Alzheimer’s Disease: An Overview of Recent Developments and a Look to the Future

There is a growing amount of data suggesting that the prevalence of Alzheimer’s disease (AD) is increasing worldwide. A review and meta-analysis of global prevalence from 1980 to 2013 of individuals more than 60 years of age estimated that 46.8 million people worldwide were living with dementia in 2015, and this number was projected to nearly double every 20 years (1). There is also a significant trending increase in reported AD prevalence in developing countries (e.g., China, India), which tend to have younger populations compared with the United States and Western Europe (2). Interestingly, Sub-Saharan African regions had estimated prevalence rates as low as 2.07%, compared to 8.50% in Latin America in the 2011 World Alzheimer Report. However, updated data in 2015 showed prevalence in African regions had increased to about 5.5% (1). Researchers did not attribute the increase to greater prevalence of AD in Africa; rather it was a result of more available data to produce better prevalence estimates since 2011 (1). In East Asian countries, a decreased prevalence of vascular dementia and an increase in AD diagnosis have been observed, and although this shift is thought to be attributed, at least in part, to risk-factor modification, it is also attributed by methodological differences between studies (3). China, for example, has been challenging to study because of its large geographical area and widely diversified population. This lack of data also limits the ability to establish incidence rates, which have been reported as high as 24 per 1,000 people in China (4). To address this issue, a large multiregional prospective cohort dementia incidence study in China published in 2015 found that the age-specific incidence rate of dementia in China may in fact be similar to rates in developed countries, with the rate of dementia exponentially increasing in older age cohorts (4).

The methodological variance in studying at-risk populations will likely continue to change as pathological processes seen in AD are now identified prior to the development of actual dementia symptoms. With early detection possible because of the development of technology using specific cerebral spinal fluid markers, magnetic resonance imaging (MRI) volumetric techniques to measure atrophy, 18F-fluorodeoxyglucose (FDG) tracer, and amyloid positron emission tomography (PET), the National Institute on Aging–Alzheimer’s Association (NIA-AA) has revised the diagnostic criteria for AD. The NIA-AA’s “preclinical” phase is currently defined by three stages with key biological changes in the central nervous system. The first stage includes the identification of PET amyloid tracer retention and low cerebrospinal fluid (CSF) β-amyloid 1–42. The second phase is further defined by neuronal dysfunction measured through FDG, PET, and functional MRI as well as cortical thinning and hippocampal atrophy. The third phase includes the criteria described earlier with either very subtle or no clinical symptoms for up to decades prior to the development of mild cognitive impairment (MCI) and AD (5). Although the International Working Group (IWG) also addresses asymptomatic and symptomatic stages of AD, the IWG approaches terminology somewhat differently. For example, the IWG calls the MCI-AD stage “prodromal AD” and does not recognize the equivalent of NIA-AA’s stage 3, which accounts for subjective cognitive impairments. Also, the IWG’s criteria strongly emphasize an impairment of encoding memory, which is not always present in atypical AD (6). Both NIA-AA and IWG rely strongly on biomarker pathology, although, as could be expected with a three-stage system, the NIA-AA is more specific on how these biomarkers are combined (6). DSM-5 takes a yet different approach with its terminology for minor neurocognitive disorder (equivalent to MCI) and major neurocognitive disorder (dementia).

Biomarker testing has been a growing diagnostic approach in the field of dementia. The focus in the clinic has until recently primarily been on tau protein and β-amyloid 1–42 peptides, their relative concentrations, and metabolic and volumetric brain imaging studies. In the CSF, a ratio of β-amyloid 1–42 to total tau of less than 1.0 was reported to be 85%−94% sensitive and 83%−89% specific in distinguishing AD from non-AD within a population, and a phosphorylated tau level of >61 pg/ml has a sensitivity of 72%−85% and specificity of 74%−85% in distinguishing AD from other dementias (7). In addition, CSF biomarker testing has proven to be a relatively safe procedure for most older adult patients and has been reported to have more than 80% accuracy in identifying conversion from MCI to AD (8). However, these studies tended to be in relatively homogeneous samples; therefore, the accuracy rates would be lower in diverse clinic samples, and accuracy can vary widely with type of assay used. The emergence of assays that measure soluble amyloid oligomers (the presumed toxic species) may help make this test more precise. Positive amyloid PET and spinal markers have also been linked to higher rates of conversion from MCI to dementia, but to date there is no validated or Food and Drug Administration (FDA)–approved test to predict future AD among at-risk individuals. Likewise, although spinal and imaging tests of β-amyloid and tau are highly promising for clinical differential diagnosis, none of these tests are fully validated to diagnose any specific disease such as AD.

PET amyloid imaging is FDA approved to detect the presence or absence of fibrillary amyloid, which, in turn, can aid the clinician in lowering the probability for AD if the test is negative. PET amyloid imaging has a high negative predictive value but has only a modest positive predictive value for AD dementia (9). The standardization of CSF assays remains a work in progress, and tau-PET has yet to be fully validated in autopsy studies (10). Other biomarkers such as whole genomic sequencing, metabolomics, and proteomics are not yet validated for diagnosis or prediction. Because none of these tests alone give definite results among at-risk patients or patients with MCI, the combination of FDG-PET, amyloid PET, volumetric hippocampal MRI, and CSF studies leads to higher negative and positive predictive values in differentiating development of AD from typical aging (2).

At the same time, there are limits to the newer diagnostic approaches, including lack of clinical access to advanced technology, interlaboratory variability in testing, and patient ability to tolerate procedures. Knowing that AD is associated with inflammatory processes and that abnormal tau and amyloid depositions are found in blood, saliva, skin, and other tissues, researchers are focusing their efforts on peripheral biomarkers for AD testing. However, this technology is still early in development and is difficult to replicate (11). Another focus has been the use of mobile technology to test cognitive function in the aging population. For example, the popular Saint Louis University Mental State Examination has been made into a mobile app, but the lack of normative data to make this and similar applications valid and useful diagnostic options remains a great barrier (12).

In addition to research efforts toward earlier diagnosis of AD, considerable efforts have aimed at the development of new treatment options. Today, there is no proven therapy for disease modification of AD dementia or delaying the onset among at-risk individuals. Over the past several years, therapies have remained limited to cholinesterase inhibitors and memantine, both of which have evidence for only modestly improving cognitive function symptomatically. Memantine is approved only for moderate to severe stage dementia, whereas cholinesterase inhibitors are approved for mild through severe stages. A new dose formulation for donepezil, 23 mg, was approved by the FDA for moderate to severe disease in 2010 because it showed superior cognitive benefits in comparison with 10-mg dosing and a Severe Impairment Battery score mean change of 2.11, p<.001 (13). However, how much this change actually translates into functional improvement is up for debate. In the same study, the higher dose of donepezil also had more significant gastrointestinal side effects, which led to higher discontinuation within the first month of treatment in comparison with the 10-mg dose. Memantine also now has an extended-release formulation of 28 mg, which has shown significant advantage over placebo, but it has not yet been compared head to head with standard 10-mg dosing (14). The extended-release formulation has been combined with 10 mg donepezil for treatment of moderate to severe dementia, with evidence for overall additive (marginal) clinical benefits compared with monotherapies (15). It should be noted that these new more expensive formulations are generally not viewed as any more effective or useful than generic versions. Furthermore, the evidence for memantine-donepezil combination is based on pooled analysis of studies looking at only a relatively short period (six months) as well as some naturalistic studies with selection biases; hence, the long-term clinical effectiveness is unknown. Off-label use of memantine in MCI and mild AD is widespread but, in our view, is not supported by available evidence.

Given the strong evidence for β-amyloid aggregation in AD, many companies have directed therapies toward preventing amyloid plaque formation or increasing clearance. There have been several recent trials of passive immunotherapies—monoclonal antibodies to β-amyloid. These immunotherapies include solanezumab, bapineuzumab, gantenerumab, crenezumab, and ponezumab. Solanezumab is the intervention in the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease study, which targets cognitively typical populations at risk for AD between 65 and 85 years of age who have a positive amyloid PET scan. None of these therapies have demonstrated significant positive results, and some of them, including gantenerumab and crenezumab, have been associated with vasogenic edema, microhemorrhages, and other adverse effects (14). Other approaches—such as multiple active immunotherapy agents, γ-secretase inhibitors of Aβ synthesis, and β-secretase inhibitors targeting amyloid precursor protein (APP) cleavage—are either undergoing trials or have been interrupted for various reasons, including adverse events. Very early trials are also under development for reduction of tau oligomers and neurofibrillary tangles, including the vaccines AADvac1 and ACI-35 (16). Companies are also testing other strategies, such as 5HT6 antagonists, gene therapy, brain stimulation, nutraceuticals, aerobic exercise, and regenerative (e.g., stem cells) therapies (17).

In addition to the strategies listed above, there are several ongoing AD prevention trials focused on identifying correct target populations and modifying risk factors. The Alzheimer’s Prevention Initiative is currently collecting composite cognitive test score data from unimpaired known carriers of the presenilin mutation, with the goal of better identifying adults at risk for developing AD (18). The ENLIGHTEN trial focuses on lifestyle modification, and it aims to address the independent effects of exercise and nutritional interventions on cognition of adults at risk for cognitive impairment. Started in 2011, its objectives are to measure cardiovascular function and cognition among patients with MCI, cardiovascular risk factors, and sedentary lifestyles. Participants were randomly assigned for six months to one of four arms—including an aerobic exercise arm, a Dietary Approaches to Stop Hypertension (DASH) diet arm, a combined arm with the two interventions, and a health education control group—with primary outcome measures including executive function and blood pressure (19).

The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) is another randomized controlled trial (RCT) focused on lifestyle modification that differs from the ENLIGHTEN trial in that it not only includes diet, exercise, and cardiovascular risk monitoring but also provided cognitive training for the intervention group (20). The population studied had cardiovascular risk factors for dementia and mean cognitive functioning that was either typical or slightly lower than expected for age. The primary endpoint of FINGER was a change in cognition measured by a comprehensive neuropsychological test battery (NTB). Difference in NTB scores between the intervention and care-as-usual groups was .02, and this result was statistically significant (p=.030; [20]). Results of this large RCT are encouraging, and the interventions tested in the FINGER trial could help maintain or even improve cognition in the growing at-risk population. The TOMMORROW study (www.tommorrowstudy.com) also focuses on prevention in that it identifies at-risk adults between 65 and 83 years of age who test positive for APOE and TOMM40 genes; the study is also testing whether the two genetic polymorphisms together have greater predictive value. The intervention in this study is a low-dose oral hypoglycemic agent, pioglitazone, which appears to reduce amyloid deposition in some animal models.

In cases of autosomal-dominant early-onset AD (EOAD), an approach to prevention through advanced reproductive genetics has been reported. Patients who have inherited a mutation in the genes for APP or the genes for presenilin protein (PSEN1 and PSEN2) have a 50% chance of passing these genes to their offspring. Unfortunately, these genes have near 100% penetrance, meaning that offspring who inherit the genes will develop EOAD. New techniques now allow the selection of embryos without inheritable Alzheimer genes for in vitro fertilization, which may give the option to parents carrying the abnormal gene to have unaffected biological offspring. This process is called Preimplantation Genetic Diagnosis (PGD), and it has been applied for several fatal single-gene mutations, such as Huntington’s disease and prion dementia (21). The first case of PGD for EOAD was reported about a decade ago, and several additional cases have recently been reported. Because familial EOAD affects multiple generations in a family system (and the children of symptomatic parents are often at a young reproductive age), it is important to refer family members to genetic counselors to discuss risks and benefits associated with finding out one’s genetic status and to discuss family planning options.

Family systems are strongly adversely affected by AD. This result may be due in part to the genetic burden of EOAD and late-onset AD but also to neuropsychiatric symptoms that patients develop during the course of their illness. Symptoms including agitation, paranoia, and hallucinations are not uncommon and are accompanied by behavioral disturbances such as violence or refusal of care. These disturbances are the primary cause of institutionalization of patients with AD, yet there are very limited data supporting the use of psychotropic medications to treat such symptoms (22). As an added concern, the use of antipsychotics carries a significant mortality risk among patients compared with placebo (23). For these reasons, it is also important to recognize in a timely fashion caregiver burnout (e.g., stress, depression, insomnia, physical health changes) and refer them for practical help and psychiatric treatments when indicated.

Currently, the medications quinidine and dextromethorphan in combination, citalopram, antipsychotic brexpiprazole, and cannabinoid nabilone are undergoing clinical trials for use by patients with AD who have behavioral disturbance (24). In September 2015, dextromethorphan-quinidine, a drug combination approved for pseudobulbar affect, was shown to decrease agitation and aggression scores among patients with AD in a placebo-controlled trial. However, treatment was associated with increased risk of adverse effects, such as falls (25). A modified version of this combination with a lower dose of quinidine, called AVP-786, has received fast-track designation and is undergoing two 12-week, phase 3 clinical trials for the treatment of agitation and aggression among patients with AD. The Citalopram for Agitation in Alzheimer Disease study found that patients who responded to citalopram (up to 30 mg/day) were more likely to have moderate agitation at baseline and lower levels of cognitive impairment (26). In small trials, medications targeting the endocannabinoid system, such as nabilone and dronabinol, have shown antioxidative and anti-inflammatory effects that may be beneficial in agitation and aggression among patients with AD. Recruitment is underway for a trial using nabilone. Brexpiprazole is a derivative of aripiprazole, and it is being used in a trial to treat agitation of participants with AD.