Tumor-associated peptide–human leukocyte antigen complexes (pHLAs) represent the largest pool of cell surface–expressed cancer-specific epitopes, making them attractive targets for cancer therapies. Soluble bispecific molecules that incorporate an anti-CD3 effector function are being developed to redirect T cells against these targets using 2 different approaches. The first achieves pHLA recognition via affinity-enhanced versions of natural TCRs (e.g., immune-mobilizing monoclonal T cell receptors against cancer [ImmTAC] molecules), whereas the second harnesses an antibody-based format (TCR-mimic antibodies). For both classes of reagent, target specificity is vital, considering the vast universe of potential pHLA molecules that can be presented on healthy cells. Here, we made use of structural, biochemical, and computational approaches to investigate the molecular rules underpinning the reactivity patterns of pHLA-targeting bispecifics. We demonstrate that affinity-enhanced TCRs engage pHLA using a comparatively broad and balanced energetic footprint, with interactions distributed over several HLA and peptide side chains. As ImmTAC molecules, these TCRs also retained a greater degree of pHLA selectivity, with less off-target activity in cellular assays. Conversely, TCR-mimic antibodies tended to exhibit binding modes focused more toward hot spots on the HLA surface and exhibited a greater degree of crossreactivity. Our findings extend our understanding of the basic principles that underpin pHLA selectivity and exemplify a number of molecular approaches that can be used to probe the specificity of pHLA-targeting molecules, aiding the development of future reagents.
Christopher J. Holland, Rory M. Crean, Johanne M. Pentier, Ben de Wet, Angharad Lloyd, Velupillai Srikannathasan, Nikolai Lissin, Katy A. Lloyd, Thomas H. Blicher, Paul J. Conroy, Miriam Hock, Robert J. Pengelly, Thomas E. Spinner, Brian Cameron, Elizabeth A. Potter, Anitha Jeyanthan, Peter E. Molloy, Malkit Sami, Milos Aleksic, Nathaniel Liddy, Ross A. Robinson, Stephen Harper, Marco Lepore, Chris R. Pudney, Marc W. van der Kamp, Pierre J. Rizkallah, Bent K. Jakobsen, Annelise Vuidepot, David K. Cole
Usage data is cumulative from February 2024 through February 2025.
Usage | JCI | PMC |
---|---|---|
Text version | 1,975 | 1,102 |
266 | 173 | |
Figure | 498 | 15 |
Table | 262 | 0 |
Supplemental data | 114 | 15 |
Citation downloads | 84 | 0 |
Totals | 3,199 | 1,305 |
Total Views | 4,504 |
Usage information is collected from two different sources: this site (JCI) and Pubmed Central (PMC). JCI information (compiled daily) shows human readership based on methods we employ to screen out robotic usage. PMC information (aggregated monthly) is also similarly screened of robotic usage.
Various methods are used to distinguish robotic usage. For example, Google automatically scans articles to add to its search index and identifies itself as robotic; other services might not clearly identify themselves as robotic, or they are new or unknown as robotic. Because this activity can be misinterpreted as human readership, data may be re-processed periodically to reflect an improved understanding of robotic activity. Because of these factors, readers should consider usage information illustrative but subject to change.