Problems and solutions to the development of vaccines in the elderly☆

Vaccine research and development 

Although Jenner's vaccination experiment was performed more than 200 years ago, the optimal vaccine strategy is still an issue of debate. It is self-evident that the goal of vaccination is to protect the host against disease. However, the types of immune responses to the vaccine that confer protection are not so obvious. Although neutralizing antibody has been the gold standard in terms of evaluating the efficacy of different vaccines, it is becoming clear that for many pathogens, other components of the immune system may be required to confer long-term protection. Among the researchers who are involved in vaccine development, there is an increasing realization of the importance of eliciting cell-mediated immunity in addition to, or in place of, production of antibody [1], [2], [3], [4], [5].

A second challenge is the heterogeneous nature of the individuals who require inoculations with respect to overall physiology, history of previous infections, and immune system variables. The requirements for vaccines for infants are much different from those that are optimal for the elderly, and the process by which vaccines are developed does not necessarily take these factors into account. Prehuman experiments that are aimed at optimizing vaccine preparations rarely include aged animals. Even when aged animal models are incorporated into vaccine research, it is not possible to reproduce in old mice the unique and complex immunologic history of an aged human, which includes such variables as the spectrum and frequency of pathogen encounters, specific infectious episodes, and previous vaccinations. Indeed, cumulative immunologic experience is one of the key factors in determining the quality of the T-cell pool in aged humans.

Diminished response to vaccines in the elderly 

A variety of studies have documented age-related reduced responsiveness to vaccines. Protection of the aged population from influenza outbreaks has been a major concern, and annual vaccination of all persons older than 65 is recommended. During influenza epidemics, a mortality rate as high as 90% can occur in the elderly [6]. In young adults, immunization provides 65% to 80% protection against illness caused by an influenza virus that is present in the vaccine [7]. Influenza vaccination of the elderly only affords 30% to 50% protection against disease [8]. The risk for influenza after immunization is highest among elderly individuals without antibody or cell-mediated responses to the vaccine [8]. Even in elderly persons who show responsiveness, the antibody titer and T-cell proliferation to the vaccine antigens are reduced compared with that in younger individuals [8]. Increasing the antigen dose leads to higher titers of immunoglobulin G (IgG) and IgA antibodies, but these titers are not necessarily reflected in the hemagglutination-inhibition response [9]. The antibody response to vaccination also is impacted negatively by higher levels of preimmunization titers [10].

The immunization route may be critical, as demonstrated by enhanced local IgA titers after administration of intranasal versus intramuscular influenza vaccines, although the vaccine preparations were not precisely comparable [7]. Aging is associated with reduced IgA responses after oral immunization, suggesting a down-regulation of mucosal immunity [11]. Increased research on the specific changes in the mucosal immune system is critical for improvement of strategies for intranasal vaccination and for development of emerging technologies in the area of edible vaccines.

Frequency of vaccination, inclusion of adjuvants, and increasing antigen dose have been explored as strategies to improve efficacy of vaccination in the elderly. In older mice, improved protection from infection was observed with an intranasal vaccine that contained cholera toxin adjuvant [12]. An adjuvant effect also is achieved through incorporation of CpG motifs into vaccines. These oligonucleotides mimic bacterial DNA sequences and are associated with enhanced vaccine responsiveness in older mice [13]. The success of this adjuvant may be based on its ability to promote T helper cell type 1 (Th1) responses, thus reversing the shift to T helper cell type 2 (Th2) responses that is associated with aging, which is particularly evident in frail, elderly individuals [14]. Combining a cold-adapted live attenuated vaccine with the trivalent subunit inactivated vaccine also leads to increased responsiveness in elderly subjects [15]. Dietary supplementation of aged mice with antioxidants or with known activators of the α-isoform of the peroxisome proliferator activated receptor (PPAR-α) enhances mucosal and systemic immune responses to diphtheria vaccination [16].

Vaccines directed at other pathogens that affect the health of the elderly also are under development. Varicella zoster virus (VZV), which is latent in the dorsal route ganglia throughout life in previously infected persons, can reemerge as painful shingles when cell-mediated immune function is reduced. Vaccination with a live attenuated form of VZV induces a robust cell-mediated immune response (even in elderly subjects) based on the criterion of interferon γ (IFN-γ) secretion. The ultimate test of the vaccine's success (ie, its protective effect against VZV reactivation) has not been analyzed adequately [17].

Skin tests to evaluate the need for treatment or vaccination are also problematic in the elderly. Tuberculin reactivity decreases with age despite epidemiologic evidence that the elderly are more likely to be infected [18]. Tetanus toxoid delayed-type hypersensitivity reactivity similarly is diminished in some elderly persons, even after revaccination [18].

Respiratory syncycial virus (RSV) represents an important area of vaccine research. Originally viewed as primarily a pediatric challenge, RSV now is recognized as the cause of significant illness in elderly persons, ranking second only to influenza as a major viral pathogen in this age group [19]. No effective vaccine against RSV is available, and evidence shows that, as with influenza virus, repeated infections with antigenically related virus strains are common throughout an individual's lifetime. Unlike influenza virus, for which protective levels of neutralizing antibody have been defined, immunity to RSV in humans in not understood well, and there are no unequivocal correlates of protection [20]. Experimental vaccination of mice demonstrates that high cytotoxic T lymphocytes (CTL) activity correlates with resistance to infection, but this resistance declines within 2 months of infection, even in younger animals [21]. In aging humans, the ability of T cells to produce IFN-γ in response to in vitro stimulation with RSV is reduced significantly [22]. This finding may explain the increased morbidity that is caused by actual infection in this age group, and it suggests that vaccine preparations should enhance RSV-specific cellular immunity. Indeed, normal humoral immunity to RSV is observed in the elderly even in the context of severe clinical disease [19].

Infection with HIV is a newly recognized health issue in the elderly population. Individuals older than 50 comprise over 11% of patients with AIDS in the United States [23]. In fact, among the risk factors for rapid progression to AIDS in persons with HIV infection, aging represents an independent risk factor. Many of the changes associated with immunologic aging are accelerated in persons with HIV infection. In aging and HIV, there is an increased proportion of memory CD8 T cells with shortened telomeres that lacked expression of the CD28 costimulatory molecule. Increased proportions of CD28− cells are associated with loss of control over the infection and progression to AIDS. Because the size of the immune-cell pool is controlled by homeostatic mechanisms that independently regulate populations of naı̈;ve and memory T cells, high proportions of CD28− CD8 T cells can result in reduced numbers of more functional CD8 T cells. Even in the absence of HIV infection, the increased proportion of CD8 T cells that lack CD28− expression in elderly persons is associated with reduced numbers of naı̈;ve CD8 T cells, indicating a negative regulatory feedback mechanism that is subset-specific [24]. These results suggest that HIV vaccine protocol that target CD8 T cells needs to take into account the substantial increase in CD28− T cells that accompanies infection.

With the increased mobility of the elderly population, travel vaccines present a new challenge to immunologists. Individuals who travel to developing countries are advised to receive the hepatitis A vaccine, which is important for the elderly because of the more severe disease in these individuals and the approximate 2% mortality rate from hepatitis A among those older than 40. The recommendation of the Centers for Disease Control and Prevention to undergo vaccination 4 weeks before travel may not be appropriate for older individuals, because their antibody response often is delayed [25]. Elderly persons also show increased incidence of adverse effects related to certain vaccines (eg, yellow fever vaccine) [26], underscoring the need to balance the risk for severe illness and death caused by actual infection against the risk for vaccine-induced illness.

Determination of the magnitude of the antibody response after vaccination does not necessarily provide information on the degree of protection that is conferred against clinical infection. Differences in the avidity of the antibody response and variations in cellular responses contribute to protection. Assays to determine these aspects of the response to vaccines are not usually evaluated during vaccine development [25]. Markers of T-cell immunity after vaccination may be important predictors of protection. One of the potentially most informative markers of protection may be granzyme B, which is a member of the family of lymphocyte granule serine proteases. In a prospective study of institutionalized elderly persons, there was a significant correlation between reduced levels of granzyme B produced by T cells stimulated in vitro with influenza virus and symptomatic, laboratory-confirmed influenza [27]. Because Granzyme B is a key component of the lytic pathway of CTL, these findings underscore the importance of developing vaccines that induce CD8 T-cell immunity and of incorporating assays for cellular immune responses in the evaluation of vaccines.

Although vaccines may not necessarily confer full protection against the relevant pathogen in the elderly, exposure to vaccines may influence seemingly unrelated diseases of aging. Previous exposure to vaccines against diphtheria, tetanus, polio, and influenza is associated with a reduced risk for Alzheimer's disease (AD) [28]. It is unclear which aspect of the immune response that is elicited by the vaccine affects the development of AD, but the association merits more extensive investigation, particularly with respect to the development of an AD therapeutic vaccine. Other indications of immune involvement with AD have emerged from observational studies. Telomere length of T cells, but not B cells or monocytes, was associated with disease status in a group of community-dwelling patients with AD [29]. In another study, T cells from patients with AD showed reduced proliferative responses to β-amyloid, but not to other antigens [30]. These findings suggest that altered baseline immune status in patients with AD may be an important factor in the development of therapeutic vaccines that are aimed at retarding disease progression. Initial trials of a human therapeutic vaccine, which were based on promising results in a murine genetic model of AD, were, in fact, interrupted because of unexpected negative effects.

Mucosal vaccines 

The gastrointestinal tract represents more than 50% of the human immune system. Accessibility of many mucosal surfaces and the efficacy of oral and nasal vaccine delivery in inducing mucosal and systemic immune responses have encouraged development of vaccine strategies that capitalize on this immunization route. Many infections enter the body through mucosal routes, such as the intestines or lungs, and there is evidence that administration of a vaccine at the infection site gives better protection than does inoculation of a vaccine into the arm [31]. For pathogens that are transmitted through sexual contact, achieving protection through induction of mucosal immunity is essential.

Edible vaccines are appealing alternatives to injectable vaccines for practical and immunologic reasons [31], [32]. The noninvasive aspect, which may decrease potential adverse effects, offers a particular advantage to the elderly. Although the effect of aging on mucosal immunity has not been thoroughly investigated, the ease in production, handling, and administrations make edible vaccines an ideal low-budget means of prophylaxis, which could result in reduced medical budgets for the rapidly expanding elderly population. The challenge in designing edible vaccines is to identify the most appropriate vehicle to produce and mucosally deliver protective antigens or immune modulators, such as interleukins. Given the potential advantages of mucosally delivered vaccines, a more precise characterization of the age-related alterations in mucosal immunity would yield immediate gains in the improvement of vaccine efficacy [33].

The adjuvants currently used in mucosal vaccine strategies may not be sufficiently stimulatory to the aged immune system. There is an age-associated reduction of immune responses to cholera toxin and Escherichia coli enterotoxin, which are adjuvants that frequently are used in mucosally delivered vaccine preparations. The major immunoglobulin class that is produced in mucosal tissues, IgA, also has reduced levels with age, underscoring the notion that changes in the mucosal immune system may have broad implications on overall health and preventative medicine in the elderly [34].

Immunologic barriers to vaccine efficacy in the elderly 

The increased morbidity and mortality that is caused by infections, which is a central motivating factor in vaccinating the elderly, can be traced to diminished immune function. In addition to the diminished immune control over actual infections, skin tests to detect previous pathogen exposure also are blunted in a large proportion of elderly persons, leading to interference with appropriate and timely treatment. Older adults also show reduced immune responses to vaccines that are intended to prevent infection. A variety of changes in the immune system might contribute to the diminished immune response to vaccines, including alterations in B cells, antigen presentation, T-cell activation, and alterations in the proportions of memory versus naı̈;ve T cells. A thorough understanding of these changes is essential to addressing some of the challenges in tailoring vaccine preparations to the elderly population.

Replicative senescence and the immune system 

Replicative senescence (also called cellular senescence) has been studied extensively by cell biologists in relation to organismic aging and cancer. The natural barrier to unlimited proliferation that characterizes all normal human somatic cells has been hypothesized to function as a tumor-suppressor mechanism [64]. Although the term replicative senescence highlights the proliferative aspect of the process, cells that reach the irreversible stage of cell-cycle arrest in cell culture after extensive replicative activity also show striking changes in function and gene expression. In fibroblasts, the most extensively studied cell type, replicative senescence involves the down- or up-regulation of dozens of genes [64], [65]. Senescent fibroblasts are altered in their intrinsic characteristics and in the profound effect they exert on the in vitro and in vivo growth of epithelial tumor cells [66].

Over the past 10 years, using a variety of different culture protocols, several laboratories have demonstrated that CD4 and CD8 subsets of human T cells undergo replicative senescence in cell culture. The signature genetic change identified in senescent cultures is the complete loss of CD28 expression [67]. Another fundamental change associated with T-cell senescence is the inability to up-regulate the activity of the telomere-extending enzyme telomerase. In general, telomerase activity is present in tumor cells and absent in normal somatic cells, but cells of the immune system are exceptions. B cells and T cells can up-regulate telomerase during activation, albeit in a stringently controlled manner [68]. During primary stimulation with antigen, mitogen, or activatory antibodies, T cells exhibit telomerase activity that is comparable with that of tumor cells [69], [70]. With repeated exposure to the same antigen, however, memory T cells lose the ability to up-regulate telomerase, and this process parallels the loss of CD28 expression [71].

In cell culture, CD4 and CD8 T cells show markedly divergent patterns in CD28 and telomerase dynamics. By the fourth encounter with antigen, CD8 T-cell cultures are predominantly CD28− and show no telomerase activity, whereas CD4 T-cell cultures from the same donor that are subjected to identical antigenic stimulation protocols show high CD28 expression and telomerase activity, even by the 10th encounter with antigen. This fundamental difference between the two T-cell subsets may explain the preponderance of CD28− T cells within the CD8 subset that occurs with aging and chronic HIV infection.

The increased proportion of putatively senescent T cells within the CD8 subset in aged persons may be related to the stronger, more persistent nature of the antigenic drive for CD8+ T cells that respond to chronic systemic intracellular pathogens, in contrast to the more discrete localization of extracellular pathogens recognized by CD4+ T cells [72]. The maintenance of specific CD8+ T-cell receptor clonotypes at high circulating frequency in humans has been documented in chronic viral infections, such as cytomegalovirus, HIV, and Epstein-Barr virus, and in repeated infections with influenza virus, which contains conserved CD8+ T-cell epitopes [72]. In mice, the presence of clonal populations of CD8+ T cells correlates with chronic hepatitis virus infection [73].

A significant outcome of the age-related oligoclonal expansions that comprise an increasing proportion of the T-cell pool is that the remaining naı̈;ve-cell repertoire is narrowed. The poor response of elderly persons to neoantigens, including those present in vaccine preparations, may be a clinical manifestation of the more restricted naı̈;ve-cell repertoire. Another negative influence of the high proportion of putatively senescent memory T cells is its possible contribution to the age-associated increase in the serum levels of proinflammatory cytokines, which have been implicated in numerous age-related pathologies. This notion is consistent with the marked increase in tumor necrosis α (TNFα) and IL-6 production by T cells that reach replicative senescence in cell culture (CS Spaulding, RB Effros, PhD, unpublished observations).

Organismic aging and vaccine efficacy 

Some immune changes that influence vaccine success in the elderly may be secondary to age-related alterations that occur on an organismic level. Characteristics such as sex and parity influence the production of immune cytokines and the composition of spleen cells in early life suggesting that age-associated decline in sex hormones exerts an influence on immune function. Altered lipid homeostasis during aging [74] may also have profound influences on the immune system. There is an inverse correlation between membrane viscosity and T-cell responses to mitogen [75], and lymphocytes from Senieur protocol-screened elderly display a higher membrane viscosity than do lymphocytes from younger controls.

A hallmark of aging is the reduced capacity to respond to a variety of stressors. The response to thermal stress by rapid production of heat shock proteins is reduced with age in a variety of cell types, including those of the immune system. [76]. Oxidative stress, another well-documented physiologic change that occurs during aging, may influence such diverse immune-related parameters as telomere shortening [77], lymphocyte lipid peroxidation [78], protein damage [79], lung viral load [80] and the PPAR family of nuclear steroid hormone receptors, which are expressed in immune cells and influence the signal transduction of nuclear factor κB (NF-κB) [81]. Oxidative stress may have a role in the diminished pulmonary function that serves as a biomarker of mortality in the elderly [82].

Psychologic stress may alter immune control over pathogens and the response to vaccines in the elderly. Even in younger individuals, acute stressors can affect vaccine-induced immunity. In the elderly, whose immune systems already are compromised, acute and chronic stress impacts immune function. Markedly reduced antibody response to influenza vaccination has been documented in caregivers of spouses with AD, who often are subjected to chronic stress [83]. An inverse relationship between the levels of salivary cortisol, a biomarker of stress, and IgG antibody titers to influenza vaccine has been reported in elderly caregivers of spouses who have other forms of dementia [84].

Proteasomes are involved in a host of cellular functions, including the heat shock response, degradation of cyclins involved in cell-cycle regulation, regulation of the pleiotropic transcription factor NF-κB, and breakdown of proteins into peptides for antigen processing. An age-related decline in proteasome function has been described in hepatocytes, neural tissue, and T cells. In elderly individuals, the reduced capacity of proteasomes to degrade IκB leads to inhibition of NF-κB. A critical age-related proteasome change is the reduced capacity to generate antigenic peptides in APCs [85]. These changes also may contribute to the reduced number of HLA molecules per cell in monocytes, B cells, and T cells in elderly persons [86]. All of these factors could affect vaccine efficacy, especially if some level of antigen processing is required for an optimal protective response.

Summary 

Scientists involved in vaccine research and development face the challenge of protecting the ever-increasing elderly population from a broad spectrum of infectious diseases. The optimal vaccine-induced immune response to confer protection is undefined for many pathogens, and the field of vaccine research is undergoing a gradual shift from the original focus on humoral immunity to a focus that incorporates cellular and innate immune components. The age-related changes in various aspects of immune function, including an increase in a population of T cells that shows signs of replicative senescence, underscore the need to enhance research aimed at designing vaccines to meet the unique requirements of the elderly population.