Modifiable risk factors and colorectal adenomas among those at high risk of colorectal cancer

Epidemiological studies have identified several modifiable risk factors for colorectal neoplasms in the general population. However, associations between modifiable risk factors, including body mass index (BMI), smoking, alcohol consumption and dietary patterns, and colorectal neoplasms in two groups at high risk of colorectal cancer, Lynch syndrome patients and sporadic adenoma patients, have been sparsely studied.

This thesis presents two cohort studies, one of 486 Lynch syndrome patients (the GEOLynch cohort study) and one including data from 565 persons with sporadic adenomas (the POLIEP follow-up study), in which we assessed whether a high BMI, smoking, high alcohol consumption and specific dietary patterns influenced colorectal adenoma development. We also assessed whether the association between BMI and recurrence of sporadic adenomas was modified by polymorphisms in the insulin-like growth factor (IGF) genes.

First, we observed that excess body weight increased the risk of incident colorectal adenomas in men with Lynch syndrome. Secondly, we showed that current smoking increased the risk of colorectal adenomas in Lynch syndrome in both sexes. Former smokers still showed an elevated risk, but lower than current smokers. Number of years smoked, among ever smokers, was positively associated with colorectal adenomas. A clear association with alcohol consumption was not observed. Thirdly, we identified four dietary patterns in the Lynch syndrome cohort; i) ‘Prudent’, ii) ‘Meat’, iii) ‘Snack’, vi) ‘Cosmopolitan’. The ‘Snack’ pattern was associated with increased adenoma occurrence. The other patterns showed Hazard Ratios in the expected directions based on similar studies in the general population but these were not statistically significantly associated with adenoma occurrence. Additionally, among 565 sporadic adenoma patients, we found that BMI was not associated with adenoma recurrence (n=165), nor with recurrence of advanced adenomas (n=37) after a median of 4.7 years of follow-up. Variation in IGF-axis genes (rs1520220 in IGF1 and rs3213221 in IGF2) influenced the likelihood of colorectal adenoma recurrence. Furthermore, we observed that the association between BMI and adenoma recurrence was modified by variation in the IGF2 gene (rs1004446 and rs1003483). Finally, the three dietary patterns identified (‘Low meat’, ‘Cosmopolitan’, or ‘Refined foods’) among the sporadic adenoma patients did not show marked associations with adenoma recurrence, although the ‘Low meat’ pattern might reduce the risk of advanced recurrences. No significant associations were seen for smoking and alcohol consumption.

Overall, the results of our Lynch syndrome cohort suggest that modifiable risk factors, e.g. high BMI and smoking, influence colorectal adenoma development in Lynch syndrome patients. On the other hand, these risk factors do not appear to influence recurrence of sporadic colorectal adenomas.