Hetairos: AI Predicts Brain Tumor Subtypes from Routine Slides

Molecular classification of central nervous system (CNS) tumors via methylation profiling offers the highest diagnostic precision but is expensive, time-consuming, and not widely available. Standard H&E histology slides, by contrast, are routine and globally accessible.

Hetairos is a new AI model that predicts 102 methylation-based CNS tumor subtypes directly from digital H&E slides. Trained and validated on over 11,000 slides from 9,606 patients across 11 centers on four continents, it achieves strong accuracy — especially on high-confidence predictions (0.87–0.88) — and significantly outperforms board-certified neuropathologists when limited to histology alone.

In prospective real-world use, Hetairos provided high-confidence predictions in a majority of cases requiring molecular testing, agreed with final integrated diagnoses in over 90% of those cases, and reduced diagnostic turnaround time from ~12 days to just 12 minutes at a fraction of the cost.

This Ledger Entry expands how readers think about AI in precision oncology by demonstrating that advanced models can reliably infer epigenetic tumor subtypes directly from routine H&E histology, enabling faster, cheaper, and more accessible diagnostics while helping prioritize molecular testing in CNS tumors.