Abstract Breast cancer (BC) represents a major clinical hurdle, with metastasis being the most devastating events in patients’ history. Traditional lineage tracing approaches have been widely employed to study the metastatic progression of BC. However, these approaches were limited to the investigation of clonal dynamics and did not provide detailed molecular information underlying the pro-metastatic phenotype. In this work, we exploited a library of degenerated barcodes which were both integrated in the genome and expressed as synthetic transcripts. Coupling this barcoding strategy to scRNA-seq analysis, we tracked the clonal and transcriptional evolution of more than 37,000 MDA-MB-231 (a human triple-negative BC cell line) cells in vivo. Clonal distribution analysis revealed that BC clones display proliferative heterogeneity in primary tumors, with a small fraction of clones outcompeting all the others. Furthermore, primary breast tumors turned out to be composed of two phenotypically distinct groups of clones: highly proliferative, on one side, and migratory/stressed/stem-like, on the other. Then, we showed that the striking majority of metastatic cells derived from a single clone which was under-represented in its primary tumor. Importantly, these pro-metastatic clones upregulated genes and pathways associated with extracellular matrix (ECM) interaction, migration, and type-I interferon (IFN) response. In this scenario, independent pro-metastatic clones heterogeneously upregulated these pathways, thus suggesting a multi-faceted functional phenotype. Despite this variability, we detected a set of genes whose expression was statistically associated with the pro-metastatic phenotype, and we validated the relevance of these molecular features both at the biological and clinical level. First, we observed a reduced metastatic burden upon the in vivo knock-down of genes involved in ECM interaction (e.g., ANGPTL4, KCNQ1OT1, and ITGB4) and type-I IFN response (e.g., LY6E and IFI6). Second, we derived two compact gene signatures showing robust predictive power of BC patient survival, independently of other known clinical factors. In conclusion, our work revealed previously unappreciated cellular mechanisms of BC metastasization, providing new actionable therapeutic targets and new molecular predictors of BC prognosis. Citation Format: Niccolò Roda, Andrea Cossa, Federica Ruscitto, Chiara Priami, Giada Blandano, Alberto Dalmasso, Francesco Bertolini, Pier Giuseppe Pelicci. Single-cell transcriptional lineage tracing reveals complexity and plasticity of breast cancer pro-metastatic phenotypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 3793.