Vorschau

PDF, Englisch Download (8MB) | Nutzungsbedingungen

Abstract

Colorectal cancer (CRC) is among the most frequently diagnosed cancers worldwide. Recent research focused on the association of CRC with an altered microbiome. More specifically, two bacteria, Fusobacterium nucleatum (F. nucleatum) and Streptococcus gallolyticus subspecies gallolyticus (S. gallolyticus) were individually brought in context with CRC. F. nucleatum is predominantly present in oral plaques and was found to be abundant in stool and tumor tissue of CRC patients. S. gallolyticus is a rare commensal in the human intestine and inducer of infective endocarditis that is associated with presence of CRC. The aim of this thesis was to explore potential serological associations of F. nucleatum and S. gallolyticus with CRC using multiplex serology, a high-throughput technology that allows the analysis of large seroepidemiological studies. Multiplex serology was to be developed for F. nucleatum and S. gallolyticus and applied in a retrospective case-control study to analyze potential serological associations with CRC. Prospective studies were to be analyzed to give information on temporality of the association: if serological associations are present prior to diagnosis, these antibodies might serve as early marker for risk of developing CRC. Eleven proteins for each, F. nucleatum and S. gallolyticus, were selected, recombinantly expressed and applied in multiplex serology. Serological validation of the assays was possible only to a limited extent due to a lack of a gold standard assay for comparison. Cut-offs for antibody-positivity to the individual proteins were arbitrarily defined to allow for 10% of controls as positive. Antibody responses to F. nucleatum and S. gallolyticus were analyzed in a retrospective case-control study conducted in Germany and two independent case-control studies nested within multi-center prospective cohorts from Europe and southern United States. Positivity to any of the F. nucleatum proteins was not associated with CRC, neither retro- nor prospectively. In contrast, odds for prevalent and incident CRC in the German case-control study as well as the European prospective study were significantly 2-fold increased with positivity to two or more proteins of a S. gallolyticus 6-marker panel. However, this association was not found in the southern United States study. In conclusion, antibody responses to S. gallolyticus, but not F. nucleatum, were significantly associated with CRC prior to diagnosis and might serve as marker for CRC development. A causal relationship of S. gallolyticus with CRC cannot be inferred from the generated data, however, results of this thesis might stimulate research on the involvement of S. gallolyticus in CRC development as well as risk factors leading to S. gallolyticus colonization.