Preview

PDF, English Download (8MB) | Terms of use

Abstract

Microsatellite-unstable (MSI) cancers occurring in the context of the hereditary Lynch syndrome or as sporadic cancers elicit pronounced tumor-specific immune responses. The pronounced immune response was shown to be closely associated with frameshift peptides (FSP) that are generated as a result of deficiency in DNA mismatch repair system leading to insertion/deletion mutations in coding microsatellites (cMS). FSP neoantigens are long antigenic amino acid stretches that bear multiple epitopes to be presented. There is no central tolerance against FSPs, and shared FSPs derived from driver mutations are promising candidates for vaccination approaches to treat or prevent MSI cancers. In the present thesis, the main goals were to set up a mouse model for the immunology of MSI cancers and to systematically identify immune evasion mechanisms in MSI cancers. A murine model is essential to characterize alterations of immune responses over time, in all stages of cancer and pre-cancerous stages. In addition, it allows testing an FSP vaccine for efficacy in tumor prevention and treatment, either as a single agent or on combination with other immune-modulatory drugs. To establish such a model, the complete mouse genome was screened and genes bearing cMS were detected. After mutation and expression analysis by using murine Lynch tumors, epitopes of the most promising potential FSP candidates were predicted by using the Syfpeithi and netMHC algorithms. Immunogenicity of the 10 FSPs with the highest ranks was analyzed by vaccinating C57BL/6 mice and analyzing immune responses using IFNg ELISpot. Four FSPs were identified that were highly immunogenic and inducing spot numbers higher than Ova control peptides: Maz (-1) and Senp6 (-1) induced only CD4 T cell responses, Xirp1 (-1) induced only CD8 T cell and Nacad (-1) induced both CD4 and CD8 T cell responses. Peptide-specific IgG Elisa demonstrated that three of the peptides Senp6 (-1), Maz (-1) and Nacad (-1) also induced humoral immune response. Immunogenic regions of the peptides could be mapped to the C-terminus of Senp6 (-1) and Xirp (-1) and to the N-terminus of Nacad (-1), whereas the antigenic region for Maz (-1) spanned almost the entire peptide. These results suggest that the Lynch mouse model is well suitable for evaluating the efficacy of FSP vaccination to treat and even prevent tumors in Lynch syndrome. II The second aim of this thesis was to systematically analyze immune evasion mechanisms in MSI cancer. We first analyzed mutations of genes related to MHC class I antigen presentation in publicly accessible mutation databases. The mutation data of 91 MSI patients in the DFCI cohort showed that 72% of all MSI CRC tumors had defects in MHC class I presentation; displayed by at least one mutation in the corresponding genes (B2M, TAP1, TAP2, HLA-A, HLA-B, HLA-C and NLRC5). Mutual exclusivity analysis revealed that mutations affecting B2M were negatively related to HLA-B mutations, whereas there was a strong positive correlation between HLA class I heavy chain mutations. These results indicate that there is a strong immunoselection in MSI tumorigenesis, leading to immune evasion through mutations of MHC class I-related genes in more than two-thirds of MSI cancers. We identified NLRC5 mutations as a potential novel immune evasion mechanism in the database analysis; therefore, potential consequences of NLRC5 inactivation were further analyzed in MSI colorectal cancer samples. We detected cMS mutations of NLRC5 in 4 out of 95 tumor samples (4.2%), three of them being one-basepair deletions and one silent mutation. Importantly, we detected low levels of MHC class I antigen expression in NLRC5- mutated tumors. One tumor showed partial reduction of MHC class I expression, which colocalized with the NLRC5 mutation. These results suggest NLRC5 mutations as a novel potential mechanism of immune evasion in MSI cancer. Taken together, the present thesis led to the establishment of the first model to evaluate the immune biology of MSI cancers and Lynch syndrome in the murine system. Moreover, it has established a comprehensive overview of immune evasion in MSI cancers, thus contributing to the development of better treatment strategies and potentially to the first cancer-preventive vaccine for non-viral human cancers.