A spatial AI-based biomarker with independent prognostic value in early low- and intermediate- risk breast cancer
- 1Karolinska Instittutet
Abstract
BackgroundPatients with ER-positive/HER2-negative early breast cancer of low or intermediate grade are generally considered to have a favourable prognosis, yet a number will recur. This risk is poorly captured by standard clinicopathological factors. AI-based medical devices for whole slide image (WSI) analyses used for risk-stratification often predict on slide-level thereby overlooking spatial localisation of the biomarkers. We hypothesise that the spatial distribution of a previously reported AI-based marker provides independent prognostic value beyond its global score.
Patients and MethodsThis Swedish multicentre cohort study includes 6106 patients with primary invasive breast cancer: 3078 from the Stockholm region (training) and 3028 from Sk[a]ne (independent test). Tile-level high-grade morphology scores were generated with ViT_DG, a vision-transformer adaptation of DeepGrade, and the presence of clusters of high-risk tiles localised at the invasive tumour front was defined as a binary biomarker. Primary analyses were pre-specified in the ER-positive/HER2-negative/lymph-node-negative and Grade 1-2 subgroup. Association with breast-cancer recurrence-free interval (BCRFI) and distant metastasis were assessed using multivariable Cox proportional hazards models (HR).
ResultsIn the primary subgroup (n=1536 training; n=1403 test), tumour front clusters were present in 37% and 39% of patients. Cluster-positive status was associated with a 2.30-fold increased risk in the external test. Cluster-negative patients achieved a 10-year BCRFI of 95.3%, equivalent to Grade 1 patients and better than Grade 2 (92.4%), whereas cluster-positive patients (90.2%) did worse than the Grade 3 reference. The association was retained in patients classified as low-risk by ViT_DG (test set multivariable HR 2.86), identifying high-risk patients not captured by the global score.
ConclusionsThe absence of high-risk morphology clusters at the tumour front is an independent prognostic indicator, identifying a low-risk group independently of the global AI biomarker classification. More broadly, preserving the spatial distribution of AI-derived predictions enables prognostic stratification beyond slide-level aggregation.
ClinicalTrials.gov identifier: NCT02306096
Related
- Baseline serum carcinoembryonic antigen (CEA) is an independent prognostic biomarker for metastatic castration-resistant prostate cancer (mCRPC) patients receiving 177Lu-PSMA therapy
- Joint neutrophil-to-lymphocyte ratio-systemic inflammation response index score as an independent prognostic biomarker for immunotherapy-treated intrahepatic cholangiocarcinoma.
- Unveiling GSPT1 as a crucial prognostic biomarker and therapeutic vulnerability in breast cancer
- Transcriptomic Profiling of Mouse Mammary Tumors Enables Prognostic and Predictive Biomarker Discovery for Human Breast Cancer
- ABCF2 is a prognostic and immunological pan-cancer biomarker with growth-promoting function in breast cancer: a multi-omics study
BibTeX
@misc{boissin2026spatial,
title = {{A spatial AI-based biomarker with independent prognostic value in early low- and intermediate- risk breast cancer}},
author = {C. Boissin and Y. Xiang and B. Liu and J. Vallon-Christersson and E. Walker and J. Hartman and M. Rantalainen},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.07.26364961},
url = {https://www.medrxiv.org/content/10.64898/2026.10.07.26364961v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - A spatial AI-based biomarker with independent prognostic value in early low- and intermediate- risk breast cancer
AU - C. Boissin
AU - Y. Xiang
AU - B. Liu
AU - J. Vallon-Christersson
AU - E. Walker
AU - J. Hartman
AU - M. Rantalainen
PY - 2026
DA - 2026/10/07
PB - medRxiv
DO - 10.64898/2026.10.07.26364961
UR - https://www.medrxiv.org/content/10.64898/2026.10.07.26364961v1
AB - BackgroundPatients with ER-positive/HER2-negative early breast cancer of low or intermediate grade are generally considered to have a favourable prognosis, yet a number will recur. This risk is poorly captured by standard clinicopathological factors. AI-based medical devices for whole slide image (WSI) analyses used for risk-stratification often predict on slide-level thereby overlooking spatial localisation of the biomarkers. We hypothesise that the spatial distribution of a previously reported AI-based marker provides independent prognostic value beyond its global score. Patients and MethodsThis Swedish multicentre cohort study includes 6106 patients with primary invasive breast cancer: 3078 from the Stockholm region (training) and 3028 from Sk[a]ne (independent test). Tile-level high-grade morphology scores were generated with ViT_DG, a vision-transformer adaptation of DeepGrade, and the presence of clusters of high-risk tiles localised at the invasive tumour front was defined as a binary biomarker. Primary analyses were pre-specified in the ER-positive/HER2-negative/lymph-node-negative and Grade 1-2 subgroup. Association with breast-cancer recurrence-free interval (BCRFI) and distant metastasis were assessed using multivariable Cox proportional hazards models (HR). ResultsIn the primary subgroup (n=1536 training; n=1403 test), tumour front clusters were present in 37% and 39% of patients. Cluster-positive status was associated with a 2.30-fold increased risk in the external test. Cluster-negative patients achieved a 10-year BCRFI of 95.3%, equivalent to Grade 1 patients and better than Grade 2 (92.4%), whereas cluster-positive patients (90.2%) did worse than the Grade 3 reference. The association was retained in patients classified as low-risk by ViT_DG (test set multivariable HR 2.86), identifying high-risk patients not captured by the global score. ConclusionsThe absence of high-risk morphology clusters at the tumour front is an independent prognostic indicator, identifying a low-risk group independently of the global AI biomarker classification. More broadly, preserving the spatial distribution of AI-derived predictions enables prognostic stratification beyond slide-level aggregation. ClinicalTrials.gov identifier: NCT02306096
ER -