Subtype‐specific micro‐RNA expression signatures in breast cancer progression

• Published in: International journal of cancer Vol. 139; no. 5; pp. 1117 - 1128
• Main Authors: Haakensen, Vilde D., Nygaard, Vegard, Greger, Liliana, Aure, Miriam R., Fromm, Bastian, Bukholm, Ida R.K., Lüders, Torben, Chin, Suet‐Feung, Git, Anna, Caldas, Carlos, Kristensen, Vessela N., Brazma, Alvis, Børresen‐Dale, Anne‐Lise, Hovig, Eivind, Helland, Åslaug
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.09.2016
• Subjects: biomarker; Biomarkers, Tumor; Breast cancer; breast cancer invasion; Breast Neoplasms - genetics; Breast Neoplasms - pathology; Cancer; Carcinoma, Ductal, Breast - genetics; Carcinoma, Ductal, Breast - pathology; Carcinoma, Intraductal, Noninfiltrating - genetics; Carcinoma, Intraductal, Noninfiltrating - pathology; Chromosome Mapping; Cluster Analysis; DCIS; Disease Progression; Female; Gene Expression Profiling; Gene Expression Regulation, Neoplastic; Humans; Mammography; Medical research; MicroRNAs; MicroRNAs - genetics; miRNA; Multigene Family; Neoplasm Invasiveness; Reproducibility of Results; subtype; Transcriptome; miRNA; biomarker; subtype; DCIS; breast cancer invasion
• ISSN: 0020-7136; 1097-0215
• DOI: 10.1002/ijc.30142

Robust markers of invasiveness may help reduce the overtreatment of in situ carcinomas. Breast cancer is a heterogeneous disease and biological mechanisms for carcinogenesis vary between subtypes. Stratification by subtype is therefore necessary to identify relevant and robust signatures of invasive disease. We have identified microRNA (miRNA) alterations during breast cancer progression in two separate datasets and used stratification and external validation to strengthen the findings. We analyzed two separate datasets (METABRIC and AHUS) consisting of a total of 186 normal breast tissue samples, 18 ductal carcinoma in situ (DCIS) and 1,338 invasive breast carcinomas. Validation in a separate dataset and stratification by molecular subtypes based on immunohistochemistry, PAM50 and integrated cluster classifications were performed. We propose subtype‐specific miRNA signatures of invasive carcinoma and a validated signature of DCIS. miRNAs included in the invasive signatures include downregulation of miR‐139‐5p in aggressive subtypes and upregulation of miR‐29c‐5p expression in the luminal subtypes. No miRNAs were differentially expressed in the transition from DCIS to invasive carcinomas on the whole, indicating the need for subtype stratification. A total of 27 miRNAs were included in our proposed DCIS signature. Significant alterations of expression included upregulation of miR‐21‐5p and the miR‐200 family and downregulation of let‐7 family members in DCIS samples. The signatures proposed here can form the basis for studies exploring DCIS samples with increased invasive potential and serum biomarkers for in situ and invasive breast cancer. What's new? Current debate about the overtreatment of noninvasive and indolent breast carcinomas is being addressed through multiple streams of research. Key among them is the identification of diagnostic measures of malignant potential, which could help reduce overtreatment. Here, analyses of data from two different studies on ductal carcinoma in situ (DCIS) and invasive breast carcinoma revealed subtype‐specific microRNA signatures. The miRNA profiles were associated with transitions from normal tissue to DCIS and from DCIS to specific subtypes of invasive carcinoma. The miRNA signatures may be capable of predicting malignant potential in a clinical setting.