Pardoll, D. M. — Page 2 | Search bioRxiv

bioRxiv2018

DeepTCR: a deep learning framework for revealing structural concepts within TCR Repertoire

Deep learning algorithms have been utilized to achieve enhanced performance in pattern-recognition tasks, such as in image and vocal recognition 1,2. The ability to learn complex patterns in data has tremendous implications in the genomics and immunology worlds, where sequence motifs become learned features that can be used to predict functionality, guiding our understanding of disease and basic biology 3-6. T-cell receptor (TCR) sequencing assesses the diversity of the adaptive immune system, where complex structural patterns in the TCR can be used to model its antigenic interaction. We present DeepTCR, a broad collection of unsupervised and supervised deep learning methods able to uncover structure in highly complex and large TCR sequencing data by learning a joint representation of a given TCR by its CDR3 sequences, V/D/J gene usage, and HLA background in which the T-cells reside. We demonstrate the utility of deep learning to provide an improved featurization of the TCR across multiple human and murine datasets, including improved classification of antigen-specific TCRs in both unsupervised and supervised learning tasks, understanding immunotherapy-related shaping of repertoire in the murine setting, and predicting response to checkpoint blockade immunotherapy from pre-treatment tumor biopsies in a clinical trial of melanoma. Our results show the flexibility and capacity for deep neural networks to handle the complexity of high-dimensional TCR genomic data for both descriptive and predictive purposes across basic science and clinical research.

bioinformatics↗

bioRxiv2018

A second generation IL-2 receptor-targeted diphtheria fusion toxin exhibits anti-tumor activity and synergy with anti-PD-1 in melanoma

Denileukin diftitox (DAB1-389-IL-2, Ontak(R)) is a diphtheria toxin-based fusion protein that depletes CD25-positive cells including regulatory T cells (Tregs) and was approved for the treatment of persistent or recurrent cutaneous T cell lymphoma. However, the clinical use of denileukin diftitox was limited by vascular leak toxicity and production issues related to drug aggregation and purity. We found that a single amino acid substitution (V6A) in a motif associated with vascular leak induction yields a fully active, second-generation biologic, s-DAB1-386-IL-2(V6A), which elicits 50-fold less HUVEC monolayer permeation and is 3.7-fold less lethal to mice by LD50 analysis than s-DAB1-386-IL-2 Additionally, to overcome aggregation problems, we developed a novel production method for the fusion toxin using Corynebacterium diphtheriae that secretes fully-folded, biologically active, monomeric s-DAB1-386-IL-2 into the culture medium. Using the poorly immunogenic mouse B16F10 melanoma model, we initiated treatment 7 days after tumor challenge and observed that, while both s-DAB1-386-IL-2(V6A) and s-DAB1-386-IL-2 are inhibitors of tumor growth, the capacity to treat with higher doses of s-DAB1-386-IL-2(V6A) could provide a superior activity window. In a sequential dual therapy study in tumors that have progressed for 10 days both s-DAB1-386-IL-2(V6A) and s-DAB1-386-IL-2 given prior to checkpoint inhibition with anti-PD-1 antibodies inhibited tumor growth, while either drug given as monotherapy had less effect. s-DAB1-386-IL-2(V6A), a fully monomeric protein with reduced vascular leak, is a second-generation diphtheria toxin-based fusion protein with promise as a cancer immunotherapeutic both alone and in conjunction with PD-1 blockade.\n\nSignificance StatementRegulatory T cells (Tregs) infiltrate tumors in various cancers and promote an immunosuppressive microenvironment that hinders anti-tumor immunity. Denileukin diftitox, a diphtheria toxin-based fusion protein that depletes Tregs, was approved for the treatment of T cell malignancies, but its clinical use was limited due to the presence of protein aggregates and toxicity associated with vascular leakage. Here we report the production of a second generation IL-2 receptor-targeted, fully-folded, monomeric diphtheria fusion toxin, and a V6A mutant variant which showed reduced vascular leak in vitro and reduced lethality in mice. In a mouse model of melanoma, we found significant decrease in tumor growth associated with reduction in Tregs when the protein was tested as monotherapy or in combination with checkpoint blockade.

cancer biology↗