Immunotherapy of CT26 murine tumors is characterized by an oligoclonal response against the AH1 tumor rejection antigen

Abstract

The possibility to cure immunocompetent mice bearing murine CT26 colorectal tumors using cytokinebased therapeutics allows to study the tumor rejection process at a molecular level. Following treatment with L19-mIL12 or F8-mTNF, two antibody fusion proteins which preferentially concentrate a murine cytokine payload at the tumor site, CT26 tumors could be cured in a process that crucially relies on CD8+ T cells. In both settings, the AH1 peptide (derived from the gp70 envelop protein of murine leukemia virus) acted as the main tumor rejection antigen and ~50% of CD8+ T cells in the tumor mass are AH1-specific after therapy. In order to characterize the clonality of the T cell response after successful antibody-cytokine immunotherapy, we isolated CD8+ T cells from tumors and submitted them to T cell receptor (TCR) sequencing. As expected, different TCR sequences were found in different mice, as these molecules originate from a stochastic rearrangement process. CD8+ T cells featuring the ten most abundant TCR sequences represented more than 60% of total CD8+ T cell clones in the tumor mass, but less than 10% in the spleen. Looking at sorted CD8+ T cells from individual animals, AH1-specific TCRs were consistently found among the most abundant sequences.Collectively, these data suggest that the antitumor response driven by two different antibody-cytokine fusions proceeds through an oligoclonal expansion and activation of tumor-infiltrating CD8+ T cells.

Comprehensive investigation of T and B cell receptor repertoires in an MC38 tumor model following murine anti‑PD‑1 administration

“…as for cdr3 lengths, the present study showed that applying anti-Pd-1 had no effects on the cdr3 lengths of the Tcr/Bcr repertoires. Similarly, cytokine-based immunotherapy of murine cT26 colorectal tumors did not lead to significant changes of CDR3 length distribution (38). In addition, it has been reported that there is no correlation between the cdr3 length distributions and genetic variability of the cdr3 region (39), suggesting that the repertoire divergence between the treatment and control group in this study is not represented in terms of cdr3 length.…”