The epigenome remembers and primes for cancer risk

Through Cancer Grand Challenges team PROSPECT is funded by Cancer Research UK, the US National Cancer Institute, the Bowelbabe Fund for Cancer Research UK and the French National Cancer Institute.

Tell us about the starting point for this work?
Surya: We wanted to understand how exposures and experiences throughout a lifetime influence the risk of cancer. We know that things like smoking and radiation cause DNA damage and increase mutagenesis. But there's growing evidence that things like diet, metabolism and obesity, are strongly connected to cancer, but aren’t thought to be strongly mutagenic. We wanted to understand if they act through epigenetic mechanisms. And these are temporary exposures— your diet in college doesn't necessarily reflect your diet now. So how does your body remember those exposures?

How did you go about studying this?
Surya: We chose to study one of the strongest risk factors for cancer in any tissue, particularly the colon, which is chronic inflammation, using a mouse model of colitis. We also developed a new technology, SHARE-TRACE, that allowed us to simultaneously in single cells measure their clonal history, so their past familial relationships, their current gene expression states, and their epigenetic states.

What did you find?
Surya: When the mice recover, they're clinically normal and the tissue looks normal to the eye. But we found that they still have changes within their chromatin, in their epigenome, that lasts for over 100 days. The most remarkable part of that finding is that the colon and the intestinal epithelium is one of the most rapidly turned over tissues in the body. Using our new technologies, we found that this memory could be passed clonally, meaning a mother cell could pass it to its daughter cells.

Jason: 100 days is quite a long time for a mouse and similar processes to be true in humans, that these inflammatory memories that are inscribed in the epigenome will persist for months, if not years.

What does this mean for cancer risk?
Surya: The implication for cancer is that these epigenetic changes are essentially behaving somewhat like a somatic mutation, where you get clonal lineages, which all have the same memory, and all are potentially primed for cancer. We went on to show that mice that had recovered from colitis formed bigger tumours than healthy controls.

How does this change the way we think about how cancer starts?
Surya: We think that if the tissue has this epigenetic memory of colitis and damage, that when it gets a DNA mutation, it’s primed to turn on the same regenerative programmes it turned on in response to inflammation, which causes increased growth. We've created somewhat of a new framework to think about cancer risk factors— the field has traditionally focused on how exposure to risk factors changes the genome and expansion of mutations. But now we can say there's also a persistent, durable epigenetic memory that these cells have which is causal in promoting tumour growth and will work hand in hand with genetic mutations.

Jason: We found that the cell of origin is an experienced one. It's seen inflammation. It's adapted to that inflammation. And it's inscribed those memories in its epigenome. And those memories, just like memories in our brain, they aren’t forgotten very quickly.

What’s next?
Jason: The epigenome is a really big place. At every base pair in the genome —
about 3 billion in humans and 2.8 billion in mice — we could be encoding epigenetic information. Now that we have tools and technologies that allow us to find the “needle in the haystack”, we can track their gain and loss in response to diverse human experiences. With PROSPECT we are investigating whether other sorts of influences like obesity, alcohol and diet are affecting the same parts of the genome, or whether they are priming cells in other ways. And we’re also looking in patients with early-onset disease.

What could the impact be for patients?
Surya: If we can identify epigenetic signatures that are predictive of developing cancer, we would have the ability to track risk in people.

Jason: Genetic mutations often catch disease too late. To give patients time and allow clinicians to successfully intervene, we need to catch diseases much earlier in life. We think that we’ve revealed an important mechanism describing disease risk, and we hope to inspire more efforts to translate epigenetics to the clinic.

Portfolio perspective

The PROSPECT team’s work puts forward the concept of epigenetic signatures, akin to the mutational signatures made famous by team Mutographs, which can also be tracked to ascertain cancer risk. Mutographs work also showed mutations are required but not sufficient for tumour formation, with team PROMINENT now investigating promotional events that tip the balance. PROSPECT’s work also identifies mechanisms by which tumour promoters can drive cancer formation. PROSPECT’s findings also expand the SPECIFICANCER team’s concept of tissue specificity. SPECIFICANCER showed that the tissue specific ability of oncogenic mutations to drive carcinogenesis was dependent on the epigenetic landscape of the tissue, laid down during development. PROSPECT has now shown that the epigenetic landscape is durably shaped by cancer risk factors, and that these epigenetic changes can be inherited by daughter cells, fuelling clonal expansion.

Nagaraja, S., Ojeda-Miron, L., Zhang, R. et al. Epigenetic memory of colitis promotes tumour growth. Nature (2026). https://doi.org/10.1038/s41586-026-10258-4