Abstract Introduction: The current molecular classification of breast cancer has significantly improved treatment targeting and disease prognosis. However, the heterogeneity among subtypes also poses a limitation when proposing new targets that integrate common biological processes, hindering their applicability across the entire spectrum of the disease. In that sense, it is essential to gain a deeper understanding of the similarities and common mechanisms involved in the molecular subtypes of breast cancer. Therefore, in this study, we aimed to identify shared differentially expressed genes (DEGs) and biological processes among the different breast cancer subtypes that could potentially impact the presentation of clinically significant variables in Colombian patients diagnosed at the National Cancer Institute (NCI). Methods: We included 97 women diagnosed with breast cancer at the NCI in Colombia. From each patient, we collected tumor biopsy and adjacent non-tumoral tissue for RNA extraction and sequencing (RNA-seq). A comparative analysis of DEGs and gene set enrichment (GSE) was performed between tumor tissue and its corresponding adjacent non-tumor tissue for each subtype independently to identify common DEGs and enriched biological processes. Shared over- and under-expressed genes in at least two subtypes were tested for their association with clinically significant prognostic variables. Results: Among breast cancer subtypes, ER-/HER2+ and ER-/HER2- tumors showed the highest number of shared DEGs (22 genes), of which 19 were over-expressed and 3 under-expressed. In contrast, GSE analysis revealed a high proportion of shared biological processes (5), molecular functions (3), and cellular component classification (11) between ER+/HER2+ and ER-/HER2+ tumors, corresponding to mechanisms involved in tissue homeostasis and extracellular matrix structural constituents. Furthermore, the association with clinically significant prognostic variables was assessed for 22 over-expressed genes and 9 under-expressed genes shared in at least two subtypes. Of this set of 31 genes, 9 were significantly associated with the degree of tumor differentiation measured by the Bloom-Richardson (BR) score. Interestingly, tumor suppressor genes like ADAMTS18 showed a decline in expression as the BR score increased (fold-change (FC) in BR I: 14 vs BR II: 11; BR III: 5, p-value: 0.028). Conversely, oncogenes like EFNA4 showed the opposite trend, with an increase in expression as the BR score also increased (FC in BR I: 9 vs. BR II: 11 vs. BR III: 17, p-value: 0.019). Another identified shared over-expressed gene, JPT1, which is involved in β-catenin signaling, was found to be associated not only with higher BR scores but also with a higher proliferation index Ki67 (FC in <20%: 16.5 vs. ≥20%: 23; p-value: 0.05). Conclusions: Despite the high degree of molecular heterogeneity among subtypes, it is still possible to identify shared over and under-expressed genes in at least 2 subtypes with significant prognostic implications in breast cancer subtypes. Citation Format: Laura Rey-Vargas, Lina Maria Bejarano, Diego Felipe Ballen-Lozano, Silvia J. Serrano-Gomez. Shared gene expression pattern among breast cancer subtypes and its association with prognostic variables in Colombian patients [abstract]. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr C101.
Read full abstract