An estimated 6.2 million Americans aged 65 and older have Alzheimer’s disease (AD), and that number is on track to triple by 2050. Given the lack of therapeutics to prevent or delay AD, as well as the limited success of therapeutics to modify disease, any reports of risk reduction strategies are worth a close look, which is why I was intrigued when I came across a recent retrospective study by Avram Bukhbinder and colleagues indicating that an annual flu vaccine was associated with lower risk of AD. As with all observational studies, this research has a number of shortcomings that limit interpretation, so what can we take away from these data?
What does the flu vaccine have to do with dementia?
Apart from preventing severe disease and complications from influenza infection, flu vaccination is hypothesized to promote broader changes in the immune system, which in turn impacts AD pathogenesis. The immune system is generally conceptualized as two branches: innate immunity, with which we are born, and adaptive immunity, a system in which specialized cells develop “memory” of previous diseases, leading to faster and more specific reactions to pathogens in the future. While vaccination has traditionally been associated with activating adaptive immune responses for a target disease, vaccines have more recently been recognized for their ability to impact innate immunity and induce non-specific protection against a broad range of pathogens – part of a phenomenon known as “trained immunity.” In addition to increasing the body’s ability to resist infection, these innate immune system changes may lower the risk for chronic inflammation and decrease neuroinflammation.
Trained immunity – which, notably, is not exclusive to the flu vaccine – may in turn impact AD pathogenesis. Aging is associated with a chronic, low-grade, systemic elevation of proinflammatory cytokines and other biomarkers of inflammation, including in the brain. One hallmark of this “inflammaging” process is the increased activation of microglia, the resident innate immune cells in the central nervous system, responsible for clearing cell debris, misfolded proteins, and dying neurons. Activated microglia are a common feature of AD and are observed concentrating around amyloid plaques in the brain from autopsies of AD patients. Proinflammatory cytokines secreted by activated microglia are also upregulated in the brains of AD patients (reviewed here). Therefore, an immune response that modulates the activity of innate immune cells such as microglia has the potential to modulate disease, and vaccines may elicit just such a response.
About the study
Though others have previously reported reduced rates of dementia among seniors who received flu vaccines, the retrospective study by Bukhbinder et al. is by far the largest investigation on the topic to date, as well as the most rigorous in controlling for covariates known to increase AD risk, including history of traumatic brain injury, vitamin B12 deficiency, type 2 diabetes, congestive heart failure, anxiety disorder, depression, and others. The investigators adopted a matched pair design, in which a total of just under 2 million flu-vaccinated and unvaccinated participants age 65 and older (mean age = 73.7±8.7 years, 56.9% female) were matched based on similar covariates, demographics, medication usage, and comorbidities. Participants were included in the flu-vaccinated group if they had received between 1-6 flu vaccines in the six years leading into the study. The authors then followed patients for 4 years and measured the incidence of AD (excluding any unvaccinated participants who received flu vaccines during the follow-up period).
The results showed that seniors who received at least one flu vaccine were 40% less likely than their non-vaccinated peers to develop AD in the 4-year follow-up period. A total of 5.1% of the vaccinated subjects and 8.5% of the non-vaccinated subjects were diagnosed with AD during the follow-up, corresponding to a relative risk (RR) of 0.6 (95% CI: 0.59-0.6) and an absolute risk reduction (ARR) of 3.4% (95% CI: 3.3 -3.5%) for flu vaccination. The number needed to treat was 29.4, meaning that for every 29 seniors who receive the flu vaccine, there will be one fewer who develops AD as compared to a group of 29 unvaccinated seniors.
Additionally, the inverse association between vaccination and AD incidence appeared to be dose-dependent, with more vaccines associated with higher protection from AD. Secondary analysis of the number of vaccines received by subjects in the 6-year look-back period compared the incidence of AD over the 4-year observation period in unvaccinated subjects, those receiving 3 flu vaccinations, and those receiving 6 flu vaccinations (modeled in the figure below). The lowest incidence of AD was associated with 6 flu vaccines, or annual vaccination.
Figure: Estimated cumulative incidence function of AD incidence by number of influenza vaccinations. Source: Bukhbinder et al.
Flu Vaccines or Flu Infection?
Of course, this is an observational epidemiology study, and as I’ve discussed in the past, it cannot address causality. The existence of a plausible (albeit still fairly vague) mechanism in activating trained immunity bolsters the argument for causality, as does the apparent dose-dependency of the effect. But despite the authors’ attempts to match participants and correct for dozens of covariates, the study still leaves ample room for alternative interpretations.
For one, this study leaves ample room for the influence of healthy user bias, a recurring theme in epidemiology studies. Those who regularly receive flu shots are likely more health-conscious overall, so perhaps they are less likely to develop AD for reasons unrelated to vaccination status – such as exercise or diet habits. Though the investigators adjusted for diseases such as diabetes and heart disease, there is no guarantee that healthy lifestyle variables would track with those disease metrics, and the researchers did not take these lifestyle variables into account separately as potential confounds.
But perhaps the most glaring confound stems from the very purpose of flu vaccines: preventing the disease of influenza. If those who receive the vaccine are less likely to get the flu, could it be possible that, instead of flu vaccines reducing risk of AD, flu infection raises risk of AD? Injury to the brain can occur during an influenza infection from direct viral invasion or as a result of collateral damage from the systemic immune response, providing a potential pathway by which influenza could increase risk of AD, but the effect of flu infection on AD risk has received little attention. The authors of this study did not include influenza infection in their analysis due to concerns over misclassification – a major limitation of this study. Common infections are reported to be associated with increased risk of dementia, but again, these data come from association studies that cannot establish causality. So the question remains open: could some – or all – of the benefits of vaccination be due to preventing influenza infection and associated complications?
Another reason to get your flu shot
No therapies have emerged that can reverse or cure AD, so identifying measures that may reduce the risk of onset is critical. While the study by Bukhbinder et al. comes with many potential caveats and confounds, it does show that a link between flu vaccination and reduced risk of AD is at least plausible. And unlike expensive or invasive interventions, an annual flu vaccination is cheap, easy, and low-risk.
More work needs to be done to understand if and how vaccines might protect against neurodegenerative disease. A randomized controlled trial would certainly provide more insight into the potential causality between flu vaccines and AD, but given the mortality risk of influenza, such a study is unlikely to be conducted for ethical reasons. Further, even randomized trials would leave open the question of flu vaccine vs. flu infection. So for now, we can simply make our best guesses based on mechanistic data and epidemiology studies such as the one by Bukhbinder et al.
But while the full picture may not be clear, the message we can take away from this study is this: what do you have to lose? At the very least, the flu vaccine will help to avoid a bad case of influenza and complications that can arise from that virus. If there’s a chance it helps prevent AD, then all the better. Just another reason to get the annual flu shot.