Breast Cancer Treatment: AI Tool Predicts Side Effect Risks

“It is an explainable AI tool, which means that it shows the reasoning behind its decision-making. This makes it easier for doctors to make decisions and provide data-backed explanations to their patients,” stated Dr. Rattay, an associate professor and consultant breast surgeon at the University of Leicester’s Leicester Cancer Research Centre (UK).

While a few of the possible risk variables associated with side effects have been identified previously, the PRE-ACT project (Prediction of Radiotherapy side Effects using explainable AI for patient Communication and Treatment modification) aims to provide physicians and patients with comprehensible explanations along with more precise forecasts for each patient.

“Thankfully, long-term survival rates from breast cancer continue to increase, but for some patients, this means having to live with the side effects of their treatment. These include skin changes, scarring, lymphoedema, which is a painful swelling of the arm, and even heart damage from radiation treatment. That’s why we are developing an AI tool to inform doctors and patients about the risk of chronic arm swelling after surgery and radiotherapy for breast cancer. We hope this will assist doctors and patients in choosing options for radiation treatment and reduce side effects for all patients,” said Dr. Rattay.

The researchers from six European nations trained several machine learning algorithms to anticipate arm swelling up to three years following surgery and radiation using data from three European and French datasets (REQUITE, Hypo-G, and CANTO) on 6,361 patients with breast cancer.

Dr. Guido Bologna, co-investigator on the project and associate professor at the University of Applied Sciences and Arts of Western Switzerland in Geneva, gave a speech before EBCC14. He described, “The final, best-performing model makes predictions using 32 different patient and treatment features, including whether or not patients had chemotherapy, whether sentinel lymph node biopsy under the armpit was carried out, and the type of radiotherapy given.”

In the three datasets, 6% of patients had considerable lymphoedema. In 81.6% of cases, the AI tool accurately predicted lymphoedema, while in 72.9% of cases, it properly identified people who would not acquire it. The model’s total prediction accuracy was 73.4%.

Dr. Rattay said, “Patients at higher risk of arm swelling could be offered additional supportive measures, such as wearing an arm compression sleeve during treatment, which has been shown to reduce arm swelling in the long term. Clinicians may also use this information to discuss options for lymph node irradiation in patients, where its benefit may be fairly borderline. We will test the effect of the prediction model on clinician and patient behavior and use of the prophylactic arm sleeve in the proposed clinical trial.”

The researchers will integrate the present AI model into software that may give physicians and patients assessments and forecasts. Later this year, when the PRE-ACT-01 clinical trial begins, this will be put to the test. Additionally, the program is being developed further to forecast other negative effects, such as harm to the skin and heart. Although they will not be utilized to generate predictions in the PRE-ACT study, the researchers will gather information on genetic markers and imaging data as part of the experiment to increase the precision of the AI tools.

“We hope to recruit approximately 780 patients by early 2026, with a follow-up period of two years,” said Dr. Rattay.

Source link