Transition-Based Analysis Distinguishes Stage-Associated and Subtype-Specific lncRNA Signatures in Breast Cancer

Molecular profiling of cancer samples has traditionally emphasized protein-coding genes and microRNAs, whereas long noncoding RNAs (lncRNAs) remain challenging to characterize because of their generally low expression levels and poorly understood biological functions. Thousands of lncRNAs are expressed in cancer, but their contributions to tumor heterogeneity and clinical characterization have largely been overlooked. Here, we systematically analyzed lncRNA expression in the breast cancer cohort from TCGA. We developed a transition-based framework to distinguish lncRNAs consistently expressed across samples from those exhibiting statistically significant expression differences between consecutive clinical stages. Using a stringent (|z|>3) threshold, we identified four sets of transition-associated lncRNAs (Tr-lncRNAs) with directional expression changes across clinical stages. An orthogonal analysis identified subtype-associated lncRNAs (St-lncRNAs) based on maximal expression variance, distinguishing Luminal A, Luminal B, HER2-positive, and basal-like tumors. The 30 most informative lncRNAs revealed that subtype-associated patterns remained broadly consistent across clinical stages. Genomic-context analyses revealed frequent antisense relationships between lncRNAs and their counterpart protein-coding genes, raising the possibility of regulatory interactions. In addition, statistical integration of Ribo-seq evidence with translation-initiation criteria identified several lncRNA encoding micropeptides, which could expand the repertoire of candidate cancer antigens. Finally, ST8SIA6-AS1, MAPT-IT1, and LINC01087 were associated with patient survival across independent breast cancer studies. Together, these findings establish a framework for distinguishing stage-associated expression changes from subtype-associated patterns and prioritizing understudied transcripts as candidate indicators of breast cancer characteristics and progression.