Question

Refer to the open-access Radiology article titled “Predicting Breast Cancer by Applying Deep Learning to Linked Health Records and Mammograms” and answer the following questions.

Name TWO (2) clinical features that can be used as potential breast cancer biomarkers.

Answer 1

- Breast cancer is a heterogeneous disease. For the past decades, new technical tools have been developed for biomarkers at the DNA, RNA and protein levels to better understand the biology of breast cancer. Like for example, as researchers proved in the above article, how machine learning and deep learning technologies are used to accurately predict breast cancer using health records and mammograms can be considered as a good example. Novel biomarker development is paramount to deliver personalized cancer therapies.

- Breast cancer is a complex disease in which a number of cellular pathways involving cell growth and proliferation are altered. A number of studies have tried to predict molecular patterns for breast cancer recurrence. This include studies in various breast cancer subtypes wherein breast cancers are characterized by the presence of receptors such as estrogen receptor (ER), progesterone receptor (PR), and HER2/ErbB2 receptor (HER2). Some studies suggested that ER-negative breast cancers are associated with a higher risk of recurrence during the initial 5 years after diagnosis, compared to ER-positive breast cancers. Further research revealed that over-expression of HER2 gene is found almost exclusively in breast cancer patients which is associated with higher recurrence rates and lower response to chemotherapy or hormone therapy. Hence, early detection and better surveillance plays a pivotal role in reducing the recurrence of breast cancer and mortality rates.

- Indication of adjuvant systemic treatment for breast canceris based on standard prognostic characteristics as age, menopausal status, tumor size, lymph node status, histological grade, estrogen receptor (ER) and HER2 status. During the last few years, research has focused on the identification of potential markers (specified DNA sequence, RNA levels or expressed protein) to improve sub-group classification and correlate it with clinical outcome and therapy response.

- Although mammography is widely used to detect incipient breast cancer, this technique does not provide any indication of eventual disease outcome. Two breast cancer susceptibility genes BRCA1 and BRCA2 were discovered after research on families with very strong patterns of breast cancer in 1994 and 1995. These genes are present in all men and women and are considered to be tumor suppressors. Women who have a mutation in the BRCA1 or BRCA2 gene have a significantly elevated risk of developing breast cancer. Genetic testing for mutations in BRCA1 and BRCA2 genes can be used for predicting the risk of individual developing breast cancer. The major concern regarding the use of these two genes in accurate prediction of the eventual onset of breast cancer lies in the variability between individuals.

- Her-2 oncogene expression in tissues or serum is the most commonly used predictive biomarker for breast cancer. Expression of Her-2 is significantly related to positive lymph nodes, poor nuclear grade, lack of steroid receptors and high proliferative activity. Her-2 expression in association with proliferative activity has been shown to identify a subgroup of node-specific breast cancer patients with poorer prognosis. Her-2 is a true predictive marker in that its presence not only provides an indication of disease outcome but also can lead to the selection of appropriate cancer treatment such as the use of Herceptin^® to treat patients with Her-2 positive tumors. The combination of Her-2 and estrogen/progesterone receptor status can provide a greater indication of disease outcome in breast cancer.