Abstract B70: Generation of MAGE-A4 230-239 peptide-specific cytotoxic T lymphocytes using an alloreactive approach

• Published in: Cancer research (Chicago, Ill.) Ill.), 2013-01, Vol.73 (1_Supplement), p.B70-B70
• Main Authors: Lövgren, Tanja, Mao, Yumeng, Böschen, Maxi-Lu, Allerbring, Eva B., Achour, Adnane, Poschke, Isabel, Kiessling, Rolf
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.TUMIMM2012-B70

The aim of this project was to generate cytotoxic T lymphocytes (CTL) specific for the weakly immunogenic MAGE-A4 antigen, by an alloreactive approach. CTL may recognize and eliminate tumor cells. However, to avoid generation of autoreactive CTL, most CTL specific for tumor-associated antigens (TAA) are eliminated by negative selection during their development in the thymus. Thus, compared to virus-specific CTL, TAA-specific CTL precursors are often rare and more challenging to expand from peripheral blood. One TAA that is difficult to raise specific CTL against is the MAGE-A4 antigen, a cancer-testis antigen expressed by many tumors but not normal tissues. A MAGE-A4 230-239 peptide is processed (Duffour et al. Eur J Immunol 1999;29;3329–37) and presented by the HLA-A2 complex (Hilling et al. J Mol Biol 2001;310:1167-76). Still, rarely have MAGE-A4 230-239 specific CTL been found after in vitro stimulation or in vivo vaccination. We speculated that the reason was elimination of basically all precursors by negative selection. To get around this, we transfected HLA-A2 into mature dendritic cells (moDC) generated from HLA-A2 negative donor monocytes. In contrast to previous studies using this approach for other TAA (Stronen et al. Scand J Immunol 2009;69:319-28), we used the HLA-A2 gene in a plasmid instead of the HLA-A2 mRNA for transfection. This approach excludes in vitro transcription and avoids RNA stability issues. The transfection efficiency with our method was between 25 to 50% after 24 hours. We pulsed the HLA-A2 transfected moDC with peptide and co-cultured them with autologous CD8 T cells. As these CD8 T cells have not encountered HLA-A2 during development, they should contain higher frequencies of precursors specific for peptides presented by this HLA allele. Thus, the setting is allogeneic only for peptides presented by HLA-A2, and will from hereon be referred to as the HLA-A2 allogenic setting. As a control, we used the optimized MART-1 26-35 (A27L) peptide, known to generate potent in vitro responses even in CD8 T cells from healthy HLA-A2 positive donors using an autologous setting. We managed to expand MART-1 specific CTL in vitro from 50% of healthy HLA-A2 positive donors (2/4), when using PBMC and in every donor when using moDC (3/3) as antigen-presenting cells (APC). Thus it was not surprising that we could also generate MART-1 specific cells from all donors with the HLA-A2 allogeneic setting (3/3). In the case of MAGE-A4, we have never su