Abstract Text: High-throughput T-cell receptor sequencing is a rapidly growing technology that provides new data-driven insights into the adaptive immune system and its role in various diseases. While much of this work focuses on identifying antigen-specific clones, there is also interest in more general metrics that can be used to summarize differences between T-cell repertoires. For example, various measures of repertoire diversity have been shown to correlate with clinical outcomes and therapeutic responses in a variety of settings. However, there remains little consensus as to which quantitative metrics are the most robust and salient for use in various investigational settings. In this study, we used the immunoSEQ® Assay (Adaptive Biotechnologies, Seattle, WA) to sequence and compare TCRB repertoires of healthy individuals with a variety of disease states including various solid and hematological malignancies, autoimmune, or neurodegenerative conditions. Repertoire properties were characterized using multiple antigen-agnostic metrics, including repertoire diversity metrics such as Simpson’s clonality, clone distribution slope, and convergence. Each metric was evaluated for its robustness to sampling depth (based on input DNA and the number of sequences detected) and its correlation with the other metrics. We also characterized the variation of these metrics both within and between indications, disease states and tissues. This study establishes 1) the strengths and limitations of the currently available quantitative metrics used to summarize various attributes of immune repertoires; 2) the relationships between these metrics and the overlapping or complementary repertoire properties they quantify; and 3) the observed ranges of these metrics within specific disease settings.
