Advancements in Breast Cancer Treatment at Lund University

Development of Prediction Models

Researchers at Lund University in Sweden have created innovative prediction models aimed at reducing the necessity for lymph node surgeries in breast cancer patients. These models were developed using machine learning techniques, utilizing data from over 3,000 breast cancer patients.

Understanding Breast Cancer Statistics

Breast cancer ranks as one of the most prevalent cancers among women, with estimates indicating that one in every eight women in Western countries will be diagnosed with the disease. The involvement of axillary lymph nodes is a critical factor that physicians consider when forecasting the disease’s progression.

Limitations of Current Surgical Practices

Currently, nearly all breast cancer patients undergo surgery to remove and examine lymph nodes, despite research indicating that approximately 70% of these patients have healthy lymph nodes.

A Non-Invasive Approach

The Lund University research team aimed to determine whether a non-invasive machine-learning prediction model could eliminate the need for lymph node surgery. They accomplished this by analyzing gene expression profiles from over 3,000 breast cancer patients, alongside other relevant data such as tumor size and patient age.

Machine Learning Models and Data Analysis

The research involved employing a machine-learning algorithm that processed data from the patients. A total of seven different models were developed, categorizing the data into three groups: Clinical (based on tumor size and age), GEX (gene profile data), and Mixed (a combination of both). The researchers ultimately selected the most effective training model for presenting their findings.

Results and Accuracy of the Prediction Models

The prediction model that integrated both the genetic profile of tumors and tumor characteristics (Mixed data set) demonstrated the highest accuracy in identifying patients with healthy lymph nodes.

Future Implications and Research

According to the researchers, “The translational approach (of using prediction models) holds promise for the development of classifiers to reduce rates of surgery for patients at low risk of nodal involvement.” Lisa Rydén, a professor of surgery with a focus on breast cancer at Lund University, noted, “The results indicate that we may be a step closer to more personalized surgical treatment by using the prediction models as a decision support tool. To confirm the reliability and precision of these models for clinical application, further studies involving additional patient data are necessary.”

References

– Prediction of lymph node metastasis in breast cancer by gene expression and 3 clinicopathological models: Development and validation within a population-based cohort
– EurekAlert! – Fewer lymph node operations for breast cancer patients with new prediction models, Lund University
– [EurekAlert Article](https://www.eurekalert.org/pub_releases/2019-09/lu-fln092019.php)