Alzheimer’s Disease and Immunotherapy

Abstract

Alzheimer’s disease (AD) is a growing health care epidemic. It is the most common cause of dementia and its incidence is rising. Age, which influences the oxidative and inflammatory states of the brain, is the most important risk factor. Currently there is no disease modifying treatments available for this irreversible, progressive debilitating disease. Immunotherapy represents an emerging, potentially disease modifying strategy aimed at reducing the pathological lesions of AD and facilitating cognitive improvement. Many clinical trials are currently underway. This literature review highlights current knowledge regarding the physiology of aging and how it relates to the pathogenesis of AD. In addition, immunotherapy is discussed in the context of its mechanism, current studies and future goals.

Statistics and Significance

Dementia is a progressive debilitating syndrome of dysfunction in several intellectual domains including memory, language, visual spatial ability, praxis, gnosis and executive functioning. Alzheimer’s disease (AD), the most common cause of dementia, affects 18 million people worldwide including 5.3 million Americans. It is an irreversible disease that causes severe functional impairment and personality changes that lead to the patient being completely dependent on their caretakers. AD is a growing health care epidemic. The incidence is expected to approach nearly a million people per year, with a total estimated prevalence of 11 to 16 million people in the USA in 2050 [1]. The major reason for this rapid growth is that the people at greatest risk are the elderly, the fasting growing segment of the population. AD is the sixth leading cause of all deaths in the United States, and the fifth leading cause of death in Americans aged 65 and older. It is estimated that nearly 20% of the population between ages 65 to 75 has AD (http://psycnet.apa.org). Significant cost implications related to AD and other dementias include an estimated 148 billion dollars annually.

Neurological Physiology due to Aging

Age is the most important risk factor for AD. Aging is the natural process that reduces the ability to cope with physiological and psychological insults. There has been a growing interest to understand the process of aging and how this relates to the neurodegeneration commonly observed in the elderly. In particular research is evaluating how brain aging is related to increasing mortality and morbidity. Aging is associated with mitochondrial dysfunction which causes increased oxidative stress [2]. Recent evidence indicates there is an elevated inflammatory profile in the aging brain consisting of an increased population of reactive microglia cells [3]. These factors set the stage for age-related neurodegenerative diseases such as AD and Parkinson disease and cause the aging brain to be more vulnerable to immune and ischemic insults [4, 5]. More cognitive and functional decline occurs after a stroke in patients aged 65 or older compared to the younger population [5]. The aging brain exhibits hyperactive responses to stimulations such as lipopolysaccaride (LPS). This evidence provides an explanation to the clinical observation that the elderly do not recover well from systemic infections. Systemic infections can trigger irreversible cognitive decline and cause exacerbation of neurodegenerative diseases [6]. Mood can also be affected. Chronic inflammatory diseases such as coronary artery disease and rheumatoid arthritis have been linked to depression in the elderly, which may be triggered by chronically elevated cytokines [7].

Prolonged illness in the elderly is associated with cognitive decline which is an independent risk factor for morbidity and mortality [8].

The Immune System and Aging

It is well established that one of the intrinsic factors associated with aging is attenuated immune function or immunosenescence [3]. Both innate and adaptive immunity is impaired. Specifically there is a reduction in the number of naive T cells, which are critical for mounting both T-cell-mediated and humoral-mediated (B-cell) adaptive immune responses to novel antigens [9]. Instead the T-cell response is shifted towards memory cells. The lack of T-cell response results in diminished B-cell proliferation and low antibody titers during vaccinations. This suppressed response increases the risk of severe infections complicated by lack of appropriate clinical signs. There is increased immune activity and inflammation in the aging brain mediated by astrocytes and microglia [10]. These reactive glial cells produce cytokines inducing the pro-inflammatory state of illness which contributes to neurodegenerative decline. During a normal inflammatory response cytokines regulate communication between the immune system and the brain via physiological changes of the blood-brain barrier (BBB) [11]. Toll-like receptors are expressed in the brain and respond to peripheral inflammation [12]. Alterations in the permeability of the BBB and chemotaxis may permit the recruitment and passage of peripheral cells into the brain parenchyma. Determining the detailed mechanism of this process is an active area of research. Investigators are exploring the processes involved in both the passage of inflammation into, and the effect of cytokines on, the central nervous system (CNS) [13]. Within the CNS cytokines mediate the signs and symptoms of illness, including fever, increased sleep, reduced appetite, lethargy, and decreased social behavior [10]. Once activated, microglia cells, which are derived from the macrophage lineage, produce cytokines and activate complement. Exaggeration of this response or inability to attenuate it contributes to the heightened neuroinflammation associated with the aging brain. Cytokines have been implicated to play important roles cognitive decline in aging animal models [14].

AD Pathogenesis

The molecular mechanisms leading to AD are largely unknown. The macroscopic manifestation is diffuse brain atrophy. Microscopically, there are two neuropathological lesions. Extracellular beta-amyloid (A-beta) peptide deposits (senile plaques) are the early AD lesions. A-beta is a 40 residue membrane-spanning protein whose physiological function is unknown. It is cleaved from a larger precursor protein. The significance of this processing in normal brain physiology is unknown. Intracellular neurofibrillary tangles are the late AD lesions. These contain hyperphosphorylated tau, a microtubule associated protein. Both of these abnormalities involve the accumulation of insoluble aggregated proteins due to inappropriate protein conformations. A better understanding of this pathology can be helpful in the development of early intervention therapies.

A-beta plaques accumulate in the cortex leading to cellular damage and cerebral amyloid angiopathy. Mitochondrial dysfunction, which is associated with the accumulation of reactive oxygen species (ROS), appears to play a role in these early events [2]. There is evidence that ROS initiates erroneous A-beta processing [15]. In addition, A-beta deposits lead to more mitochondrial damage and ROS production initiating a viscous cycle [16]. At the later stages of AD, these plaques trigger hyperphosphorylation of tau leading to neurofibrillary tangles which cause more cellular damage.

A hypersensitive reaction, involving the innate immune system, has been linked to AD pathology [17, 18]. Increased expression of inflammatory cytokines may induce beta-amyloid deposits and senile plaques. These plaques are rich in activated microglia and complement proteins underscoring the importance of inflammation in the pathogenesis. Microglia release a wide variety of pro-inflammatory mediators including cytokines, complement components and free radicals, all of which potentially contribute to further neuronal dysfunction and eventually death [19]. This neuroinflammatory process leads to ROS production, synapse dysfunction and loss of calcium regulation, which is known to be important for neurotransmission. Due to the changes in the inflammatory process that naturally accompany aging, these insults are more likely to lead to a chronic progressive response [20]. Loss of blood-brain barrier integrity may lead to protein translocation into the cerebrospinal fluid, exacerbating the pro-inflammatory state of the cerebrum [20]. These processes can go on for 10–15 years prior to the onset of symptoms [21].

Although specific mutations involving the amyloid protein have been identified in early-onset familial AD, they account for less than 1% of the cases. It has been known for decades that apolipoprotein E allele variations are associated with the non-familial late-onset AD. The e4 allele confers an increased risk whereas the e2 allele reduces the risk. There is growing evidence implicating the complement pathways in the pathogenesis of AD [22, 23]. CR1, the receptor for the complement fragments C3b and C4b, plays a role in the clearance of amyloid plaques [24]. Recent genome-wide association studies have indicated that genetic variations in CR1 are associated with global cognitive decline and higher burden of AD brain pathology [25]. Animal models indicate that blocking the complement cascade can result in slower cognitive decline [22]. Genetic analysis is important because these alleles may affect and possibly predict a patient’s response to immunotherapy.

AD Diagnosis

Currently, AD is a purely clinical diagnosis, based on a set of observable criteria that evaluate progressive cognitive and functional decline. Age is the predominant risk factor for AD with incidence rising exponentially above age 65. The short term memory loss that commonly accompanies normal healthy aging is not a progressive process nor is it associated with impairment of daily activities, therefore it must be differentiated from AD. AD is commonly preceded by a condition called mild cognitive impairment, which is characterized by mild memory loss with preservation of other cognitive and functional activities. Mild cognitive impairment is an important component in the continuum from healthy cognition to dementia. Predictors that mild cognitive impairment will lead to AD include carrying the e4 allele, brain atrophy and the presence beta-amyloid deposits [26].

AD is a diagnosis of exclusion. Reversible causes of dementia such as B12 deficiency, CNS infections and thyroid disorders need to be ruled out. Neuroimaging should be done to detect neoplasms and vascular disease. Other forms of dementia such as Lewy body disease, frontotemporal degeneration and Parkinson disease are also included in the differential. Recently proposed diagnostic revisions incorporate imaging and the presence of CSF biomarkers to confirm AD but these are not yet routinely recommended [27]. A recent observational study supports the use of CSF biomarkers for diagnosing AD highlighting their good positive predictive value [28].

Biomarkers and imaging will likely play a larger role in the management of AD in the future. Screening and prevention will likely play a larger role in the near future, as early treatment is most closely associated with clinical outcome. The Alzheimer’s Disease Neuroimaging Initiative (ADNI) is an ongoing, longitudinal, multicenter study designed to develop clinical, imaging, genetic, and biochemical biomarkers for the early detection and monitoring of AD [29]. Over 200 papers have been published by this initiative. Biomarkers from both the CSF and plasma have been investigated [30]. These include complement proteins, fibrinogen and phospholipids [31, 32]. Screening high risk patients with PET scans has also shown some promise. Using a subset of patients from the randomized controlled clinical trials one study suggested that PET scans be useful in assessing the effects of treatment on cortical fibrillar amyloid-beta load [33]. Biomarkers and imaging hold promise for detecting pre-clinical disease, monitoring disease course and response to therapy and facilitating the development of new therapies.

Immunotherapy for AD

The primary role of the A-beta peptide in initiating AD pathology has lead to multiple drug design efforts targeting its production and accumulation [34]. Despite this there are still no effective therapies that modify the disease. Immunotherapy represents a new strategy aimed at reducing the pathological lesions of AD and slowing or reversing cognitive decline. Both active and passive therapies are currently under clinical investigation. Active immunization or vaccination involves administration of fragments of A-beta peptide conjugated to an adjuvant that stimulates the host immune response. The goal is to initiate host production of antibodies against A-beta that can potentially help clear away or prevent A-beta plaques [35–37]. A safe and effective vaccine requires therapeutic levels of anti-A beta antibodies without causing adverse autoimmune responses. This is particularly challenging in the elderly due to immunosenescence. T-cell mediated inappropriate immune responses can be an adverse reaction from the adjuvant [38]. Passive immunization involves direct administration of anti-A-beta monoclonal antibodies. The advantages of passive immunization are greater dose control and easy withdrawal of treatment if adverse effects arise. The disadvantages are the need for repeated injections and the fact that this therapy can never be preventative since the development of clinical symptoms proceeds treatment. The remainder of this review will highlight recent pre-clinical and clinical trials involving both active and passive immunization therapy.

Pre-clinical Studies

Animal models have been a valuable tool for advancing both passive and active immunizations for AD [39, 40]. Mouse models carry a mutation in the amyloid precursor protein gene that results in early development of amyloid deposits in the brain. One significant disadvantage of early mouse models is that they do not generate significant tau pathology [38]. The first transgenic mouse model that showed tau pathology was developed in 2000 [41, 42]. Subsequent models have improved. Wilcock et al wrote an extensive review of AD mouse models [38].

Pre-clinical studies showed that immunization against A-beta can provide protection against and reversal of amyloid molecular pathology and cognitive decline in animal models [43]. In 1999, Schenk et al first showed that active immunization lowered cerebral A-beta deposits in transgenic mice [44]. Mice immunized after the development of significant amyloid pathology showed impressive reductions in plaque deposits. In 2000, Morgan et al showed that A-beta peptide vaccination protects mouse models from age-related memory deficits [45]. Furthermore, Janus et al showed that vaccination is associated with reduction in plaque burden and cognitive deficits in the murine AD model [46]. Lemere et al were the first to demonstrate reduction in cerebral A-beta after vaccination in non-human primate model [47]. DNA vaccines are also being tested in animals and have shown promising results in mice and rabbits. Antibodies generated by these DNA vaccines bind well to amyloid plaques in brain sections. Safety and efficacy studies are underway in Rhesus monkey models [37, 48].

Passive immunization has also shown some success in pre-clinical trials. Bard et al were the first to use passive immunotherapy in mouse models [49]. The antibodies were injected intraperitoneally and cerebral amyloid deposits were significantly reduced indicating that antibodies can cross the blood-brain barrier. Kotilinek et al observed a reversal of memory loss in mice using monoclonal antibodies against A-beta [50]. Behavioral improvement and dramatic reduction in amyloid deposits were observed in transgenic mouse models that achieved significant circulating antibody levels [51]. Proposed mechanisms of A-beta clearance include microglial mediated phagocytosis, disaggregation of amyloid deposits after antibody binding and absorption of A-beta from the brain into the bloodstream [52]. A recent study suggested that both active and passive immunization change inflammatory markers and gene expression which is consistent with these cellular level phenotype changes [39]. Antibodies that target the N-terminus of A-beta were most effective at reducing cerebral amyloid [53]. The significant adverse effect associated with passive immunization is cerebral amyloid angiopathy causing micro-hemorrhaging. The incidence of this hemorrhaging pathology was dependent on the location of the epitope. N-terminus and C-terminus epitopes had a higher incidence than mid-domain epitopes [54]. Subsequent deglycosylation of monoclonal antibodies also decreased the incidence of cerebral amyloid angiopathy [55]. Intranasal vaccination is also being tested in mouse models [56]. Currently, immunotherapy against tau pathology is also being investigated but is not yet in clinical trials [57]. Mouse models indicate that active immunization against a phospho-tau epitope slows progression of the tangle-related behavioral phenotype, including cognitive impairment [58]. Troquier et al showed that active immunization against the pathological epitope phospho-Ser422 improves cognitive deficits promoted by Tau pathology in a well-defined Tau transgenic model [59]. Table 1 summarizes recent pre-clinical investigations.

Table 1.

Summary of recent pre-clinical studies

Mechanism	Study Description	Reference
DNA vaccine	Modified vaccine shows immunogenicity in rabbits. Safety and efficacy studies that are currently being conducted in Rhesus monkeys	[48]
Passive	Describes the development of a plaque-specific antibody which shows robust clearance of pre-existing plaque without causing microhemorrhage in mice.	[89]
Intranasal	Describes an intranasal adenovirus vector vaccine using CpG molecular adjuvant that reduces plaque burden and improves cognition in mice.	[56]
Tau pathology	Describes active immunotherapy targeting Tau showing Tau clearance and improves cognitive deficits in Tau transgenic model.	[59]

Clinical Studies for active immunization

Active immunization involves injection of A-beta peptides combined with an adjuvant to stimulate the immune system. In 2000, the first randomized placebo-controlled phase 1 study evaluated the safety and immunogenicity of human aggregated A-beta peptide immunization in patients with mild to moderate AD. A dose dependent antibody response was observed but its correlation with plaque clearance and short-term clinical outcomes were disappointing. Disability Assessment for Dementia scores showed less decline among active compared with control patients after 84 weeks during phase 1 trials [60]. Phase 2 studies showed clearance of amyloid plaques but aseptic meningoencephalitis was observed in about 6% of the patients, causing the study to halt after 2 years [36, 61–63]. This reaction was attributed to the infiltration of Th1-cells into the CNS causing an autoimmune neuroinflammatory response to the adjuvant. A cohort study followed 80 patients who entered the phase 1 trial for 3 years. The data showed that successful immunization is associated with reduction of plaque burden with insignificant evidence of any effect on cognitive function [64]. The authors concluded that plaque removal alone is insufficient to modify the disease progression of AD. Phase 2 trials are currently underway for ACC-001-QS-21, a conjugated N-terminus peptide attached to an immuno-stimulatory carrier protein[65].

More recent vaccines are designed to reduce the autoimmune Th1-cell response while still inducing a humoral response mediated by Th2-cells. CAD106 is a novel virus-like particle based vaccine. This class of vaccines can be used to enhance immunogenicity against self proteins without the use of an adjuvant [66]. The viral component consists of foreign polypeptides thus circumventing the autoimmune T-cell response. Phase 1 studies of CAD106 reported adverse effects of nasopharyngitis and injection site reactions but no cases of meningoencephalitis. Phase 2 clinical trials have shown a better safety and profile and antibody response in patients with mild to moderate AD [67, 68]. 82% of the treatment arm developed the appropriate antibody response. These studies encourage larger, longer studies for B-cell targeted immunotherapy. Due to the slow and variable course of AD, very large and extended clinical trials will likely needed before a clinical benefit can be accurately assessed. Fragments of the A-beta peptide are being used to develop vaccines. These synthetic peptides are too short (6 amino acids long) to activate T-cells and are associated with less cross reactivity. These peptides mimic the unmodified N-terminus of A-beta and are currently undergoing phase 2 clinical trials [65, 69].

Clinical Studies for passive immunization

Due to the success with transgenic mice in pre-clinical studies passive immunization has advanced to phase 3 clinical trials. Monoclonal antibodies target different regions of the A-beta peptide [70]. Bapineuzumab, the first humanized monoclonal antibody which targets the N-terminus, is currently undergoing phase 3 trials [33, 71]. It has passed safety trials but has not yet been associated with clinical benefits. Vasogenic edema and microhemorrhages were reported during phase 2 clinical trials but the incidence and evidence for causation were not significant enough to hamper further investigations [72]. Interestingly cases of vasogenic edema were closely associated with ApoE4 carriers and the sites of microhemorrhages were associated with amyloid angiopathy. Clinical results were equivocal in terms of cognitive benefit, although there was evidence in reduction of plaque burden and CSF Tau biomarkers [71]. Solanezumab is a monoclonal antibody directed at the middle domain of the A-beta peptide. Phase 2 studies showed a good safety profile and reduction in plaque burden. It is currently undergoing phase 3 trials. Some cognitive benefits in patients with mild AD are reported [73]. Gantenerumab is the first fully human anti-A-beta monoclonal antibody and phase 3 clinical trials are underway [74, 75]. Ponezumab is a humanized monoclonal antibody, designed to reduce T-cell response, with an acceptable safety profile in animal studies [76]. It has currently completed phase 1 clinical trials [77].

One major advantage of the monoclonal antibodies is lack of cellular response. Disadvantages include cost and the need for continuous therapy. Determining the safest and most appropriate epitope is an ongoing are of research [78]. Intravenous immunoglobulin is also being studied in phase 3 randomized controlled clinical trials for AD and mild cognitive impairment [79]. Cerebrospinal fluid A-beta decreased and significant cognitive improvement was observed after 6 months of treatment in one small clinical trial, encouraging larger trials [80]. These larger trials will focus on outcomes in the cognitive, functional and behavioral domains. Table 2 summarizes recent and active clinical trials for both passive and active immunotherapy (www.clinicaltrials.gov).

Table 2.

List of recent and active randomized control clinical trials assessing immunotherapy treatment for AD

Biological	Mechanism	Status	Estimated Completion	Summary	Ref.
AFFITOPE AD02	Active Synthetic A-beta mimetic of the N-terminus	Phase 2	March, 2014	Double-blind, multi-center, study to assess the immunological activity, safety and tolerability The study involves 300 patients with mild AD being followed for 1 year.	[69]
AAB-003 (PF-05236812)	Active	Phase 1	October, 2014	Double-blind, multi-center study to assess the safety, tolerability and dose response pharmacokinetics. The study involves 104 patients with mild to moderate AD for 52 weeks.
ACC-001 with QS-21	Active 7 residue amino acid sequence from N-terminus conjugated to adjuvant	Phase 2	September, 2014	Double-blind, multi-center study to evaluate safety and immunogenicity in patients with mild to moderate AD. The study involves 160 patients, monitoring dose response for 78 weeks.	[65]
CAD106	Active A-beta vaccine against N-terminus designed to have less T-cell response	Phase 2	Completed December 2012	Double-blind, multi-center to investigate the safety, tolerability and antibody response following injections of CAD106. The study invo1ved 77 patients with mild AD who were followed for 90 weeks.	[67, 68]
Intravenous immunoglobulin	Passive	Phase 3	December, 2014	Multicenter, double-blind study to evaluate the efficacy and safety of IV immunoglobulin [10% IVIG] infusion every 2 weeks for 18 months in 402 patients mild to moderate AD.	[80]
Bapineuzumab	Passive N-terminus	Phase 2	January, 2013	Double-blind, multi-center to assess safety, tolerability, pharmacokinetics and capacity of subcutaneous injections to reduce amyloid load as measured by PET scan. The study involved 147 patients with mild to moderate AD given monthly injections for 24 months.	[33, 71]
Bapineuzumab	Passive N-terminus	Phase 3	April, 2012	Multicenter, double-blind, to evaluate efficacy and safety in patients with mild to moderate AD who carry the ApoE4 gene. The study involved 1121 patients given infusions every 13 weeks for 6 infusions over 1.5 years.	[33, 71]
Solanezumab	Passive Mid-domain	Phase 3	June, 2016	Open label extension study in patients who completed participation in previous trials to assess side-effects after receiving solanezumab. The study involves 1275 patients to be followed for 104 weeks.	[75, 90]
BAN2401	Passive Protofibrils	Phase 2	April, 2016	Multi-national, double-blind, parallel-group study across placebo or 5 study arms to determine clinical efficacy and dose response. The study involves 800 patients with mild cognitive impairment to mild AD getting biweekly to monthly injections for 18 months. PET scans are used to monitor amyloid burden and vMRI monitor hippocampal volume.	[91]
Gantenerumab	Passive	Phase 3	September, 2016	Multi-center, double-blind, placebo controlled study to evaluate the cognition, functioning, safety and pharmacokinetics in patients with prodromal AD. PET scans are being used to assess brain amyloid. The study involves 770 patients getting subcutaneous doses every 4 weeks for 104 weeks.	[74]

Summary & Future Prospects:

AD is the most common cause of dementia and the incidence is currently on the rise. It has a progressive, debilitating course of which there is no current effective preventative or disease modifying therapy. Acetylcholinesterase inhibitors and N-methyl-D-aspartate (NMDA) glutamate receptor antagonists are the current standard of therapy. These may mildly improve quality of life but have generally been disappointing [81]. Despite inflammation being at the heart of AD pathogenesis, a recent meta-analysis concluded that neither aspirin, NSAIDs nor steroids significantly affects cognitive decline in AD patients [82].

Immunotherapy is emerging as the leading strategy for disease modifying intervention. The main goal is to reduce the production of and/or enhance the clearance of cerebral A-beta. Although mouse models of AD have shown promising results for both passive and active immunotherapy, more investigations are needed before this approach can be applied in practice [35]. Clinical trials have not yet shown a significant effect on cognitive decline for A-beta immunotherapy despite a reduction is plaque burden [83, 84]. How the severity of plaque burden translates into the clinical manifestations of AD is currently unclear. Thus the significance of plaque burden reduction and its use as a primary outcome needs further clarification. The neurofibrillary tangles appear to be more associated with clinical symptoms [57]. This is likely because it has a later onset.

As immunotherapy for AD continues to develop several issues still need to be addressed. One issue is the most efficient and safest target epitope [85]. Antigens involved in the early steps of the amyloid cascade are being targeted. Newer vaccines need to avoid the anti-Abeta Th1 autoimmune response. Some strategies to overcome this include the development of DNA vaccines, new adjuvants and the use of foreign T cell epitopes. These strategies are currently being investigated [86].

Another issue is how to monitor therapeutic progress. Biomarkers are important to detect pre-clinical disease, select individuals with mild cognitive impairment and predict which patients may benefit most from therapy. New trials should target this population. Complement proteins show promise as possible biomarkers and seem to be linked to AD pathology [31]. There are several types of peripheral biomarkers under investigation, but more work is required before they can be deemed clinically useful [87]. Future studies also need to focus on the timing of treatment. Evidence suggests that effective vaccination occurs early in the disease process [57].

A-beta clearance may not affect cognitive deterioration once functional impairments are evident, thus an agent that has minimal efficacy in AD dementia may be beneficial at the preclinical stage. Most patients diagnosed with AD already have significant irreversible damage. This explains the recent interest into stem cell based therapies [88].