1284

Aging and Cancer

Cancer special section

The Biology of Aging

The Current Research Agenda

William B. Ershler, M.D.1,2

Dan L. Longo, M.D.2
1

Glennan Center for Geriatrics and Gerontology,

Eastern Virginia Medical School, Norfolk, Virginia.
2

Gerontology Research Center, National Institute on Aging, Baltimore, Maryland.

There has been and will continue to be a dramatic shift in the composition of our
population with regard to age. Over the next several decades, the percentage of
the population that is older than 65 years will nearly double. This has obvious and
striking implications, particularly for health care. In anticipation of this change,
there has been increased research activity in an attempt to understand the basic
biology of aging and the mechanisms whereby older individuals become susceptible to disease. The authors reviewed MEDLINE data from the past 2 decades
and key articles from the literature to develop a comprehensive overview of the
background and current status of research in biomedical gerontology. They found
that research in the mechanisms of aging has expanded strikingly during the past
20 years, and that there is now a more precise understanding of age-associated
changes in cellular and molecular functions. Scientific overlap of these age-associated changes with the biology of cancer is now recognized. An appraisal of our
current understanding of aging (or lack thereof) is made in this article. Selected
key research questions are also presented, some of which are particularly germane
to cancer biology and clinical oncology. Cancer 1997;80:128493.
q 1997 American Cancer Society.

KEYWORDS: aging, theories of aging, cancer and aging, immune senescence, free
radicals, telomeres, immunosurveillance.

Presented at Oncology Geriatric Education Retreat: Integrating Geriatrics into Oncology Training, San Juan, Puerto Rico, February 2126,
1997.
Address for reprints: William B. Ershler, M.D.,
Gerontology Research Center, National Institute
on Aging, National Institutes of Health, 4940
Eastern Avenue, Baltimore, MD 21224.
Received May 8, 1997; revision received July
7, 1997; accepted July 7, 1997.

ellular senescence and malignant transformation share certain basic pathways. In fact, in examining genes that control cellular
proliferation and cell death, molecular biologists have discovered that
mutations or alterations in the functions of some of these genes can
result in immortalized cells with malignant properties (i.e., growth
factor independent proliferation and invasiveness). Thus, at the cellular level, aging and cancer can be considered distinct steps on a
spectrum of cell behavior. Recently, there has been an increased
awareness of this paradigm, and those who study the basic sciences
of aging and cancer are opening a dialogue that has been mutually
beneficial.
In this review, currently prevalent theories of aging are discussed
and presented in the context of their applicability to cancer biology.
These include theories regarding programmed (genetic) aging, protein
and DNA glycation, and free radicals. Experimental models are reviewed herein, with mention of the influence of dietary restriction on
both aging and cancer development. Several key research questions
are presented, to which both oncology and gerontology perspectives
are critical. These include questions about the increase in cancer
incidence with advancing age and changes in tumor aggressiveness
that occur in older hosts. Other questions, about the frequency of
multiple primary cancers in older individuals and the apparent resis-

q 1997 American Cancer Society

/ 7b71$$ge4D

09-10-97 11:25:47

cana

W: Cancer

The Biology of Aging/Ershler and Longo

tance to malignancy in the oldest patients, are also

offered as subjects of importance to biomedical gerontology.

Aging versus Disease

It is a central dogma of geriatrics that aging is not a
disease and being old does not mean being infirm.
The functional declines that accompany normal aging
(discussed by Lipschitz1) have been well characterized,
but under normal circumstances they do not account
for symptoms of disease. For example, kidney function
decline with age is well recognized; 2 in fact, it has
proven to be a useful biologic marker of aging. Yet,
clinical consequences of this change in renal function
are not common in the absence of disease or exposure
to an exogenous nephrotoxic agent. Similarly, bone
marrow changes with age. Bone marrow stem cells are
fewer and proliferative potential of progenitor cells is
less.3,4 However, anemia does not occur spontaneously
in elderly individuals with no disease, and neither do
neutrophil or platelet abnormalities. In fact, in rodents, serial transplantation of bone marrow has indicated that there is enough regenerative capacity to
sustain life for several generations.3 Distinct changes
in measurable immune functions with age have also
been described (as reviewed by Miller5), but the clinical consequences of these are minimal or even nonexistent in the absence of disease. Whether these
changes contribute to a heightened susceptibility to
infection is a subject of debate.
Aging is not a disease process, but the changes
that come with aging may make an individual susceptible to disease. For example, those changes observed
in the immune system, although not primarily a problem, may render an individual susceptible to reactivation of tuberculosis6,7 or herpes zoster8 and less capable of responding to influenza vaccine with protective
titers of antibody.9,10 The immune decline, however,
is not of sufficient magnitude or duration to account
for the increased incidence of cancer among older
people.11 In fact, we12,13 and others14 17 have shown in
experimental models that immune senescence, paradoxically, may contribute to the observed reduction in
tumor growth and spread in a variety of tumors.
Life Span (Median and Maximum Survival)
From the perspective of those who study aging, there
is an important distinction between median survival
(which determines life expectancy) and maximum survival. Over the past several decades, with the advent
of modern sanitation, refrigeration, and other public
health measures (including vaccination and antibiotics), there has been a dramatic increase in median
survival.18 Early deaths have been diminished and

/ 7b71$$ge4D

09-10-97 11:25:47

cana

1285

more individuals are reaching old age. In the United

States today, life expectancy now approaches 80
years.19 Median survival is what concerns public health
officials and health care providers. In contrast, maximum survival is the focus of those gerontologists interested in the biology of aging and longevity. It is worthwhile to note that it has been estimated that if atherosclerosis and cancer were eliminated from the
population as a cause of death, about 10 years would
be added to the average life span, yet there would be
no change in the maximum life span.20
The age of the oldest human being alive today is
approximately 120 years. What is intriguing is that the
record has remained stable, unchanged by the public
health initiatives mentioned above. In fact, there has
been some recent data presented that the maximum
survival is actually declining in the United States.21,22
In the laboratory, similar limits have been established
for a variety of species. Drosophila, which are free of
predators, can live 30 days, whereas C57BL/6 mice
that live in a laboratory environment and are allowed
to eat a healthy diet ad libitum may survive 40 months.
What is interesting is that, unlike the public health
initiatives for humans, experimental interventions for
lower species have been associated with a prolongation of maximum survival. In Drosophila, for example,
transgenic offspring producing extra copies of the free
radical scavenging enzymes superoxide dismutase and
catalase survived about 33% longer than controls.23
However, there has been some criticism of this work
based on the claim that the controls were unusually
short-lived. In mammalian species, the only experimental intervention that characteristically prolongs
maximum survival is the restriction of caloric intake.
In fact, dietary restriction (DR) has become a common
experimental paradigm exploited in the investigation
of primary processes of aging (for a review, see Weindruch24).
Briefly, DR typically involves a reduction of 30
40% in caloric intake, with careful attention paid to
the provision of adequate amounts of essential nutrients. It is associated with both a delay in the acquisition of age-related diseases and a reduction in the rate
of achieving certain established biomarkers of aging
(i.e., a retardation in primary aging). Furthermore, DR
significantly reduces the incidence of cancer in cancer-prone animals whether the carcinogen is viral or
chemical. The critical questions remain, What is the
mechanism of the DR effect, and will it be applicable
to higher species? With regard to the latter, there are
now at least four comprehensive and interactive studies being conducted in the United States in which DR
is being examined in nonhuman primates. Although
it appears that the monkeys in these studies whose

W: Cancer

1286

CANCER October 1, 1997 / Volume 80 / Number 7

caloric intake is restricted are assuming a more youthful phenotype in a variety of physiologic measures,25,26
it is clearly too early to predict whether maximum
survival will be affected.

Aging versus Time

Another distinction at the heart of gerontology is between time and aging. Time itself is the standard by
which we predict aging, but in fact it is commonly
appreciated that cells, tissues, organs, and organisms
age at different rates. In contrast, aging is thought
of as the phenotypic change that occurs over time and
results in alteration of function or appearance. Byand-large, these changes result in limitations in functional reserve and result in increased susceptibility to
disease. However, it is currently nearly impossible to
predict what the rate of aging will be in any particular
system. There is no question that humans age at different rates, yet at birth there is no way to know for
certain what the aging pattern of an individual will be.
Genetic composition is critical but does not tell
the whole story. For example, genetically identical
mice may acquire biologic hallmarks of advancing age
at different rates and may even have remarkably different life spans. Nonetheless, there are genes that portend life span, and inbred species have been very instructive in their elucidation. Thus, there are longlived and short-lived strains, and their genetic compositions are subjects of intense investigation with regard
to the identification of longevity-associated genes and
biologic aging.
Some short-lived strains have shortened survival
due to predisposition to certain diseases, and these
are not considered good models for normal aging. The
New Zealand black (NZB) mouse, for example, acquires a dysregulated immune system and a disease
similar to lupus erythematosus that leads to premature
death.27 Although an excellent model for lupus, the
NZB mouse dies young (at about 12 months) without
achieving disease-independent markers of older age;
thus, it is not a good animal for research in aging. In
contrast, it appears that the heartiest strains (those
with the longest survival and the least predisposition
to early disease) are genetic hybrids that have a demonstrated maximum survival of 4 years or longer when
subjected to DR paradigms.28
It is important to note that within a specific strain
there is some homogeneity in patterns of aging. Accordingly, C57BL/6 mice fed ad libitum are likely to
live 24 26 months, and half at this age have or will
develop lymphoma.29 Fisher 344 rats have a slightly
longer survival but also have a propensity for developing renal failure late in life.30 Biomedical gerontologists have become familiar with these patterns of aging

/ 7b71$$ge4D

09-10-97 11:25:47

cana

and disease and, by necessity, have selected appropriate animal models for their investigations. A general
rule, however, is that short-lived strains, particularly
those with a genetic susceptibility to life-threatening
diseases, are not good models for research in aging.
In contrast, factors that influence maximum survival
in robust hybrid strains probably do so by interfering
with primary aging processes; thus, these strains make
excellent models.

Cellular versus Organismal Aging

There has been much written about cellular senescence and the events that lead up to cell death (see
the review by Cristofalo and Pignolo31). After a finite
number of divisions, normal somatic cells invariably
enter a state of irreversibly arrested growth, a process
termed replicative senescence.32 In fact, it has been
proposed that escape from the regulators of senescence is what oncologists term malignant transformation. However, the role of replicative senescence
in organismal aging remains a subject of vigorous debate. The controversy relates in part to the fact that
certain organisms (e.g., Drosophila and C. elegans) undergo an aging process, yet all of their adult cells are
postreplicative.
What is clear is that the loss of proliferative capacity of human cells in culture is intrinsic to the cells
and not dependent on environmental factors or even
culture conditions.32 Unless transformation occurs,
cells age with each successive division. The number
of divisions turns out to be more important than the
actual amount of time that passes. Thus, cells that are
held in a quiescent state for months and then allowed
back into a proliferative environment will continue
approximately the same number of divisions as cells
that were allowed to proliferate without a quiescent
period.33
The question remains whether this in vitro phenomenon is relevant to animal aging. One suggestive
observation is that fibroblasts cultured from samples
of old skin undergo fewer cycles of replication than
those from samples of young skin.34 Furthermore,
when various species are compared, replicative potential is directly and significantly related to life span.35
Recently, an unusual b-galactosidase whose activity
peaks at pH 6, proved to be a useful biomarker of in
vitro senescence because it was expressed by senescent but not presenescent or quiescent fibroblasts.36
This unusual b-galactosidase isoform was found to
have the predicted pattern of expression in skin from
both young and old donors.36 Thus, there was an ageassociated increase in pH 6 b-galactosidase present in
dermal fibroblasts and epidermal keratinocytes, providing an in situ correlate of replicative senescence.

W: Cancer

The Biology of Aging/Ershler and Longo

TABLE 1
Theories of Aging
Types of models

Theories

Intrinsic/Stochastic

Somatic mutation40,41
Intrinsic mutagenesis45
Impaired DNA repair46
Error catastrophe47
Ionizing radiation40,41,43,49
Free radical damage52,53
Neuroendocrine91
Immune68

Extrinsic/Stochastic
Genetically determined

The nature of the expression of this in vivo biomarker

of aging in other tissues will be important to discern.

Biomedical Gerontology: Key Research Areas

Although we by no means provide a comprehensive
list, in the paragraphs below we mention topics that
are currently of investigative interest to biogerontologists.

Theories of aging
Providing a rational, unifying explanation for the aging
process has been the subject of a great number of
theoretic expositions. Yet, no single proposal has sufficed to account for the complexities observed. Studies
that present theories of aging are listed in Table 1.
That genetic controls are involved seems obvious
when one considers that life span is highly species
specific. For example, mice generally live approximately 2 years and humans approximately 90 years.
However, the aging phenomenon is not necessarily a
direct consequence of the primary DNA sequence.
Mice and bats, for instance, have a 0.25% difference
in their primary DNA sequence, but bats live for 25
years 10 times longer than mice. Thus, regulation
of gene expression seems likely to be the source of
differences in longevity between species.
Although within a species there is considerable
variation in longevity, this variability is much less
among inbred strains or monozygotic twins than
among dizygotic twins or nontwin siblings. Furthermore, various genetically determined syndromes have
remarkable (albeit incomplete) features of accelerated
aging. These include Hutchinson Gilford syndrome
(early onset progeria), Werners syndrome (adult onset
progeria), and Downs syndrome.37 Although no progeria syndrome manifests a complete phenotype of
advanced age, the identification of the genes responsible for these particular syndromes is beginning to pay
dividends by providing clues to the molecular mechanisms involved in the aging process. For example, Wer-

/ 7b71$$ge4D

09-10-97 11:25:47

cana

1287

ners syndrome is now known to be caused by mutations in a single gene on chromosome 8 that encodes
a protein containing a helicoid domain.38,39 The future
functional characterization of this specific protein will
undoubtedly increase our level of understanding of
the aging process.
Examination of aging in yeast has also been informative with regard to the genetic controls of aging.
These single-cell organisms follow the replicative limits of mammalian cells, and it has been observed that
life span is related to silencing large chromosomal
regions. Mutations in these silencing genes lead to
increased longevity.40
Thus, if there are certain genes that regulate normal aging or at least are associated with the development of a phenotype in the aged, it stands to reason
that acquired damage to those genes might influence
the rate of aging. Over the years, several theories have
been proposed that relate to this supposition. In general, they hypothesize a stochastic accumulation of
damage, to either DNA or protein, that leads eventually to dysfunctional cells, cell death, subsequent organ dysfunction, and ultimately organism death.
Prominent among these is the somatic mutation theory,41 which predicts that genetic damage (from background radiation, for example) accumulates, produces
mutations, and results in functional decline. A variety
of refinements to this theory have been suggested, invoking the importance of mutational interactions,42
transposable elements,43 and changes in DNA methylation status.44
A related hypothesis is Burnets intrinsic mutagenesis theory,45 which proposes that spontaneous or endogenous mutations occur at different rates in different species and that this accounts for the variability
observed in life span. Closely related to this notion
is the DNA repair theory.46 Initially, there was great
excitement about this, as it was found that long-lived
animals had demonstrably greater DNA repair mechanisms than shorter-lived species.46 However, longitudinal studies of particular species have not revealed
that repair mechanisms decline consistently with age.
This, of course, does not rule out the possibility that
repair of certain specific and critical DNA lesions is
altered with advancing age. We now understand that
there are multiple DNA repair mechanisms, including
base excision repair, transcription-coupled repair,
and, as was discovered most recently, even DNA repair
mechanisms based in mitochondria. Disorders involving one or a subset of repair mechanisms could lead
to accumulation of DNA damage and dysfunction.
In yet another intrinsic/stochastic model, the error catastrophe theory proposed by Orgel,47 it was suggested that random errors in protein synthesis occur

W: Cancer

1288

CANCER October 1, 1997 / Volume 80 / Number 7

and that when the proteins involved are those responsible for DNA or RNA synthesis, there is resultant DNA
damage with consequences for daughter cells. Although this model has appeal, there has been no reported evidence that protein synthesis machinery becomes impaired or inaccurate with advancing age.
However, a candidate protein that may eventually be
shown to be so affected is telomerase. This critical
enzyme is comprised of protein and an RNA template;
the template is necessary for maintaining telomere
length and cell replicative potential. As cells become
senescent in vitro, telomerase activity declines, telomeres shorten, and ultimately replicative potential
is lost.48
Evidence that exogenous factors are involved in
the acquisition of age-associated damage to DNA and
protein is derived from a number of observations,
many of which are circumstantial or correlative but
nonetheless provocative. It now appears that the accumulation of abnormal protein within senescent cells,
as predicted by the error catastrophe theory, actually
reflects postranslational events, such as oxidation or
glycation and resultant cross-linking. There is theoretic appeal in the concept that key proteins, such
as collagen or other extracellular matrix proteins, and
DNA become dysfunctional with age due to the impairment produced by these cross-links.49 51
One mechanism that produces cross-links is
called glycation, the nonenzymatic reaction of glucose
with the amino groups of proteins. Presumably, glycation would occur more readily in the presence of
higher serum levels of glucose; thus, this theory fits
well with the observed age-associated dysregulation
of glucose metabolism and prevalent hyperglycemia
in geriatric populations. Of course, this also points out
its deficiency as a unifying mechanism, as there is no
question that individuals with well-maintained glucose levels throughout their life span will still be subject to acquired changes that are typical of aging.
Another mechanism held responsible for crosslinking is the damage produced by free radicals, and
this forms the basis of the free radical hypothesis initially promoted by Harman.52,53 This theory suggests
that aging is the result of DNA and protein damage
(e.g., mutagenesis or cross-linking) by atoms or molecules that contain unpaired electrons (free radicals).
These highly reactive species are produced as by-products of a variety of metabolic processes and are normally inhibited by intrinsic cellular antioxidant defense mechanisms. If free radical generation increases
with age, or the defense mechanisms that scavenge
free radicals (e.g., glutathione) or repair free radical
damage decline, the accumulated free radical damage
may account for altered DNA and protein function.

/ 7b71$$ge4D

09-10-97 11:25:47

cana

Evidence to support this widely held notion is incomplete. It is known that free radical generation in mammals correlates inversely with longevity; 54 similarly,
levels of free radical inhibiting enzymes, such as superoxide dismutase, have been shown to be higher in
species with longer life spans.55 However, efforts at
enhancing antioxidant mechanisms with dietary vitamin E have resulted in only a modest enhancement
of median survival in mice and no affect on maximum
life span.56
Recently, there has been much attention focused
on mitochondrial function in the context of free radical damage, because oxidative metabolism and the
production of reactive oxygen species mostly occur in
these organelles. Although mitochondrial DNA codes
for antioxidant enzymes in addition to enzymes involved in energy production, it is currently believed
that energy production declines with age due to mitochondrial DNA damage by those reactive products.
Indeed, mitochondrial damage has been shown to increase with age in experimental models,57 59 and the
shortened survival of knockout mice that are deficient
in mitochondrial antioxidant enzymes has supported
the potential importance of this mechanism.60
The most compelling data to date in support of
the free radical hypothesis come from the experiments
of Orr and Sohal, in which transgenic Drosophila producing enhanced levels of superoxide dismutase and
catalase had a maximum survival 33% greater than
controls.23 Furthermore, it is known that flies produce
high levels of free radicals associated with their impressive metabolic requirements and that their survival is enhanced dramatically when their ability to
fly is experimentally hindered.61 However, the idea of
generalizing these findings has been questioned. Some
have criticized the transgenic Drosophila experiment,
claiming that the controls were rather short-lived. Furthermore, transgenic mice overexpressing free radical
scavenging enzymes have produced very modest effects on life span.62 Thus, the conclusion that augmentation of free radical scavenging mechanisms increases
longevity cannot be considered a proven fact.
From a different perspective, there is also very
good evidence implicating a nonrandom, perhaps genetically regulated endogenous mechanism in the
aging process. For example, the neuroendocrine theory suggests that the decrements in neuronal and associated hormonal function are central to aging. It has
been suggested that age-associated decline of hypothalamic-pituitary-adrenal axis function results in a
physiologic cascade leading ultimately to the frail
phenotype. This hypothesis is appealing because it is
well established that this neuroendocrine axis regulates much of the development and also the involution

W: Cancer

The Biology of Aging/Ershler and Longo

of ovarian and testicular function. Furthermore, ageassociated declines in growth hormone and related
factors,63 dehydroepiandrosterone,64 and secondary
sex steroids65 have been implicated in age-associated
impairments, including a reduction in lean body mass
and bone density. Furthermore, pharmacologic reconstitution using these or related hormones has met with
some success in reversing age-associated functional
decline.66,67
Similarly, it has been argued that involution of
the thymus gland and subsequent decline in immune
function is a key regulator of aging.68 The argument is
based on the observation that the decline in immune
function occurs in all mammalian species but occurs
later in those with longer survival.69 Furthermore, dietary restriction is associated with maintained thymic
mass and measurable immune function as well as prolonged survival, suggesting an association of a decline
in immunity with primary aging processes.
This possibility is highlighted by the observation
that differences in the maximum survival of different
mouse strains has been associated with specific alleles
in the major histocompatibility complex, which in turn
code for immunologic determinants.70 This hypothesis, although not without its appeal, is not widely accepted as a major explanation for aging. Perhaps this
relates to the fact that biologic aging is a universal
phenomenon and certain features of it are held in
common, even among organisms with primitive or no
immune function. (The same could also be said for the
neuroendocrine theory.) It is obvious that the immune
system is of great importance in minimizing the
chance of early death, particularly from infectious diseases. Accordingly, median life span is clearly influenced by competent immunity. However, immunologic reconstitution of middle-aged or old animals has
not been shown to prolong survival.71

Biologic markers of aging

A major concern for researchers in the field is the
ability to measure aging, distinct from disease, in an
animal or human. Accordingly, the concept of a panel
of species specific biologic markers of normal aging
has been espoused (for a review, see Baker and
Sprott),72 and the National Institute on Aging has
sponsored a major program initiative to develop such
a panel of markers. The ideal biomarker would be one
that changes in a characteristic pattern with the passage of time, is not (or is only minimally) altered by
age-associated disease, and allows a prediction of both
physiologic age and expected life span. Candidate
biomarkers in various species are listed in Table 2.

/ 7b71$$ge4D

09-10-97 11:25:47

cana

1289

TABLE 2
A Sampling of Candidate Biomarkers of Aging
Subjects

Markers

Cells in Culture

b-galactosidase36
telomere length48
marrow stromal colony size92
protein carbonyl content93
mitochondrial H2O2 release61
striatal dopamine reeceptors94
Ca// ionophore lymphocyte response95
tail tendon fiber strength96
mitochondrial DNA damage57
presbyopia97
fingernail growth rate98
interleukin-6 level99
glycation end products103
mitochondrial DNA damage59
b-galactosidase in skin biopsy36
lenticular glutathione100
presbyopia101
DNA unwinding rate102

Flies
Mice, Rats

Monkeys

Humans

Immunity and aging

There is a well-characterized deficit in immune function with advancing age (for a review, see Miller5);
however, as mentioned above, the consequences are
not fully established. It is apparent that otherwise
healthy older individuals are more susceptible to reactivation of tuberculosis6,7 or Herpes zoster,8 and their
responses to vaccines, such as the commercially available and widely used influenza hemagglutinin, are
lower.9,10 However, it has been postulated that other
age-associated diseases, such as cancer,73 athersclerosis,74 diabetes,75 and even Alzheimers disease,76,77
have been related to the immune decline with age. Yet,
more recent evidence would argue that inflammation
may contribute to Alzheimers disease.
What can be said with confidence is that there
are changes in T-cell function with age that result in
decreased proliferation when measured in vitro.78
When studied as a population, there appears to be an
accumulation of T cells that have cell surfaces characteristic of memory cells; in contrast, there is a relative
decrease in nave T cells.79 B-cell function including
the capacity to make antibodies remains intact, although certain intrinsic alterations have been noted.80
Immunoregulatory functions are affected by the aging
process and paraproteinemia, and autoantibodies are
observed with increasing frequency with each passing
decade. In general, the paraproteinemia is an indicator
of dysregulated immunity, but it is considered not to
be the antecedent of multiple myeloma.81 83 However,
myeloma does increase in incidence in geriatric populations, and it must be distinguished from the benign

W: Cancer

1290

CANCER October 1, 1997 / Volume 80 / Number 7

paraproteinemia of aging. Typically, this is accomplished by examination of bone marrow, skeletal Xrays, and renal function.84
Other indications of dysregulated immune function are the alterations in certain key cytokines that
are measured in plasma, culture supernatants, or the
appropriate tissue microenvironment. Notably and
consistently, interleukin (IL)-2 levels and function decrease with age,85 whereas IL-6 levels increase.86 The
decline in IL-2 may account for a significant component of the measured decline in T-cell function, and
the increased IL-6 has been implicated in the pathogenesis of certain age-associated diseases, including
osteoporosis, Alzheimers disease, and cancer.87

FIGURE 1. Immune forces, tumor antigenicity, and age are represented.

Immune senescence and tumor growth

There is a curious association of immune competence
and experimental tumor growth that has been of some
interest to both gerontologists and oncologists. It has
long been recognized that certain tumors, particularly
of the breast and prostate (but also, to a lesser extent,
of the lung and colon), appear clinically less aggressive
in older people.88 Although this has proven difficult to
establish in large clinical data bases, there is no question in that certain experimental animal models,
slower tumor growth, fewer metastases, and longer
survival are observed in older hosts.12,16 Several explanations have been postulated, including less vigorous
angiogenic response or other microenvironmental or
extracellular matrix soil factors, but experimental
evidence has supported the paradoxical role of immune senescence. Thus, old animals reconstituted
with the syngeneic immune systems of young donors,
either by thymus and spleen17 or bone marrow13 transplantation, develop more rapid and aggressive tumor
growth, whereas young animals that have been irradiated and reconstituted with old immune cells have
tumor growth characteristics that resemble old control
animals. These observations have led to the reevaluation of the immune-enhancement hypothesis (see
Fig. 1), which was originally proposed by Prehn
and Lappe.89,90

Common Ground: Cancer and Aging

By virtue of demography, cancer may now be considered a geriatric syndrome, and clinicians trained to
treat adults with cancer should be aware that their
patient population will gradually become older. The
focus on common pathways in malignant transformation, cellular immortality, and senescence by both gerontologists and oncologists is a clear indication that
the disciplines of gerontology and oncology have
much in common. Thus, free radicals, telomeres and
telomerase, tumor suppressor genes, cellular prolifer-

/ 7b71$$ge4D

09-10-97 11:25:47

cana

This figure illustrates the proposed counterimmune forces that might influence tumor growth and explain the apparent reduction in tumor growth
that occurs with age. Shaded figures of different sizes represent tumors
growing at different rates. Tumors 2 and 4 represent weakly antigenic
tumors, whereas 1 and 3 represent strongly antigenic tumors. The two
figures at the top represent tumors in an immunocompetent (or young)
host, wheras the figures below represent tumors in an immunodeficient
(or older) host. The strength of immune-facilitating forces (for each tumor,
the set of arrows within the tumor) and of immunosuppressive forces (for
each tumor, the external arrow on the left side) are represented by the
sizes of the arrows. The strongly antigenic tumors induce strong immunosuppressive responses only in immunocompetent hosts, and tumor growth
is therefore inhibited. Immunodeficient hosts produce a weak immunosuppressive response to a strongly antigenic tumor, and tumor growth is
correspondingly greater (Tumor 2 vs. Tumor 1). Weakly antigenic tumors
induce little immunosuppressive response in either immunocompetent or
immunosuppressed hosts. The immunocompetent host will produce a
greater immune-facilitating response (see text), and tumor growth will be
greater than in the immunosuppressed host (Tumor 2 vs. Tumor 4). This
hypothesis predicts that in an immunodeficient (aged) host with a weakly
antigenic tumor, the tumor growth would be slow compared with the
growth of the same tumor in an immunocompetent young host.

ation, and genetic control are among the many common research fronts. With the aging of the population
and the increased frequency of cancer, a research
agenda that relates both to primary aging and cancer
biology is of obvious importance. Among the important basic clinical research questions to address
are the following:
1. What determines whether the accumulation of genetic damage in a particular cell leads to death,
growth arrest/senescence, or neoplasia?
2. Which genes regulate cellular aging, and do they
relate to organismal aging?
3. To what extent does free radical damage contribute

W: Cancer

The Biology of Aging/Ershler and Longo

4.

5.
6.

7.

8.

to aging and/or cancer, and can this damage be

prevented?
How can we explain the increased frequency of
multiple primary tumors that occurs with advancing age?
What is the influence of comorbidity on cancer outcome in the elderly?
What functional assessments will best predict how
an individual patient will respond to treatment (i.e.,
toxicity vs. response)?
What cancers are biologically similar across the life
span, and what cancers seem biologically distinct
as a function of age?
Is there an inherent resistance to cancer development in the oldest age group (i.e., those older than
85 years)? If so, what is the mechanism?

REFERENCES
1.
2.
3.

4.

5.
6.
7.
8.
9.

10.

11.

12.

13.

14.

15.

Lipschitz D. Physiologic changes with age. Cancer 1997. In

press.
Lindeman RD. Overview: renal physiology and pathophysiology of aging. Am J Kidney Dis 1990;16:27582.
Harrison DE. Proliferative capacity of erythropoietic stem
cell lines and aging: an overview. Mech Ageing Dev
1979;9:40926.
Harrison DE, Astle CM, Stone M. Numbers and functions of
transplantable primitive immunohematopoietic stem cells:
effects of age. J Immunol 1989;142:383340.
Miller RA. Aging and immune function. Int Rev Cytol
1991;124:187209.
Dubrow EL. Reactivation of tuberculosis: a problem of aging.
J Am Geriatr Soc 1976;24:4817.
Nagami PH, Yoshikawa TT. Tuberculosis in the geriatric patient. J Am Geriatr Soc 1983;31:35663.
Gelato MC. Aging and immune function: a possible role for
growth hormone. Hormone Res 1996;45:469.
Arden NH, Patriarca PA, Kendal AP. Experiences in the use
and efficacy of inactivated influenza vaccine in the nursing
home. In: Kendal AP, Patriarca PA, editors. Options for the
control of influenza. New York: Alan Liss, Inc., 1986;15568.
Powers DC, Sears SD, Murphy BR, Thurmar B, Clements
ML. Systemic and local antibody responses in elderly subjects give live or inactivated influenza A virus vaccines. J
Clin Microbiol 1989;27:266671.
Kaesberg PR, Ershler WB. The importance of immune senescence in the incidence and malignant properties of cancer
in hosts of advanced age. J Gerontol 1989;44:636.
Ershler WB, Stewart JA, Hacker MP, Moore AL, Tindle BA.
B16 melanoma and aging: slower growth and longer survival
in old mice. J Natl Cancer Inst 1984;72:1614.
Ershler WB, Moore AL, Shore H, Gamelli RL. Transfer of ageassociated restrained tumor growth in mice by old to young
bone marrow transplantation. Cancer Res 1984;44:567780.
Stjernsward J. Age-dependent tumor host barrier and effect
of carcinogen-induced immunodepression on rejection of
isografted methylcholanthrene-induced sarcoma cells. J
Natl Cancer Inst 1966;37:50512.
Yuhas JM, Pazmino NH, Proctor JO, Toya RE. A direct relationship between immune competence and the subcutaneous growth rate of a malignant murine lung tumor. Cancer
Res 1974;34:7228.

/ 7b71$$ge4D

09-10-97 11:25:47

cana

1291

16. Rockwell SC. Effect of host age on transplantation, growth

and radiation response of EMT6 tumors. Cancer Res 1981;
41:52731.
17. Tsuda T, Kim YT, Siskind GW, et al. Role of the thymus and
T cells in slow growth of B16 melanoma in old mice. Cancer
Res 1987;47:3097102.
18. Christensen K, Vaupel JW. Determinants of longevity: genetic, environmental and medical factors. J Intern Med 1996;
240:33341.
19. Guyer B, Strobino DM, Ventura SJ, MacDorman M, Martin
JA. Annual summary of vital statistics, 1995. Pediatrics 1996;
98:100719.
20. Greville TNE. U.S. life tables by cause of death, 19691971.
U.S. Decennial Life Tables for 19691971 1971;1:515.
21. Riggs JE. Longitudinal Gompertzian analysis of adult mortality
in the U.S., 19001986. Mech Ageing Dev 1990;54:23547.
22. Hirsch HR. Can an improved environment cause maximum
survival to decrease? comments on lifespan criteria and longitudinal Gompertzian analysis. Exp Gerontol 1994;29:11937.
23. Orr WC, Sohal RS. Extension of life-span by overexpression
of superoxide dismutase and catalase in Drosophila melanogaster. Science 1994;263:112830.
24. Weindruch R. Caloric restriction and aging. Sci Am 1996;274:
4652.
25. Nakamura E, Lane MA, Roth GS, Ingram DK. Evaluating measures of hematology and blood chemistry in male rhesus monkeys as biomarkers of aging. Exp Gerontol 1994;29:15177.
26. Kemnitz JW, Weindruch R, Roecker EB, Crawford K, Kaufman PL, Ershler WB. Dietary restriction of adult male rhesus
monkeys: design, methodology, and preliminary findings
from the first year of study. J Gerontol 1993;48:B1726.
27. East J, Prosser PR. Autoimmunity and malignancy in New
Zealand black mice. Proc R Soc Med 1967;60:8235.
28. Weindruch R, Walford RL. Dietary restriction in mice beginning at 1 year of age. Science 1982;215:14158.
29. Pattengale PK, Taylor CR. Experimental models of lymphoproliferative disease: the mouse as a model for human nonHodgkins lymphomas and related leukemias. Am J Pathol
1983;113:23765.
30. Weindruch R, Masoro EJ. Concerns about rodent models for
aging research. J Gerontol 1991;3:B878.
31. Cristofalo VJ, Pignolo RJ. Replicative senescence of human
fibroblast-like cells in culture. Physiol Rev 1993;73:61738.
32. Hayflick L. The limited in vitro lifetime of human diploid
cell strains. Exp Cell Res 1965;37:61436.
33. Cristofalo VJ, Palazzo R, Charpentier RL. Limited lifespan of
human fibroblasts in vitro: metabolic time or replications?
In: Adelman RC, Roberts J, Baker GT, editors. Neural regulatory mechanisms during aging. New York: Alan R. Liss, Inc.,
1980.
34. Schneider EL, Mitsui Y. The relationship between in vitro
cellular aging and in vivo human age. Proc Natl Acad Sci U
S A 1976;73:35848.
35. Rohme D. Evidence for a relationship between longevity of
mammalian species and life-spans of normal fibroblasts in
vitro and erythrocytes in vivo. Proc Natl Acad Sci U S A 1976;
78:500913.
36. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C,
et al. A biomarker that identifies senescent human cells in
culture and in aging skin in vivo. Proc Natl Acad Sci U S A
1995;92:93637.
37. Martin GM. The genetics of aging. Hosp Pract (Off Ed)
1997;32:4750.

W: Cancer

1292

CANCER October 1, 1997 / Volume 80 / Number 7

38. Yu CE, Oshima J, Fu YH, Wijsman EM, Hisama F, Alisch R,

et al. Positional cloning of the Werners syndrome gene.
Science 1996;272:25862.
39. Yu CE, Oshima J, Wijsman EM, Nakura J, Miki T, Piussan C,
et al. Mutations in the consensus helicase domains of the
Werner syndrome gene. Am J Genet 1997;60:33041.
40. Kennedy BK, Guarente L. Genetic analysis of aging in Saccharomyces cerevisiae. Trends Genet 1996;12:3559.
41. Szilard L. On the nature of the aging process. Proc Natl Acad
Sci U S A 1959;45:305.
42. Morley AA. Is aging the result of dominant or co-dominant
mutations? J Theor Biol 1982;98:46974.
43. Cummings DJ. Mitochondrial DNA in Podopora anserina: a
molecular approach to cellular senescence. Monogr Dev Biol
1984;17:25466.
44. Fairweather S, Fox M, Margison BP. The in vitro lifespan of
MRC-5 cells is shortened by 5-azacytidineinduced demethylation. Exp Cell Res 1987;168:1538.
45. Burnet M. Intrinsic mutagenesis: a genetic approach for
aging. New York: John Wiley & Sons, Inc., 1974.
46. Hart RW, Setlow RB. Correlation between deoxyribonucleic
acid excision repair and lifespan in a number of mammalian
species. Proc Natl Acad Sci U S A 1974;71:216973.
47. Orgel LE. The maintenance of the accuracy of protein synthesis and its relevance to aging. Proc Natl Acad Sci U S A
1963;49:51721.
48. Allsop RC, Vaziri H, Patterson C, Goldstein S, Younglai EV,
Futcher AB, et al. Telomere length predicts replicative capacity of human fibroblasts. Proc Natl Acad Sci U S A 1992;
89:101148.
49. Bjorksten J. Cross linkage and the aging process. In:
Rothstein M, editor. Theoretical aspects of aging. New York:
Academic Press, 1974:43.
50. Kohn RR. Aging of animals: possible mechanisms. In: Principles of mammalian aging. 2nd edition. Englewood Cliffs, NJ:
Prentice-Hall, 1978.
51. Kreisle RA, Stebler B, Ershler WB. Effect of host age on tumor
associated angiogenesis in mice. J Natl Cancer Inst 1990;
82:447.
52. Harman D. Aging: a theory based on free radical and radiation chemistry. J Gerontol 1956;11:298300.
53. Harman D. The aging process. Proc Natl Acad Sci U S A
1981;78:71248.
54. Sohal RS, Svensson I, Sohal BH, Brunk UT. Superoxide radical production in different animal species. Mech Ageing Dev
1989;49:12935.
55. Sohal RS, Sohal BH, Brunk UT. Relationship between antioxidant defenses and longevity in different mammalian species. Mech Ageing Dev 1990;53:21727.
56. Sohal RS, Allen RG. Oxidative stress as a causal factor in
differentiation and aging: a unifying hypothesis. Exp Gerontol 1990;25:499522.
57. Lee CM, Chung SS, Kaczkowski JM, Weindruch R, Aiken JM.
Multiple mitochondrial DNA deletions associated with age
in skeletal muscle of rhesus monkeys. J Gerontol 1993;48:
B2015.
58. Schwarze SR, Lee CM, Chung SS, Roecker EB, Weindruch R,
Aiken J. High levels of mitochondrial DNA deletions in skeletal muscle of old rhesus monkeys. Mech Ageing Dev 1995;83:
91101.
59. Melov S, Shoffner JM, Kaufman A, Wallace DC. Marked increase in the number and variety of mitochondrial DNA
rearrangements in aging human skeletal muscle. Nucleic
Acids Res 1995;23:41226.

/ 7b71$$ge4D

09-10-97 11:25:47

cana

60. Li Y, Huang YY, Carlson EJ, Melov S, Ursell PC, Olson JL, et
al. Dilated cardiomyopathy and neonatal lethality in mutant
mice lacking manganese superoxide dismutase. Nature
1995;11:37681.
61. Sohal RS. Hydrogen peroxide production by mitochondria may
be a biomarker of aging. Mech Ageing Dev 1991;60:18998.
62. Epstein CJ, Avraham KB, Lovett M. Transgenic mice with
increased Cu/Zn-superoxide dismutase activity: animal
model of dosage effects in Downs syndrome. Proc Natl Acad
Sci U S A 1987;84:80448.
63. Harris TB, Roubenoff RR, Langlois J, Hannan M, Havlik R,
Wilson P. Association of insulin-like growth factor-I with
body composition, weight history, and past health behaviors
in the very old: the Framingham Heart Study. J Am Geriatr
Soc 1997;45:1339.
64. Birkenhager-Gillesse EG, Derkson J, Lagaay AM. Dehydroepiandrosterone sulfate (DHEAS) in the oldest old, aged 85
and over. Ann N Y Acad Sci 1994;719:5432.
65. Rudman D, Drinka PJ, Wilson CR, Scherman F, Cuisinier
MC, Schults S. Relations of endogenous anabolic hormones
and physical activity to bone mineral density and lean body
mass in elderly men. Clin Endocrinol 1994;40:65361.
66. Rudman D, Feller AG, Nagraj HS, Gergans GA, Lalitha PY,
Goldberg AF, et al. Effects of human growth hormone in
men over 60 years old. N Engl J Med 1990;323:16.
67. Hobbs CJ, Plymate SR, Rosen CJ, Adler RA. Testosterone
administration increases insulin-like growth facto-1 levels
in normal men. J Clin Endocrinol Metab 1993;77:7769.
68. Walford R. The immunologic theory of aging. Copenhagen:
Munksgard, 1969.
69. Makinodan T, Kay MMB. Age influence on the immune system. Adv Immunol 1980;29:287330.
70. Smith GS, Walford RL. Influence of the main histocompatibility complex on aging in mice. Nature 1977;270:7279.
71. Hirokawa K, Albright JW, Makinodan T. Restoration of impaired immune function in aging animals. I. Effect of syngeneic thymus and bone marrow grafts. Clin Immunol Immunopathol 1985;5:3716.
72. Baker GT, Sprott RL. Biomarkers of aging. Exp Gerontol
1988;23:22339.
73. Gatti RQ, Good RA. Aging, immunity and malignancy. Geriatrics 1979;25:15868.
74. Bulychev VV. Longevity, atherosclerosis and cellular immunity. Klin Med (Mosk) 1993;71:514.
75. Lehuen A, Bendelac A, Bach JF, Carnaud C. The nonobese
diabetic mouse model: independent expression of humoral
and cell mediated autoimmune features. J Immunol 1990;
144:214751.
76. Hull M, Strauss S, Berger M, Volk B, Bauer J. The participation of interleukin-6, a stress-inducible cytokine, in the
pathogenesis of Alzheimers disease. Behav Brain Res 1996;
78:3741.
77. Hull M, Fiebich BL., Lieb S, Berger SS, Volk B, Bauer J. Interleukin-6associated inflammatory processes in Alzheimers disease: new therapeutic options. Neurobiol Aging
1996;17:795800.
78. Gillis S, Kozak R, Durante M, Weksler ME. Decreased production and response to T cell growth factor by lymphocytes
from aged humans. J Clin Invest 1981;67:93742.
79. Miller RA. The aging immune system: primer and prospectus. Science 1996;273:704.
80. Stephan RP, Sanders VM, Witte PL. Stage-specific alterations
in murine lymphopoiesis with age. Int Immunol 1996;8:
50918.

W: Cancer

The Biology of Aging/Ershler and Longo

81. Radl J, Sepers JM, Skvaril F. Immunoglobulin patterns in humans over 95 years of age. Clin Exp Immunol 1975;22:8490.
82. Radl J. Animal model of human disease: benign monoclonal
gammopathy (idiopathic paraproteinemia). Am J Pathol
1981;105:913.
83. Radl J. Age-related monoclonal gammopathies: clinical lessons from the aging C57BL/6 mouse. Immunol Today
1990;11:2346.
84. Kyle RA. Monoclonal gammopathy of undetermined significance and solitary plasmacytoma: implications for progression to overt myeloma. Hematol Oncol Clin North Am
1997;11:7187.
85. Thoman M, Weigle WO. Lymphokines and aging: interleukin-2 production and activity in aged animals. J Immunol
1981;127:21026.
86. Ershler WB, Sun WH, Binkley N. Interleukin-6 and aging:
blood levels and mononuclear cell production increase with
advancing age and in vitro production is modifiable by dietary restriction. Lymphokine Cytokine Res 1993;12:22530.
87. Ershler WB. Interleukin-6: a cytokine for gerontologists. J
Am Geriatr Soc 1993;41:17681.
88. Holmes F. Clinical evidence for a change in tumor aggressiveness with age. Semin Oncol 1989;16:3440.
89. Prehn RT, Lappe MA. An immunostimulation theory of tumor development. Transplant Rev 1971;7:2654.
90. Prehn RT. The immune reaction as a stimulator of tumor
growth. Science 1972;176:1701.
91. Finch CE, Landfield PW. Neuroendocrine and autonomic
functions in aging mammals. In: Finch CE, Schneider, EL,
editors. Handbook of the biology of aging. New York: Van
Nostrand Reinhold, 1985:567.
92. Jiang D, Fei RG, Penergass WR, Wolf NS. An age-related
reduction in the replicative capacity of two murine hematopoietic stroma cell types. Exp Hematol 1992;20:121622.
93. Sohal RS, Agarwal S, Dubey A, Orr WC. Protein oxidative

/ 7b71$$ge4D

09-10-97 11:25:47

cana

1293

damage is associated with life expectancy of houseflies. Proc

Natl Acad Sci U S A 1993;90:72559.
94. Roth GS, Joseph JA. Age-related changes in transcriptional
and posttranscriptional regulation of the dopaminergic system. Life Sci 1994;55:20315.
95. Franklin RA, Arkins S, Kelley KW. The proliferative response
of rat T cells to calcium ionophores with age. Cell Immunol
1990;130:41628.
96. Higgins KA, Stout JT, Heller DA, Parker RF. Individual variability in tail tendon fiber break time in three age cohorts
of different strains of mice. Exp Gerontol 1991;26:146777.
97. Bito LZ, De Rousseau CJ, Kaufman PL, Bito JW. Age-dependent loss of accommodative amplitude in rhesus monkeys:
an animal model for presbyopia. Invest Ophthalmol Vis Sci
1982;23:2331.
98. Williams DD, Short R, Bowden DM. Fingernail growth rate
as a biomarker of aging in the pigtailed macaque (Macaca
nemestrina). Exp Gerontol 1990;25:42332.
99. Lane MA, Reznick AZ, Tilmont EM, Lanir, A, Ball SS, Read
V, et al. Aging and food restriction alter some indices of bone
metabolism in male rhesus monkeys (Macaca mulatta). J
Nutr 1995;125:160010.
100. Weale R. Why does the human visual system age in the way
it does? Exp Eye Res 1995;60:4955.
101. Tamm S, Tamm E, Rohen JW. Age-related changes of the
human ciliary muscle: a quantitative morphometric study.
Mech Ageing Dev 1992;62:20921.
102. Hartnell JM, Morley JE, Mooradian AD. Reduction of alkaliinduced white blood cell DNA unwinding rate: a potential
biomarker of aging. J Gerontol 1989;44:B12530.
103. Schleicher ED, Wagner E, Nerlich AG. Increased accumulation of the glycoxidation product N (epsilon)(carboxymethyl) lysine in human tissues in diabetes and aging. J Clin
Invest 1997;99:45768.

W: Cancer