Abstract:
Cancer immunotherapy has presented itself as a promising approach in the treatment of cancers. Actively incorporating the immune system in cancer treatment has proven to be successful in many different therapeutic approaches, ranging from CAR-T therapy or adoptive cellular transfer to peptide or RNA-based vaccines or biologics.
One particularly attractive target for modifications in adoptive cellular transfer or adoptive cell therapy is the T cell receptor. This molecule expressed by T cells can bind to major histocompatibility complex molecules on target cells where they present peptide ligands derived from extra- or intracellular proteins by degradation. Successful recognition can lead to activation of the T cell and an immune response. Thus, introducing TCRs with optimal binding affinity for activation that are specific for a particular peptide-MHC known to be expressed on cancer cells has emerged as an attractive strategy.
In anti-cancer biologics, TCR variable domains have also been used in place of antibodies in bispecific constructs that combine a tumor targeting moiety with a T cell engager like an anti-CD3 antibody. Here, the goal is to specifically bind to tumor cells and broadly activate T cells within the tumor microenvironment. In both approaches, reaching the optimal affinity may require increasing
candidate TCRs affinity through maturation. Since the TCR repertoire seems to have evolutionally evolved to cover the very large number of possible peptide-MHC complexes by cross-reactivity of individual TCRs, these maturations may entail unwanted affinity changes towards peptide-MHCs that are not the intended target and express on healthy tissues as well for example, severely compromising the clinical safety of these approaches.
To counteract these tendencies during development of these approaches, strategies were evaluated to determine the binding motif and potential cross-reactivity of either cellularly expressed or soluble TCRs with affinities which can be encountered in the natural TCR repertoire, as well as highly affinity maturated TCRs in soluble bispecific constructs. The different screening strategies were evaluated both with well-established cell biological approaches to determine T cell activation and more recent biochemical methods like bio-layer interferometry to directly measure binding kinetics between TCRs and peptide-HLA molecules. To facilitate screenings with the latter platform, a newly developed empty
HLA-A*02:01 mutant stabilized by a disulfide bridge is introduced that greatly facilitates the generation of large peptide-HLA libraries for screenings. The different combinations of screening strategies and measurement method were compared within each TCR group with respect to sensitivity and usefulness. In addition, an approach is demonstrated to convert a determined binding motif into a search motif and used to identify off-target binders for high-affinity bispecific TCR constructs.