Using AI to uncover the causes of late recurrence in ER-positive breast cancer

Recurrence. It’s the last word anyone who’s been through breast cancer treatment wants to hear.

And thankfully, most people who've had treatment for ER-positive breast cancer never will. But for some, the cancer returns years after their original diagnosis, even after surgery and many years of hormone therapy. At the moment, doctors still cannot reliably tell who remains at risk of late relapse.

It’s a story we hear far too often. You’ve completed 5 or 10 years of hormone therapy. Persevering. Dealing with the side effects. And then one day, that dreaded word: “Recurrence”.

This can happen because a small number of breast cancer cells are able to resist treatment and keep growing in the body, but at first, there are too few to be detected. Over time, these cells can build and cause the cancer to recur. When that happens, the disease is often harder to treat. And if it has spread to other parts of the body, it may no longer be curable.

That is why this work is so important. We need better ways to detect early signs of treatment resistance, understand how treatment-resistant breast cancer cells survive long-term and intervene before recurrence occurs. That’s why we’re researching now, so no one is left to face an uncertain future. And no one is unknowingly at risk.

The team is drawing on tumour samples collected over time from more than 4,000 people with ER-positive breast cancer who took part in the POETIC clinical trial. Funded by Cancer Research UK, POETIC is managed by the Institute of Cancer Research Clinical Trials and Statistics Unit and has follow-up for more than 10 years after surgery. This creates a uniquely valuable resource for understanding why some breast cancers return. Maggie is the current Translational Lead for the trial.

In this Breast Cancer Now funded study, the researchers are focusing on a rare and valuable set of samples from 30 trial participants whose tumours were collected at several key stages: before treatment, after a short course of hormone therapy, and again when the cancer returned after surgery and long-term treatment. This gives the team a rare chance to see how breast cancer changes over time and why it sometimes comes back.

Using advanced imaging, Maggie and her team will study these tumours in remarkable detail, down to individual cancer cells and the surrounding tissue. This will help them see how the cells are organised, how they interact with their surroundings, and identify small groups of cancer cells that survive treatment.

They will also study the tumours in depth to understand what makes some cancer cells resistant to treatment and able to keep growing, even when they are too few to be detected.

To make sense of these large and complex data, the team will use artificial intelligence to look for patterns that might otherwise be missed. These patterns could reveal the earliest signs that a cancer is becoming resistant to treatment or is more likely to return later.

The team will then compare tumour samples from participants whose breast cancer came back in the same area, came back soon after treatment, or returned years later. They’ll be making use of the incredible biological and clinical data already collected through the POETIC trial. They’ll also compare these findings with samples from people whose breast cancer did not return. And by doing so, they hope to identify important biological differences that could help guide future treatment strategies.

Many people live in fear that their breast cancer will return. Worrying about every ache or pain. Not knowing what the future holds. This research could transform the prediction, prevention and management of breast cancer recurrence, so people can live life beyond treatment with confidence.

By identifying the early warning signs, the project could give doctors the practical tools to estimate someone’s risk of recurrence and tailor follow-up care. And it could pave the way for new treatments that will stop recurrence in its tracks.

Thousands. Around 44,000 people are diagnosed with ER-positive breast cancer each year in the UK. This research could help every single one of them by identifying who’s most at risk of recurrence and improving long-term care. And in time, it could mean fewer people living with the fear of their breast cancer coming back.