Doctors and scientists have developed an artificial intelligence tool that can accurately predict how likely tumors are to grow back in cancer patients after treatment.
The breakthrough, described as “exciting” by clinical oncologists, could revolutionize patient monitoring. Although advances in treatment in recent years have increased the chances of survival, the risk remains that the disease will return.
Monitoring patients after treatment is vital to ensure that any recurrence of cancer is treated urgently. At present, however, doctors tend to rely on traditional methods, including those focused on the initial amount and spread of cancer, to predict how the patient may cope in the future.
Now the world’s first study by the Royal Marsden NHS Foundation Trust, the Institute for Cancer Research, London and Imperial College London has identified a model using machine learning – AI – that can predict the risk of cancer recurrence and do better. of existing methods.
“This is an important step forward so that we can use AI to find out which patients are most at risk of cancer recurrence and to detect this recurrence earlier so that re-treatment is more effective.” said Dr. Richard Lee, a consultant physician in respiratory medicine and early diagnosis at the Royal Marsden NHS Foundation Trust.
Lee, lead researcher of the OCTAPUS-AI study, told the Guardian that it could be vital not only to improve outcomes for cancer patients but also to alleviate their fears, with relapse being a “key source of anxiety” for many. . “We hope to push the boundaries to improve the care of cancer patients, help them live longer and reduce the impact of the disease on their lives.
The AI tool can lead to earlier detection of recurrence in patients considered high risk, ensuring that they receive treatment more urgently, but it can also lead to less unnecessary follow-up scans and hospital visits. for those considered low risk.
“Reducing the number of scans required in this setting can be beneficial, but it can also reduce radiation exposure, hospital visits and make more efficient use of NHS valuable resources,” Lee said.
In a retrospective study, doctors, scientists and researchers developed a machine learning model to determine if it could accurately identify patients with non-small cell lung cancer (NSCLC) at risk of relapse after radiation therapy. Machine learning is a form of AI that allows software to automatically predict results.
Lung cancer is the world’s leading cause of cancer deaths, accounting for just over one-fifth (21%) of cancer deaths in the United Kingdom. NSCLC accounts for almost five-sixths (85%) of lung cancer cases, and when caught early, the disease is often treatable. However, more than a third (36%) of patients with NSCLC experience a relapse in the United Kingdom.
The researchers used clinical data from 657 NSCLC patients treated at five hospitals in the UK to fuel their model – and added data on various prognostic factors to better predict the patient’s chances of relapse.
These include the patient’s age, gender, BMI, smoking status, intensity of radiation therapy, and tumor characteristics. The researchers then used the AI model to categorize patients at low and high risk of relapse, how long they could survive before relapse, and overall survival two years after treatment.
It has been found that the tool is more accurate in predicting the results of traditional methods. The results of the study, supported by the cancer charity Royal Marsden and the National Institutes of Health Research, were published in The Lancet’s eBioMedicine.
“There is currently no established framework for monitoring patients with non-small cell lung cancer after radiation therapy in the UK,” said study leader Dr Sumeet Hindocha, a clinical oncologist, registrar at Royal Marsden and Imperial College London. “This means that there are variations in the type and frequency of follow-up that patients receive… Using AI with health data may be the answer.
“Because this type of data can be easily accessed, this methodology can be replicated in different health systems.
The study is an “exciting first step” towards the introduction of a national and international tool to guide the monitoring of cancer patients after treatment, Hindocha added.
Add Comment