Umeå universitets logga

Colorectal cancer (CRC) is a heterogenous disease consisting of multiple molecular subtypes, each of which has diverse treatment responses and prognoses. The importance of the gut microbiota in CRC development and progression has undergone increasing recognition in recent years, with a structural segregation in terms of microbial composition between CRC patients and healthy controls. However, many questions remain before a full understanding of the impact of the gut microbiota on CRC is reached. The overall aim of this thesis was to explore the role of gut microbes in CRC, including their potential as CRC biomarkers and associations with clinicopathological, immunological, and molecular traits of CRC. A particular focus was the CRC-associated oral pathogen Parvimonas micra.

To investigate faecal microbiota as a potential biomarker for CRC, we studied the presence of specific bacteria in faeces from CRC patients and controls using qPCR. We found higher levels of P. micra in faecal samples from CRC patients than from control patients. A test for high levels of P. micra was able to identify CRC with a specificity of 87.3% and a sensitivity of 60.5%. Adding the oral pathogen Fusobacterium nucleatum, colibactin-producing bacteria, and faecal haemoglobin to the test enhanced its sensitivity.

We further aimed to explore the associations of P. micra and F. nucleatum with molecular subtypes of CRC and the tumour immune response. The levels of P. micra and F. nucleatum, as analysed by qPCR in both faeces and tumour tissue from CRC patients, were found to be positively correlated. High levels of intratumoural P. micra and F. nucleatum were associated with tumours of the microsatellite instable subtype and BRAF-mutated tumours. For F. nucleatum, an additional association with right-sided tumours was found. Moreover, both P. micra and F. nucleatum in tumour tissue were associated with the immune-activated consensus molecular subtype (CMS) 1 subtype of CRC. In line with this finding, we found novel associations between intratumoural P. micra and specific immune traits, which were evaluated by flow cytometry, indicating an active immune response in CRC. These results were further confirmed using transcriptomics. However, no associations with specific immune traits were found for F. nucleatum.

We also investigated associations between faecal and intratumoural levels of P. micra and F. nucleatum and survival in CRC patients. CRC patients with high levels of intratumoural P. micra and F. nucleatum showed reduced five-year cancer-specific survival. This association remained significant for P. micra in multivariable analysis. No associations with cancer-specific survival were found for levels of P. micra and F. nucleatum in faeces.

To investigate the faecal microbial landscape of CRC patients on a larger scale, we used 16S rRNA sequencing. Network analysis revealed a cluster of associated bacteria, including P. micra and F. nucleatum, as well as other CRC-related pathogens such as Bacteroides fragilis, Peptostreptococcus stomatitis, and Porphyromonas spp. Furthermore, beta-diversity analysis indicated a significantly different gut microbial composition depending on tumour location and microsatellite instability status. Interestingly, three of the six annotated species most strongly associated with microsatellite instable tumours were also present in the cluster: P. micra, F. nucleatum, and P. stomatis.

In conclusion, our results suggest P. micra as a putative candidate for a future non-invasive microbial test panel for detection of CRC. Moreover, our results indicate that intratumoural P. micra and F. nucleatum are associated with immune-active subtypes of CRC, including microsatellite instable tumours and tumours of the CMS1 subtype, as well as decreased patient survival. Furthermore, P. micra and F. nucleatum were found to be associated with a cluster of other CRC-related oral pathogens. An improved understanding of the role of the gut microbiota in tumour progression may lead to the identification of important biomarkers for CRC disease and outcome, as well as potential targets for future therapy.