A post claiming inflammation is "at the root of autism" is circulating, and this time there is a genuinely large study behind the general idea, though the actual paper is far more careful than the headline suggests.
The largest study of its kind
Published in Molecular Psychiatry in May 2026, this meta-analysis pooled 98 studies covering 4,236 autistic people and 3,333 non-autistic controls, examining 54 different cytokines, the small proteins immune cells use to signal to each other. It is by far the largest and broadest analysis of this question to date, and unlike many earlier reviews, the authors explicitly accounted for the risk of bias in each study before pooling results.
What it actually found
Autistic people showed elevated levels of several pro-inflammatory cytokines compared to non-autistic controls, including IL-1beta, IL-6, IL-8 and TNF-alpha, among others. These are well-known signaling molecules involved in the body's general inflammatory response, not autism-specific markers. The pattern is consistent with, and considerably strengthens, a body of smaller studies going back over a decade that had found similar but less consistent signals.
Why "at the root of" overstates it
The study's own authors are notably cautious, stating that higher-quality studies are still needed before firm conclusions can be drawn. That matters for a specific reason: this is a meta-analysis of correlational studies, comparing blood cytokine levels between two groups at a single point in time. It cannot establish that inflammation causes autism, only that the two are associated. The elevated cytokines could reflect a contributing factor, a downstream consequence of autism-related stress or co-occurring conditions like gastrointestinal issues or anxiety, or something correlated with both. A meta-analysis of associations, however large, is not evidence of a root cause.
What this all adds up to
This is a real, large, carefully conducted study that meaningfully strengthens the case that immune signaling looks different, on average, in autistic people, an important lead for future research into biology and treatment. But the "at the root of autism" framing goes further than even the study's own authors are willing to go. The honest version: elevated inflammatory markers are a consistent finding worth taking seriously, not a discovered cause.
If you are exploring your own neurotype, our free autism screening is a starting point for self-understanding, not a substitute for this kind of biological research.
Source: Volk, Lukito, Radua, Luksch, Roessner and Beyer, "Atypical cytokine profiles in people on the autism spectrum: a comprehensive systematic review and meta-analysis including 54 cytokines," Molecular Psychiatry (2026). Educational summary - not a diagnosis or medical advice.