Tumor neoantigen recognition is crucial for successful cancer immunotherapies. Patients with high tumor mutational burden (TMB) and numerous neoantigens benefit more from immune checkpoint blockade and adoptive cell therapy (ACT) using tumor-infiltrating lymphocytes (TILs). TIL therapy has also shown success in low TMB cancers. Unlike TAA-recognizing T cells, neoantigen-specific T cells avoid negative thymic selection, potentially leading to higher functionality and antigen sensitivity. T cell functionality is partly determined by TCR avidity for cognate PMHC, referred to as structural avidity (measured by monomeric pMHC-TCR dissociation kinetics). Antigen sensitivity, or functional avidity, reflects TCR properties and cell state. Studies show structural and functional avidities correlate and influence T cell performance. In TIL-ACT, clinical efficacy correlates with in vivo persistence of transferred TIL clones, linked to unique gene expression patterns. However, the impact of avidity on tumor engraftment of tumor-specific T cells is poorly understood, making it a key parameter for successful T cell-based immunotherapy. This study evaluates structural avidities and antigen sensitivities of neoantigen-specific T cells to understand their correlation with tumor infiltration and homing. We profile a large library of CD8 T cells specific for neoantigens, TAAs, and viral epitopes from tumors and blood of melanoma, ovarian, lung, or colorectal cancer patients and healthy donors. We investigate whether high avidity is associated with tumor residence, CXCR3 expression, and tumor engraftment after ACT in mice. We aim to develop an in silico model to predict TCR avidity and its relationship to tumor infiltration, thus providing a functional parameter for screening neoantigen-specific T cells for ACT.

The literature extensively supports the importance of neoantigens in cancer immunotherapy. Studies have demonstrated a correlation between high tumor mutational burden (TMB) and the success of immune checkpoint blockade and adoptive cell therapies (ACT). The unique nature of neoantigens, not being subject to thymic selection, suggests that T cells targeting them might exhibit superior functionality. Previous research has highlighted the role of T cell receptor (TCR) avidity, both structural and functional, in determining T cell effectiveness. However, the relationship between avidity and tumor infiltration remains unclear, despite the clinical correlation between TIL persistence and ACT success. Existing studies also suggest unique gene expression patterns in successful TILs. This study builds on this existing body of knowledge by rigorously investigating the relationship between TCR avidity and the ability of T cells to infiltrate and eliminate tumors.

This study utilized a comprehensive approach to characterize the avidity of T cells targeting various antigens (neoantigens, tumor-associated antigens (TAAs), and viral antigens). A library of 371 CD8 T cell clones was generated from 16 cancer patients (melanoma, ovarian, lung, colorectal) and 6 healthy donors. These clones were isolated using double-fluorescent reversible PMHC multimers (NTAmers) which avoid the loss of high-avidity cells. Structural avidity was assessed using monomeric PMHC-TCR dissociation kinetics, measuring the pMHC-TCR half-life (T1/2). Functional avidity (antigen sensitivity) was measured using IFNγ ELISpot, determining the EC50. A correlation between T1/2 and EC50 was observed, although variability existed, possibly due to the influence of T cell intrinsic regulatory mechanisms on functional avidity. TCRs were sequenced to analyze their repertoire. An in vitro pMHC refolding assay validated peptide affinity. The researchers found poor correlations between measured avidity and in silico predictors of pMHC affinity, stability, or processing. However, correlations were found with immunogenicity predictors (PRIME) and PMHC Dissimilarity-to-Self (DisToSelf). To investigate the link between avidity and tumor infiltration, TILs and PBLs recognizing the same epitopes were compared. TILs showed significantly higher antigen sensitivity and structural avidity than PBLs. TCR sequencing of TILs and PBLs targeting the same neoepitope revealed a correlation between structural avidity and clonal frequency in the tumor. An in vivo experiment in mice using NY-ESO-1-specific TCRs with varying avidities confirmed that high structural avidity was linked to better tumor infiltration and control. The expression of chemokine receptors, especially CXCR3, was analyzed. High-avidity T cells showed increased CXCR3 expression, and blocking CXCR3 impaired tumor control in mice. Finally, an in silico model was developed using TCR biophysicochemical properties to predict TCR structural avidity, successfully discriminating high- and low-avidity TCRs.

This study revealed significant heterogeneity in the avidity of neoantigen-specific CD8 T cells. Neoantigen-specific T cells demonstrated higher structural avidity compared to TAA-specific T cells. TILs consistently exhibited superior antigen sensitivity and structural avidity compared to their blood counterparts (PBLs). High structural avidity was associated with increased tumor residence, CXCR3 expression, and effective tumor infiltration in mice models. The developed in silico model accurately predicted TCR structural avidity based on biophysicochemical properties, with an AUC of 0.96, indicating a strong predictive power. Furthermore, the enrichment of high-avidity TCRs in tumors compared to blood was consistently observed in multiple patients. Blocking CXCR3 significantly impaired tumor control by high-avidity T cells in mice, highlighting the importance of CXCR3-mediated tumor infiltration.

The study findings strongly support the idea that the clinical effectiveness of neoantigen-specific T cells is not solely determined by their tumor specificity, but also by their heightened functionality and superior ability to infiltrate and reside within tumors. The substantial heterogeneity of TCR avidities within neoantigen-specific T cells underscores the importance of selecting high-avidity clones for personalized immunotherapies. The study's use of structural avidity, a robust biomarker independent of cellular activation states, offers a valuable approach for identifying potent T cells. The association of high avidity with CXCR3 expression suggests a mechanistic link between TCR strength and tumor homing. The predictive in silico model developed in this study provides a powerful tool for identifying high-avidity TCRs without prior knowledge of their specificity, streamlining the selection process for personalized cancer therapies.

This study demonstrates a strong link between neoantigen recognition, T cell functionality (structural avidity), and tumor infiltration. High structural avidity T cells, particularly those found in tumors (TILs), are more effective at controlling tumor growth and exhibit a higher propensity to home to tumors via mechanisms involving CXCR3. The development and validation of an in silico model for predicting TCR avidity offers a significant advance in selecting the most effective T cells for personalized cancer immunotherapy, potentially improving outcomes for patients receiving ACT. Future research could focus on further refining the predictive model, investigating the role of other chemokine receptors and integrins, and exploring the interplay between high avidity and T cell exhaustion.

While the study demonstrates a clear correlation between structural avidity and tumor infiltration, it does not definitively establish causality. Further research is needed to confirm that high avidity is the sole determinant of effective tumor infiltration and control. The study's focus on specific cancer types may limit the generalizability of its findings to other cancers. The in silico model, although robustly validated in this study, requires further validation in larger, independent datasets to ensure its broad applicability.