Deep learning reveals predictive sequence concepts within immune repertoires to immunotherapy

T-cell receptor sequencing has been used to characterize the immune response to cancer. However, most analyses have been restricted to quantitative measures such as clonality that do not leverage the CDR3 sequence. We use DeepTCR, a framework of deep learning algorithms, to reveal sequence concepts that are predictive of response to immunotherapy. We demonstrate that DeepTCR can predict response and utilize the model to infer the antigenic specificities of the predictive signature and their unique dynamics during therapy. The predictive signature of non-response is associated with high frequencies of TCRs predicted to recognize tumor-specific antigens and these tumor-specific TCRs undergo a higher degree of dynamic changes on therapy in non-responders vs. responders. These results are consistent with a biological model where the hallmark of non-responders is an accumulation of tumor-specific T-cells that undergo turnover on therapy, possibly because of the dysfunctional state of these T-cells in non-responders