NUS Breast-Cancer DNA Signatures Research Tool 2026

Study and open-access tool

Researchers led by Dr Jason Pitt of the National University of Singapore's Cancer Science Institute analysed nearly 2,800 breast-cancer genomes and identified eight de novo DNA gain-and-loss signatures. NUS reported on 18 May 2026 that the team had also launched the open-access CNA Visualizer tool to help researchers interact with and decode structural genomic patterns. The underlying study was published in Cancer Research on 14 May 2026. (NUS, 18 May 2026, accessed 10 September 2026.)

The tool is designed for scientific exploration rather than direct patient use. NUS describes it as an open-access web portal through which scientists can visually explore the large dataset and examine copy-number alterations across cancer genomes. The public release does not describe CNA Visualizer as a hospital diagnostic service, a patient portal or a treatment-selection system. Its value in this announcement is that it makes the research framework available for further studies across cancer types (NUS, 18 May 2026, accessed 10 September 2026).

What the signatures represent

The work profiles changes in the number of copies of DNA in breast-cancer cells and relates those structural changes to mechanisms of tumour development and clinical outcomes. The analysis used the open-access The Cancer Genome Atlas (TCGA) and METABRIC breast-cancer databases. By breaking broad genetic patterns into more detailed, disease-specific categories, the study examined how genome instability interacts with the tumour immune microenvironment. A signature is a research pattern for genomic analysis, not a diagnosis by itself or a standalone test that determines treatment for an individual patient (NUS, 18 May 2026, accessed 10 September 2026).

The eight signatures also separated distinct genomic effects associated with BRCA1 and BRCA2 mutations. NUS reported that patients with relatively stable or “quiet” genomes and low macrophage infiltration tended to have better survival outcomes in the analysed data. This is an association observed in the study’s datasets, not a rule that predicts an individual’s survival or shows that a patient will respond to a particular medicine. The research pattern must be interpreted alongside clinical information and validated for the use being considered (NUS, 18 May 2026, accessed 10 September 2026).

Clinical-development boundary

NUS said the signatures and tool could support future diagnostic refinement and matching of patients to targeted therapies, including research into homologous recombination deficiency and therapies such as PARP inhibitors. Those are potential research and translation uses; the announcement does not establish clinical validation, regulatory approval, or that the tool is currently used to make treatment decisions in Singapore. (NUS, 18 May 2026, accessed 10 September 2026.)

The next phase described by NUS is validation in clinical settings to assess whether the signatures reliably predict patient responses to targeted therapy. The team also plans to examine further how genome instability and the tumour microenvironment influence longer-term clinical outcomes. Until that work is completed, the appropriate interpretation is a research framework and exploratory tool. “Could help refine” does not mean that CNA Visualizer is a clinically validated test, that the eight signatures are approved biomarkers, or that a patient should choose treatment from the article alone (NUS, 18 May 2026, accessed 10 September 2026).