Colorectal cancer (CRC) is the second cause of cancer-related death in western countries. CRC prevention and screening are major public health issues. Better knowledge of colorectal carcinogenesis could lead to better prevention. Gut microbiota (GM) is a complex community of bacteria, fungi, protozoa, viruses and bacteriophages which live in a symbiotic and epigenetic relationship with the host. GM can promote either digestive health or CRC through inflammatory and proliferative effects.

Lynch syndrome (LS) is the most common familial CRC syndrome with a lifetime CRC risk of 52% in women and 69% in men. The risk of CRC depends upon type of altered mismatch-repair gene and environmental factors (diet, exercise, obesity, tobacco and alcohol intake, etc.). Regular surveillance including annual or biannual colonoscopy is recommended in LS patients.

Chemoprevention has the potential to represent a cost-effective intervention in these high-risk patients and could allow a delay in colonoscopy surveillance. Regular low dose aspirin use is associated with a 20 to 30% reduction in the risk of sporadic colonic adenomas and CRC. The real benefit of aspirin is still to be consolidated. AAS-Lynch trial is an ongoing prospective multicenter (n=37), double-blind, placebo-controlled, randomized clinical trial, designed to investigate whether daily aspirin, at a dose of 100 or 300 mg compared with placebo, would decrease the occurrence or recurrence of colorectal adenomas in LS patients. The primary endpoint is the number of patients with at least one adenoma detected by chromo-endoscopy 48 months after initial colon clearance. At randomization and at the end of study, stool collection, blood collection, quality of life questionnaire, validated food frequency questionnaire (SU-VI-MAX2) and physical activity questionnaire are performed.

The ongoing AAS-Lynch study allow accessing to a unique fecal collection in very well characterized LS patients including a comprehensive dietary evaluation at high risk for colorectal neoplasia and planned colonoscopy surveillance during a 48 months follow-up, exposed or not exposed to chronic low dose aspirin. The expertise of the scientific consortium with state of the art microbiota analysis, the comprehensive collection of data and the prospective design of the study will allow the evaluation of the true role of gut microbiota in CRC carcinogenesis.

Background and originality of the project with regards to the state of the art :

Colorectal cancer (CRC) is the second leading cause of cancer-related death in western countries and the third most prevalent cancer worlwide. CRC prevention and screening are major public health issues. Better knowledge of colorectal carcinogenesis could lead to better prevention and screening. Recent observational data in human and experimental data in vitro or in rodents showed that gut microbiota could contribute to CRC pathogenesis.

The proposal will allow the evaluation of the role of gut microbiota in CRC carcinogenesis in a cohort of Lynch syndrome (LS) patients at high risk for colorectal neoplasia and with well characterized environmental risk factors particularly dietary factors.

Microbiota and CRC Gut microbiota is a complex community consisting of bacteria, fungi, protozoa, viruses and bacteriophages which live in a symbiotic and epigenetic relationship with the host. Gut microbiota can promote either health or tumor progression through its inflammatory and proliferative effects likely dependent on the context and genetic factors of the host. Specific microorganisms or variability of the microbiome have been associated with CRC over the past years. Fusobacterium nucleatum and Porphyromonas gingivalis sequences are over represented in CRC versus matched normal control tissue, and a positive association with lymph node metastasis has been characterized. In patients with sporadic CRC, Fusobacterium nucleatum overexpression is associated with specific CRC carcinogenesis pathways. Studies have also identified several other bacteria as Escherichia coli, Bacteroides fragilis, and Enterococcus faecalis increase in CRC patients, whereas the Clostridiales, Faecalibacterium, Blautia, Bifidobacterium, or Lactobacillus genus were low or absent. In vitro, Fusobacterium nucleatum promotes CRC cell proliferation; in mice, the presence of F. nucleatum inpatient-derived CRC xenografts (where patient CRC samples are implanted in mice) increases tumor growth rates. A possible mechanism to explain these findings is that the F. nucleatum adhesin, FadA, binds to E-cadherin on the CRC cell surface and activates oncogenic Wnt/β-catenin signaling. F. nucleatum can also alter the function of tumor-infiltrating lymphocytes and natural killer (NK) cells by binding to the inhibitory immune receptor TIGIT (T cell immunoreceptor with Ig and ITI Mdomains) through another adhesin, Fap2. Fap2 also binds a disaccharide sugar motif [galactoseN-acetyl-D-galactosamine (Gal-GalNAc)] which is expressed at high levels on the surface of many tumor cell types, as well as other cell types, and facilitates F. nucleatum binding to CRC cells. Enterotoxigenic Bacteroides fragilis (ETBF) potentiates colorectal carcinogenesis in mice, and ETBF was detected in biofilms coating human CRCs and precancerous colonic lesions. Escherichia coli expressing the genomic island polyketide synthase (pks+) enhance tumorigenesis in preclinical CRC models and are enriched in human CRC tissues. pks+ E. coli produce the genotoxin colibactin, which alkylates DNA, resulting in DNA adducts in colonic epithelial cells.

Bacterial metabolites have also been seen to have an effect. Butyrate which can only be produced by specific members of the Firmicutes phylum through the fermentation of dietary fiber and resistant starches can modulate inflammation, epithelial proliferation, and apoptosis. Hydrogen sulfide is also an important regulatory metabolite.

The complexity of microbiota and the relationship between microorganisms in the gut does not underline yet a particular bacterial signature for tumor progression so far. Then, how gut microbiota really contributes to CRC pathogenesis in the host is not fully understood.

CRC and Lynch syndrome LS is the most common familial CRC syndrome. Cancer family history is an important way to identify these at-high risk individuals. Inheritance of a germline alteration in one of the DNA mismatch repair (MMR) genes MLH1, MSH2, MSH6, PMS2 or in EPCAM genes causes a high risk of CRC and other extra-colonic related cancers. Recent studies have suggested a median age of CRC diagnosis of 61 years and a lifetime CRC risk of 52% in women and 69% in men. Specific carcinogenesis pathways are characterized in LS neoplasia.

LS patients presents precancerous lesion as non-advanced or advanced adenomas and serrated polyps that should be detected and removed. The risk of CRC depends upon which mismatch-repair gene is mutated. MSH6 and PMS2 germline mutations are associated with a lower risk with late onset of CRC.

Environmental factors (dietary factors, exercise, obesity, tobacco and alcohol intake, etc.) are also involved in LS carcinogenesis. Regular surveillance including annual or biannual colonoscopy is recommended in LS patients.

Aspirin CRC chemoprevention Chemoprevention has the potential to represent a cost-effective intervention in these high-risk patients and could allow a delay in colonoscopy surveillance. Several epidemiological studies have shown that regular use of low dose aspirin (75 to 300 mg/d) is associated with a 20 to 30 % reduction in the risk of sporadic colonic polyps and CRC. Five out of seven randomized controlled trials (RCT) showed a significant decrease in neoplasia recurrence and a meta-analysis of four of these studies shown a decrease in sporadic colorectal polyp recurrence. However, in a recent large randomized study in diabetic patients, aspirin was not associated with a lower CRC incidence.

An observational large study suggests also a preventive aspirin effect in LS. The primary analysis of the Colorectal Adenoma/Carcinoma Prevention Program 2 (CAPP2) study in LS patients receiving 600-mg daily aspirin (300 mg b.i.d) showed no significant decrease in CCR neoplasia. However, a preplanned secondary analysis after extended follow-up (mean 56 months) showed a reduced risk of CRC in the aspirin group (HR=0.41; 95% CI, 0.32-0.99). In this study, the endoscopic follow-up was not optimal with a low detection rate of colorectal neoplasia. Therefore the real effect and clinical benefit of aspirin are still to be consolidated in LS patients since aspirin chronic use is associated with possible side-effects (bleeding).

AAS-Lynch Study AAS-Lynch trial is an ongoing prospective multicenter (n=37), double-blind, placebo-controlled, randomized clinical trial, designed to investigate whether daily aspirin, at a dose of 100 or 300 mg compared with placebo, would decrease the occurrence or recurrence of colorectal adenomas in LS patients under 75 years of age. The primary endpoint is the number of patients with at least one adenoma detected by chromo-endoscopy 48 months after initial colon clearance with complete polyp removal and after treatment onset. The secondary end-points are the delay between the appearance of the first adenoma and treatment onset, adenomatous polyp burden measured by chromo-endoscopy, the number of patients who presented an adenoma during follow-up according to the affected gene (MLH1, MSH2, MSH6, PMS2 or other), sessile serrated adenoma burden, colon cancers diagnosed at scheduled surveillance colonoscopies, interval colon cancer (diagnosed between two colonoscopies). The number of colonoscopies performed during follow-up, the quality of bowel cleansing, the observance assessed by remaining tablets in blisters and adverse events are monitored. At randomization and at the end of study, stool collection, blood drops collection on blotting paper, quality of life using the SF36 questionnaire, validated food frequency questionnaire (SU-VI-MAX2, U557 INSERM) and physical activity questionnaire are performed.

Hypothesis, main objective(s) and endpoint(s) Gut microbiota seems to be a major player in CRC carcinogenesis. Better knowledge on microbiota role in CRC carcinogenesis could lead to better prevention and surveillance particularly in high risk LS patients but also in standard risk patients.

The main objective of this study is to evaluate the role of the gut microbiota using metagenomic analysis in CRC carcinogenesis in LS patients. The AAS-LYNCH study design design will allow evaluation of the role of the gut microbiota in adenoma apparition or recurrence in patients according to well characterized associated dietary risk factors and other life style factors, involved genes and neoplasia involved pathways.

Primary objective To compare fecal microbiota diversity in LS patients with no previous adenoma vs LS patients with an history of adenoma or adenoma at inclusion in AAS-Lynch study.

Secondary objectives

To describe differences in the gut microbiota diversity using metagenomic analysis according to :

• involved LS genes (MLH1, MSH2, MSH6, PMS2 or in EPCAM genes)
• adenoma type (advanced or non-advanced adenoma) and presence of serrated polyp
• diet, using dietary patterns (a priori and a posteriori patterns)
• other life style factors : BMI, physical activity, alcohol, tobacco

Perspectives (not included in the present application) :

With an expected 30% adenoma recurrence after colon initial clearance, the 48 months follow-up will allow to explore differences in gut microbiota characterized by metagenomic analysis according adenoma recurrence. The 48 months follow-up with new fecal collection will also allow describing differences in gut microbiota according to aspirin exposition. No additional funding will be required for these evaluations.

A study evaluating the chemopreventive aspirin effect in LS patients coordinated by the UK-CAPP group is ongoing with a similar design but without placebo group and different endpoints. Final shared analysis was discussed. An evaluation of international variation in fecal microbiota in relation with LS will be discussed with this group.

Primary endpoint (linked to the primary objective) This study will compare fecal microbiota richness and composition differences between LS patients with no previous adenoma at inclusion in AAS-Lynch study and LS patients with history of adenoma or adenoma at inclusion in AAS-Lynch study.

Secondary endpoints (linked to the secondary objectives)

Gut microbiota composition(taxonomic, functional and richness assessed by metagenomic analysis) will be described according to :

• involved LS genes (MLH1, MSH2, MSH6, PMS2 or in EPCAM genes)
• adenoma type (advanced or non-advanced adenoma) and presence of serrated polyp
• diet, using dietary patterns (a priori and a posteriori patterns)
• other life style factors : BMI, physical activity, alcohol, tobacco

Project plan describing the methodology and work to be performed :

• WP 1. Preparation of the project, regulatory issues and fecal handling
• WP 2. Verification of socio-demographic, lifestyle and nutritional data quality, nutritional data management.
• WP 3. Microbiota metagenomic study and data analysis
• WP 4. Final meeting and report, submission for publication

Expected results and scientific and medical potential impact of the translational research:

Colorectal cancer (CRC) is the second leading cause of cancer-related death in western countries. CRC prevention and screening are major public health issues. Better knowledge of the role of gut microbiota in CRC carcinogenesis could contribute to a better description of CRC risk with at a putative better surveillance.

Interventional therapies as prebiotic or probiotic therapies using specific microbes or genetically modified microbes could also be considered as preventive tools in near future.

Inclusion Criteria:

• Patients included in the AAS-Lynch study (NCT02813824 on clinicaltrials.gov)
• Fecal samples taken for the AAS-Lynch study (NCT02813824 on clinicaltrials.gov)
• Food questionnaires collected as part of the AAS-Lynch study (NCT02813824 on clinicaltrials.gov)

Exclusion Criteria:

• Patients who did not included in the AAS-Lynch study (NCT02813824 on clinicaltrials.gov)
• Patients who did not consent for the Fecal samples in AAS-Lynch study (NCT02813824 on clinicaltrials.gov)
• Patients who did not do the Food questionnairy in AAS-Lynch study (NCT02813824 on clinicaltrials.gov)

• Patient with Lynch syndrome bearing an alteration of "mismatch repair" genes or,when no characteristic alteration has been found, with a personal or family history of Lynch syndrome according to modified Amsterdam criteria
• Aged more than 25 years, et aged more than 18 years with an early familial history and any reason to perform a colonoscopy every 2 years
• Aged less than 75 years