14 January 2013. That sagacious beast may be able to read your mind now, but there's a good chance Fido will lose his smarts and develop Alzheimer's-like dementia as he ages. Three out of four old dogs will deposit amyloid plaques in the brain. Brain atrophy, phosphorylated tau, and changes in cerebrospinal fluid amyloid-β are common in old dogs, as well, creating a confluence of AD biomarkers that is fairly rare in the animal world. As treatment success in mouse models time and again has given way to disappointment in clinical trials, some researchers think the field has been barking up the wrong tree. Now, three companies are partnering to develop a canine model as a pre-competitive tool for the field. InterVivo Solutions, Inc., C2N Diagnostics, and inviCRO intend to develop, standardize, and validate a preclinical model of AD in dogs. Researchers in the pharmaceutical industry welcomed the approach, but many consider the cost and time needed for testing to be impediments to widespread use of dogs for drug development, and held off on public comment for the time being.
InterVivo develops animal models for various human diseases, including AD. C2N specializes in measuring fluid biomarkers, including protein synthesis rates, while inviCRO provides in-vivo preclinical imaging analysis. As outlined in a press release, the companies will collaborate to assess dogs in a longitudinal study modeled on the Alzheimer's Disease Neuroimaging Initiative (ADNI).
"Many pharmaceutical companies want to know when those [pathological] changes begin in dogs, how they vary, and how at-risk canines can be selected for study," said Joseph Araujo of InterVivo Solutions. Araujo anticipates a four-year study with the potential to expand if the data warrant it.
The study will use three groups of beagles at different ages. Following young, middle-aged, and older dogs will give the researchers an opportunity to mark what pathological changes occur and when. Araujo told Alzforum that amyloid usually begins to deposit in dogs around late middle age, when they are eight to 10 years old. Brain structure changes at that age as well. "We plan to have a good number of subjects in that group, and within four years we'll be able to study how those change relate to each other," said Araujo.
"I think this is a great initiative," said Elizabeth Head, University of Kentucky, Lexington. Head has been studying dementia in dogs for more than 15 years. "We've been doing this kind of work in the field, but putting it all together in one large study will be incredibly informative," she told Alzforum. She added that using three groups of animals will allow the researchers to combine cross-sectional and longitudinal studies with the possibility of getting useful data within a year.
Joel Braunstein of C2N Diagnostics told Alzforum that the consortium will take quarterly measures of six to eight biomarkers, including Aβ and phospho-tau. For now, the companies are not saying exactly what biomarkers they plan to look at. C2N was set up to capitalize on a stable isotope labeling method developed by Randall Bateman at Washington University, St. Louis, Missouri, to measure the dynamics of Aβ in the CSF (see ARF related news story). Braunstein said C2N will use this methodology to follow the appearance of newly synthesized proteins in dog CSF as well.
Beagles will be tested for cognition every six months using a variety of protocols to measure visuospatial working memory (delayed non-matching to position task), executive function, selective attention, and visuospatial perception. InterVivo has expertise in this area. It was spun off from CanCog, a company co-founded by William Milgram, University of Toronto, who worked to develop canine models of cognitive decline. Imaging analysis at inviCRO will measure functional and structural changes in the brain on a yearly basis using FDG-PET and MRI, respectively. They do not plan to image amyloid plaques at this time. Magnetic resonance spectroscopy, which measures metabolite concentrations, will be performed yearly as well.
Whether drug developers will adopt a canine model is unclear. "At $600-$700K per year for a 40-dog study, the cost may seem prohibitive," said Head. She added that this is a drop in the bucket compared to the cost of clinical trials. "This could be a good model for human AD, but the tradeoff is the expense and the time needed."
The companies are looking for partners to work with this model. "We will solicit proposals from various parties and evaluate how well they would work going forward," said Araujo. Partners could be from industry or they could be public or private foundations, said Braunstein. Though some proprietary information could be kept private, the troika's overriding goal is to share data from this project through the peer-reviewed literature, Braunstein said.—Tom Fagan.