Treatment and Drug Development
Current treatment approaches for AD in humans are focused on helping people maintain mental function, manage behavioral symptoms, and slow or delay the symptoms of the disease. Unfortunately, there is no effective treatment for AD. Drugs that are used today in the management of AD can only alleviate the symptoms, and even that only temporarily. Cholinesterase inhibitors (donepezil, rivastigmine, and galantamine) have been approved for the management of AD in humans. Drugs acting as acetylcholinesterase inhibitors reduce the activity of the acetylcholine esterase which degrades acetylcholine, thus increasing the amount of acetylcholine available in the brain and therefore stimulate brain cells which receive more synaptic inputs (reviewed in Winblad et al., 2016). This very unspecific treatment can slow the disease progression for 6–12 months, and even achieve temporary improvement as shown for donepezil (Birks and Harvey, 2018), but does not cure the disease which will eventually progress. Another drug used for the treatment of AD is memantine that acts on the glutamatergic system by blocking NMDA (N-methyl-D-aspartate) receptors. A recent random-effects network meta-analysis of 41 randomized controlled trials reveled the most suitable dosages of cholinesterase inhibitors and memantine to treat patients with mild, moderate and severe AD (Dou et al., 2018). Although these pharmacological interventions have beneficial effects on cognition, function and global clinical impression, these treatments do not alleviate the neuropsychiatric symptoms (Dou et al., 2018). Unfortunately, there have been numerous clinical trials with many candidate drugs, but most of these had negative outcomes (Cummings et al., 2018).

Treatment of Canine Dementia
Current treatment options for CCD target prevention, slowing and/or improving the cognitive decline in dogs. Some drugs or food supplements are available for senior dogs and might act neuroprotective. Some enhance the blood flow into the brain, others work as antioxidants and more effort is now directed to slowing the progression of the disease instead of providing only symptomatic treatment. One commonly prescribed drug for cognitive impaired dogs is selegiline, which acts as an inhibitor of monoamine oxidase B (MAOB), thus reducing degradation of several neurotransmitters in the brain, and may have neuroprotective effects on dopaminergic, noradrenergic and cholinergic neurons (Landsberg, 2005; Magyar, 2011). Another drug that is occasionally used is nicergoline, which increases the blood flow through the brain. It may enhance neuronal transmission and act neuroprotective, increase dopamine and noradrenaline turnover and inhibit platelet aggregation (Landsberg, 2005). Propentofylline also has a neuroprotective role as it inhibits the production of free radicals and reduces the activation of microglial cell, thus acting anti-inflammatory (Frampton et al., 2003). Antidepressants such as selective serotonin reuptake inhibitors fluoxetine and sertraline, amitriptyline, paroxetine and anxiolytics benzodiazepines, gabapentin, valproic acid and buspirone can also be used to treat the anxiety and aggression which may accompany CCD. Clomipramin is an antidepressant commonly prescribed for dogs with anxiety (Landsberg, 2005), but these are all symptomatic treatments and do not treat the disease itself. S-adenosylmethionine tosylate supplementation was reported to be safe and effective in improving signs of age-related mental decline in dogs (Rème et al., 2008).
There are also some nutraceutical preparations available for dogs, which are based on natural products and/or supplement formulations. Behavioral enrichment alongside with antioxidant-rich diet and exercise is an approach for maintaining cognitive function and slowing the progression of CCD in senior pets. As means of preventative intervention, aging beagles were fed a diet rich in antioxidant, which improved cognition, maintained cognition and reduced oxidative damage and Aβ pathology in treated dogs (Milgram et al., 2004; Dowling and Head, 2012). Another longitudinal survey in beagles looked at the proteomic changes following administration of antioxidant-rich diet in combination with behavioral enrichment (Opii et al., 2008). Following treatment, the levels of oxidative stress biomarkers decreased and the increased expression levels of Cu/Zn superoxide dismutase, fructose-bisphosphate aldolase C, creatine kinase, glutamate dehydrogenase and glyceraldehyde-3-phosphate dehydrogenase correlated with improved cognition (Opii et al., 2008). In addition, some other studies implicated the nutrition as cognition modifying factor in dogs (Araujo et al., 2005a; Siwak et al., 2005; Osella et al., 2007; Christie et al., 2009; Head et al., 2009; Snigdha et al., 2011, 2012; Katina et al., 2016; Chapagain et al., 2018), highlighting combination of nutraceutical supplements directed at several mechanisms of pathological aging, in combination with behavioral enrichment, as more effective (Araujo et al., 2005a, 2008). Dogs receiving both an antioxidant-rich diet and environmental enrichment showed increased levels of brain-derived neurotrophic factor (BDNF) mRNA when compared to untreated aged dogs. As a result of increase in BDNF mRNA, the cognitive performance improved and the amount of cortical Aβ deposits decreased (Fahnestock et al., 2012). Interesting is the finding that neuronal loss in the hippocampus, occurring in the aged dogs, could be partially reversed by more engagement with the dog, i.e., with stimulation of brain function (Siwak et al., 2000).

Animal Models for Developing Novel Treatments
In comparison to transgenic mouse models, natural animal models better represent the pathophysiology of AD. Models of “physiologically” aged rats, degus and dogs are useful for studying mechanistic aspects of AD, which are also very valuable in the development of therapeutics that would alleviate age-related declines in cognitive function. Mouse models for AD research carry mutations, found in familial AD, and are artificially accumulating Aβ plaques and NFTs. Whether mice are good models have been thoroughly discussed elsewhere (Götz et al., 2018). Several drugs have cleared the amyloid load in mice but failed to do so in people with AD. Recently the genetic background and environmental factors have been demonstrated as the variability in AD development, which was partially recognized by incorporating genetic diversity into mouse models of AD (Neuner et al., 2018). Transgenic minipigs expressing APP695 or PSEN1 have also been developed but have not shown the histopathological nor the cognitive impairment signs (Holm et al., 2016). Therefore, natural animal models of species with spontaneously occurring neurodegeneration are potentially more useful in developing and testing novel treatments for such diseases.
To date, most of the drugs in development for AD treatment have been directed toward the removal of amyloid plaques or NFTs, not taking into the account the multifactorial causation of the disease. Several experimental drugs that have successfully removed plaques from mouse brains have not lessened the symptoms of AD in people. For instance, drugs acting as BACE1 (beta-site amyloid precursor protein cleaving enzyme-1) inhibitors had failed in Phase II/III clinical trials (Hawkes, 2017; Dobrowolska Zakaria and Vassar, 2018; Egan et al., 2018). A BACE1 inhibitor verubecestat successfully blocked the accumulation of amyloid protein in mice (Villarreal et al., 2017), rats and monkeys (Kennedy et al., 2016), but did not reduce cognitive or functional decline in patients with mild to moderate AD (Egan et al., 2018). Although decrease in Aβ biomarkers in CSF and brain has been noticed after treatment with BACE1 inhibitors, the failure to prevent cognitive decline might have been due to irreversible neurotoxic accumulation of Aβ that occurred prior to the start of the treatment. For this reason, focus is on the development of therapies that commence at presymptomatic stage (preclinical stage and the stage of mild cognitive impairment) although for this, development of novel, useful biomarkers, is also crucial.
As canine cognitive decline and human Alzheimer’s disease show neuropathological, cognitive and behavioral parallels, the testing of products for the treatment of AD in canine model could be useful to determine the efficacy of these compounds in humans, and also to develop novel therapeutic agents for the treatment of senior dogs. Several drugs have been tested in elderly dogs and their suitability and effectiveness correlated with results obtained in human trials, when available. Drugs tested in dogs are listed in Table 3.
TABLE 3  Pharmacological interventions tested in dogs with cognitive decline.
Name/type   Mode of action   Testing in dog   Results/outcomes   References
LY2886721  BACE1 inhibitor  Pharmacology study in dogs and clinical trial in healthy volunteers  Aβ lowering in plasma and CSF  May et al.,2015
Selegiline (L-deprenyl)  MAOB inhibitor  Longitudinal study  Higher life expectancy (cognitive status not monitored)  Ruehl et al.,1997
Selegiline (L-deprenyl)  MAOB inhibitor  Performance studies  Improved visuospatial working memory (in only a subset of dogs)  Head et al., 1996; Campbell et al., 2001; Studzinski et al.,2005
Atorvastatin  Reduction of Aβ and BACE1  Longitudinal study  Neuroprotective  Barone et al.,2012
Adrafinil  A wakefulness-promoting agent (eugeroic) with nootropic effects  Longitudinal study; pharmacological study  A significant increase in locomotion; improved learning; impaired working memory  Siwak et al., 2000; Studzinski et al.,2005
Ampakine  Positive modulator of AMPA receptors (enhance excitatory glutamatergic neurotransmission)  Pharmacological study  Decrease of performance accuracy; may have memory impairing effects  Studzinski et al.,2005
CP-118,954  Acetylcholinesterase inhibitor  Pharmacological study  Minimal cognitive enhancing effects  Studzinski et al.,2005
Phenserine  Acetylcholinesterase inhibitor  Pharmacological study; performance study  Enhancing effects on memory and learning; improved performance (only in a subset of dogs treated)  Studzinski et al., 2005; Araujo et al.,2011a
Donepezil  Acetylcholinesterase inhibitor  Performance study  memory enhancement  Araujo et al.,2011a
CNP520  BACE1 inhibitor  Dogs, human  Safe to use in dogs; tolerated in healthy humans; ongoing clinical trials  Neumann et al.,2018
Antioxidant-rich diet with cognitive enrichment  /  Dogs  Improved cognition  Opii et al., 2008; Dowling and Head,2012
Anti-Aβ immunotherapy  Passive vaccination with injections of antibodies against Aβ42  Dogs  Reduced amyloid plaques and reduced astrogliosis  Head et al., 2008; Neus Bosch et al., 2015; Davis et al.,2017 BACE1 is a protease that controls the formation of Aβ and most likely plays an important role in the development of pathogenesis in AD. The usefulness of BACE1 small-molecule inhibitor LY2886721 has been tested in a dog model and in humans in the first clinical phase (May et al., 2015). It significantly reduced plasma and CSF Aβ levels both in dogs and healthy volunteers (May et al., 2015). After administration of two BACE1 inhibitors (cyclic sulfoxide hydroxyethylamine NB-B4 and oxazine derivative NB-C8) a unique pattern of secreted Aβ peptides was observed in canine CSF (Mattsson et al., 2012). Besides the expected reduced levels of Aβ40 and Aβ42, reduced levels of Aβ1–34 and increased levels of Aβ5–40 were detected, which were proposed as prognostic markers of BACE1 inhibition therapies (Mattsson et al., 2012). A recent survey, using BACE1 inhibitor CNP520, demonstrated reduced brain and CSF Aβ in rats, dogs, and reduced CSF Aβ in humans and was assessed to be well tolerated in healthy adults and further clinical trials are ongoing (Neumann et al., 2018). A cardiovascular disease drug, atorvastatin, has been investigated in a canine model of dementia whether it could reduce Aβ plaques, BACE1 protein levels and oxidative stress (Barone et al., 2012). It does offer some neuroprotective role through the up-regulation of enzyme biliverdin reductase-A (Barone et al., 2012). Selegiline (L-deprenyl), a type B monoamine oxidase inhibitor, has been reported to prolong the survival of aged dogs (Ruehl et al., 1997) and in some cases improve visuospatial working memory (Head et al., 1996). However, in subsequent studies only a small subset of dogs seemed to show improvement, pointing to a minor clinical relevance of selegiline administration (Campbell et al., 2001). Furthermore, several human AD trials provided no evidence of clinically meaningful benefits of selegiline administration for people with AD (Birks and Flicker, 2003). Interestingly, although selegiline was shown to be fairly ineffective in the treatment of CCD, it is still the only FDA-approved treatment for CCD.
Testing the same compounds in dogs and humans provided similar findings also in other studies. CP-118,954, an acetylcholinesterase inhibitor, showed minimal cognitive enhancing effects in dogs and human clinical trials were discontinued (Studzinski et al., 2005). Donepezil administration enhanced memory in dogs (Araujo et al., 2011a). Phenserine, an acetylcholinesterase inhibitor which also inhibits synthesis of Aβ and acts as a cognitive enhancing therapeutic, improved learning and working memory in geriatric dogs (Studzinski et al., 2005; Araujo et al., 2011a). Its administration showed similar results in Phase II human clinical trials, suggesting its effectiveness in improving memory in AD patients, but in Phase III human clinical trial, with patients with mild to moderate AD, no significant differences between the phenserine treated and placebo groups were noticed and the trial was discontinued (Thatte, 2005). There has been further discussions if this trial, and several other AD trials, failed due to procedural errors rather than due to a lack of drug efficacy (Winblad et al., 2010). In more recent studies patients with mild AD showed improvement in cognition following phenserine treatments, this was mirrored also by an increase in CSF Aβ40 (Kadir et al., 2008; Nordberg et al., 2015). There is another mode of action of phenserine in AD, namely inhibition of neuronal self-induced preprogrammed cell death, which further clinical trials might pursue (Becker et al., 2018). Ampakine (drug that alters the glutaminergic system) also failed in human AD trials and in dogs with age-associated memory disorder and dementia showed insignificant memory-enhancing effects after treatment with ampakine (Studzinski et al., 2005). Dogs, treated with adrafinil, a mild central nervous system stimulant used to relieve excessive sleepiness and inattention in elderly patients, were more attentive (Siwak et al., 2000). Cholinergic hypofunction might play a role in age-dependent cognitive decline in dogs (Araujo et al., 2011b), this was determined by administration of muscarinic cholinergic receptor antagonist scopolamine, which induces transient memory impairment. Similar findings were presented for aged and demented humans (Schliebs and Arendt, 2011). All these suggest that dogs might be a very good preclinical model for developing and testing new drugs for AD in humans.
Besides BACE1 and cholinesterase inhibitors several attempts have been made to develop immunotherapy treatments directed against Aβ or TAU (Bittar et al., 2018). A range of TAU-targeted immunotherapies have entered clinical development (recently reviewed in Novak et al., 2018). Clinical trials in patients with AD, which were administered Aβ peptide with conjugate to stimulate the immune response (i.e., active immunization) or anti-Aβ antibodies (i.e., passive immunization) have shown modest results and also serious side effects (reviewed in detail in Wisniewski and Drummond, 2016).
In dogs, Aβ immunotherapy could reduce the presence of amyloid plaques and astroglial reaction in aged individuals (Neus Bosch et al., 2015). A vaccine directed against fibrillary Aβ (anti-Aβ42) reduced the presence of Aβ plaques in canine prefrontal cortex and improved the function of the frontal cortex, thus resulting in cognitive benefits (Head et al., 2008). Anti-Aβ42 vaccination of dogs with CCD, in combination with behavioral enrichment, resulted in reduced presence of Aβ plaques in the brain, a significant maintenance of learning abilities over time and cognitive maintenance with no improvement in cognition (Davis et al., 2017). This anti-Aβ vaccine, however, also increased the frequency of brain microhemorrhages (Davis et al., 2017) which is commonly observed in the development of these immunotherapies also in human trials and could be very serious side effect. Similarly to pharmacological compounds, immunotherapies in both dogs and humans have provided modest improvements at most, yet similarities between human and canine patients again highlight the parallels between diseases and usefulness of dogs as a preclinical model for AD.