为肿瘤学和自身免疫领域下一代T细胞衔接器解锁pMHC靶点空间

Peptide-MHC (pMHC) complexes represent a compelling and emerging class of targets for oncology and autoimmune diseases. Their presentation on the cell surface enables the targeting of peptides from non-membrane bound proteins which could be highly specific to pathogenic cells. The lack of disease cell-surface specific targets is currently limiting the potential of bispecific T cell engagers (TCEs) for solid tumours and autoimmune diseases. Despite their promise, pMHC targeting TCEs face two significant challenges. Firstly, a pMHC targeting entity needs to bind to the pMHC of interest with high affinity and specificity to ensure drug efficacy and safety. Secondly, since cell surface density of pMHC is generally low (typically ~10 -1,000 copies per cell), the bispecific format and anti-CD3 moiety need to work concordantly to mediate efficient killing with minimal cytokine release. We have developed unique technology platforms to overcome these major challenges for delivering first-in-class pMHC directed TCEs for solid tumours and autoimmunity. To harness the natural ability of TCRs to bind pMHC with high specificity and overcome the constraints for TCR discovery against self antigens in humans due to negative selection in T cell development, we generated a humanised transgenic mouse, OptiMus ® mouse, by introducing human TCRalpha, TCRbeta, CD8alpha, CD8beta, HLA- A02, and beta2M genes in situ by precision genome engineering. Over a decade, 1.6 million bases of human DNA were introduced into the mouse to fully humanise the entire TCRalphabeta repertoire, CD8 co-receptor and MHC class I. The immune compartment of this mouse strain is phenotypically normal, and importantly, it allows TCR discovery against human antigens without the self-antigen constraint. The diverse TCR repertoire produced by OptiMus ® mouse enables the discovery of hundreds of antigen-reactive TCRs as a starting point for drug discovery. Drug efficacy at low pMHC target density requires a TCE format that is optimised for immune synapse engagement. The anti-CD3 moiety must be fine-tuned in terms of epitope selection and affinity modulation to maximize the therapeutic window. We have explored over 500 formats to identify a novel design in which a TCR, anti-CD3 and Fc are accommodated to allow efficient immune synapse formation and target cell killing, and which has an antibody-like developability and pharmacokinetics profile. Furthermore, we discovered de novo anti-CD3 moieties that are optimised with our format for desired properties: durable serial killing, high in vivo potency, low cytokine release and no T cell mediated drug disposition. By maximising the power of our technology platforms, from human TCR discovery to TCE engineering, we are advancing a pipeline of first-in-class TCEs for oncology and autoimmune indications. Our approach unlocks previously inaccessible targets and redefines the potential of TCEs in precision immunotherapy.