Blood Test Could Detect Molecular Barcodes Capable of Distinguishing Cancer Types

Some cancers are difficult to classify, track, and monitor after treatment, posing a major clinical challenge. Many tumors shed little DNA into the bloodstream, making it hard to detect minimal residual disease or predict relapse using current liquid biopsy tools. At the same time, cancer cells activate molecular programs that are absent in normal tissues, but many of these signals remain unexplored. Researchers have now shown that a previously overlooked class of cancer-specific small RNAs can both drive tumor behavior and be detected through simple blood tests, offering a new way to classify cancers and monitor patients over time.

In research led by Arc Institute (Palo Alto, CA, USA), the team focused on “orphan” non-coding RNAs, small RNA molecules not found in normal tissue but repeatedly detected in cancer genomes. Using large-scale computational mining of cancer sequencing datasets, the researchers systematically mapped these RNAs across tumors and termed them oncogenic RNAs or oncRNAs.

To understand their relevance, the team combined cancer genome analysis with machine learning, functional screening, and animal models. They developed classifiers that use oncRNA expression patterns as molecular barcodes to distinguish cancer types, subtypes, and cellular states. Parallel functional screens used lentiviral libraries to overexpress or suppress hundreds of oncRNAs in cancer cells, allowing the researchers to identify which of these molecules actively drive tumor growth and metastasis.

By analyzing small RNA sequencing data from The Cancer Genome Atlas across 32 cancer types, the researchers identified approximately 260,000 oncRNAs present across all major cancers. Each cancer type showed a distinct oncRNA pattern, enabling machine learning models to classify tumor identity with 90.9% accuracy, which was independently validated at 82.1% accuracy in nearly 1,000 additional tumors. Subtype-specific patterns were also observed within cancers, such as differences between basal and luminal breast tumors.

Functional screening revealed that about 5% of oncRNAs directly promoted tumor growth in mouse xenograft models. Detailed analysis of two breast cancer oncRNAs showed activation of pathways linked to epithelial-mesenchymal transition and cell proliferation. These findings, published in Cell, were consistent with pathway activity observed in patient tumor datasets.

The most clinically significant finding was that many oncRNAs are actively secreted into the bloodstream by cancer cells. Analysis of cell-free RNA from cancer cell lines showed that roughly 30% of oncRNAs enter circulation. In serum samples from 192 breast cancer patients enrolled in the I-SPY 2 neoadjuvant chemotherapy trial, changes in total oncRNA burden after treatment strongly predicted outcomes. Patients with persistently high oncRNA levels had nearly fourfold worse overall survival, independent of standard clinical measures.

These results suggest oncRNAs could address a major gap in cancer care by enabling sensitive blood-based monitoring of minimal residual disease, especially in cancers that shed little DNA. Ongoing work aims to expand validation in larger prospective trials and explore how oncRNA dynamics could guide treatment decisions, detect recurrence earlier, and refine patient stratification.

“We think oncRNAs represent a new class of cancer-emergent molecules that function as both drivers and biomarkers. We hope that this resource, which we’ve made open source, opens new directions for the field,” stated the authors.

Related Links:
Arc Institute