Comment from Maxime Taquet

Recombinant shingles vaccine and dementia We welcome FDA's initiative to systematically identify repurposing candidates for unmet medical needs, and wish to submit evidence relevant to neurodegenerative disease (an FDA-identified priority area) concerning the recombinant herpes zoster (shingles) vaccine and its potential to reduce dementia risk. Using US electronic health record data (TriNetX, >100 million patients) and using a natural experiment created by the step change from the use of the live to the use of the recombinant shingles vaccine, we recently showed that the recombinant zoster vaccine is associated with a 17% relative reduction in dementia burden over 6 years compared with the now-discontinued live zoster vaccine (RMTL ratio 0.83, 95% CI 0.79–0.87, P<0.0001; 164 additional diagnosis-free days), with consistent effects in men and women. This builds on independent natural-experiment evidence from Wales and Australia, each exploiting age-based eligibility cutoffs for the live vaccine, and together these studies represent a convergent signal across three countries using two distinct quasi-experimental designs. Unlike conventional vaccinated-versus-unvaccinated cohort comparisons, which are vulnerable to "healthy vaccinee" bias (individuals who seek vaccination differ systematically from those who do not), our design compared vaccine recipients before and after a step-change in vaccine availability, so that both groups actively sought vaccination, removing this major source of confounding. All-cause mortality and negative-control-outcome comparisons showed no differences between cohorts, further arguing against residual confounding. We would highlight natural-experiment designs applied to large real-world/EHR datasets (e.g., step-changes in formulary or guideline-driven exposure, age-eligibility cutoffs) as a scalable, generalisable methodology FDA could encourage for surfacing repurposing candidates for already-approved products, particularly vaccines and other interventions unlikely to attract commercial sponsorship for label expansion. This approach can approximate causal inference without a trial and can be pre-registered and replicated across independent data sources, as demonstrated here. The recombinant vaccine has essentially replaced the live vaccine in the USA, UK, Australia and EU, meaning that the distinction between vaccine types is not merely of historical interest but is central to public health relevance going forward. We would encourage the FDA, alongside the NIH, to formally evaluate this candidate, given that the existing evidence, while not yet meeting the substantial-evidence standard for a labeling change, is sufficiently convergent to justify prioritised further study. We would be pleased to provide further input to support FDA's evaluation of this candidate. Reference Taquet M, Dercon Q, Todd JA, Harrison PJ. The recombinant shingles vaccine is associated with lower risk of dementia. Nat Med 2024;30:2777–2781.