Background The amount of neoantigen (neoAg)-reactive tumor-infiltrating T lymphocytes (TIL) infused to patients is associated with clinical response.^{1 2} However, no biomarkers can currently predict whether neoAg-reactive T cells can be expanded from a tumor, and bystander T cells often dominate products administered to patients unlikely to benefit. We have tested a manufacturing method to enrich TIL in neoAg-reactive T cells and investigated baseline immune variables associated with end product neoAg reactivity.

Methods To obtain neoAg-reactive TIL, T cells were primarily expanded from ten colorectal cancer liver metastases, six lung cancers, four melanomas, and two ovarian cancers, that were also sequenced to identify non-synonymous mutations predicted to generate neoAg peptides. For each tumor, up to 191 peptides (25-mers) representing top ranking neoAgs were synthesized, pulsed on autologous antigen-presenting cells and co-cultured with TIL. Reactive TIL were FACS sorted and expanded into end products. Deep TCR Vb chain sequencing and multiparameter FACS served to analyze the immune features of pre-operative circulating T cells and baseline TIL to test associations with end product neoAg reactivity.

Results Out of 22 tumors tested, enrichment of neoAg-reactive T cells in end products, assessed by de novo expression of CD134 or CD137 upon neoAg pooled-peptide stimulation, ranged from 0% to 89.5% (7 highly reactive [>25%], 7 intermediate [5 to 25%], and 8 low [<5%]). The capacity to expand neoAg-reactive TIL was not correlated with predicted neoAg load, initial TIL abundance, T cell clonality or phenotype. However, the percent overlap in TCR repertoire between the tumor and the circulating blood (median 2.7%; range 0.5% to 8.0%) best correlated with end product neoAg-reactivity (spearman r 0.527; p = 0.014), especially in CD4+ TIL (figure 1). High TIL product neoAg-reactivity (>25%) was only obtained if tumor and blood TCRs overlapped by at least 2.6% (median 4.2%). The percent tumor and blood TCR overlap correlated, in circulating CD4⁺ and CD8⁺, with CD45RA^neg (non-naïve, non-terminally differentiated) T cells expressing CXCR5 (the CXCL13 receptor), and in tumors, with CD103⁺CXCR6⁺ expressing TIL. Unbiased tSNE analysis of baseline TIL phenotype that yielded high versus low neoAg-reactivity also supported these findings.

Conclusions Our results suggest that the degree of baseline TCR overlap between blood and tumor repertoire could help identify patients from which neoAg-reactive TIL can be expanded. We are using single T-cell sequencing methods to understand the underlying biology of this crosstalk, which could inform future trial designs and new manufacturing processes.

References

1. Levi ST, Copeland AR, Nah S, et al. Neoantigen identification and response to adoptive cell transfer in anti-PD-1 naive and experienced patients with metastatic melanoma. Clin Cancer Res 2022;28(14):3042–3052.

2. Kristensen NP, Heeke C, Tvingsholm SA, et al. Neoantigen-reactive CD8+ T cells affect clinical outcome of adoptive cell therapy with tumor-infiltrating lymphocytes in melanoma. J Clin Invest 2022;132(2).

Ethics Approval This study was approved by the Centre hospitalier de l’Université de Montréal Ethics Board; approval number 21.299.

• Download figure
• Open in new tab
• Download powerpoint

Abstract 443 Figure 1

Correlation between CD4+ TIL end product neoAg-reactivity and the overlap in TCR repertoire between circulating blood and tumor