Abstract Introduction: Breast cancer is the most common tumor among women and is responsible for a high mortality rate worldwide. Gene expression profiling is vastly used in cancer research, and in breast cancer in particular it is used for diagnostic and prognostic characterization. The most frequent subtype, invasive ductal carcinoma (IDC), is considered a heterogeneous group of tumors with different alterations at the molecular level. Perou et al. in 2000 initially described five groups (Luminal A, Luminal B, HER-enriched, normal and basal-like) which have distinct gene expression patterns with different clinical implications. The use of genomics tools has contributed greatly to the understanding of the biology of breast tumors and the identification of new biomarkers. In recent years, long transcripts named long non-coding RNAs (lncRNAs) have been identified that do not encode proteins and are derived from intronic and intergenic sequences. Analysis of lncRNAs has shown differences in transcript expression in specific tissues and some tumors, showing that this area is promising for a better comprehending of cancer. In this context, the present study aims to evaluate the expression of coding and non-coding transcripts, in epithelial tumor cells of invasive ductal carcinoma. Experimental procedure: Determination of gene expression was performed by hybridization to a customized microarray slide with 244,000 probes, of which 44,000 are directed against protein-coding transcripts and 200,000 against lncRNAs. RNA samples derived from laser microdissected epithelial cells (103 IDCs of all subtypes, characterized by immunohistochemical markers) was twice amplified, labeled and hydribized to the microarray slide. The microarrays were analyzed using GeneSpring GX 12.1 software. Probes close to the background, saturated, or with uncertain reading were eliminated, resulting in 23.215 probes for further analysis. Clustering of the samples was performed according to Perou′s intrinsic groups following the PAM50 UNC microarray database guidelines. Coding and non-coding probes were analyzed separately, searching for differentially expressed transcripts in hormone receptor positive and negative subgroups, using t-test (unpaired, with FDR correction) considering fold change ≥|2| and p≥0.05. We compared the lists searching for genes and non-coding anti-sense transcripts within the same locus and performed Pearson correlation considering sense and anti-sense (SAS) pairs with coefficient ≥|0.5| and p ≥ 0.05. Partial Results: Robustness of microarray data was shown by our ability to replicate clustering of molecular profiles using only PAM50 genes. We identified SAS pairs with inverted expression pattern within the same locus, associated with hormone receptor status in breast tumors. We found 2386 coding probes and 358 non-coding probes differentially expressed, of which 56 were from the same locus and 15 had inverted expression (Pearson correlation ≥|0.5|). We are performing qRT-PCR validation of two pairs of candidate genes (LCOR and RFWD2) and their non-coding transcripts on the opposite strand of the same locus, which were inversely expressed with greater correlation coefficient. Perspective: The influence of non-coding antisense transcripts on the sense transcript′s coding is not yet fully understood, but it is suggestive that there is a role in regulating the expression of the gene encoding protein. Citation Format: Mabel Fernandez Pinilla, Alex de calvalho Fiorini, Dirce Maria Carraro. Evaluation of the expression of coding and noncoding transcripts in epithelial cells of invasive ductal breast cancer of patients with and without hormone receptor. [abstract]. In: Proceedings of the AACR Special Conference on Advances in Breast Cancer Research: Genetics, Biology, and Clinical Applications; Oct 3-6, 2013; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Res 2013;11(10 Suppl):Abstract nr B018.