Advanced biological age is associated with improved antibody responses in older high-dose influenza vaccine recipients over four consecutive seasons

Lower respiratory tract infections are one of the leading causes of total disease burden in the world, especially so for older adults [1]. Prior to the COVID-19 pandemic, influenza accounted for more than 70% of all disability-adjusted life years associated with communicable diseases in European adults over 65 [2], and between 2010 and 2020, influenza caused approximately 4 million hospitalizations and 340,000 deaths in the United States alone [3]. Seasonal vaccination remains our best preventative measure against influenza, although there is significant heterogeneity in its effectiveness. For example, while protection against influenza-like illness is mostly similar across ages for all viral subtypes [4], vaccination is less effective against hospitalization for older adults [5], especially so for A/H3N2 and when there is a mismatch with the circulating strain [6]. Underlying health conditions can also contribute to lower vaccine effectiveness in older adults, evident by reduced protection against hospitalization with increasing frailty [7]. It is not clear why vaccination becomes less efficacious with age, although impairments in the ability to induce strong antibody-responses to the vaccine are likely a major contributor [8]. Nonetheless, growing evidence indicates that immune deficits most greatly contributing to an increased risk of hospitalization, disability and death involve cell- as opposed to antibody-mediated systems, the latter of which is typically evaluated as an easily measurable predictor of sterilizing immunity [9].

Given the complexity of the immune system, it is not surprising that the manner in which immunity changes with age is also complex and multifactorial. For example, innate immune cells such as monocytes tend to exhibit a hyperinflammatory phenotype with reduced signalling capacity [10], while the proliferative capacity of T- and B-lymphocytes decline [11]. The architecture of primary and secondary lymphoid organs are also broadly affected, compromising the efficiency of intracellular communication and the magnitude and balance of cellular output [12]. Modern theories in aging research suggest that a single fundamental mechanism underlying these alterations is unlikely and, instead, propose a network of interrelated age-related biological phenomena [13‐15]. Consistent with this theoretical perspective, several hallmarks of aging [14] are related to impaired immune function, including disruptions to the proteostasis network impacting efficient antigen presentation [16], dysregulated metabolism on T/B-cell function [17, 18], and reduced telomere length and the decline in T-cell clonal expansion [19]. Further, proxy measures of biological aging, which represents the overall loss of physiological integrity due to the accumulation of these hallmarks, have been shown to be positively correlated with the likelihood of SARS-CoV-2 infection [20] and the severity of disease [20‐22].

We hypothesized that biological aging is an important determinant of vaccine effectiveness in older adults, and specifically, vaccine immunogenicity will be most impaired in those with the most advanced biological age. To test this, we employed data and biosamples from our 4-year clinical trial comparing standard- to high-dose influenza vaccine in adults 65 and older and estimated the association between biological age and antibody titres pre- and post-vaccination. Given that CMV infection is often correlated with weaker influenza vaccine responses [23], we also investigated a potential interaction between CMV and biological age in associations with vaccine immunogenicity.

There is considerable heterogeneity in vaccine-induced protection against serious outcomes of influenza in older adults, which is typically studied in terms antibody-responses providing sterilizing immunity which are proven correlates of protection [25]. Although a number of immunological factors have been postulated as mediators of impaired vaccine-responses with age, the upstream biological mechanisms are poorly understood. We tested the hypothesis that variation in the extent of biological aging – the age-related loss of integrity and functionality of multiple physiological systems – would explain differences in vaccine immunogenicity in community-dwelling adults aged 65 and older. We quantified biological age using the previously validated KDM algorithm and a panel of 10 blood biomarkers reflecting integrity of several organ systems. In both the CLSA training sample and the vaccine trial sample, the measure of biological age was positively associated with chronological age (r ~ 0.6) and with frailty (Cohen’s d [26]~ 3–7.5). However, in contrast to our expectation, those with older biological age showed improved A/H1N1, A/H3N2 and B antibody titres post-vaccination, although this association was present only for high-dose participants and was diminished in those who were CMV seropositive. Even in this group, effect-sizes were small (d = 0.16–0.26). We also investigated the association between biological age and the incidence of break-through infections, but did not observe any trends that would support a relationship.

Given the lack of data pertaining to biological aging and immunity, it is difficult to explain why vaccine responses would increase with biological age. Aspects of T- and B-cell function, including activation, chemotaxis and proliferation [27, 28], and the frequency of naïve T- and B-cell populations [29] tend to decrease with age. However, IFN-? producing CD4⁺ T-cells, which promote strong vaccine responses through dendritic cell activation and enhanced antigen presentation [9], have been shown to be elevated in older adults [30‐32]. This has been similarly shown for cell subsets that produce interleukin-10, a cytokine that promotes B-cell survival, proliferation and isotype switching [33], as well as germinal centre B-cell responses to vaccination [34]; specifically, IL-10 secreting T-regulatory cells increase with age [35, 36], as does monocyte IL-10 production following influenza vaccination in older adults [37]. It is also important to consider that these age-related changes may not be sufficient to strengthen antibody responses to standard-dose vaccine, hence why we only observed biological age to be associated with immunogenicity in high-dose recipients. This is supported by recent work showing that the relative effectiveness of high-dose over standard-dose vaccine is consistently higher in adults 85 and over compared to those 65 to 84 [38], as well as our own findings pertaining to frailty [39] and chronic inflammation (Picard et al., 2022: manuscript in press). Interestingly, and in line with aforementioned literature, the high-dose influenza vaccine is particularly effective at inducing the generation of antibody-producing plasmablasts [40], a process that is known to be enhanced by IFN-? producing CD4⁺ T-cells [41] and IL-10 [42].

Our observation of the association between biological age and vaccine responses being diminished in CMV seropositive individuals may also shed light on the underlying cellular mechanisms involved. Although the literature is inconsistent regarding the impact of CMV infection on influenza antibody responses [23], its influence on the T-cell compartment is well-documented. Following primary infection, which eventually occurs in up to 90% of adults worldwide, CMV enters a latent phase and over time T-cells that are specific to immunodominant epitopes on the virus can accumulate in great numbers; this is commonly referred to as memory inflation [43]. Specifically, CMV positivity is associated with an increased frequency of memory and terminally-differentiated CD8⁺ cells [43] and reduced frequency of influenza-specific IFN-? producing CD4⁺ cells [44, 45]. These effects, especially in the context of CD4⁺ cells, would likely counteract any beneficial synergism between high-dose vaccination and biological aging in improving vaccine antibody responses. Additional consequences of CMV infection, such as the reduction of switched memory B-cells [46], would also be expected to hamper possible benefits. Finally, in contrast to studies measuring biological age from DNA methylation [47, 48], we did not observe an association between biological age and CMV seropositivity. DNA methylation measures of biological aging are greatly influenced by changes in T- and NK-cell pools that also change with CMV infection [49], such as more frequent naïve and less frequent activated and memory subsets [43]. Hence, more studies are required to determine whether CMV actually plays a role in the multisystem physiological breakdown that represents biological aging.

Our study has notable strengths. It was nested within a previously published clinical trial conducted over 4 consecutive seasons where dose was randomized each year and a large number of participants were CMV-seropositive. Also, our biological age algorithm was based on a previously validated toolkit, and employed a biomarker panel with broad physiological context and a relatively large, separate training cohort of adults of similar demographics. Unfortunately, we are unable to draw conclusions on adults younger than 65, were limited in our ability to test associations between biological age and breakthrough infections due to a relatively low sample size, and did not pursue the cellular mechanism behind our primary findings. This is vital and future work should focus on the dynamics of important immune cell populations, particularly changes from pre- to post-vaccination, in relation to biological aging, high-dose vaccine and CMV-serostatus. The frequency and function of B-cells, CD4⁺ follicular helper cells, and monocytes would be excellent candidates.

Our findings suggest that advanced biological age further improves the positive effect of high-dose influenza vaccine on antibody titres in older adults, which is suppressed in the presence of CMV positivity. This three-way interaction between biological age, vaccine dose and CMV serostatus implicates a common immunological mechanism, although further study is needed to confirm this, and identify which cells are actually involved.