A Conversation on Aging Research

Richard Hodes, MD ’71, is the director of the National Institute on Aging, one of the twenty-seven institutes and centers of the National Institutes of Health. The NIA and its researchers take an encompassing scientific approach to understanding the nature of aging and investigating how to extend the number of healthy active years humans can enjoy. Established by the U.S. Congress in 1974, the NIA is the nexus of research on aging conducted throughout the Department of Health and Human Services. For twenty-eight years, the institute and the work of its scientists have been directed by Hodes.

Harvard Medicine talked with Hodes earlier this year about research at the NIA, his interest in the field of aging research, and what aging, and healthy aging, mean. What follows is an edited version of that conversation.

Harvard Medicine: Let’s start with some basics. How would you define aging?

Richard Hodes: There are several levels to the definition of aging. There’s chronological age, which is measured by the passage of time and the changes that typically occur over time. There’s physiological aging, some of which has to do with appearance, and there’s functional aging. There’s also what is collectively known as aging processes. An example of aging processes is the so-called epigenetic clock of aging, which involves changes in gene expression over time.

The goal is to improve not simply life span but what is now referred to as health span: a high quality of life throughout life.

The concept of an epigenetic clock of aging developed because of experiments that followed individuals over time and monitored genetic changes in tissue such as peripheral blood. When analyzed, researchers found changes that correlated with age, but, interestingly, they also found that these changes varied by individual and by the tissue being monitored. The conclusion of a great many such analyses led to the realization that physiological aging differs from person to person. If you fine-tune the parameters of what aging means at a molecular level, you find that molecular variations also differ from person to person and that these interpersonal variations get greater with age.

The relationship between what happens normally with age, with aging, and what is associated with pathology, disease, and unhealthy aging has led to an appreciation of how conditions or diseases that occur with increased frequency with age reflect some common ongoing processes that accelerate or function abnormally over time. The model for interpreting this is that age-related changes—aging—can be a risk factor for so many diseases and conditions.

HM: Then what is healthy aging?

RH: Healthy aging has meant, and still does mean, growing older chronologically while remaining healthy, without disease or disability. To put it more positively, to grow older while maintaining independence and the ability to bounce back from medical, physical, or other types of stress. So, it’s chronologic age without disease and with the maximum preservation of function. The goal is to improve not simply life span but what is now referred to as health span: a high quality of life throughout life.

HM: Does this evolving understanding of aging reflect the arc of research at NIA during your career there?

RH: Absolutely. The technical, even conceptual, advances for how we analyze aging and the causes of aging have yet to be fully appreciated. Just a few years ago, for instance, we couldn’t measure epigenetic changes in individual cells. We are now able to measure many changes in particular genes or proteins or the folding of proteins. However, we must be careful not to be too reductive; we must remember to bring our understanding back around to how the changes interact with other cellular and organic mechanisms.

HM: Is the lens provided by the Baltimore Longitudinal Study on Aging useful when bringing a more systems-based view to lab-based discoveries?

RH: That study was launched fifty-plus years ago and was the first really organized attempt to understand what happens with aging. Fundamentally, it allows us to consider how aging differs from disease and whether aging processes can be distinguished from disease.

But it also provides an important distinction between the scientific approach of longitudinal versus cross-sectional studies. In the 1950s when it was just getting underway, studies of aging usually compared young people and old people. It was an understandable approach but one we now know is flawed. Who were in these studies? Well, the easiest way to find young volunteers was to approach students. And where were older people found? Nursing homes. So, a lot of the comparisons were of young healthy students and nursing home residents, a group that does not exhibit typical aging.

The Baltimore Longitudinal Study on Aging, which is this country’s longest-running study on aging, was designed to intensely follow participants every year or every couple of years depending upon how old they are and what parameter is being studied. Participants stay at the research center for two or three days, have a series of tests done—blood, cognitive, physical functioning, and capacity. Later, brain imaging was added to the testing battery. Researchers begin to see what happens to individuals over time, and what determines who ages well by avoiding disease disability and who doesn’t.

Everything I’ve just said about research progression from physiologic measures to more molecular ones and genetics and epigenetics has been paralleled in this longitudinal study. One of its great virtues is that early on researchers had the insight to store blood biological materials such as serum and peripheral blood cells. Now, when someone gets an idea about how something might change with age, they can, in many cases, do a retrospective longitudinal study and have twenty or thirty years of data available. Analyses of blood sugar levels, cholesterol levels, and, more recently, epigenetic and proteomic analyses using stored white blood cells can be retrospectively reconstructed. Researchers can also determine the changes that occur at increasingly fine molecular levels.

The study is showing us that aging is not the same as disease and that disease is not inevitable. It also is allowing us to investigate what it is about individuals that’s responsible for their aging differently. Is it genetic? Is it environmental exposures? How much is modifiable? All these questions are part of our ongoing study of the three thousand or so participants.

HM: Have you found any surprises? And have they led to interventions?

RH: Well, there have been some surprises, and they vary over a spectrum of characteristics with phenotypes. Take personality.

HM: Personality?

RH: Yes! We haven’t talked about that, not as a molecular characteristic. It had long been proposed that personality changed with age. Again, a lot of this thinking was based on cross-sectional studies comparing young college students with old people in nursing homes. Real differences were found—and were attributed to age. But longitudinal studies found that, to a large degree, personality characteristics were constant over time. And they could be predictive of health outcomes—for example, optimism was associated with some positive health outcomes. Such associations were interesting because no one really had a sense of those before.

Now, can you translate this insight into an intervention? Well, people have talked about whether you can compensate for what you understand about personality characteristics to improve our health behaviors. There was an appreciation that blood pressure was a risk factor for heart disease or stroke. Back when I was at HMS we were focused on, for example, the diastolic blood pressure as the real risk factor. Studies conducted in the past decades, however, have shown that the systolic pressure is critically important and that when you control blood pressure by bringing the systolic pressure down to levels defined by criteria determined through clinical trials, you decrease the probability of stroke, heart disease, and congestive heart failure.

A study called SPRINT—and SPRINT MIND, a companion study to SPRINT—sought to determine whether a more aggressive control of blood pressure, taking people with a 140 systolic to 120 instead, could be beneficial and, if so, would it be beneficial if you were a 30-year-old, a 40-year-old, a 60-year-old, or an 80-year-old.

That study was ended early for ethical reasons, but that was a very positive reason, for the study found that individuals who had their blood pressure controlled to a level of 120—remember 140 used to be a target—did much better. They survived better and they had fewer cardiovascular events, so the study was stopped. Because SPRINT ended, SPRINT MIND, which was investigating any changes in cognitive ability as a function of systolic pressure, also ended. When the available data from SPRINT MIND were analyzed, however, researchers found trends indicating that the probability of developing mild cognitive impairment, often an early stage of dementia or a precursor of dementia, decreased in people who had more vigorous control of their blood pressure.

But it’s an oversimplification to think that the immune system dampens with age, that it becomes less capable of responding. Changes to the immune system are specific to an individual, and they are not neutral.

Because of this work, our recommendations for blood pressure control, not just in midlife but in older individuals, have changed. These changes have translated into public health recommendations that address other identified risk factors that could potentially make a difference in multiple diseases and conditions of aging.

HM: These studies do point to the benefits of carefully evaluating medical measures, risk factors, and medications used in clinical practice. As you look ahead, do you see equally promising developments coming from research on novel therapeutics?

RH: In fact, I can give you examples of how translation from more molecular science to potentially promising interventions is already occurring.

Some of the changes to the so-called pillars of aging, the components of the growing field of geroscience, pertain to the function of mitochondria, the energy-generating organelles in cells. In experimental models, studies have shown that interventions aimed at improving or sustaining mitochondrial function through aging make a difference in health and outcomes. Those findings are making their way into the first stages of experimental trials.

Another example, which has received a lot of attention lately, is the phenomenon of cellular senescence.

Not so long ago, relatively speaking, we thought cells were immortal. Now we know that’s not the case. The initial definition of senescence referred to replicative senescence—that the cells can’t divide anymore. Well, we’ve learned there’s much more that happens. Senescent cells are not passive; they have an abnormal phenotype that includes the production of a lot of inflammatory proteins.

Genetic manipulation studies involving animal models have provided some very striking results. Specific manipulations of genetic material can selectively eliminate the small percentage of senescent cells that normally exist in an experimental animal. Doing so improves its musculoskeletal integrity, muscle mass, ability to run, cognitive function, function of multiple organs, and overall life-span expectancy.

You can’t use those genetic procedures to directly eliminate senescent cells in humans, but as a corollary, drug interventions in animal models can eliminate the protective mechanisms that senescent cells have that keep them around. Drugs or drug combinations developed from therapeutics used in cancer regimens or used as anti-inflammatories have had a similar effect. If you give such drugs to test animals and selectively eliminate senescent cells, the animals live longer and have better-conserved function overall: brain function, cardiovascular function, musculoskeletal function. Studies of drugs that control senescent cells are in their early stages, with some now being taken into clinical trials. We don’t know if they’re going to be effective, but they do represent an example of translation from basic science.

HM: It’s also an interesting example of how pervasive the effects of inflammation and inflammatory proteins can be in the body. It seems that changes in immune response are increasingly being associated with disease in aging. True?

RH: Multiple changes occur in the immune system with aging. But it’s an oversimplification to think that the immune system dampens with age, that it becomes less capable of responding. Changes to the immune system are specific to an individual, and they are not neutral. COVID-19 provides an example. The immune system in older adults may be less responsive to vaccines or other immunizations, but it may also respond to infection with hyper-responsiveness and inappropriate and prolonged inflammation. The strikingly selective vulnerability of older adults to morbidity and mortality with COVID has been linked to the hyperinflammatory response to the viral infection.

But from COVID we are also learning other things about the effects disease can have on older individuals. These go beyond infection. What must be done to provide care to people who are older, who are institutionalized, who have had their social contacts modified? These are questions we are increasingly asking. Our strategy for caring for older adults will need to include helping them be more socially engaged and more connected with their health care system. It’s not just important to, for instance, gain knowledge of the clinical aspects of immune system function in older people, it’s also vital to promote and sustain the social, behavioral, and economic aspects of their lives.

Ann Marie Menting is the editor of Harvard Medicine magazine.

Image: Stephen Voss