A novel structural modeling magnitude and orientation radiomic descriptor for evaluating response to neoadjuvant therapy in rectal cancers via MRI
Abstract With advances in neoadjuvant therapies for rectal cancer, accurately evaluating tumor regression and response is increasingly critical for enabling personalized follow-up, including non-operative management. Given the lack of reliable assessment methods, there is an opportunity to develop c...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-07-01
|
| Series: | npj Precision Oncology |
| Online Access: | https://doi.org/10.1038/s41698-025-01007-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract With advances in neoadjuvant therapies for rectal cancer, accurately evaluating tumor regression and response is increasingly critical for enabling personalized follow-up, including non-operative management. Given the lack of reliable assessment methods, there is an opportunity to develop computerized image-based markers for identifying early responders. We present a novel radiomic signature, the STructural mOdeling magnituDE and Orientation (StODeO) descriptor for quantifying image-based surrogates of tumor shrinkage within and around the rectum via routine MRI scans. StODeO descriptors measure both the magnitude (how much) and direction (inward/outward) of structural displacements in the diseased rectal wall, with respect to a newly constructed, first-of-its-kind image-based reference atlas of the non-diseased rectum. Combined with traditional radiomic texture features, StODeO showed superior performance in distinguishing pathologic response groups of rectal cancer patients post-chemoradiation (hold-out validation AUC 0.77) as well as segregating composite response groups in a clinical trial of immune checkpoint inhibitors (external validation accuracy 0.71; NCT02688712). The integrated StODeO-texture signature demonstrated robustness across annotation sources and imaging variations, intuitive modeling of therapy-associated structural changes, and statistical associations with tumor-immune biology, including macrophages and CD8+ T-cells from multiplexed biopsy analysis. StODeO offers a novel image-based surrogate of tissue displacements from treatment effects and residual disease after neoadjuvant therapy in rectal cancers. |
|---|---|
| ISSN: | 2397-768X |