Translating the biology of aging into novel therapeutics for Alzheimer disease

Introduction

Advances in medicine, public health, and education have resulted in increased human lifespan, and the elderly population has grown dramatically worldwide. The increase in lifespan, however, has been accompanied by an increase in age-related chronic diseases, including type 2 diabetes mellitus, cardiovascular disease, cancer, osteoporosis, and neurodegenerative diseases.¹ In fact, morbidity rates increase steadily to middle age, then increase at a much steeper rate such that it is twice as high in people over 80 compared to 60- to 64-year-olds, a phenomenon related to the Gompertz curve.²

Geroscience is a multidisciplinary field that examines the relationship between biological aging and age-related diseases.³ The trans-NIH Geroscience Interest Group Summit discussed 7 processes that contribute to biological aging: macromolecular damage, epigenetic changes, inflammation, adaptation to stress, and impairments in proteostasis, stem cell regeneration, and metabolism.¹ Intriguingly, these 7 processes are highly intertwined with one another. Thus, targeting the common biological processes of aging may be an effective approach to developing therapies to prevent or delay age-related diseases.

The leading risk factor for sporadic Alzheimer disease (AD) is also aging. Processes that are altered with aging that have been implicated in AD include inflammation, impaired autophagy, mitochondrial dysfunction, vascular problems, epigenetic changes, and synaptic loss (figure). An increased incorporation of extensive knowledge regarding biological gerontology into research on AD would likely increase our productivity in developing new drugs for AD.

• Download figure
• Open in new tab
• Download powerpoint

Figure Age-related changes in biological processes contribute to neurodegeneration in Alzheimer disease (AD) and other dementias

Processes that are altered with aging that precede neurodegeneration include inflammation, impaired autophagy, vascular dysfunction, synaptic loss, mitochondrial and metabolic dysfunctions, and epigenetic changes. These processes provide numerous novel targets for new drug development for AD.

Pathologic hallmarks of AD

Pathologic hallmarks of AD include senile plaques comprising β-amyloid (Aβ) proteins along with many other misfolded proteins and neurofibrillary tangles formed by hyperphosphorylated tau protein aggregates. The accumulation of plaques and tangles occurs over many decades, in association with aging. While Aβ can be detected in early adulthood even in cognitively healthy people, elevated amyloidosis begins in the 6th decade of life and increases linearly into old age.⁴ Medial temporal tauopathy also begins in middle age and overt cognitive impairment coincides with increased neurofibrillary tangle burden.⁵ Clearly, pathologic and clinical findings indicate that sporadic AD and related dementias are age-related diseases.

Although Aβ plaques and neurofibrillary tangles are pathologic markers of AD, it is not known if these pathologies represent valid drug targets or if these targets alone are sufficient to treat AD. The current drug development pipeline strongly reflects focus on these 2 major pathologic proteins, with 32.5% of the currently ongoing 126 AD clinical trials targeting either Aβ (30/126, 23.8%) or tau (11/126, 8.7%). Of phase 3 trials, 52% (13/25) are pursuing these targets (12 targeting amyloid and 1 targeting tau).⁶ Therapeutic attempts to remove or decrease the production of Aβ have appeared promising in preclinical and early-phase trials, but have been largely unsuccessful in altering the progression of AD in later-phase clinical trials.⁷

Because the mechanisms underlying AD and related dementias are complex and multifactorial, greater exploration of targets beyond these 2 pathologic markers is warranted.

Inflammation

Inflammation is a hallmark of aging. Inflammaging refers to the low-grade, chronic, systemic inflammation associated with aging in the absence of overt infection and is a significant risk factor for morbidity and mortality in the elderly.⁸ Furthermore, in middle-aged and older adults, higher levels of systemic inflammatory markers (e.g., C-reactive protein, interleukin [IL]–6, fibrinogen) are associated with cortical thinning,⁹ lower and greater declines in regional cerebral blood flow,¹⁰ and poorer cognitive functions, including executive function and learning and memory.¹¹ However, randomized clinical trials of broad-spectrum anti-inflammatory drugs such as nonsteroidal anti-inflammatory drugs (e.g., naproxen, aspirin, celecoxib) and others (e.g., prednisone, statins, rosiglitazone) have thus far failed to improve cognitive outcomes in patients with AD.¹² Currently, there are 12 ongoing clinical trials testing anti-inflammatory interventions (table 1), 7 of which are repurposed drugs (e.g., antiviral, antibiotic, and rheumatoid arthritis medications) and 1 is a biologic (plasma).⁶

It is worth emphasizing a few recent efforts that target specific aspects of inflammation. For example, GC021109, which was shown to be safe in a phase 1 trial of patients with AD, promotes microglial phagocytosis by binding to the microglial P2Y6 receptor.^13,14 Other approaches attempt to reduce microglial cytokine production. The NLRP3 inflammasome is a multiprotein complex involved with the innate immune system and activates procaspase-1, which in turn induces production of proinflammatory cytokines IL-1β and IL-18.⁸ The NLRP3 inflammasome significantly contributes to neuroinflammation and age-related cognitive decline and is potently activated by Aβ.¹⁵ Novel compounds that specifically inhibit the NLRP3 inflammasome are currently under development, with companies such as IFM Therapeutics (Boston, MA) expecting to have NLRP3 inhibitors for AD in clinical studies by 2020 (ifmthera.com/pipeline/). Resolution is the final stage of the inflammatory response when immune cells, apoptotic cells, and debris are cleared from the site of insult and proinflammatory mediators are catabolized such that the tissue can reach functional homeostasis.¹⁶ With chronic inflammation, there is incomplete resolution of the initial response, resulting in chronic and increased levels of tumor necrosis factors, interferons, and IL-6. Instead of globally inhibiting inflammation, it may be fruitful to promote resolution of inflammation by using specialized proresolving mediators.¹⁷

Senescent cells are thought to fuel aging and age-related pathologies as they release proinflammatory cytokines, chemokines, and tissue-damaging proteases and can negatively affect the surrounding tissue microenvironment.¹⁸ Senolytic drugs are drawing attention as they selectively induce apoptosis of these senescent cells. While senolytic drugs appear to provide beneficial effects in rodent models of aging, they may also be associated with slowed wound healing.¹⁹ Further work is needed to establish a causal link between senescent cell accumulation and AD before these drugs can be translated into the clinic. In addition, drugs targeting specific phenotypes of myeloid cells or select elements of the complement pathway are in preclinical and clinical development.^20,21 It is critical that these agents inhibit the detrimental aspects of inflammation while preserving the defense mechanisms against pathogens and tissue damage and allowing for continued phagocytosis by microglia.

Impairment in autophagy and clearance of misfolded proteins

Autophagy is a process that involves the degradation and recycling of damaged/aggregated proteins, lipids, and larger cellular components including organelles. Autophagic activity decreases with aging, and genes that promote autophagy have been associated with increased lifespan in model organisms.²² In AD models, impaired autophagy is tightly linked to the accumulation of plaques and tangles. Since Aβ and tau are not the only proteins that misfold or form toxic aggregates in the aging brain, broadly promoting or restoring the impaired autophagic mechanisms may be a more parsimonious way to prevent accumulations of the many various damaged proteins that occur with aging.²³ Nilotinib, a Food and Drug Administration (FDA)–approved drug for the treatment of adult chronic myeloid leukemia, is currently being tested in a phase 2 trial in AD.²⁴ Nilotinib boosts the autophagic machinery by increasing levels of parkin, which is an E3 ubiquitin ligase that plays a critical role in ubiquitination and clearance of misfolded and damaged proteins.²⁵ Perhaps the greatest challenge in restoring clearance mechanisms is to concurrently promote each stage of its machinery, from autophagosome biogenesis, lysosomal fusion, to the final degradation of cargos within autolysosomes.²⁶ This is important because simply inducing autophagy would lead to accumulation of autophagosomes and undigested autolysosomes, which can block axonal trafficking and lead to axonal swelling. A combination of drugs may be necessary in restoring this autophagic flux. Additional challenges include off-target effects of agents that promote autophagy as well as the lack of validated assays to measure autophagic flux in humans.²²

Mitochondrial and metabolic dysfunctions

The free radical theory of aging, the idea that free radicals produced in the course of cellular metabolism contribute to aging and degenerative diseases, was first introduced in 1956 by Harman.²⁷ Indeed, mitochondrial dysfunction and oxidative stress increase with aging and are tightly linked to neurodegeneration. High metabolic demands along with low levels of antioxidative defense mechanisms make the brain especially vulnerable to oxidative damage.²⁸ Aging and AD are associated with inefficient mitochondria and an imbalance between pro-oxidants and antioxidants, which in turn, induce oxidative damage to DNA, proteins, and lipids in the brain. To date, antioxidant agents have not been successful in treating AD,²⁹ partly due to their low bioavailability and poor blood–brain barrier penetration. They may also block oxidative signaling that is necessary for normal cellular functioning.

There are 14 interventions targeting mitochondrial or metabolic dysfunctions that are currently in clinical trials for AD (table 2).⁶ Many of these are repurposed drugs and 7 are antidiabetics. Type 2 diabetes mellitus is a significant risk factor for AD and both diseases share pathologies including insulin resistance and oxidative stress.³⁰ Thus, insulin, its derivatives, and drugs designed to improve insulin sensitivity such as liraglutide and exenatide are currently being tested for AD (table 2).

Agents that specifically target mitochondria to decrease mitochondria-derived reactive oxygen species may also hold promise for preventing neuronal dysfunction and death. For example, CP2 is a cell-permeable tricyclic pyrone that crosses the blood–brain barrier and accumulates in neuronal mitochondria.³¹ It mildly inhibits the mitochondrial complex I, producing a mito-hormetic effect, augmenting respiratory capacity and reducing proton leak in wild-type mice. CP2 also prevents cognitive impairment in mouse models of AD (APP, PS1, and APP/PS1) while reducing amyloid plaques and phosphorylated tau. SS-31 is another example of a compound with mitochondria-specific actions. It is a small peptide that binds to cardiolipin, a lipid exclusively expressed on the inner mitochondrial membrane that plays a structural role in organizing the components of the electron transport chain into supercomplexes for more efficient oxidative phosphorylation.^32,33 SS-31 also inhibits the opening of the mitochondrial permeability transition pore that forms under mitochondrial stress, which can lead to mitochondrial swelling and apoptosis.³⁴ Preclinical studies have shown that SS-31 prevents anesthesia-induced cognitive impairment and promotes mitochondrial and synaptic health in models of AD.^34,35 SS-31 (elamipretide) is currently being tested in clinical trials for rare disease indications, including mitochondrial myopathy, Barth syndrome, Leber hereditary optic neuropathy, and Huntington disease (stealthbt.com/clinical). A repurposing opportunity for SS-31 exists for AD.

Vascular dysfunction

Aging is associated with vascular disease, which contributes to vascular cognitive impairment, vascular dementia, and other dementias including AD. Studies have shown that the total burden of vascular pathology correlates with cognitive deficits,³⁶ and 79.9% of people with AD were reported to have vascular pathology at autopsy.³⁷ The brain requires continuous and well-regulated blood flow due to the high energy demand and inability to store energy. Vascular impairment can result in hypoperfusion, dysfunction of the neurovascular unit, oxidative stress, or inflammation, which in turn can lead to endothelial damage, small vessel disease, blood–brain barrier breakdown, demyelination, infarcts, hemorrhages, and cerebral atrophy.³⁸ There are currently 11 clinical trials testing interventions that target the vascular system (table 3), of which 9 are repurposed drugs such as antihypertensives and antihyperlipidemic agents and 2 are omega-3 fatty acids.⁶

While managing vascular risk factors such as hypertension, dyslipidemia, and diabetes is likely to improve long-term brain health and lower dementia risk, additional efforts to target specific vascular pathobiologies are needed. For example, upon blood–brain barrier disruption or vascular breakdown, fibrinogen leaks into the CNS, resulting in formation of clots comprising insoluble fibrin and inflammation.³⁹ Extravascular fibrin(ogen) then interacts with microglia via the integrin receptor CD11b/CD18, which induces secretion of cytokines and chemokines and stimulates recruitment of peripheral monocytes and macrophages.⁴⁰ While fibrinogen is undetectable in the healthy brain, it is present in the brains of patients with AD as well as in older people without AD pathology.^41,42 Aβ also interacts with fibrinogen and coagulation factor XII, which can further increase clotting, fibrin deposition, and proinflammatory molecules.⁴³ Therefore, a higher plasma level of fibrinogen is associated with an increased risk for AD and vascular dementia.⁴⁴ Thrombin, an enzyme in blood plasma that causes blood clotting by converting fibrinogen to fibrin, is also elevated in the brains of patients with AD, and produces proinflammatory effects on endothelial cells, microglia, and astrocytes.⁴⁵ Targeting the coagulation system may reduce fibrin formation, neuroinflammation, and neurodegeneration; however, it is important to specifically target the pathogenic properties of coagulation proteins without affecting the beneficial processes of blood clotting.

Epigenetic changes

Aging is also accompanied by important epigenetic changes including histone modifications, DNA methylation, and microRNA expression, resulting in alterations in gene expression and genome architecture.⁴⁶ As epigenetic mechanisms are critical for normal brain function, including learning and memory processes, epigenetic dysregulations may contribute to the onset of cognitive dysfunction and AD.⁴⁷ Because age- and disease-related epigenetic disturbances are potentially reversible, they have drawn attention as attractive targets for pharmacologic interventions particularly in cancer therapies, but they are also gaining traction in neurodegenerative diseases. There are a few epigenetic drugs currently being tested in clinical trials for AD. Vorinostat, a pan-histone deacetylase (HDAC) inhibitor approved for the treatment of cutaneous T-cell lymphoma, is currently in phase I trial in patients with AD.⁴⁸ ORY-2001 is a selective dual LSD1-MAO-B inhibitor that has shown marked cognitive improvement in transgenic AD models, and phase I results showed safety, tolerability, and brain penetrance.⁴⁹ LSD1 is an epigenetic modulator that regulates histone methylation. The major advantage of epigenetic drugs is that they can regulate expression of multiple genes critical for neuronal functions. However, studies are required to confirm that these drugs aimed to counter neurodegeneration in the brain do not also increase gene expression associated with harm, such as tumorigenesis. Preclinical efforts addressing these safety issues are also underway, including selective inhibitors of HDAC2 that show functional efficacy and reduced side effects in animal models (rodintherapeutics.com/our-approach/#literature).

Synaptic loss and dysfunction

Normal age-related cognitive decline is not accompanied by the extensive neuronal loss seen in AD and other dementias; however, it is associated with dysfunctions and loss of cortical synapses.⁵⁰ These synaptic alterations occur in the same circuits that degenerate in AD. Based on a postmortem study, synapse loss is observed in a large proportion of people with mild cognitive impairment and appears to be an early event in the disease process that precedes neuronal loss in AD.⁵¹ Because age-related synaptic alterations may render neurons more vulnerable to degeneration, maintaining synaptic health in the face of aging may be an important strategy for preventing neurodegeneration.

There is growing interest in therapies that promote neuronal and synaptic health for preventing and treating dementia. Currently 19 neuroprotective interventions are being tested in clinical trials for AD (table 4), of which 5 are cell therapies (e.g., stem cells), 3 are natural products, and 2 are biologics.⁶ In particular, nerve growth factor signaling has drawn interest as a potential neuroprotective target for AD. LM11A-31-BHS, a small molecule ligand for the p75 neurotrophin receptor (p75^NTR), is currently being tested in a phase 2 trial in patients with AD. Treatment with this molecule reversed cholinergic neurite degeneration in AD mouse models even when it was initiated after significant pathology was already present.⁵²

Combination therapies

As discussed above, multiple processes go awry with aging, many of which negatively affect neuronal and cognitive health and ultimately contribute to the onset and progression of AD. Therefore, a combination of drugs to address these systems may be necessary to effectively treat it. There are currently 11 combination therapy clinical trials (table 5), of which 5 are testing an add-on therapy to an already approved AD drug (e.g., donepezil), and several others are using multiple drugs to target Aβ.⁶ In 2018, a phase 2 clinical trial to test AMX0035, a combination of sodium phenylbutyrate and tauroursodeoxycholic acid (TUDCA), is expected to begin. Sodium phenylbutyrate is an FDA-approved therapy for urea cycle disorders and it is a Class I and II HDAC inhibitor. Preclinical studies have shown that it also acts as a chemical chaperone and ameliorates ER stress, prevents neuronal loss, increases clearance of intraneuronal Aβ, and restores cognitive functions in AD mouse models.^53,–,55 TUDCA is a bile acid that supports mitochondrial energetics by reducing mitochondrial permeability and increasing apoptotic thresholds of cells.⁵⁶ Studies in an AD mouse model (APP/PS1) have shown that TUDCA decreases Aβ deposits, reduces glial activation, and restores cognitive functions.^57,58 Together, the 2 drugs are designed to restore epigenetic changes, reduce cellular stress, and protect against neurodegeneration, mitochondrial dysfunction, and neuroinflammation.

Combination trials present some challenges as each drug needs to be tested for safety, and multiple biomarkers are often required to ensure target engagement and to monitor disease progression and response to treatment. Some evidence of efficacy should be demonstrated for each drug, though benefits of each may be incremental. Combination therapies that target multiple age-related dysfunctions with synergistic activities are especially promising for treating AD. More such combination studies are needed.

Future directions

Biological aging is the leading risk factor for the major debilitating chronic diseases of old age that cause morbidity and mortality, including AD and other dementias. Drugs that treat fundamental biological mechanisms of aging have been proposed to be useful for most prevalent chronic diseases of aging.¹⁸ In fact, many repurposed drugs are used to treat other age-related diseases. Despite over 75 years of accumulated research on biological aging,⁵⁹ the current drug development pipeline is dominated by therapeutics targeting Aβ and tau, and there has been proportionately less translation of biological gerontology into our efforts to develop drugs for AD. Nevertheless, aging biology provides numerous novel targets for new drug development for AD (figure).

Because of the multifaceted nature of biological aging, it is unlikely that drugs addressing a single target will be very successful in effectively treating AD. Nevertheless, single drug clinical trials may be needed to demonstrate incremental benefits, even if modest, before combination trials can be pursued. As interventions that target one aberrant system tend to also attenuate others, ultimately, combination therapies that target multiple age-related dysfunctions may produce synergistic activities. Combination therapies are already the standard of care for other diseases of aging, including heart disease, cancers, and hypertension, and will likely be necessary in treating AD and other dementias. And because the same biological aging mechanisms underpin the common diseases of aging, repurposing drugs already on the market is a rational strategy for testing new therapies for AD and related dementias, including the sporadic forms of frontotemporal dementia and vascular dementia. Novel therapeutics for new and relevant targets will clearly also be needed.

In addition to combination therapies, addressing the multifaceted nature of the relationship between biological aging and AD with drugs possessing pleiotropic effects (simultaneously producing more than one effect) will be advantageous. In clinical areas such as oncology, psychiatry, and cardiovascular medicine, many effective drugs act on multiple targets while single-targeted approaches seldom progress to the final stages of clinical trials.⁶⁰ For example, statins are widely used to lower cholesterol levels in patients with dyslipidemia, but statins also have pleiotropic effects that are independent of their effects on cholesterol, including improved endothelial function, inhibition of vascular inflammation, stabilization of atherosclerotic plaques, and immunomodulation.⁶⁰ To effectively treat AD, pleiotropic drugs may need to hit the right nodes of relevant biological networks affected by aging such that they positively influence those networks and interconnected pathways.

Finally, a parsimonious approach to drug discovery and development with regard to translating knowledge from biological aging to AD is needed. For example, due to the plethora of misfolded proteins that accumulate with aging in the brain, biologics that attempt to address a single misfolded protein may be far less efficacious than drugs that enhance autophagy and clearance of all misfolded proteins. Similarly, age-related inflammation, vascular disease, epigenetic dysregulation, mitochondrial/metabolic dysfunction, and synaptic failure may be upstream causes of neuronal dysfunction and death leading to the classic pathologic hallmarks that have been historically among the first drug targets in AD. A better understanding and translation of the systemic, cellular, and molecular processes of biological aging that precede and increase vulnerability to AD will help identify new strategies and therapeutic targets for drug discovery and development.