aducanumab (vs placebo), and the global-function outcome was statistically significant for the high dose.

In addition, these clinical outcomes paralleled a reduction in the brain’s beta-amyloid burden (via SUVR, which was assessed by using Lilly’s amyloid PET tracer, florbetapir) at 26 and 54 weeks.

But balance these efficacy data, such as they are, with the dose-dependent rates of vasogenic edema (ARIA-E)–the same adverse event that brought down bapineuzumab (in addition to a lack of efficacy in phase 3 trials). (03/26/2015 addendum: The average composite SUVR change at 54 weeks for 1 mg/kg was provided in Biogen’s press release: -0.56 [NS].)

By comparison, the rates of ARIA-E with bapi (0.5 mg/kg) were 15% (vs 0.2% with placebo) in ApoE4 carriers (who generally show a greater amyloid burden on PET images). In ApoE4 noncarriers, a higher dose of bapi (1.0 mg/kg) was associated with an ARIA-E rate of 9.4% (vs 0.2% with placebo).

Given Biogen’s efficacy and safety results with the low and high doses of aducanumab, it appears that the company is now trying to “thread the needle” (to quote one analyst) by continuing to assess an intermediate dose of the MAb: 6 mg/kg IV. The ostensible hope is to generate positive clinical outcomes with this intermediate dose, while producing an acceptable rate of vasogenic edema in treated patients. Nevertheless, the rates of ARIA-E with aducanumab, particularly in ApoE carriers who received the higher doses, are astonishing (43% with 6 mg/kg; 55% with 10 mg/kg).

I would also offer a caveat regarding the cognitive outcome. In addition to the low n values, the 30-point MMSE is not a particularly sensitive outcome measure, especially when compared to the more usual clinical-trial barometer of the 70-point ADAS-cog11 or the 85-point ADAS-cog13 (the latter of which may be a more sensitive measure of cognitive change in earlier stages of AD). Biogen might counter that their results were so robust that they were observable even with a crude measure like the MMSE in early disease. Obviously there are at least 2 ways to spin this cognitive outcome–which remains to be reproduced.

Regardless, Biogen is charging forward with phase 3 development of aducanumab in AD (and leaping right over phase 2); although the trial (or trials) has not yet been registered at clinicaltrials.gov.

ADAS-cog = Alzheimer’s Disease Assessment Scale-cognitive subscale; ApoE = apolipoprotein E; ARIA-E = amyloid-related imaging abnormalties, vasogenic edema; CDR-SB = Clinical Dementia Rating Scale Sum of Boxes; MAb = monoclonal antibody; MMSE = Mini-Mental State Examination; PET = positron emission tomography; SUVR = standardized uptake value from 2 different regions.