CT-Translate flags immunotherapy responders: top-ranked third hits 61% vs 43% baseline
david_van_dijk · x · 2026-10-22
CellType shared retrospective results for CT-Translate, a model that predicts which cancer patients will respond to immunotherapy.
- Across 5 published cohorts and 451 patients, the overall response rate was 43%
- Within the model's top-ranked third of patients, the response rate rose to 61%
- Every prediction came from a model that never saw that patient, using five-fold cross-validation
While retrospective, the results show meaningful discriminative power for AI-based patient stratification in immunotherapy.
Related event: CT-Translate predicts immunotherapy response, lifting rate to 61%(3 posts)→
More from Research
- We need an L0.5 for continual learning: agent state stacks up in 4 layers — willcb · 2026-10-22
- JOLT trains the teacher too, with penalties when its predictions diverge from the student — perplexity_ai · 2026-10-22
- Perplexity's JOLT matches GRPO with 6.5x fewer training tokens, JOLT+ wins on all five benchmarks — perplexity_ai · 2026-10-22
- Parallel Decoding Distillation Adopted by Alibaba, NVIDIA, Amazon Weeks After Release — ArashVahdat · 2026-10-22
- RFD4-Proteina hits 3.5% DNA-binder pass rate vs 0.8% for BoltzGen — AllThingsApx · 2026-10-22
- RFD4-Proteina passes computational checks on all 77 small-molecule targets — AllThingsApx · 2026-10-22