Aging: therapeutics for a healthy future

Abstract

Increased healthcare and pharmaceutical understanding has led to the eradication of many childhood, infectious and preventable diseases; however, we are now experiencing the impact of aging disorders as the lifespan increases. These disorders have already become a major burden on society and threaten to become a defining challenge of our generation. Indications such as Alzheimer’s disease gain headlines and have focused the thinking of many towards dementia and cognitive decline in aging. Indications related to neurological function and related behaviors are thus an extremely important starting point in the consideration of therapeutics. However, the reality is that pathological aging covers a spectrum of significant neurological and peripheral indications. Development of therapeutics to treat aging and age-related disorders is therefore a huge need, but represents a largely unexplored path.

Fundamental scientific questions need to be considered as we embark towards a goal of improving health in old age, including how we 1) define aging as a therapeutic target, 2) model aging preclinically and 3) effectively translate from preclinical models to man. Furthermore, the challenges associated with identifying novel therapeutics in a financial, regulatory and clinical sense need to be contemplated carefully to ensure we address the unmet need in our increasingly elderly population. The complexity of the challenge requires different perspectives, cross-functional partnerships and diverse concepts. We seek to raise issues to guide the field, considering the current state of thinking to aid in identifying roadblocks and important challenges early. The need for therapeutics that address aging and age-related disorders is acute, but the promise of effective treatments provides huge opportunities that, as a community, we all seek to enable effectively as soon as possible.

1. Aging – what are we attempting to treat?

Many consider aging as a purely chronological phenomenon; it is an immutable fact that we all get older. However, this is a simplification as individuals all age functionally in different ways and the concept of “biological aging” is more relevant than chronological aging (Khan et al., 2017). When we consider biological aging, we have a therapeutic target, not simply targeting getting old, rather treating physiological decline that is manifested by dysfunction and morbidity in late life. When biological aging becomes pathological, it can be considered as a failure of homeostasis. There is a progressive component, but ultimately a point is reached at which there is inability to counter the amassed toll on the body of time-dependent accumulation of cellular damage (DNA mutations, protein misfolding, oxidative stress etc.), which occurs throughout life. With age, cellular processes (including stress response pathways invoked by damage) become less efficient, ultimately leading to cellular death and irreversible consequences. At younger ages, the body is able to mount compensatory responses and life is healthy. During middle age, the body’s ability to maintain homeostasis declines, resulting in chronic diseases that accelerate the gradual degradation of life quality, resulting in severe detriments, frailty and, eventually, mortality. There are many ways that a homeostatic balance can be maintained, giving many opportunities for development of therapeutics.

When considering therapeutic outcomes, it is also important to consider the primary goal. Lifespan extension is an obvious target, but controversially there may be a maximum lifespan (Soto-Gamez and Demaria, 2017) and, more importantly, long life with low quality of life may be even less desirable. Instead, the concept of improving health-span, the time of life in which we are healthy, is more realistic and one that many aspire to – being healthy in old age - whether or not this also leads to a longer life. The pathways modulating lifespan and healthspan are likely overlapping but distinct. Many interventions extend both in animal models, but different genes have also been associated with longevity versus disease-free aging in human studies (Erikson et al., 2016).

Age is a major risk factor for many chronic diseases, including those with huge impact on individuals and society such as Alzheimer’s disease (AD), cancer, cardiovascular disease, and diabetes (Niccoli and Partridge, 2012). Aging is also a complex, multifactorial process. Key “pillars” or processes of aging have been identified, including inflammation, macromolecular damage, and breakdown of proteostasis, epigenetic changes, and accumulation of senescent cells (Kennedy et al., 2014; Lopez-Otin et al., 2013). These pillars of aging link organ systems too, the brain clearly impacted by neuroinflammation, deficits in proteostasis leading to accumulation of aggregate-prone proteins and loss of stem cells in key areas involved in memory formation. Also in the periphery, from bone to skin to core organs, the biology of aging has widespread impact (Fig. 1). These mechanistic processes are also highly intertwined, so that perturbation of one process can ramify throughout the whole system. This complexity and interdependence could lead to the inability to impact the system by intervening with the targeting of only one gene or with one single drug. However, it seems likely that a single event or small number of events can push “normal” aging over the edge into chronic disease. Thus, specific genetic mutations and single drugs can increase lifespan, including rapamycin and sirtuin-activating compounds (Pan and Finkel, 2017) across a range of animal models including C. elegans, D. melanogaster and M. musculus (Gao et al., 2017; Bitto et al., 2015).

Fig. 1.

The brain and peripheral organs share common biological mechanisms of aging. The well described pillars of aging (Lopez-Otin et al., 2013) are evident systemically, linking the neurological and behavioral consequences of aging and peripheral organs whose function also declines with age. A strong hope for therapeutics of aging is that a cross-therapeutic area benefit will be achieved.

There is still a lot that we don’t know about aging, with organisms within a species often showing considerable variability in both lifespan and healthspan. For example, some older people maintain good cognitive function throughout age, even as centenarians, whereas others develop dementia in middle age. This variability is due to stochastic as well as genetic and environmental factors; even isogenic individuals of a single species (C. elegans or M. musculus) raised in the same environment show wide differences in lifespans and ages of onset of degenerative pathologies (Gladyshev, 2016; Lucanic et al., 2017a; Wyss-Coray, 2016). Understanding why these differences occur and how they contribute to pathological aging is an essential to guide the development of novel therapeutics. We also need to understand if there are mechanistic differences between normal and pathological biological aging, or if pathological aging is just accelerated biological aging. Likely, it is a combination, with these complex disorders of old age being an accumulation of multiple disease specific risk factors, environmental exposures and aging processes, where modifying even one component prior to overt disease may be able to alter the onset and/or course of multiple diseases (Podtelezhnikov et al., 2011; Villeda et al., 2011). Attractively, if you can delay negative aging processes, keeping people healthier as they age, an array of serious age-related disorders could be prevented. It is clear that increased understanding of the biology of aging can lead to wholly new therapeutic approaches of great potential impact.

2. Issues involved in developing anti-aging therapies

A therapeutic that halts or prevents aging, putting off the onset of a debilitating disorder, could lead to a longer life. Extending life on the face of it is highly positive; the vast majority of the population looks forward to a long and fruitful life. However, if anti-aging therapies only increase lifespan without increasing healthspan it will just increase the population of diseased elderly, increasing healthcare spending and providing an undesirable, unsustainable, future for society. If healthspan is increased, thus delaying the onset of an array of age-related disorders there could be a highly positive societal and economic impact. People could work longer, contribute more economically and have a more positive, productive life with a concomitant reduction in chronic health-related disorders. This would potentially lead to later retirement (an event likely to happen anyway with the aging of the population), and changes in population distribution, as longer lifespan correlates with reduced populations. The ethical implications must always be kept in mind, but the advantages and impacts of developing anti-aging drugs to improve the quality of life in old age are highly significant (Vaiserman and Lushchak, 2017).

2.1. Multiple challenges of developing anti-aging therapeutics

With the lofty goal to counteract aging, development of novel therapeutics will encounter a series of challenges, many of which have little or no precedent in prior clinical efforts. In particular, it is not clear how to design a clinical trial to assess strategies to slow the progression of aging per se. For diseases with a degenerative component, such as AD, trials often last many years, which is necessary to observe therapeutic effect over a slow rate of neurodegenerative decline. Independent of neurodegenerative disease, the rate of purely age-related decline may even be slower, and more variable between individuals, likely requiring even larger trials of longer duration. Moreover, therapies that require intervention earlier, in pre-symptomatic states, will require trials of extremely long duration, uncertain outcome and include immense heterogeneity. Designs will be based on outcomes of animal studies, where the aging time course differs dramatically from human aging and is likely different in many ways.

Mouse models will be critical to identify new links between aging and disease, but it is important to note that while inbred mouse strains are likely to have relatively more homogeneous cohorts, outbred mice that are more representative of aging in human populations show more variation In aging processes (Koks et al., 2019). Thus, how rodent outcomes translate to human trials remains unclear and the question of when it is best to start an intervention will be to an extent empirical from multiple human studies. New approaches to animal model development can allow researchers to evaluate the effect of a single genetic variant or therapeutic candidate on a genetically diverse background population in mice, an approach that would require larger, more complicated animal studies, but might improve the translational validity to a diverse human population.

The identification of biomarkers for biological aging could dramatically alter clinical and epidemiological studies as they could be used to monitor short and medium term interventional strategies. Classical biomarkers include sarcopenia (Kalinkovich and Livshits, 2015), physical frailty and cognitive frailty (Ruan et al., 2017). Aging also impacts biological functions that are too numerous to include in their entirety in a single clinical study, thus endpoint selection will need to be well-informed by the underlying biological hypothesis for the treatment and the animal data generated with a therapeutic. More recently, a range of other biomarkers of aging (described below) have been developed largely using AI-based strategies exploring deep biologic datasets. These measures potentially permit measures of biological aging across a wider life trajectory and may be dynamic in nature, responding to interventions in a shorter timeframe. In addition, wearable devices and other objective and data-rich measurements are being tested and are likely to provide much better and cleaner assessments of the aging process, improving the quality of clinical trials and our ability to back-translate to similar objective endpoints in non-human species. While the best biomarkers remain a matter of debate, the development of anti-aging therapeutics can still proceed through the analysis of specific indications, at least in the short term. As age is a common risk factor for the majority of disorders, it is expected that an anti-aging therapy would either delay onset or slow the decline observed in multiple pathological conditions.

2.2. Rationale to develop anti-aging therapeutics

Development of drugs in recent years has focused almost entirely on selectivity and specificity with a primary goal to reduce the risks of any side effects. Such an approach is highly valid when considering specific indications targeting selected organs. Aging is different; the body ages systemically, although not necessarily uniformly, and it may indeed be preferential to have broad, systemic anti-aging effects. Consideration of systemic therapies, broadly distributed targets or organ systems that can have wide impacts may be strong and viable strategies to take. Approaches of potential broad value are to modulate processes that are ubiquitous, systemic and that potentially have multiple impacts such as targeting the plasma proteome, cellular senescence, proteostasis and metabolic processes (Table 1).

Table 1.

Key therapeutic strategies for aging biology.

Process	Therapeutic Candidates	References
Pharmacological
Metabolic function	Metformin (Biguanides)	(Barzilai et al., 2016)
Cellular Senescence	Fisetin, Bcl2 inhibitors, Dasatinib, Quecertin	(Zhang et al., 2019; Yousefzadeh et al., 2018; Chang et al., 2016)
Proteostasis/autophagy	Rapamycin, Rapalogs	(Lamming et al., 2013)
Nutrient Sensing	Klotho	(Kuro-o et al., 1997)
Intercellular signaling	Plasma, Plasma Fractions, GDF11	(Smith et al., 2015; Katsimpardi et al., 2014; Kheifets and Braithwaite, 2019; Sha et al., 2019)
Stem cell exhaustion	Mesenchymal stem cells	(Brooks and Robbins, 2018)
Epigenetic reprogramming	Yamanaka factors	(Tamanini et al., 2018)
Telomere shortening	Telomerase	(Shay and Wright, 2019)
Genomic Instability	NAD +	(Michan, 2014)
Non-Pharmacological
Caloric Restriction	Multiple regimens	(Kivipelto et al., 2018; Michan, 2014; Anton and Leeuwenburgh, 2013)
Exercise	Multiple regimens	(Kramer et al., 2006; Kivipelto et al., 2018)

In a particularly broad sense, the results of heterochronic parabiosis and plasma-transfer experiments utilize this concept to demonstrate widespread effects of a complex molecular mixture. These procedures effectively take a mixture of thousands of proteins present in plasma that have been identified to either accelerate or reverse aging of organs including the brain, pancreas, heart, liver and others (Koks et al., 2019a; Kalinkovich and Livshits, 2015; Ruan et al., 2017; Katsimpardi et al., 2014; Helman et al., 2016; Smith et al., 2015), These factors display a wide array of functional mechanisms Including modulating inflammation (Sallam and Laher, 2016), stem cell proliferation and differentiation (Ahmed et al., 2017) and vascular benefit (Harvey et al., 2015). Identification of specific blood-borne factors such as eotaxin (Villeda et al., 2011), GDF11 (Katsimpardi et al., 2014), Beta-2-microglobulin (Smith et al., 2015) and TIMP2 (Castellano et al., 2017) indicate that individual factors may drive at least some of these activities and be systemic regulators of function and dysfunction, but also in aging the ability to modulate multiple mechanisms in a single therapeutic of plasma or a fraction of plasma could be very valuable.

Another broad approach capitalizes on cellular senescence - as the body ages, cellular processes become disrupted, including an arrest in the ability of cells to proliferate and differentiate leading to their ultimate death. Dying senescent cells accumulate with aging and release soluble factors that can drive further age-related diseases in a paracrine or endocrine manner, the senescence associated phenotype (SASP) (Childs et al., 2015). The removal of senescent cells has been demonstrated to reverse age-related deficits (Baker et al, 2011) and thus modulating senescence processes or senescent cells could be an anti-aging approach.

Rapamycin is one of the current prototypic anti-aging drugs and at least partly effects aging processes through modulation of proteostasis. The proteostatic systems of different cells and organs are highly interconnected at the organismal level. Cell and organ non-autonomous regulation of specific proteostatic pathways has now been demonstrated in both invertebrates and mice (Sala et al, 2017). Therefore, it may be possible to target a proteostatic regulator in a peripheral organ (e.g., mTORC pathway in muscle) and have a positive impact on the CNS or vice versa (Dubinsky et al, 2014; Williams et al., 2014). Interorgan communication regulating proteostasis may be mediated via soluble factors such as exosomes, which can transfer chaperones and other proteins between cells (Kaushik and Cuervo, 2015; Tsai et al, 2015). Studies in mice suggest it should be possible to impact systemic aging by targeting a single peripheral organ. For example, activation of 4E-BP1 (a downstream effector of mTORCl) specifically in muscle is sufficient to promote healthy aging across many tissues (Tsai et al, 2015). Of note, rapamycin-mediated mTOR inhibition also results in partial SASP inhibition, improvement of adult stem cell function with age, and robust anti-inflammatory effects, which all (in addition to improved proteostasis) may contribute to its promotion of lifespan and healthspan.

Metformin has been proposed as a broad anti-aging therapeutic through its effects on metabolic function. It is used to treat diabetes, a condition associated with accelerated aging (McDaniel, 1999). Epidemiological correlates of metformin treatment of elderly with type 2 diabetes with increased longevity have been made (Bannister et al, 2014) and health and lifespan of mice have been enhanced with the drug (Martin-Montatvo et al, 2013). The data with rapamycin and metformin indicate that there may be multiple means to broadly target processes in the body resulting in delayed aging, increased healthspan and longevity. Nonetheless, despite the hope from epidemiologic studies, properly controlled clinical trials are required to understand the specific value of both rapamycin and metformin as therapeutics for aging. Other pathways and mechanisms that have not previously been targeted can also be of great potential and thus the field can not rely on these approaches alone.

2.3. Overcoming translational challenges from animal models to humans

The development of anti-aging drugs poses new challenges. Quantitative measures to assess change and clinical impact will be essential to evaluate the effectiveness of novel therapeutics. Although, for ultimate approval, endpoints that are indicative of improved quality of life will be necessary, the development of such therapeutics can be led by surrogate biomarkers. Several biomarkers have now been identified that distinguish biological from chronological aging in various organisms. These include composite measures of health and physiological function such as the Frailty Index 34, panels of inflammatory cytokines, p16^INK4A(a marker of cellular senescence), and specific patterns of DNA methylation changes (Chen et al., 2016; Horvath, 2013). Of these, DNA methylation changes, known as the “epigenetic clock” have been the most thoroughly studied. The clock was first identified in humans, and shown to predict future mortality and other health outcomes (Horvath, 2013; Chen et al., 2016). Epigenetic clocks have now been identified in mice and respond as expected to experimental interventions - for example, caloric restriction slows the clock down, whereas a high fat diet accelerates it (Chen et al., 2016; Horvath, 2013; Stubbs et al., 2017; Wagner, 2017).

The path to developing therapeutics for disease to be tested in man usually requires the use of animal models in order to define agents, their effects and their safety. The biggest risk factors for age-related diseases are age and time, which are the major constraints to developing animal models in both a logistically and economically feasible manner. Researchers usually use mouse models with aggressive forms of disease that develop the pathology of interest on an accelerated timeline. These models then lack cellular and systemic age-related factors that likely contribute to the disease. It is not cost effective or feasible to incorporate the time variable of normal aging so research paths bias towards abbreviated timelines. It is possible to study lower organisms with much shorter lifespans, but species differences and translatability are even more challenging than with mice. A start towards this would be to have extensive, pre-existing aging mouse colonies, but these still provide complexity - animals show increasing variability in their biology and health as they age. So, in studying age-related processes large cohorts will be required, some mice will die before the time of the intended experiment, others will develop tumors or age-related issues like blindness. Furthermore, immune systems and metabolism will change in ways we don’t fully understand yet. It becomes very difficult to do meaningful mechanistic studies if one doesn’t know whether it’s aging per se or an age-related comorbidity that is driving the phenotype.

A further issue in translation is that humans have more diverse genetic backgrounds and environmental exposures than experimental mice do. It is possible to increase genetic diversity in mice: for example, the NIA Interventions Testing Program uses 4-way crosses that approximate human genetic diversity (Nadon et al., 2008; Lucanic et al., 2017). With respect to environmental factors, in mouse studies it has often been difficult to replicate results from lab to lab even with identical genetic backgrounds and phenotypic assays. Similarly, in humans, environmental conditions in clinical trials also vary among clinical centers and experimenters, and lifetime environmental exposures are far more diverse than in mice. Thus, it may be difficult to find a single intervention that works in everyone, but we should not be discouraged, for example ThioflavinT displays robust anti-aging effects across 22 strains of C. elegans (Lucanic et al., 2017a) and rapamycin extends lifespan across a range of species and in multiple mouse strains. As research advances, we will understand more about the relationship between biology in preclinical models and man and in particular the temporal translation to help us advance therapeutics.

3. What may anti-aging therapeutics look like?

Is a golden bullet of a single drug to impact aging biology a realistic scenario? The diversity of age-related disorders, the multitude of potential endpoints, the complexities of genetic risk factors and environmental challenges accumulating over a lifetime all make a single therapeutic unlikely. However, if there are fundamental underlying mechanisms, such as cellular senescence or failure of proteostatic maintenance, or a natural mixture, such as plasma or a fraction of plasma able to modulate multiple mechanisms, then potentially a single therapeutic could halt, or at least delay, most age-related disorders. It is still early for us to ascertain this, but the prospect will be tested in the clinic.

Non-pharmacological interventions (e.g., exercise and caloric restriction) may be more effective and easier to develop than pharmacological ones in the near term and certainly should be pursued aggressively. Evidence is mounting that such approaches are the most effective that we have currently against disorders such as AD and that they may act through the multimodal mechanisms that we believe are important in aging (Nadon et al., 2008) (Table 1). Many studies are comprehensively testing such approaches and conceptually it can often be easier to see how implementation for these may be, both practically and financially, much more efficient than developing a new pharmaceutical therapy. However, easy as these appear they can be difficult for people to implement – keeping exercise regimens going in the frail elderly (or even healthy adults), or embracing caloric restriction/intermittent fasting is often not a long-term solution to which most can adhere. It may be more effective to prescribe combinations of pharmacological and non-pharmacological interventions. In any event, we need to make the research and development of non-pharmacological interventions a priority to fully understand their impact and underlying biological rationales. Ultimately, we may need to look at individual genetic and environmental risk factors and design a custom-tailored combination of lifestyle and pharmacological interventions.

To some extent, perhaps aging is already being treated, with drugs targeted to specific diseases, such as Metformin for hyperglycemia and aspirin to reduce cardiovascular disease risk. These and a number of other drugs commonly used to treat chronic disease conditions have been found to extend lifespan in lower organisms and in mice (Blagosklonny, 2017; Newman et al, 2016). It may turn out that the most effective treatments for chronic disease that we’ve developed over the last 30 years (e.g., metformin, statins) are ones that impact the interface between aging and disease. On the other hand, these drugs may not affect the aging process per se. The thorough testing of these pre-existing drugs in controlled studies specifically addressing effects on aging, such as the Targeting Aging with Metformin (TAME) study (Barzilai et al., 2016), will be particularly informative.

4. Funding and regulatory issues

Associated with the treatment of aging is a new set of interpretations of clinical success that go beyond purely scientific concepts. In particular, advancing a therapeutic for aging means that we have long timelines, yet to be validated outcome measures and a lack of clarity about the regulatory environment for such therapies. These challenges impact how aging therapeutics will be funded and considered. Nonetheless, there is an appetite by the broad community to achieve these goals and thus navigating this path is feasible and worthwhile.

4.1. Private sector

Ultimately drug development is an expensive business that is driven by private organizations investing in the reward of eventual products. At the early stage, even before pharmaceutical companies are involved, venture capital needs to invest and is certainly drawn to the market demographics of anti-aging therapeutics – approaching a quarter of the population will be older than 60 by 2050. Indeed, a multi-billion dollar per year anti-aging industry already exists in the form of nutraceutical and cosmetic companies. The challenge will be to bring a scientific approach to the space, which rigorously delivers products that are tested and effective to people - not just marketing unproven remedies. New opportunities for developing anti-aging therapeutic approaches can also come from the blurring of lines between technology and biotechnology to offer novel interventions and concepts. However, the private sector needs to explore such new areas in a capital-efficient manner: there need to be clear go/no go points for testing in a reasonable amount of time. Therefore, even if not proven as regulatory endpoints, biomarkers that can be assessed in shorter timeframes will be invaluable in keeping the pathway to therapeutics advancing effectively and ensuring that capital risks arc evaluated during therapeutic development.

4.2. Public sector

The role of the public sector in driving this field is highly significant; pursuing combinations of basic and translational research is fundamental to how success can be achieved. The National Institutes of Health (NIH) is positioned to take some risks that the private sector would be less likely to engage. For example, the National Institute on Aging (NIA) is currently funding a number of studies on the potential beneficial effects of exercise on cognition, both during normal aging and in neurodegenerative disorders, as well as other non-pharmacological approaches, like caloric restriction. The NIA is soliciting applications for clinical trials for pharmaceutical and non-pharmacological interventions to treat not only AD, but the full spectrum of age-related diseases (diabetes, osteoporosis, etc.), including interventions that target multiple comorbid conditions simultaneously. Concepts such as the TAME trial, which will use as its endpoint “time to a new occurrence of composite outcome that includes cardiovascular events, cancer, dementia, and mortality” (Barzilai et al., 2016), are also indicative of the sort of innovative, commercially challenging approach that the NIH can help drive.

NIH funds and partners with industry as well as academia. For example, the Accelerating Medicines Program (AMP) is a collaborative research program co-funded by the NIH, 10 biotechnology companies and 12 non-profit disease foundations. AMP’s goal is to identify and validate new drug targets (with an initial focus on AD, type II diabetes, and rheumatoid arthritis/lupus), making data and analyses from the participating projects freely available in a public database. Both the NIH and the National Science Foundation (NSF) also fund biotechnology companies directly through their Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) grant programs. Aging is not just a domain of interest in the US, funding agencies worldwide are keenly supporting advancement in this area as the whole world’s population is aging and the value of such broad approaches are evident.

Partnerships between public and private sectors may be the most effective way to drive this field forward; with willingness and passion across all areas, we will have the best opportunities to successfully advance.

4.3. Regulatory and clinical trials issues

Regulatory issues pose some hurdles to the development of anti-aging therapeutics, but with a common goal in mind these can be navigated. Firstly, the FDA requires clinical trial endpoints be related to specifically impacting health or quality of life - survival, function, or feeling, not biomarkers. This must be kept in mind as we develop drugs, although biomarkers are going to be critical in assessing efficacy especially over extended time periods, they will not by themselves be sufficient for approval. Secondly, payers require a specific disease code for patient reimbursement, these will need much consideration as we move concepts from targeting specific indications to generalizing age-related diseases. These areas, which can be resolved by working together, should not be left too late for consideration.

5. Conclusions

We are at an exciting juncture where the realities of anti-aging therapies are upon us, and discussing how we can practically advance such approaches is a necessity. Even though a majority of research and therapeutic development focuses on individual domains such as neuroscience or behavior alone, thinking in the context of a systemic impact as we age provides wholly new opportunities, not only to tackle neurological disorders, but a spectrum of age-related ailments. The involvement of multiple disciplines, perspectives and constituents in the field will be needed to be successful. This collaborative approach must be triggered so that quality of life for all can be improved in the near future.