Vitacrystallography: Structural Biomarkers of Breast Cancer Obtained by X-ray Scattering
Discusses how XRD techniques are advancing structural biomarkers study.
With breast cancer being one of the most widespread causes of death for women, there is an unmet need for its early detection. For this purpose, we propose a non-invasive approach based on X-ray scattering. We measured samples from 107 unique patients provided by the Breast Cancer Now Tissue Biobank, with the total dataset containing 2958 entries. Two different sample-to-detector distances, 2 and 16 cm, were used to access various structural biomarkers at distinct ranges of momentum transfer values. The biomarkers related to lipid metabolism are consistent with those of previous studies. Machine learning analysis based on the Random Forest Classifier demonstrates excellent performance metrics for cancer/non-cancer binary decisions. The best sensitivity and specificity values are 80% and 92%, respectively, for the sample-to-detector distance of 2 cm and 86% and 83% for the sample-to-detector distance of 16 cm.
Overview
Breast cancer is the most common cancer among women, and current screening methods, such as mammography and biopsy, are costly and uncomfortable. This study proposes a non-invasive alternative using X-ray scattering to analyze extracellular matrix (ECM) components as structural biomarkers for cancer detection.
Key Findings:
Data & Methods: Nearly 3000 X-ray scattering measurements were collected from 107 patients (60 cancer, 47 non-cancer) from the Breast Cancer Now Tissue Biobank. Measurements were taken at two sample-to-detector distances (2 cm and 16 cm) to analyze different structural biomarkers.
Structural Biomarkers:
Changes in ECM structure, particularly lipid metabolism alterations, correlate with cancer presence.
Specific scattering peaks at q = 1.5 nm⁻¹, 13.9 nm⁻¹, and 20.2 nm⁻¹ indicate differences between cancerous and non-cancerous tissues.
Machine Learning Analysis:
Random Forest Classifier achieved high accuracy in distinguishing cancer from non-cancer samples.
Sensitivity and specificity values were:
80% / 92% (2 cm distance)
86% / 83% (16 cm distance)
Significance:
The method is non-invasive, rapid, and cost-effective, providing a patient-friendly alternative to traditional screening.
It can be extended to detect other cancers and diseases.
This study demonstrates that X-ray scattering combined with machine learning offers a promising new approach to early breast cancer detection, reducing the need for invasive biopsies and improving screening accessibility.