MSK lab uses generative AI to design cancer binders that beat FDA-approved CAR T proteins in mice
anshulkundaje · x · 2026-09-10
Caleb Lareau's lab at Memorial Sloan Kettering Cancer Center spent three years applying generative AI to design "binder" proteins that recognize cancer cells, publishing the results in Nature Biomedical Engineering.
- In mouse proof-of-concept studies, the AI-designed binders targeting BCMA significantly outperformed those used in FDA-approved CAR T therapies
- The study also explains why AI models have struggled to design binders against the CD19 blood cancer target, and why CAR T cells sometimes activate prematurely and become exhausted before reaching tumors
- The long-term goal: replace the field's traditional "guess and check" binder discovery with evidence-based design, expanding access to CAR T therapy — currently only 4% of US patients with advanced or metastatic cancer are candidates, and 3% actually benefit
Related event: MSK's AI-Designed Anti-Cancer Proteins Fully Clear Tumors in Mice(3 posts)→
More from Research
- HarnessMM cuts agent tokens per task by up to 88% while raising success rates — omarsar0 · 2026-10-26
- Ben Recht: The world needs a truly open language model, and it's easier to build than you think — beenwrekt · 2026-10-26
- New Blog Breaks Black-Box Distillation Into Five Practical Steps — chrisrohlf · 2026-10-26
- DetectorScope turns black-box AI text detectors into auditable ones with sparse autoencoders — _akhaliq · 2026-10-26
- OPST paper shows LLMs self-improve with zero human data, guided only by topic descriptions — _akhaliq · 2026-10-26
- Tencent's HiSTEVE 1.0: hierarchical VLA agent plays Minecraft with native keyboard and mouse, hits 85% SOTA — _akhaliq · 2026-10-26