Mammography is the benchmark for breast disease detection and diagnosis. However, it can miss 10-15% of early stage breast cancers, and it is unable to distinguish between benign and malignant lesions with certainty. Availability of an alternate method e.g. biomarker based, which allows early detection/precise distinction between benign disease and breast cancer will reduce mortality associated with breast cancer. We previously used DNA microarrays to screen >300 blood RNA specimens, and identified 24 RNA signatures that allowed precise identification of selected patients with breast disease. Based on the prospect of commercializing the identified biomarkers as a blood-based test for breast disease detection, we propose to validate the candidate biomarkers using a platform that is more amenable to translation into the clinic. Objective: To develop a biomarker based blood test for breast disease detection and classification.
The specific aims of the project are: 1) design and construct QuantiGene probes for multiplexed blood RNA analysis, 2) test the probes with blood RNA samples, and develop a prototype classification model for identification of individuals with benign breast disease (BD) and breast cancer (BC) and 3) validate the performance of the prototype model in identifying donor categories. Methods: QuantiGene RNA probes targeting 24 biomarkers of interest and 3 housekeeping genes will be designed and constructed. Blood will be collected with PAXGene RNA stabilization tubes, from female donors (>21y) classified as normal (n=30), with BD (n=30) and BC (n=30), and screened with the QuantiGene probes. Data will be normalized with the best housekeeping gene and analyzed. Then, a prototype classification model will be developed and validated using new samples collected from additional normal (n=10), BD (n=10) and BC (n=10) donors. Data analysis: Descriptive, graphical and non-parametric statistics will be performed to determine the pattern and significance of expression of the biomarkers. The prototype classification model will be evaluated by calculating performance evaluation measures (sensitivity, SN; specificity, SP; and accuracy), to distinguish between high-performance classifiers and the null expectation of no significant classifier. SN and SP values will be reported across a range of decision rules to generate the receiver operator characteristics (ROC). We will also assess; a) detection sensitivity, b) assay range, c) precision, d) relative accuracy and fold-change correlations among the variables. Expected outcome: If successful, the proposed blood test will augment mammography, produce faster results, reduce the time a patient has to wait before getting a conclusive diagnosis and allow screening to be performed remotely. In the long run, such a test will reduce patient mortality/morbidity and overall healthcare cost.
Although mammography is still the best tool for screening and detection of breast disease, it can miss 10-15% of early stage cancers and it is unable to differentiate between benign and malignant lesions with certainty. We propose to develop a blood based test that will complement/augment mammography, produce faster and more objective results, reduce the time a patient has to wait before getting a conclusive diagnosis and allow many more women including those in rural and medically underserved areas to routinely test for breast cancer. The health care benefit of this test will include reduced morbidity/mortality and overall healthcare cost/burden due to breast cancer. ? ? ?
