Background
The ability to identify problematic off-targets is critical for TCR-T therapies as TCRs that recognize off-targets expressed at high levels in critical organs could cause toxicities. TCR-T therapies require a fully human system to appropriately de-risk their use in clinical studies. The paucity of relevant animal models to detect potential off-target activity of TCRs has led to reliance on predictive computational algorithms guided by positional scanning mutagenesis. These methods fail to screen against potential off-targets with low sequence homology to target epitopes. To overcome these limitations, TScan has developed SafetyScan- an unbiased, genome-wide, high-throughput platform to eliminate TCR candidates that cross-react with primary human tissues.
Methods
T cells expressing a candidate TCR are co-cultured with target cells expressing a genome-wide library of protein fragments. Target cells recognized by the TCR are sequenced to reveal the natural target(s) of the TCR as well as putative off-targets, even if they have low sequence homology to the target epitope. To determine if any of these putative off-targets represent bona fide off-targets of the TCR, T cells expressing the candidate TCR are co-cultured with an array of primary and induced pluripotent stem cells (iPSC)-derived cells derived from epithelial, mesenchymal, endothelial, fibroblastic and muscle cells from vital and non-vital organs, from male and female donors that endogenously express the cognate human leukocyte antigen (HLA) and putative off-targets. Levels of IFN- in the culture supernatants are used as a measure of T cell reactivity. Bulk RNA sequencing quantifies the expression of putative off-targets and HLA in the primary cells.
Results
In a proof of capability study, an affinity-enhanced TCR from a different sponsor that led to clinical toxicity due to off-target reactivity with cardiac muscle protein titin was evaluated.1 2 The genome-wide screen identified multiple putative off-targets, including titin. Co-cultures of T cells expressing the affinity-enhanced TCR with iPSC-derived cardiomyocytes displayed significant reactivity. RNA-seq confirmed that titin was expressed in these cells, albeit at lower levels than expected for cardiac tissue. The platform was applied to assess and de-risk putative off-targets of the six TCRs currently in TScan’s ImmunoBank and being studied in the clinic, restricted to HLA-A*02:01, HLA-A*01:01, HLA-B*07:02 and HLA-C*07:02 and targeting HPV16 E7, PRAME, MAGE-A1 and MAGE-C2.
Conclusions
These data demonstrate the sensitivity of comprehensive genome-wide screening for identifying putative off-targets and characterization of primary human cell panel using RNA-sequencing, to help select appropriate and physiologically-relevant primary human cells to test for off-target reactivity.
References
Cameron BJ, Gerry AB, Dukes J, et al. Identification of a Titin-derived HLA-A1-presented peptide as a cross-reactive target for engineered MAGE A3-directed T cells. Sci Transl Med. 2013;5(197):197ra103.
Linette GP, Stadtmauer EA, Maus MV, et al. Cardiovascular toxicity and titin cross-reactivity of affinity-enhanced T cells in myeloma and melanoma. Blood. 2013;122(6):863–871.