PT - JOURNAL ARTICLE AU - Beshnova, Daria AU - Ye, Jianfeng AU - Onabolu, Oreoluwa AU - Moon, Benjamin AU - Zheng, Wenxin AU - Fu, Yang-Xin AU - Brugarolas, James AU - Lea, Jayanthi AU - Li, Bo TI - De novo prediction of cancer-associated T cell receptors for noninvasive cancer detection AID - 10.1126/scitranslmed.aaz3738 DP - 2020 Aug 19 TA - Science Translational Medicine PG - eaaz3738 VI - 12 IP - 557 4099 - http://stm.sciencemag.org/content/12/557/eaaz3738.short 4100 - http://stm.sciencemag.org/content/12/557/eaaz3738.full AB - A key goal in oncology is diagnosing cancer early, when it is more treatable. Despite decades of progress, early diagnosis of asymptomatic patients remains a major challenge. Most methods for this involve detecting cancer cells or their DNA, but Beshnova et al. suggested a different approach, focused on the body’s immune response. The authors reasoned that the presence of cancer may cause alterations in the T cell receptor repertoire, which could then be detected. They designed a deep learning method for distinguishing the T cell repertoires in the blood of patients with and without cancer, which they validated in samples from multiple clinical cohorts.The adaptive immune system recognizes tumor antigens at an early stage to eradicate cancer cells. This process is accompanied by systemic proliferation of the tumor antigen–specific T lymphocytes. While detection of asymptomatic early-stage cancers is challenging due to small tumor size and limited somatic alterations, tracking peripheral T cell repertoire changes may provide an attractive solution to cancer diagnosis. Here, we developed a deep learning method called DeepCAT to enable de novo prediction of cancer-associated T cell receptors (TCRs). We validated DeepCAT using cancer-specific or non-cancer TCRs obtained from multiple major histocompatibility complex I (MHC-I) multimer-sorting experiments and demonstrated its prediction power for TCRs specific to cancer antigens. We blindly applied DeepCAT to distinguish over 250 patients with cancer from over 600 healthy individuals using blood TCR sequences and observed high prediction accuracy, with area under the curve (AUC) ≥ 0.95 for multiple early-stage cancers. This work sets the stage for using the peripheral blood TCR repertoire for noninvasive cancer detection.