Prediction of long-term recurrence-free and overall survival in early-onset colorectal cancer: the ENCORE multi-centre study
Abstract Survivors of early-onset colorectal cancer (EOCRC, i.e., diagnosed before age 50) are likely to experience recurrence after completing treatment. In this international, multi-centric, phase I-II-III EDRN biomarker study, we identified a panel of tumor-derived biomarkers of EOCRC recurrence....
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-06-01
|
| Series: | npj Precision Oncology |
| Online Access: | https://doi.org/10.1038/s41698-025-00978-7 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Survivors of early-onset colorectal cancer (EOCRC, i.e., diagnosed before age 50) are likely to experience recurrence after completing treatment. In this international, multi-centric, phase I-II-III EDRN biomarker study, we identified a panel of tumor-derived biomarkers of EOCRC recurrence. We then trained and independently validated a machine learning model (XGBoost) to predict 5-year recurrence-free and overall survival (RFS and OS) of patients with stage I-III EOCRC. Patients with “low-risk” EOCRC demonstrated statistically higher rates of 2-, 5-, and 10 year RFS in both the training cohort (51.0 vs. 92.4%; 34.4% vs. 92.4%; 25.8% vs. 92.4%, respectively; p < 0.0001) and the validation cohort (78.9% vs. 100.0%; 75.0% vs. 100.0%; 75.0% vs. 100.0%, respectively; p = 0.0019). We also report a significant reduction in both over-treatment and missed recurrences compared to current clinically available options. This tissue-based, machine learning-powered assay was prognostic of long-term RFS and OS outcomes after curative-intent treatment of EOCRC (ENCORE was first registered on ClinicalTrial.gov [ID: NCT06271980] on February 15th, 2024). |
|---|---|
| ISSN: | 2397-768X |