Deep learning reveals prognostic information beyond AJCC stage in routine histopathology of early-stage melanoma: the population-based D-ESMEL study
- 1Department of Pathology and Clinical Bioinformatics, Erasmus MC Cancer Institute, University Medical Center Rotterdam, Rotterdam, The Netherlands
Abstract
Of all melanoma patients who develop distant metastasis over time, 41% were initially diagnosed with stage I/II. More accurate prognostic models are needed to select patients for potential (neo)adjuvant therapy. We hypothesise that histopathological features in whole slide images (WSIs) from primary early-stage melanomas provide additional information beyond American Joint Committee on Cancer (AJCC) stage. We utilised data from the population-based Dutch Early-Stage Melanoma (D-ESMEL) study. This included a discovery set (n=328) of case-control sets matched on follow-up time and known prognostic features (i.e. Breslow thickness, ulceration, age, and sex) and a validation cohort based on >5,500 patients following a nested case-control (NCC) design with stage-matched controls (n=266). We developed an end-to-end deep learning workflow to automatically extract relevant patches from WSIs and predict patient-level metastatic risk based on histopathological features independent of AJCC stage. The discovery set served as a training set for the workflow using 10-fold cross-validation. The prognostic performance of this model and its independence from clinical risk factors was evaluated in the validation cohort using odds ratios (OR) and the weighted C-index. We achieved a balanced accuracy of 0.88 (mean: 0.81; 95% CI: 0.77-0.85) in the best validation fold of the discovery set. In the validation cohort, metastatic risk predictions from the deep learning model were an independent prognostic factor (adjusted OR: 1.97; p<0.01). Combining the deep learning predictions (C-index: 0.86) with AJCC stage (C-index: 0.80) yielded the highest C-index of 0.87, significantly improving the C-index compared to AJCC stage alone (p<0.001). Our histopathology-based deep learning model was rigorously evaluated in a population-based cohort and demonstrated added prognostic value beyond AJCC stage. Further investigations of the underlying biological features could improve clinical decision making and patient outcomes.
Related
- Burden of malignant melanoma among the elderly population in East Asia from 1990 to 2023 and machine learning projection to 2040
- Prognostic value of platelet-to-white blood cell ratio in critically ill patients with non-traumatic subarachnoid hemorrhage: a machine learning-driven study of the public database and the hospital information system
- The Prognostic Role of CD8+ Tumor‐Infiltrating Lymphocyte Density in Head and Neck Mucosal Melanoma
- Multi-modal data Integration for Metastatic Risk Prediction of AJCC Stage I/II Melanomas: Results from the Population-based Dutch Early-stage Melanoma Study (D-ESMEL)
- CDKN2A c.146T>C (I49T) in Mexican-ancestry individuals: Findings from melanoma and population cohorts
BibTeX
@misc{ouwerkerk2026deep,
title = {{Deep learning reveals prognostic information beyond AJCC stage in routine histopathology of early-stage melanoma: the population-based D-ESMEL study}},
author = {J. Ouwerkerk and F. Akram and L. Hollestein and T. Kerkour and C. Zhou and A. Stubbs and A. Mooyaart and Y. Li},
howpublished = {medRxiv},
year = {2026},
month = oct,
doi = {10.64898/2026.10.29.26366599},
url = {https://www.medrxiv.org/content/10.64898/2026.10.29.26366599v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Deep learning reveals prognostic information beyond AJCC stage in routine histopathology of early-stage melanoma: the population-based D-ESMEL study
AU - J. Ouwerkerk
AU - F. Akram
AU - L. Hollestein
AU - T. Kerkour
AU - C. Zhou
AU - A. Stubbs
AU - A. Mooyaart
AU - Y. Li
PY - 2026
DA - 2026/10/29
PB - medRxiv
DO - 10.64898/2026.10.29.26366599
UR - https://www.medrxiv.org/content/10.64898/2026.10.29.26366599v1
AB - Of all melanoma patients who develop distant metastasis over time, 41% were initially diagnosed with stage I/II. More accurate prognostic models are needed to select patients for potential (neo)adjuvant therapy. We hypothesise that histopathological features in whole slide images (WSIs) from primary early-stage melanomas provide additional information beyond American Joint Committee on Cancer (AJCC) stage. We utilised data from the population-based Dutch Early-Stage Melanoma (D-ESMEL) study. This included a discovery set (n=328) of case-control sets matched on follow-up time and known prognostic features (i.e. Breslow thickness, ulceration, age, and sex) and a validation cohort based on >5,500 patients following a nested case-control (NCC) design with stage-matched controls (n=266). We developed an end-to-end deep learning workflow to automatically extract relevant patches from WSIs and predict patient-level metastatic risk based on histopathological features independent of AJCC stage. The discovery set served as a training set for the workflow using 10-fold cross-validation. The prognostic performance of this model and its independence from clinical risk factors was evaluated in the validation cohort using odds ratios (OR) and the weighted C-index. We achieved a balanced accuracy of 0.88 (mean: 0.81; 95% CI: 0.77-0.85) in the best validation fold of the discovery set. In the validation cohort, metastatic risk predictions from the deep learning model were an independent prognostic factor (adjusted OR: 1.97; p<0.01). Combining the deep learning predictions (C-index: 0.86) with AJCC stage (C-index: 0.80) yielded the highest C-index of 0.87, significantly improving the C-index compared to AJCC stage alone (p<0.001). Our histopathology-based deep learning model was rigorously evaluated in a population-based cohort and demonstrated added prognostic value beyond AJCC stage. Further investigations of the underlying biological features could improve clinical decision making and patient outcomes.
ER -