Aging and Cancer: Are Telomeres and Telomerase the Connection?
Aging is associated with the gradual decline in the performance of most organ systems, resulting in the loss of reserve capacity. In some organ systems the progressive loss of functional capacity with age is due to the loss of functional cells. As a person's cells age, they can no longer assist in maintaining and repairing human tissue, and this may be responsible for some of the degenerative disabilities that are often associated with increased age. The fundamental mechanisms regulating human aging are complex and the main theories fall into two general camps with many subdivisions: genetic/developmental aging programs and accumulation of damage theories. While both contribute to aging many of the best ideas (e.g. cellular senescence, hormonal/endocrine changes, cross-linking/glycation, mitochondrial DNA damage, free radical and oxidative damage theories) while appealing individually are not mutually exclusive. The idea commonly known as the disposable soma theory of aging states that the amount of energy invested in maintaining the somatic tissues of an organism is limited to that required to live long enough to reproduce successfully. Thus, for each species there is a trade off between investment in the maintenance of somatic cells and investment in reproduction. The disposable soma theory can also be considered an example of antagonistic pleiotropy, since optimization of health in young animals during the period of reproductive fitness could produce unselected late consequences in the post-reproductive individual who escapes an early death from disease or accidents. In addition to a decline in physiological functions with age, an indirect consequence of a limit to somatic cell growth in long-lived species, is that the finite cellular growth potential (replicative senescence) may also be a temporary barrier against uncontrolled proliferation of potential tumor cells.
Telomeres are repeated DNA sequences that protect the ends of chromosomes from being treated like a broken piece of DNA needing repair. Without telomeres, the ends of the chromosomes would be "repaired", leading to chromosome fusion and massive genomic instability. Telomeres are also thought to be the "clock" that regulates how many times an individual cell can divide. Telomeric sequences shorten each time the DNA replicates. When at least some of the telomeres reach a critically short length, the cell stops dividing (is senescent) which may cause or contribute to some age-related diseases. In cancer, a special cellular reverse transcriptase, telomerase, is reactivated and maintains the length of telomeres, allowing tumor cells to continue to proliferate. Telomeres do not contain genes but are repetitive DNA elements that are expendable and thus their loss may not appear to affect cellular function until they reach a very short length.
The mechanisms of DNA replication in linear chromosomes is different for each of the two strands (called leading and lagging strands). The lagging strand is made as series of discrete fragments, each requiring a new RNA primer to initiate synthesis. The DNA between the last RNA priming event and the end of the chromosome cannot be replicated because there is no DNA beyond the end to which the next RNA primer can anneal, thus this gap cannot be filled in (this is referred to as the "end replication problem"). Since one strand cannot copy its end, telomere shortening occurs during progressive cell divisions. The shortened telomeres are inherited by daughter cells and the process repeats itself in subsequent divisions. There may also be some nucleases that digest part of the end made by leading strand synthesis. These nucleases would create a single-stranded region to which special proteins bind. These "end binding factors" on both ends may help to hide the telomeres from the DNA repair machinery.
In contrast to tumor cells, which can divide forever (are "immortal"), normal somatic human cells have a limited capacity to proliferate (are "mortal"). In general, cells cultured from a child divide more times in culture than those from an adult. The length of the telomeres decreases both as a function of donor age and with the number of times a cell has divided. What was not known until recently is the molecular mechanism that permitted cells to keep track of their divisions. The telomere hypothesis was first suggested by Alexi Olovnikov, a Russian scientist, in a theoretical paper nearly 30 years ago before the modern molecular biology era and his ideas remained untestable for over a decade. The essence of the theory is that each time a cell divides, the telomeres become a little shorter until the cell simply can no longer divide. During the last 20 years there have been many correlative studies published suggesting, but not directly proving, that telomeres may be the counting mechanism. In 1998, direct evidence was published demonstrating that the progressive loss of telomeres is the cellular timing mechanism or clock. This work showed that cellular aging can be bypassed or put on hold by the introduction of a single gene, the catalytic component of telomerase. The cells with introduced telomerase extended the length of their telomeres, have already divided for 350 generations past the time they normally would stop dividing, and are continuing to divide. There appear to be two mechanisms responsible for the proliferative failure of normal cells. The first, M1 (Mortality stage 1), occurs when there are still at least several thousand base pairs of telomeric sequences left at the end of most of the chromosomes. M1 may be induced by a DNA damage signal produced by one or a few of the 92 telomeres that have particularly short telomeres. The M1 mechanism causes a growth arrest mediated by the tumor suppressor genes p16/pRB and p53. If the actions of p53 and p16/pRB are blocked, either by mutation or by binding to viral oncoproteins, then cells can continue to divide and telomeres continue to shorten until the M2 (Mortality stage 2) mechanism is induced. M2 represents the physiological result of critically short telomeres when cells are no longer able to protect the ends of the chromosomes, so that end-degradation and end-to-end fusion occurs and causes genomic instability and cell death. In cultured cells, a focus of immortal cells occasionally arises. In most cases, these cells have reactivated the expression of telomerase, which is able to repair and maintain the telomeres.
The expression of many cellular proteins changes. For example, in fibroblasts there are increases in the synthesis of proteases such as procollagenase, plasminogen activator, and stromelysin, while there are decreases in the synthesis of procollagen and tissue inhibitors of metalloproteases. It is thought that these changes may contribute to some of the age-related changes that occur for example in skin. It might take only 1- 5% of the fibroblasts in the dermis (which underlies the epidermis) to change protein expression to change the whole structure of skin. All of your cells do NOT have to get old in order to get wrinkled skin or for injuries not to heal as quickly. Sun protected epidermal keratinocytes of the skin is actually younger than sun damaged skin. In many instances sun damaged skin cells not only accumulate mutations but also have fewer proliferations left than the sun protected skin cells. The more sun damage to the skin, the more cell turn-over, and the more rapidly the telomere clock runs down. By using sunblock and/or staying out of the sun, you reduce the chances for both skin cancer and wrinkled skin.
While there have been many studies indicating a correlation between telomere shortening and proliferative failure of human cells, the evidence that it is causal has only recently been demonstrated. Introduction of the telomerase catalytic protein component into normal human cells without detectable telomerase results in restoration of telomerase activity. Normal human cells stably expressing transfected telomerase demonstrate extension of life span, providing more direct evidence that telomere shortening controls cellular aging. The cells with introduced telomerase maintain a normal chromosome complement and continue to grow in a normal manner. Initial concerns that the introduction of telomerase into normal cells may substantially increase the risk of cancer have not proven true. One way to think about this is that reproductive tissues maintain high levels of telomerase throughout life, and there is no increased incidence of cancers in these tissues when compared with others. Thus, the major role of telomerase is to maintain telomere stability and keep the cells dividing. These observations provide the first direct evidence for the hypothesis that telomere length determines the proliferative capacity of human cells.
The development of better cellular models of human disease and production of human products are among the immediate applications of this new advance. This technology has the potential to produce unlimited quantities of normal human cells of virtually any tissue type and may have most immediate translational applications in the area of transplantation medicine. In the future, it may be possible to take a person's own cells, manipulate and rejuvenate them without using up their life span and then give them back to the patient. In addition, genetic engineering of telomerase-immortalized cells could lead to the development of cell-based therapies for certain genetic disorders.
Initially, this new technology would be done ex vivo (e.g. in the laboratory environment). We would remove a few cells from a person by fine needle biopsy, reset the telomere clock of these cells in the laboratory (thus providing the cells with unlimited divisions), then reintroduce the cells back into the person. This would have the enormous advantage of not having to deal with immune rejection, since these are the person's own cells. In the case of a genetic disorder, we would reset the telomeres, then reinsert the corrected gene, and finally reintroduce the cells into the patient. It is important to recognize that in the future we may not need to permanently introduce the telomerase gene. We could insert telomerase to grow the telomeres and then remove it, or perhaps we will discover a small molecule that can reset the telomere clock. This would avoid potential problems with having too much growth potential (and thus increased risk for cancer). However, our results indicate that introduction of telomerase in perfectly normal cells does not increase the cancer risk very much and the cells with intact telomerase can still undergo differentiation and participate in normal tissue functions. Genetic engineering of telomerase-immortalized cells could lead to the development of cell-based therapies for certain genetic disorders such as muscular dystrophy. Other areas of cell engineering that may be possible in the future include having an unlimited supply of skin cells for grafts for burn patients or to generate products for cosmetic applications (e.g. aging skin), improving general immunity for older patients or those with blood disorders such as AIDs. In addition, producing an unlimited supply of pancreatic islet cells that are glucose responsive for the treatment of diabetes, rejuvenating the endothelial lining of blood vessels to protect against plaque formation and cardiovascular disease, the rejuvenation of bone marrow stem cells for blood transplants, in osteoprogenitor cells for rebuilding bone, and perhaps in retinal cells of the eye for the treatment of macular degeneration (a leading cause of age-related blindness) are other areas being considered.
Technically yes, but we still do not know how far we can take this. Thus, if cells are completely senescent it may not be possible to get them to grow again even if we could re-elongate the telomeres. However, if telomeres are rate limiting for growth, as we believe, then only a few divisions may be needed to reset the clock. Since all of our cells do not age at the same rate, we believe there will not be a technical barrier to reset the telomere clock in human cells that we can grow in the laboratory.
Probably, but perhaps not in all tissues. Under normal conditions most tissues can last a typical life span. However, with the improvement in sanitation, the development of antibiotics, vaccines, and modern pharmaceutical drugs, humans are living longer. This could NOT have been selected for in evolutionary terms, when the average human lived 30-40 years before the 20th century. Thus, we are observing aged-related decline in normal people who live a long time. In the past we only observed this in disease states such as AIDs where T-cells become low due to HIV killing mature cells and in muscular dystrophy where kids run out of muscle satellite cells. However, in older individuals without diseases, we are seeing immunological deficiencies, wounds that do not heal (such as pressure ulcers), wearing down of the vascular endothelium leading to coronary disease, proliferative decline of retinal pigmented epithelial cells leading to age-related blindness. etc. In these age-related disorders, telomeres and tissue engineering may have an important impact. The relationship between replicative cellular aging and these diseases is not yet proven, and this remains an important goal of future research. These applications will be seen in the future, and it would be a mistake to offer false hope at the present time, since the techniques are likely to take a bit more time to research prior to introduction into the clinic. Through our work and that of many other scientists, we have learned that telomere shortening is important in cellular aging. This major understanding now provides us a clear direction for future studies. It is also important to state that much of what we now know about telomere biology in human cells is built on many years of outstanding work that was done and continues to be done by scientists studying model organisms, such as pond protozoa and baker's yeast. Many telomere mechanisms operating in these model organisms pointed the way to our current understanding of human telomere regulation. While there is still a long way to go, it is important for us to recognize the importance of these fundamental advances in model organisms, which now allow us to begin to translate these into practical outcomes for treating human disease.
Most immediately, we are developing standard cell reagents that all scientists can share, compare, and do experiments on, as opposed to looking at primary culture of cells, which are always changing because their telomeres shorten over time. We have also introduced telomerase into patient cells with a variety of chromosome instability or genetic diseases. We can immortalize these cells, which may help us discover the nature of the genetic defect. For example, there are genetic cancer susceptibiliy syndromes that present in specific tissues. Even though all cells in these patients have the same genetic lesion, we need to understand why only specific tissue types get cancer. By using telomerase to immortalize cells from these patient we may help clarify some of these issues.
Previously, we found that the enzyme, telomerase was present in specialized reproductive cells and most cancer cells that appear to divide indefinitely. Even though telomerase is present in some stem cells of renewal tissues, the levels are generally much lower and these tissues show continuing telomere shortening throughout life. Telomerase works by adding back telomeric DNA to the ends of chromosomes, thus compensating for the losses that normally occur. Most normal cells do not have this enzyme and thus they lose telomeres with each division. A potential paradox was to explain why introduction of telomerase did not result in cancer progression. To understand this, it is important remember that cancer is caused by the accumulation of several alterations that occur over a lifetime, and which affect processes controlling cellular growth rates, responses to growth factors and the ability to invade and undergo metastasis. Telomerase only affects the counting of the number of times a cell has divided, not the rates of cell growth in the conventional way of thinking. The main function for telomerase is to maintain telomeres and permit continued cell growth that would otherwise be limited by shortened telomeres. Thus, telomerase is NOT an oncogene and does not cause cancer.
An understanding of the molecular details of the relationship between cellular senescence and cancer has begun to emerge. Cells probably need to accumulate at least 4-6 mutations to become tumorigenic, and each mutation likely requires an expansion of the mutant clone to at least a million cells (20 doublings) before there are a sufficient number of cells in which the next mutation could occur. Some of these mutations are recessive, in which case it might take two clonal expansions in order to first have the mutation and then eliminate the wild-type allele. It thus probably takes between 80 and 200 doublings for a normal cell to generate a malignant tumor. Eighty doublings is close enough to the observed life span of some cultured cells that it is possible that a fortuitously rapid accumulation of mutations could generate a mortal tumor. Furthermore, it is becoming evident that some tumor suppressors are not just cell cycle regulators but key molecules essential for the induction of cellular senescence. Alteration or mutation of p53 and pRB (or possibly other oncogenes or tumor suppressor genes) may be sufficient to result in the development of cancer, but these would be predicted to be "mortal" tumors. However, the re-expression of telomerase, the enzyme that maintains telomeres and prevents their shortening, occurs in most tumors and is probably a critical event in the sustained growth of most cancers. For example, in neuroblastoma (an adrenal tumor that is common in children) there is a special subclass of advanced disease called 4s in which there are distal metastases but in which there is often spontaneous remission after surgery. In our study, and those of others, telomerase activity is generally not detected in 4s neuroblastoma and most have spontaneous remissions of their cancer. It appears that because these tumor cells do not express telomerase activity, they eventually develop critically shortened telomeres and regress. These observations may have immediate implications in risk stratifying children with 4s neuroblastoma. Importantly, these results support the idea that telomerase activity is required for the sustained growth of most tumor cells. It also demonstrates that all tumor cells are NOT immortal.
We believe that progressive telomere shortening is halted in cancer cells by the presence of the enzyme telomerase which maintains and stabilizes the telomeres, allowing cells to divide indefinitely. Telomerase activity is detected in almost all human tumors. It is hoped that a therapy can be developed that inhibits telomerase activity and interferes with the growth of many types of cancer. Thus, almost all scientists agree that cancer cells must maintain their telomeres and any treatment that prevents telomere maintenance is a potential "Achilles heel".
One research strategy is to inhibit the activity of telomerase, forcing immortal cells into a normal pattern of permanent growth arrest (senescence) or death (apoptosis). Following conventional treatments (surgery, radiotherapy, chemotherapy) anti-telomerase agents would be given to limit the proliferative capacity of the rare surviving tumor cells in the hope that this would prevent cancer recurrence. We believe this treatment would be very selective in that only cells with an activated telomerase would be affected. As far as we know, that includes only "immortal" tumor cells and germline (reproductive cells) and at lower levels stem cells in renewal tissues. A real problem with cancer therapy today, is that we attack all dividing cells by using drugs that damage DNA in dividing cells so much that they simply die. This affects all normal cells as well as cancer cells. The new generation of anti-cancer treatments is seeking to go after more specific cancer targets. Since telomerase is expressed in almost all cancer cells but very few normal cells, we believe the side effects would be much less than standard therapy. However, there may be some side effects on some proliferative cells of renewal tissues that do express some telomerase. While we cannot predict what these side effects (toxicities) are likely to be, it is important to know that all oncologists have to deal with these toxicity problems anyway and it is our hope that a short period of treatment using anti-telomerase therapies would not cause major complications. Another problem with cancer therapy today is that while surgery and adjuvant chemotherapy and/or radiotherapy generally result in initial successes, there is often cancer relapse that occurs 1-5 years after initial treatments. Thus, one hope is that the new generation of anti-cancer therapies will be used in an adjuvant setting after surgery and perhaps in combination with standard chemotherapy and/or radiotherapy to begin attacking the small pockets of hidden cancer cells that escape this initial treatment. It is these drug resistant cells that eventually cause cancer relapse. The mechanism of action of anti-telomerase therapy is such that it would take a period of time before telomeres shorten sufficiently to cause cells to stop dividing or die. Thus, to be effective anti-telomerase drugs would need to go after small numbers of cells not large bulk tumors.
Anti-telomerase therapy could affect germline (reproductive) cells and possibly stem cells of renewal tissues (such as crypt cells of the intestine, basal cells of the skin, and certain hematopoietic cells of the blood). Our experiments to date indicate that the telomeres of such cells are generally much longer than cancer cell telomeres. Thus, we believe there is a window of opportunity to inhibit telomerase in cancer cells and cause their death without causing the stem cells to "run out" of telomeres. Since the mode of action of telomerase inhibitors may require telomeric shortening before inhibition of cell growth or induction of apoptosis, there may be a significant delay in efficacy. Thus telomerase inhibitors may be useful in early stage cancer to prevent overgrowth of metastatic cells, as well as in high-risk patients with inherited-susceptibility to cancer syndromes to prevent the emergence of telomerase-expressing cells (chemoprevention). In advanced cancer, anti-telomerase agents would most likely be given to limit the proliferative capacity of the rare surviving tumor cells in the hope that this would prevent cancer recurrence. In addition, methods may have to be devised to increase the rate of telomere shortening when telomerase inhibitors are used therapeutically. These may include placing patients in hyperbaric chambers (where there is high oxygen and which has been shown to increase the rate of telomere shortening). In addition, there are clinical trials being proposed to make cytotoxic immune cells that recognize only telomerase or using the telomerase promoter in combination with genes (such as capase) that induce cell death. These latter approaches would immediately kill tumors cells irrespective of their telomere length. The downside of this is that proliferating stem cells and reproductive cells would also be affected.
As with any new anti-cancer treatment, there needs to be clinical trials. In Phase I trials we would determine if there are immediate side effects and toxicities of such treatments. These would initially be given to patients that had failed all other treatments and the volunteers for such trials are important in understanding the potential problems we may run into in more comprehensive clinical trials. After phase I trials are completed and overt toxicities are deemed not to be a problem, then randomized phase 2 and 3 clinical trials are used to test the efficacy of the new therapy.
One can never be sure. In experiments conducted by our laboratory and others, inhibition of telomerase does result in cells eventually undergoing apoptosis (death). If there was drug resistance due to other pathways for maintaining telomeres, then one might have expected to see telomerase negative survivors (e.g. revertants) but we didn't. Thus, while there is good evidence for alternative pathways for telomere maintenance in model organisms such as yeast and such pathways can occur in human cells, the engagement of the telomerase independent program in human cells does not occur very frequently.
There have been over 1000 published studies on telomerase activity and human cancers. Approximately 85-90% have been reported to have detectable telomerase activity. This includes all stages of cancer and in some instances early cancers may not have yet immortalized and these would be "mortal tumors". In addition there is the possibility that some cancers may have an alternative pathway for maintaining telomeres. The telomeres in these types of cancers should be heterogeneous in length, varying from very big to very small in size as previously described in yeast and in human fibroblasts treated with viral oncoproteins such as SV40 T-antigen. Several thousand primary human tumors have been examined for telomeres and almost none have these types of telomeres. In addition to "mortal tumors" and alternative telomere maintenance mechanisms, perhaps most telomerase negative human tumors that have been reported can be explained by some of the following: 1) experimental human error during the loading of samples for analysis; 2) tissue which is supposed to have tumor but does not; 3) tissue frozen improperly; 4) tissue inhibitors of telomerase so that it cannot be detected; 5) necrotic samples so telomerase is inactivated; and 6) tissue sampling errors (parts of the tumor are telomerase positive and other negative). In summary, telomerase is almost certainly a universal characteristic of most human cancers.
Recent experiments have shown that antisense oligonucleotides (peptide nucleic acids and 2'-O-methyl-RNAs) directed towards binding the template region of telomerase RNA are highly effective in repressing telomerase activity in cancer cells. In addition to approaches directed at telomerase RNA, other strategies include targeting the catalytic reverse transcriptase subunit of telomerase as well as its associated proteins. Also, there is mounting evidence that telomere binding proteins and quadruplex DNA are potential telomere targets. Finally, identification of the cellular genes that regulate the telomerase repression pathway offer another independent tactic for developing telomerase antitumor drugs. In this regard there is substantial evidence that a gene on chromosome 3p contains a telomerase repressor for both renal and breast cancer.
At present we are doing xenograph testing of our antisense telomerase inhibitors (by putting human tumors in immunodeficient mice). If we can inhibit human tumors in mice this would be an important precursor to human clinical trials. The hope is that these inhibitors may not be very toxic and could prevent or result in a delay until cancer relapse and thus we will be able to manage cancer patients much more effectively. Clearly there are other interesting anti-cancer agents being developed and we may ultimately use these combinatorially (e.g. angiogenic inhibitors together with telomerase inhibitors). This approach is similar to the period when we began treating children with multiple chemotherapeutic agents and found that different combinations made considerable improvements in outcomes. Adults cannot withstand the same amount of toxic chemotherapy agents as children, and thus less toxic approaches will be needed for adults.
Telomerase activity is detected in many pre-malignant specimens (in situ lung and breast cancers), while colon and pancreatic cancer have detectable telomerase activity at later (carcinoma) stages. The ability to use almost any clinical specimen and to demonstrate telomerase may allow the detection of cancers at an earlier stage. Telomerase activity is detected in lung cells in cancer patients obtained by bronchial alveolar lavage. In addition, fine needle aspirations (breast, liver and prostate cancer), washes (bladder and colon), and sedimented cells from urine (bladder and prostate) provide minimally invasive sources of cells to detect telomerase activity and are likely to have immediate diagnostic utility as well as monitoring of minimal residual disease. In an effort to improve the diagnostic value of telomerase determinations, in situ hybridization and immunohistochemistry methods for the demonstration of telomerase on archival paraffin embedded clinical specimens have been tested and they appear to distinguish cancer from normal cells, correlate well with telomerase activity, and thus may provide added value to telomerase activity assays (which requires fresh or fresh-frozen specimens). In addition, the presence or absence of telomerase may have prognostic value and help risk stratify patients into those with favorable outcomes (to avoid unnecessary treatments for patients with low or no detectable telomerase) and those with high telomerase activity and with unfavorable outcomes (to help oncologists manage patient treatments more effectively). Much progress has been made in the development of more accurate molecular-based cancer tests to assess tissue specimens. However, most of these methods do not have sufficient specificity (ability to differentiate between normal, precancerous, and cancerous cells) and sensitivity (accuracy in detecting the presence of cancer) to identify a wide variety of cancer types. Therefore, new clinical assays, applicable to most types of cancer, are needed. Since our development of a PCR-based telomerase assays in 1994, well over 1000 clinical studies have been published on the topic of telomerase and cancer. These have almost universally shown that telomerase determinations have utility not only in screening and early cancer detection, but also as a means to indicate residual disease (monitoring) after standard surgery and/or adjuvant therapy. In addition, the level of telomerase activity may have prognostic value since, in several types of cancer, patients with high telomerase activity have poorer outcomes when compared to those cancer patients with low telomerase levels.
Since telomerase is detected in almost all tumors and in some cases early in the development of the cancer, early cancer detection using telomerase may be a promising approach. There are already a number of situations where knowledge of the presence of telomerase may have value in risk stratifying patients into those that are likely to have favorable outcomes and those that are not. In some instances we may actually change the way we treat patients if we had this information. There is a lot of interest in this aspect of early cancer detection. However, we shouldn't set our hopes up that telomerase will be the unique cancer marker so that every time when we get our cholesterol levels tested that we will also be getting a telomerase assay. While this is certainly a possibility, it is more likely that we will eventually use a panel of cancer markers for this approach. There is a major effort at the National Cancer Institute to establish a series of molecular markers for cancer and it is expected within a few years we will have perhaps one hundred or more molecular markers that may indicate the presence of cancer. We would predict within the next ten years we may be able to take a few cells, do microarray analyses or proteomic analyses and determine with reasonable assurance not only if a person has cancer but also the type of cancer, the genes that are altered, and hopefully a therapeutic protocol that was developed to specifically target this type of cancer for the specific person would be available.
A main focus of our group at the present time is to explore how knowledge of telomerase and telomere biology will have useful applications in cancer diagnostics and therapeutics as well as determining how introduction of telomerase into normal human cells may have utility in cell and tissue engineering.
To advance our knowledge about organismal aging, we need to have a much larger group of scientists working in the basic biology of aging field. In order to do this we need to expand the financial research support not only for attracting young scientists, but the best scientists into aging research. In part, this is being done by additional resources being provided to the National Institutes of Aging and private Foundations, but this is not sufficient to explore all promising research areas and additional philanthropic support is needed.
There are those in Congress who believe aging research is only about living longer, but this is not true except for a few individuals. Most scientists are interested in healthy aging so we need to make it clear that by living healthier we may be able to reduce some of the health care costs associated with age-related disorders. Thus, we need to emphasize that we are not interested in extending longevity per se, with senior citizens experiencing longer periods of degenerative diseases with the associated consequences of becoming impoverished and more dependant on social security support and requiring more health care costs. Instead, we need to focus on improving health-span so our senior citizens will continue to be productive and contributing members of society. The survival curve is clearly becoming increasingly rectangular, with most deaths occurring after 75 years. Despite the aging of the population, few people live past the age of 95 years. This implies that there is a fixed, natural life span. Thus, we need to educate our Congressional representatives that if the average age at the onset of disability and chronic disease can be increased, then the total amount of disability will most likely decrease rather than increase in a mature population. In the ideal case, the healthy citizens of a modern society will survive to an advanced age with their vigor and functional independence maintained, and morbidity and disability will be compressed into a relatively short period. As Ernest Wydner once said "It should be the function of medicine to have people die young as late as possible".