Tumor Specific Plasma DNA in Breast Cancer

In 2011, there was an estimated 233,000 cases of invasive breast cancer, and 39,970 deaths from breast cancer in the United States. The vast majority of patients are diagnosed with Stage I-III resectable and potentially curable disease, and for these patients, the most pressing questions are whether adjuvant endocrine or chemotherapy are indicated, and if so, how to determine whether these treatments are working. Adjuvant systemic therapy reduces relative recurrence rates by 30-50%, depending on the age of the patient and tumor characteristics. However, patients with early stage disease often do not bear measurable markers of disease such as an elevated cancer antigen 27-29 (CA27.29) or circulating tumor cells. Patients with early stage breast cancer are typically treated with adjuvant therapy based on historical evidence showing that such therapy prolongs survival in this population.

The contents of dying tumor cells can be detected in the bloodstream, and this may be enhanced by the leaky vasculature of solid tumors. Protein biomarkers of tumor cell death are difficult to detect due to the complex nature of plasma and the lack of technical sensitivity. In contrast, DNA is easier to detect through polymerase chain reaction (PCR) amplification. Indeed, circulating tumor DNA has been detected in plasma from patients with osteosarcoma, breast cancer, and colorectal cancer. Until recently, it was impractical to develop an assay to routinely quantify circulating tumor DNA due to heterogeneity between patients and tumors. Advances in genomic technology now permit sequencing a tumor genome to identify patient-specific genomic aberrations. Major genomic alterations (i.e., insertions, amplifications, deletions, inversions, translocations) can be readily detected using PCR primers which will recognize tumor DNA but not normal DNA.

While this strategy may be generally applicable to diverse types of solid tumors, two issues are apparent in breast cancer. Firstly, the incidence of chromosomal rearrangements varies widely. Whole-genome sequencing of 15 breast tumors revealed a range of 1-231 major genomic alterations (mean= 68), where 2 tumors had 1 alteration, and 9 tumors had > 20 alterations. Single-base point mutations are more common but difficult to reliably detect using PCR. Therefore, the investigators must consider that a small subset of patients may have a limited number of genomic alterations available for this assay. Secondly, intratumoral heterogeneity may mean that some genomic alterations are not present in every tumor cell. Such heterogeneity has been inferred from FISH and immunohistochemistry (IHC) studies for many years, and is now being verified at the genomic level. The investigators must consider that only a subpopulation of tumor cells may be sensitive to cytotoxic therapy, so changes in the levels of circulating tumor DNA may only be reflected with analysis of genomic alterations specific to the sensitive cells.