AI Predicts Triple-Negative Breast Cancer Prognosis

Researchers at Sweden’s Karolinska Institutet explored how well different AI models can predict the prognosis of triple-negative breast cancer by evaluating specific immune cells within the tumor. The study, published in the journal eClinicalMedicine, is a significant step toward employing AI in cancer therapy to improve patient health.

Tumour-infiltrating lymphocytes are immune cells that help fight cancer. When they are present in a tumor, it indicates that the immune system is attempting to attack and eliminate cancer cells.

Compared ten AI models for Breast Cancer

The researchers evaluated 10 different AI models for their capacity to detect tumor-infiltrating lymphocytes in triple-negative breast cancer tissue samples.

The results revealed that the AI models differed in their analytical capabilities. Despite these discrepancies, eight of the ten models shown high prognostic ability, which means they could predict patients’ future health in a similar manner.

Even models trained on fewer samples showed good prognostic ability, suggesting that tumor-infiltrating lymphocytes are a robust biomarker.”

Balazs Acs, researcher at the Department of Oncology-Pathology, Karolinska Institutet

Independent studies needed

The study concludes that huge datasets are required to compare different AI technologies and guarantee that they perform properly before they can be employed in healthcare. While the results are promising, further validation is required.

“Our research highlights the importance of independent studies that mimic real clinical practice,” says Balazs Acs. “Only through such testing can we ensure that AI tools are reliable and effective for clinical use.”

For more information: Vidal, J. M., et al. (2024). The analytical and clinical validity of AI algorithms to score TILs in TNBC: can we use different machine learning models interchangeably? eClinicalMedicine. doi.org/10.1016/j.eclinm.2024.102928.