8th Baltic Congress of Radiology

Breast Radiology

oral abstract

Hereditary cancer syndromes increase the chances of developing oncological disease. The majority of breast cancers are sporadic due to external (environmental) factors, but up to 10–15% of breast cancers may develope due to inherited gene mutations - BRCA1 or BRCA2 (1) . Female carriers of BRCA1 or BRCA2 gene mutations have very high lifetime risks for breast and ovarian cancers. BRCA2 positive cases are characterized by a much more aggressive form of the disease and a poorer response to treatment (2). Radiomics have been gaining a lot of attention when analyzing the texture data of radiological images, invisible to the human eye. Some imaging biomarkers have been correlated with genetic mutations, such as BRCA1 or BRCA2. There have been a number of studies that describe the imaging characteristics of BRCA1 or BRCA2- associated breast cancer. Radiogenomics combines a large volume of quantitative data extracted from medical images with individual genomic phenotypes and constructs a prediction model through deep learning to stratify patients, guide therapeutic strategies, and evaluate clinical outcomes (3,4).

Our purpose is to find imaging biomarkers by using structural analysis (radiomics) software called Olea that would help us to diagnose BRCA mutated breast cancer when a lesion is smaller than 1 cm diameter and in that way we can increase a diagnostic accuracy and a survival rate.

MRI detected breast cancer lesions for BRCA mutation carriers may have specific phenotype features: they are smaller and tends to develop in the posterior portion of the breast, hovewer time intensity curve on dynamic contrast-enhanced MRI and peritumoral edema may not significantly differ according BRCA mutation status (5). There is still a lack of research showing how image texture data are related to BRCA1, BRCA2 or other genes mutations. If significant correlations are found, potential carriers of gene mutations can be identified from image studies and genetic studies can be targeted at first-degree relatives

This study is stil undregoing - by the time of BCR 2022 we'll have at least 60 patients in our study and first significant imaging biomarkers who could help us to diagnose BRCA mutated breast cancer when a lesion is smaller than 1 cm diameter.

It is desirable to detect breast cancer with a tumor diameter of less than 1 cm that does not require postoperative chemotherapy by using breast MRI (6). Because of that, our purpose is to find imaging biomarkers by using structural analysis (radiomics) software called Olea that would help us to diagnose BRCA mutated breast cancer when a lesion is smaller than 1 cm diameter and in that way we can increase a diagnostic accuracy and a survival rate.

1266