The majority of the clinical situations suggestive of an increased genetic risk for colorectal cancer (CRC) are not explained by a simple monogenic model. Our working hypothesis is that a fraction of clinical conditions suggestive of an increased genetic risk for CRC (familial aggregation, early age of tumour onset, development of multiple primary CRC) is due to the combination of a limited number of genetic variations, each conferring a moderate risk for CRC, but whose combination results into high risk.

This study, which will be both retrospective and prospective, is an association study that will compare frequencies of selected genetic variations, alone or in combination, between patients (cases) whose clinical presentation is suggestive of an increased genetic risk but who do not present a known Mendelian form of CRC, and controls, in order to assess associations of these variations with non-Mendelian genetic forms of CRC. The inclusion criteria will be: CRC in two first degree relatives, one being diagnosed before 61 years of age; or CRC diagnosed before 51 years of age or advanced colorectal adenoma before 41 years of age; or multiple primary CRCs in the same individual, the first being diagnosed before 61 years of age The exclusion criteria will be: Lynch syndrome, adenomatous polyposis and hamartomatous polyposis. The genetic variations which will be analyzed will include (i) SNPs detected by GWAS and associated to CRC. (ii) Risk factors corresponding to functional polymorphisms within candidate genes. (iii) Imbalance of the TGFbR1 allelic expression. Sample sizes were calculated to obtain 80% power for an odds ratio of 2.5 since the aim of this study is to determine genetic variations conferring a moderate risk. In order to cover all these possibilities and to have reasonable even for the genetic variations with low frequency below 0.03 or high frequency above 0.90, the target sample size was set at 700 cases and 350 controls. The recruitment of patients will be performed at the national level by the cancer genetics departments ensuring a correct evaluation of the personal and familial history and the French molecular diagnosis laboratories network ensuring in routine the analysis of the main genes involved in CRC. The control population will be composed of healthy subjects of Caucasian origin between 45 and 60 years of age, without personal or familial history among their first-degree relatives of colorectal tumours.

Demonstration that the combination of a limited number of genetic variations, each conferring a moderate risk for CRC, results into high risk would allow to base, in selected families, the evaluation of risk in relatives and indication of colonoscopy on the analysis of gene variants the combination of which would confer a high risk. This study will confirm or invalidate the contribution of aTGFbR1 allelic expression imbalance in the determinism of early-onset CRC and familial aggregation of CRC. The demonstration of the involvement in CRC genetic variations with strong effect of specific combinations should be of interest for our general understanding of the molecular basis of CRC.

Inclusion Criteria:

• colorectal cancer (CRC) in two first degree relatives, one being diagnosed before 61 years of age
• CRC diagnosed before 51 years of age or advanced colorectal adenoma before 41 years of age
• multiple primary CRCs in the same individual, the first being diagnosed before 61 years of age.

Exclusion Criteria:

• Lynch syndrome,
• adenomatous polyposis defined by more than 10 adenomas histologically proved,
• hamartomatous polyposis defined by one hamartoma histologically proved,
• absence of informed consent.