It’s the Peptide-MHC Affinity, Stupid

Abstract

Adoptively transferred T cells can reject large established tumors, but recurrence due to escape variants frequently occurs. In this issue of Cancer Cell, Engels et al. demonstrate that the affinity of the target peptide to the MHC molecule determines whether large tumors will relapse following adoptive T cell therapy.

Misinitiation of intrathymic MART‐1 transcription and biased TCR usage explain the high frequency of MART‐1‐specific T cells

“…However, experimental models have shown that the process of negative selection is exquisitely efficient and even a postselection repertoire consisting of more than 10% auto-reactive TCRs can be efficiently purged by rare antigen-positive medullary antigen presenting cells (APCs) [7,8]. If this were also the case for the human thymus, it would argue against the notion that overt positive selection of MART-1 [26][27][28][29][30][31][32][33][34][35] -specific CD8 + T cells alone could explain their unusually high frequency. Although quantitative aspects of positive selection cannot be directly addressed experimentally in humans, the prerequisites for tolerance induction, i.e.…”

“…+ T cells specific to the peptide-MHC class I complex and MART-1 [26][27][28][29][30][31][32][33][34][35] -HLA-A2, are about 100 times more frequent in healthy individuals than T cells specific to other self-antigens [3]. This high frequency could be due either to an increased thymic output or to peripheral T cell expansion.…”

“…However, most MART-1 [26][27][28][29][30][31][32][33][34][35] -specific CD8 + T cells have a naïve phenotype in healthy individuals [3], and this suggests that their high frequency originates in the thymus. In melanoma patients, MART-1 [26][27][28][29][30][31][32][33][34][35] -specific CD8 + T cells express activation markers and can be expanded further to reach frequencies of up to 1 in 10 CD8 + T cells at the tumor site or in draining lymph nodes [3]. Yet, it is unknown whether they have tumoricidal activity at these sites [5].…”

“…Two mutually nonexclusive explanations have been offered: first, lack of central and peripheral tolerance, i.e. no deletion of MART-1 [26][27][28][29][30][31][32][33][34][35] specific CD8 + T cells owing to the absence of MART-1 expression in the thymus and its seclusion in peripheral tissues; second, unusually efficient positive selection of MART-1 [26][27][28][29][30][31][32][33][34][35] -reactive CD8 + T cells owing to the biased usage of the TRAV12-2 variable gene of the TCRα-chain that imparts broad cross-reactivity to other selfantigens including those presented by cortical thymic epithelial cells (cTECs) [6]. However, experimental models have shown that the process of negative selection is exquisitely efficient and even a postselection repertoire consisting of more than 10% auto-reactive TCRs can be efficiently purged by rare antigen-positive medullary antigen presenting cells (APCs) [7,8].…”

“…First, pGE has evolved to select for high-affinity binders to MHC molecules, assuming that immunogenicity and the risk of autoimmunity will increase with increasing affinity [31]. The MART-1 26-35 peptide binds poorly to HLA-A2 (IC 50 : 7600 nM, according to the Immune Epitope Database Analysis Resource) predicting this epitope to be a poor target [32,33]. According to this reasoning, there would be no need for central tolerance of MART-1 26-35 -reactive CD8 + T cells, MART-1-positive cells could be detected in human thymus sections, and purified complete and mature MHCII hi mTECs using a C-terminal antibody that binds downstream of the MART-1 26-35 epitope (six individual thymic samples were analyzed).…”