There is an unambiguous association of Streptococcus gallolyticus infection with colorectal cancer, although there is limited information about epidemiology or interaction between molecular and environmental factors. We performed an original quantitative analysis of S. gallolyticus in unselected colorectal cancer patients (n = 190) and their association with clinical, pathological tumor molecular profiles (microsatellite instability, hypermethylator phenotype and chromosomal instability pathways), and other biological factors in colorectal tumor and normal tissues (cytomegalovirus and Epstein-Barr virus infection). We developed a new quantitative method to assess bacterial load. Analytical validation was reached with a very high sensitivity and specificity. Our results showed a 3.2% prevalence of S. gallolyticus infection in our unselected cohort of colorectal cancer cases (6/190). The average S. gallolyticus copy number was 7,018 (range 44–34,585). No previous reports relating to S. gallolyticus infection have been published for unselected cohorts of patients. Finally, and despite a low prevalence of S. gallolyticus in this study, we were able to define a specific association with tumor tissue (p = 0.03) and with coinfection with Epstein-Barr virus (p = 0.042; OR: 9.49; 95% IC: 1.1–82.9). The prevalence data provided will be very useful in the design of future studies, and will make it possible to estimate the sample size needed to assess precise objectives. In conclusion, our results show a low prevalence of S. gallolyticus infection in unselected colorectal cancer patients and an association of positive S. gallolyticus infection with tumor tissue and Epstein-Barr virus coinfection. Further studies will be needed to definitively assess the prevalence of S. gallolyticus in colorectal cancer and the associated clinicopathological and molecular profiles.

Background

Background Lynch syndrome (LS) is a hereditary condition characterized by a high risk of colorectal cancer, endometrial cancer, and other neoplasia associated with germline alterations in DNA mismatch repair genes. The classical genetic diagnostic strategy for LS consists of the Sanger sequencing of genes associated with the suspected syndrome. Next‐generation sequencing (NGS) enables the simultaneous sequencing of a large number of hereditary cancer genes. Here, we aimed to study whether other germline pathogenic variants of hereditary cancer genes are present in patients with LS. Methods A cohort of 84 probands with a previous genetic diagnosis of LS by Sanger sequencing was reanalyzed using NGS via a commercial panel of 94 hereditary cancer genes by hybrid capture. The American College of Medical Genetics and Genomics criteria were used to classify the clinical significance of the variants. The findings of NGS were confirmed by Sanger sequencing. When possible, genetic analyses of the new findings in the proband's relatives were also performed by Sanger sequencing. Results We identified five families (6%), out of 84, with at least two germline pathogenic variants conferring to high or moderate risk in different dominant cancer‐predisposing genes: [MLH1‐BRCA2‐NBN], [MLH1‐BRCA1], [MSH2‐ATM], [MSH6‐NF1], and [MLH1‐FANCA]. Interestingly, only one out of these five families exhibited a clinical phenotype associated with the new pathogenic variants. The family with three pathogenic variants of the [MLH1‐BRCA2‐NBN] genes showed a high aggregation of tumors associated with LS and breast and ovarian cancer syndrome. Conclusions Our results showed that the co‐occurrence of more than one pathogenic variant in cancer‐predisposing genes was remarkable among cases of LS. In most cases, no clinicial manifestations were associated with the secondary pathogenic variants. Further studies are needed to confirm these findings and elucidate their clinical impact. Reanalysis of LS families should be considered only in families with mixed clinical phenotypes.

No abstract