AI Analysis of Immune Cells Predicts Breast Cancer Prognosis

Image: The study findings suggest that tumor-infiltrating lymphocytes are a robust biomarker of breast cancer (Photo courtesy of Shutterstock)

Tumor-infiltrating lymphocytes (TILs) are immune cells crucial in combating cancer. Their presence in a tumor indicates that the immune system is attempting to attack and eliminate cancer cells. TILs can be important indicators in predicting how patients with triple-negative breast cancer will respond to treatment and how the disease might progress. However, assessing these immune cells can yield inconsistent results. Artificial intelligence (AI) has the potential to standardize and automate this process, but proving its effectiveness for healthcare use has been challenging. Now, researchers have explored how different AI models can predict the prognosis of triple-negative breast cancer by analyzing specific immune cells within the tumor. This study, published in eClinicalMedicine, represents a significant step toward incorporating AI into cancer care to enhance patient outcomes.

Researchers at Karolinska Institutet (Stockholm, Sweden) tested ten different AI models to evaluate their ability to analyze tumor-infiltrating lymphocytes in tissue samples from patients with triple-negative breast cancer. The results revealed that the performance of the AI models varied, but eight out of the ten models demonstrated strong prognostic capability, meaning they could predict patient health outcomes with similar accuracy. Even models trained on smaller datasets showed promising results, suggesting that tumor-infiltrating lymphocytes are a reliable biomarker. The study highlights the need for large datasets to compare different AI models and validate their effectiveness before they can be used in clinical practice. Although the findings are promising, further validation is required.

“Our research highlights the importance of independent studies that mimic real clinical practice,” said Balazs Acs, researcher at the Department of Oncology-Pathology, Karolinska Institutet. “Only through such testing can we ensure that AI tools are reliable and effective for clinical use.”