Back

A Robust Cell-Free RNA Approach for the Early Detection of Colorectal Cancer

Monteagudo-Mesas, P.; Sanchez, L.; Asole, G.; Neto, B.; Tuni-Dominguez, C.; Gonzalez, L.; Rusu, E. C.; Cabus, L.; Panadero-Fajardo, S.; Catalina, P.; Garcia, S.; Simon-Extremera, P.; Padilla Garcia, L.; Lagarde, J.; Sanders, P.; Weber, M.

2026-07-04 oncology
10.64898/2026.07.01.26357015 medRxiv
Show abstract

Colorectal cancer (CRC) screening remains limited by patient adherence and sub-optimal sensitivity for early-stage disease. While liquid biopsy has revolutionized cancer diagnostics, cfDNA-based methods often struggle with early-stage detection due to low analyte levels. Here, we present a robust cell-free RNA (cfRNA) platform for the early detection of CRC. Using a retrospective cohort of 255 healthy controls and 250 CRC patients, we implemented an optimized workflow featuring a RUVg-based normalization strategy to remove platelet-driven transcriptomic noise. We identified differentially expressed genes enriched in key CRC-associated biological pathways, including inflammation, EMT, and metabolic dysregulation. An XGBoost classifier trained on these features achieved a mean AUC of 0.92 in cross-validation and 0.89 in a validation cohort, demonstrating 67% sensitivity at 90% specificity. Notably, our platform showed particular efficacy in identifying early stage cancer (stage I and II), achieving 73.7% sensitivity at 90% specificity. These findings suggest that cfRNA profiling offers a powerful, non-invasive orthogonal approach to CRC screening, capable of overcoming the sensitivity limitations of DNA-based assays in early-stage disease.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.