TABLE 2.
In silico analysis of two previous TCR-Ts that showed severe toxicity in clinical trials.
| Antigen | Epitope | NetMHC4.0 |
IEDB Rank (%) | NetCTL pan | Cross-Reactivity | ||
| Affinity (nM) | Ranks (%) | ||||||
| A0101/MAGE-A3 | MAGE-A3 | EVDPIGHLY | 11.43 | 0.01 | 0.12 | 0.05 | +++ |
| Titin | ESDPIVAQY | 8.07 | 0.01 | 0.17 | 0.05 | ++ | |
| A0201/MAGE-A3 | MAGE-A3 | KVAELVHFL | 16.05 | 0.25 | 0.9 | 0.2 | +++ |
| MAGE-A12 | KMAELVHFL | 3.28 | 0.01 | 0.2 | 0.01 | ++++ | |
| MAGE-A2 | KMVELVHFL | 4.61 | 0.03 | 0.3 | 0.05 | + | |
| MAGE-A4 | KVDELAHFL | 60.75 | 0.70 | 1.9 | 0.8 | − | |
| MAGE-A6 | KVAKLVHFL | 109.34 | 1.10 | 2.2 | 1.5 | + | |
| MAGE-A1 | KVADLVGFL | 165.35 | 1.50 | 2.8 | 1.5 | − | |
| MAGE-A8 | KAVELVRFL | 2082.94 | 6.00 | 7.6 | 3 | − | |
The cross-reactivity in vitro experimental data are taken from the references of 17 and 31. The shade data indicate the cross reactivity that cause severe in vivo toxicity. The rank of NetMHC4.0 binding affinity is based on 400,000 random natural peptides. The ranks of the peptide being processed and presented by NetCTL program is based on 200,000 random natural peptides.