Exogenous and Endogenous Risk Factors for Early-onset Colorectal Cancer

Background Firstly described in U.S. series, an increase in early-onset colorectal cancers (eoCRC), defined as CRC before 50 years, is confirmed globally.

CRC pathogenesis has been associated with several endogenous and exogenous associated risk factors, including family history of CRC, CRC-related germline pathogenic variants, obesity, alcohol habits, physical activity, red and processed meat and a Western diet. Two dietary patterns appear as exogenous and behavioral factors associated with either CRC prevention or predisposition: the 'healthy' pattern (high in fruits, vegetables, and whole grains or legumes, fish, and low-fat milk or dairy products) and the 'unhealthy' or 'Western dietary' pattern (high in red and processed meat, sugary drinks, refined grains, desserts). Above all, the Western diet is one of the most important CRC risk factors. In the Nurses' Health Study II, Western diet during adolescence was associated with the development of high-risk rectal adenomas later in life. In a meta-analysis of cohort studies, the Western dietary pattern conferred a relative risk (RR) of 1.12 for CRC, compared with a 0.89 RR for the healthy pattern. Conversely, the Mediterranean diet demonstrated a protective role in CRC development, confirmed in the EPIC cohort study.

Rationale Dietary, lifestyle, and anthropometric risk factors are still poorly understood in eoCRC patients, despite its established rising incidence. To date, scant studies, mostly case-control and only a few prospective ones, investigated the exogenous risk factors of eoCRCs and eoCRC precursors and identified high intake of alcohol and processed meat, sedentary lifestyle as potential eoCRC risk factors, and obesity as ambiguous one.

Primary Objective To compare the associations of specific dietary and lifestyle factors (smoking habit, alcohol intake, physical activity) and anthropometric factors between eoCRC patients and healthy age- and sex-matched controls in countries with increasing versus stable or decreasing early-onset colorectal cancer (eoCRC) incidence.

Secondary Objectives

1. To validate the semi-quantitative food frequency questionnaire (SQFFQ);
2. To evaluate the consistency of associations across cohorts in pairwise comparisons;
3. To assess whether associations differ among specific population subgroups (e.g., sex, ethnicity, smoking habits, BMI, physical activity, family history of CRC).

Type of study SQFFQ validation study: three non-consecutive 24-hour Dietary Recalls (24hDRs) will validate the ad hoc designed and shared SQFFQ, as previously performed [28-29]. The SQFFQ will be administered to the validation study group during three non-consecutive calls, including one non-weekday (30 minutes 24-h-recall computer-aided personal interview).

Case-control study: an international, multicentre, retrospective case-control study of prospectively enrolled patients will be performed to evaluate the associations of dietary, anthropometric, lifestyle factors in eoCRC patients compared to age, and sex matched healthy controls.

This study is sub-classified as a low-risk intervention study as it will perform fecal occult blood test as an additional procedure in the control group.

Inclusion criteria All sexes eligible Cases: eoCRC diagnosed between 18 and 49 years and confirmed by histology (biopsy or surgical specimen in case of surgery) Controls: negative past and present history of cancer; negative fecal occult blood test (FOBT), or negative colonoscopy.

Exclusion criteria CRC diagnosed at ≥ 50 years Diseases that can modify the dietary regimen (celiac disease, diabetes) Diseases that are known to predispose to eoCRC (personal past or recent history of inflammatory bowel disease, past history of pelvic irradiation) Unable to give written consents and to fill in the electronic questionnaire

Sample size

SQFFQ Validation study:

• 100 subjects from each country, free from overt disease and in equal numbers of each gender.

Case-control study:

• At least 300 eoCRC patients in total;
• At least 600 healthy age- (matching range ± 5 years) and sex-matched controls subjects in total.

Study Procedure and Study Flow-chart Starting from the signature of the informed consent by the subject/patient (or subject/patient's legal tutor), she/he will be considered enrolled in the study.

The PI will be the only Administrator of the online platform and the only one able to generate the access codes (username/password) assigned to the co-investigators.

Each co-investigator will register the enrolled subjects anonymously on the online platform (unique alphanumeric identification code - see below), obtaining a username/password assigned to the subject for survey completion.

The unique alphanumeric identification code (Subject ID number) will be assigned consecutively in increasing order starting from '001'. A screened subject/patient identification list will be kept by PI.

SQFFQ validation study:

One hundred volunteers from each country, free from overt disease and in equal numbers of each sex, will be enrolled for the semi-quantitative food frequency questionnaire (SQFFQ) validation study. Potential volunteers will be recruited through different recruitment strategies: mail drops, recruitment lists, and databases, primary care facilities as well as networks for young people and/or among people who work in the same hospital of the research team, regularly followed by preventive medicine.

An ad hoc-designed and shared online SQFFQ will be available to report the usual frequency of consumption of a detailed list of foods and beverages. Information collected concern types, amount and frequency of consumption of food and drinks, types of seasoning, and methods of cooking. The computer-administered instrument will allow the respondent to select the food consumed and the appropriate portion size from photographs on a screen reducing the burden of coding.

All volunteers will receive detailed oral and written information by members of the study team (following local practices).

Three non-consecutive 24-hour Dietary Recalls (24hDRs), including one non-weekday (30 minutes 24-h-recall computer-aided personal interview), will validate the SQFFQ [28-29].

Case-control study:

An international, multicentre, retrospective case-control study of prospectively enrolled patients will be performed to evaluate the associations of dietary, anthropometric, lifestyle factors in eoCRC patients compared to age, and sex matched healthy controls.

This study is sub-classified as a low-risk intervention study as it will perform fecal occult blood test as an additional procedure in the control group.

The study will recruit:

• at least 300 patients in total (100 at IRCCS San Raffaele Scientific Institute) with a recent diagnosis of eoCRC (diagnosis made within 2 years prior to enrollment), aged 18-49 years, will be prospectively recruited. eoCRC patients will be primarily enrolled as referred by the participant unit. Each participant unit shall identify eoCRC patients through recruitment lists and databases, primary care facilities, as well as networks for young people. Retrospective data (dietary and lifestyle factors) related to the 2 years prior to eoCRC diagnosis will be collected through an online platform.
• at least 600 Healthy controls (HCs) totally (200 at IRCCS San Raffaele Scientific Institute), matched by age (matching range ± 5 years) and sex with eoCRCs will be prospectively enrolled. Healthy volunteers will be mainly enrolled among workers within the participating hospital center, followed regularly by preventive medicine. This enrollment will be carried out by e-mail invitation disseminated through the hospital's official mailing list. The enrolled healthy volunteers will perform fecal occult blood test as an additional procedure and will be enrolled if this test is negative. Retrospective data (dietary and lifestyle factors) related to the 2 years prior to recruitment will be collected through an online platform.

The online platform will be divided into two separate sections: (i) one completed by cases / controls, concerning the semi-quantitative food frequency questionnaire (SQFFQ), lifestyle habits and anthropometric data; (ii) one completed by doctors, concerning clinical data.

The semi-quantitative food frequency approach will ask to report their usual frequency of consumption of each food and drink, referred to the two years before eoCRC onset, following detailed oral and written instructions from the members of the study team. Information collected will concern types, amount and frequency of consumption of food and drinks, types of seasoning, and methods of cooking. A computer-administered instrument will allow the respondent to select the food consumed and the appropriate portion size from photographs on a screen reducing the burden of coding.

All cases and healthy controls will also provide a set of covariates as part of the online platform: date of birth, sex, ethnicity, weight (kg)/height (m)/BMI (kg/m2) at the time of eoCRC diagnosis and at 18 years old, country where the patient/healthy control lives permanently, tobacco smoking at the time of eoCRC diagnosis and at 18 years old, sitting time, TV-viewing time, moderate-to-vigorous physical activity (MVPA), waist circumference (cm), home blood pressure levels (mmHg), fasting blood glucose (mg/dl), regular consumption of aspirin/NSAID, calcium and folate supplements, oral contraceptive agents, post-menopausal hormones and years of consumptions, if the filled questionnaire reflects diet for the last 5-10 years before.

Other clinical data will be collected by doctors as part of the online platform: date of eoCRC diagnosis, symptoms at diagnosis, eoCRC localization, eoCRC stage, histological diagnosis, type of surgery and date (if performed), chemoterapy and radiotherapy (if performed), vital status and duration of follow-up, family history of CRC and other cancers (uterus, ovary, stomach, small intestine, urinary tract/bladder/kidney, bile ducts, brain, pancreas, skin tumors), type of germline pathogenetic variant (if performed).

4.6 Responsibilities of the Investigator(s) The Investigator(s) undertake(s) to perform the study in accordance with this Protocol, Good Clinical Practice and the applicable regulatory requirements.

The Investigator is required to ensure compliance with the investigational schedule, and procedures required by the protocol.

The Investigator agrees to provide all information requested in the Case Report Form -CRF- in an accurate and legible manner.

ETHICAL CONSIDERATIONS Ethical conduct of the study The study will be performed according the ethical principles laid down in the latest accepted version of the Declaration of Helsinki.

Patient information and informed consent Each subject/patient will be informed about the modalities of the clinical study in accordance with the enclosed patient information. The subject/patient is to be informed both in writing and verbally by the investigating physician. The subject/patient must be given opportunity to decide whether or not to participate in this study and to ask questions concerning this. It must also be made clear to the subject/patient that he/she can withdraw from the study at any time without giving reasons and that he/she will not be in any way disadvantaged by this. The points mentioned in the information sheet must be communicated to the patient in language he/she understands. The informing physician and the patient must each personally date and sign an informed consent form with a declaration on data privacy. Any informed consent will be part of the investigator's file and retained with it. The subject/patient will retain a copy of the patient information.

Data management The personal data will be recorded in the electronic CRF designed for this study. All CRF will be checked for completeness, plausibility, and compliance with the ICH guidelines and the institutional Standard Operating Procedures (SOPs).

The personal data will be identified with a code (see section 6.7) not allowing directly trace the identity of the patients/individuals, except in case of need. In all phases of collection and processing of personal data, suitable measures will be adopted to guarantee their protection from the risks of unauthorized access, theft or loss, also through the use of encryption techniques, identification codes or other solutions that allow identify the interested parties only in case of need and to those who are expressly authorized to access them, in order to minimize the risks of accidental knowledge and unauthorized or abusive access to data. The genetic and clinical data of the interested parties are, however, treated separately from other "common" personal data (name, surname, etc.), which allow the interested parties to be directly identified.

The collection, management and analysis of data will take place in a way that guarantees the confidentiality of the subject's identity and will follow the General Data Protection Regulations (GDPR).

With regard to the possible transfer of data to third countries, the processing will take place according to one of the methods permitted by current law, such as the consent given by the patients/individuals, the adoption of Standard Clauses approved by the European Commission, the selection of subjects adhering to international programs for the free circulation of data (eg EU-USA Privacy Shield) or operating in countries considered safe by the European Commission.

All analyses will be performed within respective limitations of each ethics committee approval and in accordance with good clinical practice and the Declaration of Helsinki.

Sample size

SQFFQ Validation study:

100 subjects from each country, free from overt disease and in equal numbers of each gender.

Case-control study:

Power analysis for odds ratios calculation The odds ratios of eoCRC for different exposition factors will be calculated.

For this purpose, a power analysis is performed, based on the following parameters:

• Significance level (alpha, 1 type error probability) = 0.05 (two tails test)
• Power of the analysis = 80%
• Number of controls per case = 2
• Minimum Odds Ratio to detect = 2
• Proportion of exposed in the general population The proportion of exposed individuals in the general population depends on the considered exposition factors (such as a specific diet habit, alcohol consumption, etc.): for this reason, the sample size was calculated for different exposition proportions, varying from 5% to 95%.

Statistical analysis

SQFFQ validation Study:

The validity of the SQFFQ will be computed by calculating the Spearman correlation coefficients between individual food groups intakes from SQFFQ and the average of three 24hDR food groups intakes as the reference method.

The percentage of the agreement as the proportions of individuals who were classified correctly into the same or adjacent quintile for validity analyses will be calculated.

Case-control Study:

In the case-control study, an association between eoCRC and dietary habits (or other exposition factors such as alcohol consumption, smoking habits, etc.) will be estimated using logistic regression and odds ratio calculation. Multivariate logistic regression models will be used to examine to what extent eoCRC development can be 'explained' by variations in eating habits. The stratified analysis will also be performed by considering only a specific gender and/or other grouping variables of interest such as age, age at diagnosis. The effects of potential confounders other than matching criteria (e.g., smoking, physical activity, etc.) will also be estimated and taken into account. All analyses are based on a conservative estimate of eoCRC cancer cases, with at least 2 controls per case.