Ki-67, a protein used to measure tumour growth, may also help prevent chromosome errors that drive cancer, a study suggests.

The findings could change how scientists view Ki-67, a marker commonly used in breast cancer and other tumours to assess how quickly cancer cells are growing.

Researchers found the protein may help preserve genome stability by maintaining the structural integrity of centromeres, key parts of chromosomes that help ensure DNA is shared correctly during cell division.

The research was led by professor Paola Vagnarelli at Brunel University of London in collaboration with scientists at the University of Edinburgh and the Technical University of Berlin.

Professor Vagnarelli said: “Doctors already measure Ki-67 to see how aggressive a cancer might be. But our results suggest it is actually helping maintain genome stability.

“That means it may be more than a marker. It could potentially also be a therapeutic target.”

The study examined three proteins that attach to chromosomes during cell division and help rebuild the molecular system that tells each new cell what kind of cell it is.

Every human cell carries identical DNA. What makes a liver cell different from a brain cell is which genes are switched on and which are kept inactive.

When a cell divides, that entire system of switches must be rebuilt. The three proteins involved in this process were Ki-67, Repo-Man and PNUTS.

Vagnarelli’s team developed a method that individually removes each protein from a living cell at the precise point of division. Older techniques could not isolate that moment cleanly.

They found that cells rely on all three proteins to reset themselves after division, but each failed in a different way when removed.

Without PNUTS, gene activity spiralled out of control and thousands of genes switched on at once.

Without Repo-Man, cells escaped safety checkpoints that usually stop damaged or abnormal cells from continuing to divide.

“What we didn’t expect was how clean the separation was,” said Vagnarelli.

Each protein fails in its own specific way. There is no redundancy, no safety net. Which means there are three separate points at which this process can go wrong.

“When the system breaks down, cells can emerge with the wrong number of chromosomes. That condition, called aneuploidy, is seen in disorders such as Down syndrome and in many cancers.

“We also found that these chromosome errors can trigger inflammatory signals inside the cell.”

Aneuploidy means a cell has too many or too few chromosomes, which can disrupt normal growth and function.

Inflammatory signals are chemical messages that can make a cell behave as if it is responding to injury or infection.

“These cells behave almost as if they are under attack,” said Vagnarelli.

“The immune response switches on because the genome is unstable.

“That link between chromosome imbalance and inflammation could help explain patterns we see in several diseases.”

The researchers said the findings may help cancer scientists better understand how chromosome instability, loss of gene regulation and cells dividing before they are ready contribute to tumour growth.

They said understanding the normal machinery that prevents these errors may help researchers find ways to push cancer cells into making mistakes they cannot survive.

“We now have a clearer map of the machinery that resets the cell after division,” said Vagnarelli.

“That knowledge gives us a starting point for thinking about new therapeutic approaches.”

A common cholesterol drug could help weaken a fluid shield that helps ovarian cancer tumours survive, early lab findings suggest.

The findings do not show the drug treats ovarian cancer. But they suggest changing the environment the cancer depends on could make it more vulnerable to existing treatment.

A federally funded study at Duke University School of Medicine found that ascites, a build-up of fluid in the abdomen, may do more than cause discomfort.

Doctors can drain ascites to ease pain, improve mobility and make breathing easier, but the fluid may also help cancer cells survive and spread. It occurs in 90 per cent of people with advanced ovarian cancer.

According to the study, ascites acts as a shield, helping cancer cells evade ferroptosis, a form of cell death.

Ferroptosis is a kind of cellular rusting. It happens when iron inside a cell reacts with certain fats, causing the cell membrane to break apart.

Many metastatic cancer cells, meaning cells that float freely through the abdomen looking for new places to grow, are naturally vulnerable to this kind of damage.

“Doctors have mostly viewed ascites as a symptom rather than an active driver of disease,” said Jen-Tsan Chi, professor in the department of molecular genetics and microbiology and co-leader of the Cancer Biology Program at the Duke Cancer Institute.

“We’ve learned it gives cancer a survival advantage, which fills a major gap in understanding how ovarian cancer spreads.”

Scientists bathed cancer cell lines and patient-derived tumour cells in ascites collected from patients and watched how they responded to ferroptosis triggers.

The fluid protected cancer cells by changing how they store fats and control iron levels, effectively blocking cell death.

The protection required only trace amounts, with as little as 2 per cent immersion shielding cancer cells from destruction.

“What surprised us was how selective this effect was,” said Yasaman Setayeshpour, first author and graduate student in molecular genetics and microbiology at Duke School of Medicine.

“Ascites didn’t protect the cancer cells from other well-known types of cell death, like apoptosis or necrosis, it only blocked ferroptosis.

“To figure out why, we broke ascites down into major parts, like lipids, proteins, and small molecules, and tested what happened when each was removed.

“When we took the lipids out, the protective effect disappeared. That told us lipids are the key reason ascites helps these cancer cells survive.”

But researchers found an unexpected helper in bezafibrate, an older cholesterol drug used to lower triglycerides by altering how the body processes fats.

The cholesterol drug restored sensitivity to ferroptosis, but only when ascites was present. On its own, the drug did not trigger cell death or slow tumour growth in mice.

The drug’s impact depended on the cancer’s surroundings, in this case the fat-rich fluid bathing the tumour. Researchers found that targeting this environment, using repurposed drugs like bezafibrate, could leave cancer cells more exposed to existing cancer treatments.

Chi said the finding could have implications beyond ovarian cancer. Other cancers, including colorectal and pancreatic cancers, can also spread within the abdominal cavity.

“This work shows how much the environment around a tumour matters,” Chi said.

“Biological fluids like ascites don’t just give cancer cells a place to move. They actively help drive how cancer spreads.”

Artera has won FDA clearance for ArteraAI Breast, its breast cancer platform for patients with early-stage HR-positive, HER2-negative invasive breast cancer.

ArteraAI Breast is the first and only FDA-cleared digital pathology-based risk stratification tool for breast cancer.

These FDA milestones come alongside recent CE marking for both the ArteraAI Prostate Biopsy Assay and the ArteraAI Breast Cancer Assay in the US and Europe.

“FDA clearance for ArteraAI Breast represents a significant expansion of our FDA-cleared AI platform in oncology,” said Andre Esteva, chief executive and co-founder of Artera.

“This milestone reflects the growing role of our technology across multiple cancer types. Breast cancer care is highly nuanced, with treatment decisions that depend on individualised risk.

“Our goal remains consistent across prostate and breast cancer, and beyond: to help clinicians translate complex data into more precise, personalised treatment decisions across the cancer journey.”

ArteraAI Breast generates an AI-derived risk score showing the likelihood of distant metastasis, meaning cancer spreading to another part of the body, in patients with early-stage HR-positive, HER2-negative breast cancer.

Using digitised histopathology images, which are scanned tissue sample images, alongside patient clinical variables, the model sorts patients into low-risk and high-risk groups based on a predefined risk score cut-off.

In early-stage HR-positive, HER2-negative breast cancer, deciding the right intensity of treatment can be complex because clinical and pathological factors vary. Artera said the tool is designed to support clinicians within established decision-making frameworks.

Data presented at the 2025 San Antonio Breast Cancer Symposium evaluated the model in early-stage breast cancer and demonstrated the potential to inform chemotherapy benefit in certain patient populations.

“This clearance represents an important advance on the road to personalising treatments for patients with early-stage breast cancer,” said Eric Winer, medical oncologist and director of the Yale Cancer Center.

“Using AI and digital pathology has the potential to streamline operational workflows, while creating a strong interdisciplinary linkage between oncology and pathology. This approach may further improve the clinicians’ ability to help patients make the best treatment decisions.”

ArteraAI Breast is designed to integrate directly into standard pathology workflows using routine surgical resection samples, without requiring additional tissue or separate specimen collection.

This approach allows the software to provide same-day results, enabling pathology laboratories to give clinicians patient-specific prognostic risk information alongside standard histopathology reports.