TCR repertoire and transcriptional signatures of circulating tumour‐associated T cells facilitate effective non‐invasive cancer detection

• Published in: Clinical and translational medicine Vol. 12; no. 9; pp. e853 - n/a
• Main Authors: Ji, Fansen, Chen, Lin, Chen, Zhizhuo, Luo, Bin, Wang, Yongwang, Lan, Xun
• Format: Journal Article
• Language: English
• Published: Heidelberg John Wiley & Sons, Inc 01.09.2022; Wiley
• Subjects: Amino acids; Antigens; Biomarkers; Biopsy; Cancer; Cloning; Datasets; Lymphocytes; T cell receptors
• ISSN: 2001-1326; 2001-1326
• DOI: 10.1002/ctm2.853

Traditional serological tumour biomarkers like α-fetoprotein, prostate-specific antigens, carcinoembryonic antigens, CA19–9 and CA125 have been widely investigated in clinic, but the specificity has not reached a satisfactory stage for population level.1–3 Novel technologies utilizing tumour-derived signals from blood non-invasively provide a new tumour diagnostic strategy called liquid biopsy over the past decades.3–5 Several peripheral biomarkers such as cell-free DNA (cfDNA)6–8 especially circulating tumour DNA (ctDNA),9 circulating tumour cells (CTCs),10,11 circulating micro-RNAs,12,13 tumour-derived exosomes14 and cancer cell metabolites15 achieved great progress and showed huge prospects in tumour screening. The feasibility of using tumour-associated T cell response involved in tumour initiation and development, as a supplementary diagnosis choice has not been explored widely.16,17 Until recently, tumour-infiltrated T lymphocytes (TILs) were considered to be beneficial tumour-specific T cells.18 But due to the complex interaction of different immune components mediated by chemokines or cytokines within the tumour microenvironment (TME), the majority of passively expanding TILs cannot recognize tumour-specific antigens (TSAs) and are thus believed to be bystander T cells.19–21 These bystander T cells may dilute tumour-specific signals and make the identification of TSAs-specific T cells challenging.22–24 Programmed cell death protein 1 (PD-1) is suggested to be a biomarker for tumour-specific CD8+ T cells both in TILs and in peripheral blood mononuclear cells (PBMCs),22,25–27 but the efficacy needs to be further validated in practical applications.21,28,29 Tracking the general immunophenotype of T lymphocytes when they encounter antigens and enrichment of tumour-associated T cells over a pool of irrelevant signals during tumour development reflects the overall immune status of patients and offers opportunities for cancer prevention and therapy.30 Next generation sequencing (NGS)-based T cell receptor (TCR) repertoire quantification has provided methods for TSA recognition and now is extensively used in the identification of tumour-reactive T lymphocytes.19,31–34 The past few years have witnessed a series of studies utilizing T or B-cell repertoire to pinpoint disease-associated signatures, and evidences have demonstrated the diagnostic potential of TCR repertoire in autoimmune diseases,35 infectious diseases36,37 and even cancer.38–42 Sustained neoantigen stimulation during tumour cell development impels TCR to shift towards a tumour-specific distribution and to exhibit different amino acids motifs than those in healthy cells.38,43 Under physiological conditions, naïve T cells maturing in the thymus will flow through peripheral blood or lymphatic vessels and migrate through high endothelial venules into secondary lymphoid organs where they encounter potential tumour antigens.44,45 T cell trafficking and circulation theoretically enable these tumour-specific T cells to be detectable both in tumour sites and peripheral blood. Due to different tissue environment, the frequency of each clonotype and the degree of clonality between the two compartments may differ. [...]we specifically named the T cells among PBMCs that share TCRs with TILs as circulating Tumor Associated T cells (cTATs). The same analysis based on single-cell TCR sequencing (TCR-seq) data showed no significant differences (p > .05) in the proportion of shared TCRs at either the clone or sequence level, possibly due to the limited number of cells captured in single cell TCR-seq datasets, and many shared clones might be labelled as not shared (Figure S1A).