T. Kawaguchi1, L. Yan2, Q. Qi2, S. Liu2, J. Young1, K. Takabe1 1Roswell Park Cancer Institute,Breast Surgery, Department Of Surgical Oncology,,Buffalo, NY, USA 2Roswell Park Cancer Institute,Department Of Biostatistics & Bioinformatics,Buffalo, NY, USA
Introduction:
MicroRNAs (miRNAs) are noncoding RNAs with 19-25 nucleotides that exert its function by either degradation of coding mRNA or inhibition of mRNA translation. Dysregulations of miRNAs have been reported to play critical roles in carcinogenesis and progression of various types of cancer including breast cancer (BrCa). Some miRNAs, such as miR-31, miR-126, miR-146b, miR-206, miR-335, have been reported as tumor suppressive miRNAs targeting oncogenes. However, clinical relevance of those reports has not yet been validated using common large cohort, which provides sufficient statistical power with proved high quality genetic samples. In this study we took advantage of the high-throughput data from The Cancer Genome Atlas (TCGA) as a validation cohort to evaluate the clinical relevance of well-known nine suppressive miRNAs.
Methods:
All data were obtained from The Cancer Genome Atlas (TCGA). Expression of five suppressive miRNAs in BrCa, miR-31, miR-126, miR-146b, miR-206, miR-335, were retrieved from the GDC data portal and were analyzed using microRNA-Seq dataset. Overall survival was compared using the Cox proportional hazard model between the high and low expression groups determined by each miRNA-specific thresholds.
Results:
Among the 1097 patient breast cancer samples logged in TCGA, 1053 samples were found to contain both microRNA-seq datasets and survival data. High expression levels of miR-31 and miR-146b demonstrated significantly better survival (p = 0.032 and p = 0.025, respectively), while high expression levels of miR-206 tend to show worse prognosis (p = 0.091). The other miRNAs of interest, miR-126 and miR-335 have no significant difference between high and low expression groups. All of the miRNAs examined did not show any significant difference between high and low expression groups in clinicopathological factors such as staging, tumor size (T category), nodal metastasis (N category), distant metastasis (M category), and intrinsic subtype (ER, PR, and HER2 status), except for miR-126 that demonstrated significant association with PR and HER2 status (p = 0.003 and p < 0.001, respectively).
Conclusion:
Utilizing a big data (TCGA) with sufficient statistical power, we found that high expression of miR-31 and miR-146b was significantly associated with better overall survival. This is in agreement with previous reports that demonstrated their tumor suppressive role in BrCa. Conversely, expression of miR-126 and miR-335 did not show any survival impact, and high expression of miR-206 demonstrated a trend to worse prognosis against the previous reports. We conclude that it is essential to validate the survival impact of reported miRNAs using a large publically available data base such as TCGA.