The assumption that peripheral vaccines cannot meaningfully alter brain immunity is now under serious pressure. For adults concerned about Alzheimer's prevention, a century-old tuberculosis vaccine may be doing something that no drug has convincingly achieved: retraining the immune cells that patrol the central nervous system itself — and in doing so, shifting the biological markers that precede dementia.
Two concurrent open-label trials enrolled 23 adults aged 55 or older, split between those with and without detectable Alzheimer's-related pathology at baseline. Each participant received two intradermal BCG vaccinations one month apart, with follow-up extending across a full year. Blood and cerebrospinal fluid (CSF) were analyzed using cytokine assays and single-cell transcriptional profiling. The headline finding is that BCG induced durable, trained immunity-like changes specifically within CSF immune cells — enhanced innate responsiveness coupled with distinct transcriptional signatures — that were not mirrored in peripheral blood. This compartment-specific immune imprinting is notable. Among participants who had not yet developed Alzheimer's-related pathology, these CNS immune shifts were accompanied by measurable reductions in amyloid-β levels, a key pathological hallmark of the disease.
This work sits within a rapidly expanding framework linking immune aging — sometimes called immunosenescence — to neurodegeneration. BCG's known capacity to induce trained immunity, an epigenetic reprogramming of innate immune cells toward heightened long-term responsiveness, has been epidemiologically associated with lower Alzheimer's incidence, but the biological mechanism in the human brain was entirely uncharacterized before this study. The CSF-specific response is conceptually important: it suggests the vaccine's immunological effects penetrate the blood-brain barrier in ways that circulating biomarkers alone would never reveal. Critical limitations deserve emphasis: with only 23 participants across two arms, this is hypothesis-generating, not confirmatory. Open-label design precludes placebo comparison. The amyloid reduction finding in the pre-pathology group, while intriguing, requires replication in randomized, powered trials. Still, as a mechanistic proof-of-concept bridging trained immunity and CNS biomarker modulation, this is genuinely paradigm-challenging work.