Eight new signatures that are unique to breast cancer
At the Cancer Science Institute of Singapore, researchers have mapped the hidden grammar of breast cancer's genetic disorder, identifying eight distinct DNA signatures across nearly 2,800 patient genomes — a discovery that may finally allow medicine to treat each patient's cancer as the singular, specific disease it truly is. For decades, the chaos of cancer genetics was read in broad strokes; now, finer distinctions between mutations like BRCA1 and BRCA2 are emerging, along with clues about who will survive and who will respond to which therapies. The work, paired with an open-access tool that invites scientists worldwide to explore the data, reflects a deepening conviction that precision in diagnosis is not a luxury but a moral imperative.
- Breast cancer has long been treated as a category when it is really a constellation — and that bluntness has cost patients access to the treatments most likely to help them.
- Eight newly identified DNA signatures now distinguish how individual tumors form and behave, including a crucial separation between BRCA1 and BRCA2 mutations that clinical practice had previously collapsed into one.
- Patients whose tumors showed 'quiet' genomes and low macrophage activity were found to survive longer — a finding that could fundamentally reorder how prognosis is communicated and treatment is planned.
- The signatures point directly toward better detection of homologous recombination deficiency, unlocking more precise use of PARP inhibitors for patients whose tumors are genuinely vulnerable to them.
- A freely available web tool, the CNA Visualizer, now places this genomic dataset in the hands of researchers globally, regardless of institutional resources.
- Clinical validation trials lie ahead — the moment of reckoning where these genetic patterns must prove they can reliably predict real patient outcomes in real hospital settings.
Dr. Jason Pitt and his team at the Cancer Science Institute of Singapore have spent months inside the genetic blueprints of nearly 2,800 breast cancer patients, searching for patterns that earlier research had overlooked. What they found — eight distinct DNA signatures unique to breast cancer — could change how doctors diagnose the disease and match patients to treatments that will actually work for them.
The study, published in Cancer Research, marks a departure from how genomic instability in cancer has traditionally been studied. Rather than applying broad, cross-tumor categories, Pitt's team drew from two major open-access databases to map the specific gains and losses of DNA copies occurring in breast cancer alone. The result was a far more detailed picture — eight signatures, each telling a different story about how a tumor develops.
Among the most significant findings: BRCA1 and BRCA2, long treated as interchangeable in clinical settings, leave distinct genomic fingerprints. The team also found that patients with relatively stable, 'quiet' tumor genomes and low levels of immune macrophages tended to live longer — a distinction that could meaningfully separate patients who appear similar on the surface but face very different futures.
The practical stakes are high. These signatures could improve detection of homologous recombination deficiency — a DNA repair failure that makes tumors susceptible to PARP inhibitors — allowing oncologists to match patients to targeted therapies rather than defaulting to broad chemotherapy regimens.
To extend the reach of their findings, the team built the CNA Visualizer, a free web tool that lets researchers anywhere explore the underlying genome data and run their own analyses. The next phase of work will test these signatures in clinical settings, while the team continues investigating how genomic instability and the tumor's immune environment interact — an interplay that may yet explain why some patients survive and others do not.
Dr Jason Pitt and his team at the Cancer Science Institute of Singapore have spent months sifting through the genetic blueprints of nearly 2,800 breast cancer patients, looking for patterns that previous researchers had missed. What they found—eight distinct DNA signatures that appear across breast cancer genomes—could reshape how doctors diagnose the disease and choose which treatments will actually work for each patient.
The work, published in Cancer Research, represents a shift in how scientists think about cancer's genetic chaos. Genomic instability has long been known as a defining feature of cancer, but earlier studies tended to lump breast cancer into broad categories that applied across many tumor types. Pitt's team took a different approach. They pulled data from two massive open-access databases—The Cancer Genome Atlas and METABRIC—and systematically mapped the gains and losses of DNA copies that occur in breast cancer specifically. The result was a much finer-grained picture: eight new signatures that are unique to breast cancer, each telling a different story about how a tumor develops and behaves.
One of the study's most striking findings concerns the two most common inherited breast cancer mutations: BRCA1 and BRCA2. These genes have long been lumped together in clinical practice, but Pitt's analysis revealed they leave distinct genomic fingerprints. The team also discovered that patients whose tumors had relatively stable genomes—what they call "quiet" genomes—and low levels of immune cells called macrophages tended to survive longer. This kind of detail matters enormously in the clinic. It means that two patients with seemingly similar breast cancers might actually have very different prognoses and might benefit from very different treatments.
The practical application is already taking shape. These eight signatures could help doctors detect homologous recombination deficiency, a specific type of DNA repair failure that makes tumors vulnerable to a class of drugs called PARP inhibitors. Better detection means better matching of patients to the drugs most likely to work for them—the essence of precision medicine. Rather than giving every patient the same chemotherapy cocktail and hoping for the best, oncologists could soon use these signatures to predict which patients will respond to targeted therapies.
To make sure the findings reach beyond Pitt's lab, the team built an open-access web tool called the CNA Visualizer. Scientists anywhere in the world can now log on and explore the massive dataset of cancer genomes themselves, visualizing the patterns and running their own analyses. This democratization of data is crucial. It means researchers working in hospitals and universities without massive genomics budgets can still access the insights and build on them.
The work is not finished. The next phase will test these signatures in real clinical settings to confirm they actually predict how patients will respond to treatment. Pitt's team also plans to dig deeper into how genome instability and the tumor's immune microenvironment interact—how the chaos of the cancer's DNA and the immune cells surrounding it together shape a patient's outcome. That interplay, still poorly understood, may hold keys to why some patients thrive and others do not.
Notable Quotes
The identified signatures could help refine future diagnostic tools and better match breast cancer patients with targeted therapies— Dr Jason Pitt, Cancer Science Institute of Singapore
The Hearth Conversation Another angle on the story
Why does it matter that they found eight signatures instead of just confirming what was already known about breast cancer genetics?
Because the old signatures were broad strokes—they applied to many cancer types. These eight are specific to breast cancer. That specificity is what lets you make real predictions about an individual patient's tumor.
So BRCA1 and BRCA2 mutations aren't actually the same thing, genetically speaking?
They're not. They're both inherited mutations that increase breast cancer risk, and doctors have treated them similarly for years. But when you look at the full genome, they leave different patterns of damage. That difference could change how you treat someone.
The "quiet genome" finding—patients with stable genomes and low macrophages surviving longer—that seems counterintuitive. Shouldn't genetic chaos be worse?
You'd think so. But it turns out that a tumor with a very unstable genome might be triggering a stronger immune response, which can be harmful. A quieter tumor with fewer immune cells infiltrating it seems to give patients better odds. It's not what the simple model would predict.
What does the CNA Visualizer actually do for a researcher who isn't at a major institute?
It gives them access to the same data Pitt's team analyzed. They can explore it, test hypotheses, maybe find connections the original team didn't see. It's the difference between reading a paper about a discovery and being able to investigate the raw material yourself.
When will doctors actually start using these signatures to treat patients?
That's the validation phase ahead. They need to prove in clinical settings that these signatures actually predict treatment response. That takes time—you have to follow patients, see who responds to PARP inhibitors and who doesn't, and check whether the signatures predicted it correctly.