Generalizable features for the diagnosis of infectious disease, autoimmunity and cancer from adaptive immune receptor repertoires
Liquid biopsies based on peripheral blood offer a minimally invasive alternative to solid tissue biopsies for the detection of diseases, primarily cancers. However, such tests currently consider only the serum component of blood, overlooking a potentially rich source of biomarkers: adaptive immune receptors (AIRs) expressed on circulating B and T cells. Machine learning-based classifiers trained on AIRs have been reported to accurately identify not only cancers, but also autoimmune and infectious diseases as well. However, when using the conventional "clonotype cluster" representation of AIRs, donors within a disease or healthy cohort exhibit vastly different features, limiting the generalizability of these classifiers. This paper addresses the challenge of classifying specific diseases from circulating B or T cells by developing a novel representation of AIRs based on similarity networks constructed from their antigen-binding regions (paratopes). Features based on this novel representation, paratope cluster occupancies (PCOs), significantly improved disease classification performance for infectious disease, autoimmunity and cancer. Under identical methodological conditions, classifiers trained on PCOs achieved a mean ROC AUC of 0.893 when applied to new donors, compared to clonotype cluster-based classifiers (0.714) or the best-performing published classifier (0.777). Surprisingly, for cancer patients, we observed that some of the AIRs that were important for classification were significantly more abundant in healthy controls than in individuals with disease. These "healthy-biased" AIRs were predicted to target known cancer-associated antigens at dramatically higher rates than healthy AIRs as a whole (Z scores > 75), suggesting the existence of an overlooked reservoir of cancer-targeting immune cells that are diagnostic and identifiable from a routine blood test. Consequently, PCOs not only enhance classification of a broad range of diseases but also identify immune cells with therapeutic potential.