Abstract Complete regression of cancer has been observed in a subset of patients following adoptive T-cell transfer (ACT) of ex vivo expanded tumor-infiltrating lymphocytes (TIL). However, the low success rate of ACT presents a great challenge to broader clinical application. To provide a rational approach to TIL-based immunotherapy, we developed a three-pronged approach to comprehensively evaluate clinically relevant traits of TIL-TCR. We applied the approach to a successful case of ACT where regression was observed against tumors carrying the hotspot mutation KRAS-G12D. We show that only G12D mutant peptides (a nonamer and a decamer) but not wildtype peptides can be presented by HLA-C*08:02. Structurally, the mutated Asp formed a salt bridge necessary for HLA-C binding. Therapeutic TCRs exhibited high affinity anti-tumor binding both in solution and on reconstituted T-cells. The nonamer G12D is recognized by multiple TCR9 through conserved contacts on their shared CDR2β and CDR3α, allowing CDR3β variation to generate an oligoclonal response. The decamer G12D is recognized by TCR10 through an induced peptide conformation. Interestingly, the antigen binding affinities of these TIL-TCR inversely correlated with their persistence in vivo. Importantly, the successful ACT contained TCRs specific for two non cross-reactive conformationally distinct tumor epitopes. We conclude that successful ACT requires high affinity oligoclonal TCRs with specificity for multiple antigens to optimize tumor clearance and in vivo persistence. This approach has the potential to transform current trial and error aspect of TIL-based immunotherapy to a rational selection of TCRs with optimal therapeutic properties.