Deep learning for dual detection of microsatellite instability and POLE mutations in colorectal cancer histopathology
Problem
Detecting microsatellite instability and POLE mutations in colorectal cancer from histopathology slides to identify immunotherapy candidates.
Why it matters
MSS patients with POLE mutations may benefit from immunotherapy but are rarely screened due to cost and rarity.
What the AI did
A transformer-based deep learning classifier trained on MSI-labeled pathology slides was validated for dual detection of MSI and POLE mutations across three external datasets.
Results
not stated in source
Impact
External validation across three datasets is claimed but no sensitivity, specificity, or AUC figures are given in the abstract.
Record
| Venue | npj Precision Oncology |
|---|---|
| Published | 2024 |
| Work type | article |
| Authors | Marco Gustav, Nic Gabriel Reitsam, Zunamys Itzell Carrero, Chiara Maria Lavinia Loeffler, Marko van Treeck, Tanwei Yuan, Nicholas Paul West, Philip Quirke et al. |
| Citations | 38 (OpenAlex, at import time) |
| References | 64 |
| Model or method | transformer-based Deep Learning classifier |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1038/s41698-024-00592-z |
| Open access | gold |
| Keywords | colorectal cancer, deep learning, MSI, POLE, histopathology, immunotherapy |
Importance rating
significant — thousands to millions affected
Identifying MSI/POLE colorectal cancer patients who may benefit from immunotherapy; screening tool could expand treatment access, but no patient scale stated.