Analysis of the methylation profile of 49 cancer-related regions in human breast cancer.

MARZESE, DIEGO; GAGO, FRANCISCO; OROZCO, JAVIER; TELLO, OLGA; ROQUÉ, MARÍA; VARGAS ROIG, LAURA

Chicago, Illinois

Congreso; ASCO Annual Meeting; 2010

Analysis of the methylation profile of 49 cancer related regions in human breast cancer Background: Aberrant methylation of CpG islands is a common mechanism for suppressing gene expression in breast cancer. The methylation profile of primary tumor could serve to detect circulating tumor-specific DNA (ctDNA) in peripheral blood of cancer patients. Methods: The methylation status of 49 cancer-related regions was studied using Methyl Specific-Multiplex Ligation Probe Amplification (MS-MLPA) assay in invasive breast tumors (n=68), axillary lymph nodes (n= 21), and normal breast tissue (n=7). A double round of MS-MLPA (drMS-MLPA) was set up to identify ctDNA in pre-surgical peripheral blood samples. Blood samples from healthy individuals were used as control. Results: We observed methylation of at least 5 of the 49 studied gene regions in all the breast cancer samples. The profiles were specific for each tumor and we found significant differences in the methylation of several genes related to the molecular classification, the tumor grade, and the involvement of lymph nodes. For example, the methylation of p73 appeared in 37% of Luminal A tumors vs. 100% of HER2 tumors (p=0.0006), and in 73% of Triple Negative tumors (p=0.045). The methylation of CDH13 was present in 21% of Luminal A vs. 70% of HER2 (p=0.006). The methylation of p73 was observed in 17%, 38%, and 87% in tumor grade 1, 2 and 3, respectively (p=0.0025, grade 1 vs. 3. and p=0.001, grade 2 vs. 3). The methylation of RARβ appeared in 48% of primary tumors with positive lymph nodes whereas 14% of tumors with negative nodes (p=0.03). Using drMS-MLPA approach we were able to detect ctDNA in pre-surgical blood samples based on the epigenetic signature of primary tumors and we also could establish its sensitivity. Conclusions: The methylation of some genes correlates with pathological characteristics of breast cancer. This methylation is related with clinical outcome indicators like tumor grade or lymph nodes metastasis. The drMS-MLPA approach could be useful for disease detection and the follow up of breast cancer patients.