The immune system can be broadly divided into adaptive and innate components, both of which decline with age. A failing immune system is one of the most serious aspects of frailty in the elderly, leaving them vulnerable to pathogens that a young person would shrug off, and suffering from reduced monitoring activities aimed at the destruction of potentially cancerous, senescent, and other harmful cells.
Focusing just on the adaptive immune system, there are a range of contributing causes identified by researchers to date. Firstly, new T cells, the workers and killers of the immune system, are only created in large numbers when an individual is young. The thymus, where these immune cells mature, atrophies early in adult life, its evolved task of setting up the immune cell population done. The supply of new immune cells diminishes to a trickle thereafter, a fraction of what it was. This effectively caps the T cell population associated with the active immune system: the body only supports so many.
This soft limit on the number of T cells leads to the second cause of immune system decline, which is structural, inherent in the nature of the immune system's organization and mode of operation. The adaptive immune system remembers threats, and it does do by maintaining a converted population of memory cells, one for each threat. Unfortunately some threats, like herpesviruses, cannot be cleared from the body: they keep coming back, again and again, each time leading to more cells becoming specialized to remember them. The main culprit here appears to be cytomegalovirus (CMV), which the majority of people have been exposed to by the time they are old: aside from its effects on the immune system's memory cell contingent it is largely harmless, and you probably didn't even notice your initial infection. But ultimately your immune system becomes overpopulated by memory cells dedicated to CMV, with too few naive T cells left to do its other jobs.
These outlines are simplifications of a complex set of issues, and omit any discussion of how the array of cellular and molecular damage that accumulates with aging also impacts the immune system negatively. The important point to take away from this is that the research community has near-term options available to reverse these contributions to immune system frailty. For example: the use of tissue engineering to restore thymic tissue to the role of generating a flow of new T cells; regular infusions of fresh T cells created from the patient's own stem cells; the use of new cancer therapy technologies to target and kill memory T cells specialized to CMV, freeing up space for new T cells. None of these are beyond the capacity of today's technology - as is so often the case, it's just a matter of devoting research and development resources to the problem for a few years, in a world in which funding sources lack the vision to support even the obvious bold advances.
Here is a good, conservative, and very readable open access review of the role of CMV in immune system decline:
Aging is associated with a dysregulation of the immune response, loosely termed "immunosenescence." Each part of the immune system is influenced to some extent by the aging process. However, adaptive immunity seems more extensively affected and among all participating cells it is the T cells that are most altered. There is a large body of experimental work devoted to the investigation of age-associated differences in T cell phenotypes and functions in young and old individuals, but few longitudinal studies in humans actually delineating changes at the level of the individual.
In most studies, the number and proportion of late-differentiated T cells, especially CD8+ T cells, is reported to be higher in the elderly than in the young. Limited longitudinal studies suggest that accumulation of these cells is a dynamic process and does indeed represent an age-associated change. Accumulations of such late-stage cells may contribute to the enhanced systemic pro-inflammatory milieu commonly seen in older people.
We do not know exactly what causes these observed changes, but an understanding of the possible causes is now beginning to emerge. A favored hypothesis is that these events are at least partly due to the effects of the maintenance of essential immune surveillance against persistent viral infections, notably Cytomegalovirus (CMV), which may exhaust the immune system over time. It is still a matter of debate as to whether these changes are compensatory and beneficial or pathological and detrimental to the proper functioning of the immune system and whether they impact longevity.
Dissecting the effects on immune alterations in elderly individuals with respect to age, low grade-inflammation, disease and CMV seropositivity remains a big challenge. We are currently approaching this challenge by assessing individual variations in responses to CMV, namely antibody titer, specificity, and neutralizing activity and determination of the specific CMV cell reservoirs (e.g., monocytes) rather than just "infected vs. not infected." This approach appears to us more likely to yield informative data in populations where almost all subjects are infected with the virus, for instance elderly individuals even in industrialized countries and essentially everyone in developing countries. Furthermore, longitudinal studies are needed including young and elderly healthy individuals to dissect the effects of age vs. CMV infection.
This is the sort of data gathering exercise that I would like to see augmented in medical research by more aggressive experiments in intervention, an area in which I think the research community is far less active than it could be. In the past you could argue costs, but costs are plummeting in the life sciences as biotechnology advances rapidly. It seems to me that just as much could be learned by augmenting T cell numbers or destroying memory T cells in animal studies as by exploring human data in more detail - and moreover it would also move the world closer to working therapies should the results be positive, which won't happen in the data gathering default mode of modern research.