Share this post

Dr. Shaveta Vinayak , a researcher and clinician, works side – by – side with the BioDiscovery Professional Services team to interpret genomic data from triple – negative breast cancer (TNBC)

Triple – negative breast cancers (TNBC) account for 15 – 20% of breast cancers and are defined by lack of estrogen or progesterone receptor expression and by lack of an amplified HER2 /neu gene. They tend to have an aggressive clinical course and unlike for hormone receptor – positive or HER2/neu – positive breast cancers, there is no specific systemic treatment regimen that target s the TNBC subtype. Dr. Shaveta Vinayak, a Fellow in Oncology at Stanford University, working in Dr. James Ford’s lab oratory has undertaken a study to identify molecular signatures that would be predictive of treatment response to a specific regimen in TNBC . She used Affymetrix ® OncoScan ™ (MIP) arrays on DNA extracted from breast tumors, to detect DNA copy number changes, loss of heterozygosity (LOH), and somatic mutations. Using RNA extracted from the same samples, s he ran the Affymetrix U133A arrays to measure gene expression values .

The Goal and the Challenges

The goal of the project is to identify genomic predictors of response, using gene expression and copy number data from pre-treatment samples. Dr. Vinayak is trying to study the molecular changes occurring in the tumors to find a relationship between the given treatment and pathological outcome . In order to do this successfully and to gain maximum insight from the data , it is important to understand disease biology, clinical variables , patient cohort and therapy mechanism of action, all of which Dr. Vinayak brings to the table. Traditional approaches of data analysis have been predictably unsuccessful. For instance, a group of statisticians – cum – programmers, although skilled at building standard analysis dat a pipelines , usually do not possess the holistic understanding of disease and patient cohort that is needed to elucidate secondary insights, or for direct data exploration in order to uncover new relationships.