🔍 Intelligence Synthesis JP/EN
Ethics · Regulation · Technology — Pharma Practice Notes

AI and Cancer Research ── Cancer biology ── 2026-09-30

A glowing thread from a tissue slice's cell network reaches toward an unlit far platform: AI predictions not yet anchored in clinical use.
What this means, as an image (AI-generated, GPT Image): Most cancer AI results are retrospective or preclinical, so check the evidence stage before trusting them in clinical decisions.Download image (PNG, 2000×800)
The figure runs from varied cancer data, namely tissue, images, health records and blood, through foundation models pretrained to learn representations, to predictions of survival, progression and response to immunotherapy or drugs. Evidence is mostly retrospective or preclinical, and hurdles remain, such as minibinders in CAR use. The end point is patient stratification and drug ranking.
Image abstract — the whole article on one page (click to enlarge)
12 Cancer biology papers. Lead: SpaCEy links spatial tissue patterns to clinical outcomes using explainable graph neural networks.

Journals covered and how papers are chosen: see the index

Cancer molecular biology and AI

← Back to this issue