Natural TCRs targeting KRASG12V display fine specificity and sensitivity to human solid tumors

Abstract

BACKGROUNDNeoantigens derived from KRASMUT have been described, but the fine antigen specificity of T cell responses directed against these epitopes is poorly understood. Here, we explore KRASMUT immunogenicity and the properties of 4 T cell receptors (TCRs) specific for KRASG12V restricted to the HLA-A3 superfamily of class I alleles.METHODSA phase 1 clinical vaccine trial targeting KRASMUT was conducted. TCRs targeting KRASG12V restricted to HLA-A*03:01 or HLA-A*11:01 were isolated from vaccinated patients or healthy individuals. A comprehensive analysis of TCR antigen specificity, affinity, crossreactivity, and CD8 coreceptor dependence was performed. TCR lytic activity was evaluated, and target antigen density was determined by quantitative immunopeptidomics.RESULTSVaccination against KRASMUT resulted in the priming of CD8+ and CD4+ T cell responses. KRASG12V -specific natural (not affinity enhanced) TCRs exhibited exquisite specificity to mutated protein with no discernible reactivity against KRASWT. TCR-recognition motifs were determined and used to identify and exclude crossreactivity to noncognate peptides derived from the human proteome. Both HLA-A*03:01 and HLA-A*11:01-restricted TCR-redirected CD8+ T cells exhibited potent lytic activity against KRASG12V cancers, while only HLA-A*11:01-restricted TCR-T CD4+ T cells exhibited antitumor effector functions consistent with partial coreceptor dependence. All KRASG12V-specific TCRs displayed high sensitivity for antigen as demonstrated by their ability to eliminate tumor cell lines expressing low levels of peptide/HLA (4.4 to 242) complexes per cell.CONCLUSIONThis study identifies KRASG12V-specific TCRs with high therapeutic potential for the development of TCR-T cell therapies.TRIAL REGISTRATIONClinicalTrials.gov NCT03592888.FUNDINGAACR SU2C/Lustgarten Foundation, Parker Institute for Cancer Immunotherapy, and NIH.

Keywords: Antigen presentation; Cancer immunotherapy; Immunology; Oncology; T cell receptor.

Figure 1. mDC3/8-KRAS vaccination primes KRAS^MUT-specific T cell immunity in PAAD patients.

(A) Trial design. (B) Consolidated standards of reporting trials diagram. (C) Number of vaccine KRAS^MUT neoantigens per patient that induced IFN-γ⁺ T cells in ex vivo–expanded PBMCs collected after vaccine priming. (D) Normalized IFN-γ⁺ ELISpot counts for vaccine KRAS^MUT neoantigens after priming detected in ex vivo–expanded PBMCs. Spot counts of the nonstimulated controls were subtracted. Responses to short peptides (HLA-I) are indicated in red, and responses to long peptides (HLA-II) are indicated in blue. Symbol shape indicates specific KRAS^MUT as per legend. (E) Assessment of subject no. 2 HLA-I–restricted T cell responses against 8–16V (blue) and 7–16V (red) peptides by IFN-γ ELISpot assay following ex vivo expansion of week 2 postvaccine PBMCs. Free peptide supplemented to media bound by HLA-I expressed on donor white blood cells (HLA-A*11:01 and -A*03:01) and presented to responding T cells. Monoallelic K562 cells expressing HLA-A*03:01 (APC-A3) or HLA-A*11:01 (APC-A11) were used to identify HLA-I restriction. WT indicates WT KRAS peptide. MUT indicates mutant KRAS peptide. (F) pHLA multimer analysis to assess CD8⁺ T cell response against 8–16V/A*11:01 and 7–16V/A*11:01 following in vitro expansion of pre- (week –1) and postvaccine (week 10) CD8⁺ T cells. Successful priming of CD8⁺ T cell responses to gp100_17–25/A*03:01 and NY-ESO_60–72/B*07:02 served as positive vaccination controls. (G) Circos plot analysis following TCR-αβ RNA sequencing of FACS-sorted CD8⁺/multimer⁺ (8–16V/A*11:01) cells. Statistical differences between groups calculated using Students’ unpaired t test.

Figure 2. TCRs are specific for KRAS^G12V and exhibit distinct peptide-binding motifs with crossreactivity to KRAS^G12C.

(A) FACS profiles of TCR-engineered J^ASP90_CD8⁺ cells following 16 hours of coculture with HLA-I –matched K562 cells pulsed with KRAS^WT (black) or cognate KRAS^G12V (colored) peptide. (B) Bar graphs representing NFAT activation (specific activity, %) of J^ASP90_CD8⁺ cells following 16 hours of coculture with HLA-I–matched K562 cells pulsed with 9-mer and 10-mer KRAS^WT or KRAS^G12V peptides. (C) Peptide-binding motifs determined by X-scan analysis of TCR A3V, A11Va, A11Vb, and A11Vc depicted as heatmaps (top) and Seq2Logo plots (bottom) using J^ASP90_CD8⁺ reporter cells cocultured with HLA-I–matched K562 cells pulsed with positional peptide scanning library peptides. Heatmaps: specific activity = (GFP_Exp - GFP_Min) / (GFP_Max – GFP_Min); GFP_Min = unstimulated, GFP_Max = PMA-I. Seq2Logo plots: height of amino acid at each position corresponds to EGFP expression relative to unstimulated and PMA-I conditions. (D) Cell-reporter assay using TCR-engineered J^ASP90_CD8⁺ cocultured with K562^A*11:01 cells pulsed with titrated levels of cognate G12V versus G12C peptides. Differences in TCR functional avidities for each peptide are displayed as Δlog₁₀(EC₅₀) values. Data are representative of 2 or more experiments.

Figure 3. Assessment of TCR crossreactivity to the human proteome.

(A) Identification of candidate noncognate peptides derived from the human proteome using ScanProsite. HLA-A*03:01 and HLA-A*11:01 ligands identified using NetMHC4.0. Peptides with predicted EC₅₀ < 500 nM (NetMHC4.0) were synthesized and screened by in vitro functional assays. (B) J^ASP90_CD8⁺ reporter assay results screening noncognate peptides against A3V (n = 2) (red), A11Va (n = 5) (orange), A11Vb (n = 2) (blue), and A11Vc (n = 35) (purple). Cognate KRAS^G12V peptide used as positive control (filled bars). (C) J^ASP90_CD8⁺ reporter assay to measure A3V (red) and A11Vc (purple) functional avidity against HTR1E_42–51 vs 7–16V and RAB7B_12–21 vs 8–16V, respectively, using HLA-I–matched K562 cells pulsed with titrated levels of peptide. EC₅₀ values were determined by nonlinear regression analysis, and differences in TCR functional avidity for crossreactive versus cognate KRAS^G12V peptide are displayed as Δlog₁₀(EC₅₀) values. (D) J^ASP90_CD8⁺ reporter assay to assess A3V and A11Vc reactivity to HLA-I–matched K562 cells alone, endogenously expressing ubiquitinated (Ub) crossreactive protein, or pulsed with crossreactive or cognate KRAS^G12V peptide (filled bars). (E) J^ASP90_CD8⁺ reporter assay to assess A3V reactivity to SY5Y^A*03:01 cells in media alone or pulsed with HTR1E_42–51 or 7–16V peptide. (F) ⁵¹Cr-release assay evaluating the cytotoxic activity of primary CD8⁺ T cells engineered with A3V against SY5Y^A*03:01 cells alone or pulsed with HTR1E_42–51 peptide. (G) ⁵¹Cr-release assay evaluating the cytotoxic activity of primary CD8⁺ T cells engineered with A11Vc against Malme-3M cells alone or pulsed with RAB7B_12–21 peptide. Cytotoxicity against cognate KRAS 8–16V peptide (30:1 E:T ratio) shown as positive control. Statistical differences between groups were calculated using 2-way ANOVA followed by post hoc pairwise Student’s t test with multiple-comparison adjustment. *P < 0.05; ****P < 0.0001.

Figure 4. KRAS^G12V-specific TCRs are of high avidity and exhibit varying degrees of CD8 coreceptor independence.

(A) Cell-reporter assay using TCR-engineered J^ASP90_CD8⁺ vs J^ASP90_CD8^– cocultured with HLA-I–matched K562 cells pulsed with titrated levels of cognate 7–16V or 8–16V peptide. Statistical differences between groups calculated using 2-way ANOVA comparing specific activity of J^ASP90_CD8⁺ versus J^ASP90_CD8^– at 10^–8 M peptide (A3V) or 10^–9 M peptide (A11Va-c) followed by post hoc pairwise Student’s t test with multiple-comparison adjustment. ***P < 0.001; ****P < 0.0001. (B) EC₅₀ values of TCR-engineered J^ASP90_CD8⁺ (red) and J^ASP90–CD8^– (blue) cells as determined by nonlinear regression analysis of data presented in A. (C) Quantification of NFAT activation (specific activity, %) of TCR-engineered J^ASP90_CD8⁺ (upper) and J^ASP90_CD8⁺ (lower) cells following coculture with HLA-I–matched (colored) versus unmatched (black) KRAS^WT (BxPC3) or KRAS^G12V (CORL23, SW620, YAPC) tumor cell lines. Representative experiments of 2–4 independent evaluations are shown. Statistical differences between groups were calculated using 2-way ANOVA comparing percentages of GFP⁺ TCR-engineered J^ASP90_CD8⁺ or J^ASP90_CD8^– cocultured with HLA-I matched versus unmatched tumor cells followed by post hoc pairwise Student’s t test with multiple-comparison adjustment. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 5. CD4⁺ T cells redirected with partial CD8 coreceptor–independent TCRs exhibit cytotoxic activity.

(A) FACS plots demonstrating CD3 and TCR-αβ expression by TCR-engineered CD8⁺ (upper) and CD4⁺ (lower) T cells (colored) compared with nontransduced TCR^null cells (black). (B) FACS plots demonstrating pHLA multimer binding by TCR-engineered (colored) versus TCR^null (black) CD8⁺ (upper) and CD4⁺ (lower) T cells. Staining for A11Va-b is shown using 7–16V/ A11:01 multimer, while staining for A11Vc is shown using 8–16V/A11:01. Cytotoxic activity of TCR-engineered (C) CD8⁺ and (D) CD4⁺ T cells against HLA-I–matched (open) versus unmatched (filled) CORL23, SW620, and YAPC cell lines by 4-hour ⁵¹Cr-release assay. (E) KT₅₀, defined as time (hours) to achieved 50% cytolysis at a given E:T ratio, of TCR-engineered CD8⁺ or CD4⁺ T cells against HLA-I–matched BxPC3, CORL23, SW620, and YAPC cell lines by real-time cell analysis. (F) Cytotoxic activity of primary CD8⁺ T cells engineered with A11Va, A11Vb, or A11Vc against CORL23 cells expressing HLA-A*11:01^WT versus HLA-A*11:01^D227A/T228A. Statistical comparisons were performed comparing groups at an E:T ratio of 10:1. Statistical differences between groups were calculated using 2-way ANOVA comparing percentages of GFP⁺ TCR-engineered J^ASP90_CD8⁺ or J^ASP90_CD8^– cocultured with HLA-I–matched versus unmatched tumor cells followed by post hoc pairwise Student’s t test with multiple-comparison adjustment. **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 6. TCR-redirected CD8⁺ T cells exhibit antitumor activity against KRAS^G12V PAAD PDCs regardless of molecular subtype.

(A) Transcriptomic profiling of PDCs. (B) Cytotoxic activity by 4h ⁵¹Cr-release assay of CD8⁺ T cells engineered with TCRA3V against HLA-matched (closed circles) versus mismatched (open circles) PDCs. (C) Cytotoxic activity by 4-hour ⁵¹Cr-release assay of CD8⁺ T cells engineered with TCRA11Va against HLA-matched (closed circles) versus mismatched (open circles) PDCs. (D) Cytotoxic activity by real-time cell analysis against A*03:01 expressing PDCs by CD8⁺ T cells expressing TCRA3V versus TCRA11Va. (E) Cytotoxic activity by real-time cell analysis against A*11:01 expressing PDCs by CD8⁺ T cells expressing TCRA3V versus TCRA11Va. KT5₀ values of CD8⁺ T cells expressing (F) TCRA3V or (G) TCRA11Va against HLA-matched PDCs. Statistical significance indicated; 2-way ANOVA followed by post hoc pairwise Student’s t test with multiple-comparison adjustment.