Carbohydrate and fibre intake, short-chain fatty acid genetic modifiers, and colorectal cancer risk: a prospective analysis in the UK Biobank

Background and objectives: Wholegrains and fibre intake may decrease the risk of colorectal cancer, although evidence for intake of specific types and sources of carbohydrates and fibre remains unclear. Moreover, cellular and animal studies have shown that short-chain fatty acid (SCFA) production in the gut, which increases with wholegrains and fibre intake, may be associated with lower colorectal cancer risk. However, it is unknown if genetic factors that influence SCFA production may modify the association of carbohydrate quality with colorectal cancer risk. Our objective was to assess the associations of intakes of carbohydrate and fibre sources with risk of colorectal cancer and further explore if associations were modified by a SCFA polygenetic score. Methods: Using a minimum of 2 (maximum of five) 24-hour dietary assessments, we assessed carbohydrate types and sources and fibre intake in 114,271 UK Biobank participants who were free from cancer at dietary assessments. We used existing data to create a polygenic score to categorise participants as high or low intestinal SCFA producers, namely butyrate and propionate. Multivariable Cox proportional hazards models were used to determine the associations between intakes of carbohydrate types and fibre sources, and colorectal cancer incidence. Results: During a median follow-up of 8.8 years, 1,129 participants were diagnosed with colorectal cancer. In multivariable adjusted models, intakes of non-free sugar and fibre from wholegrains were inversely associated with colorectal cancer risk (hazard ratio (HR) per 5% energy/day: 0.92, 95% confidence interval (CI):0.87-0.98; HR per 5g/day: 0.90, 0.821.00, respectively). There was evidence of heterogeneity by host genetically predicted SCFA production; participants with high genetically predicted propionate production and consuming higher wholegrains and fibre from wholegrains had a lower risk of colorectal cancer (per 5% energy/day:0.89, 0.80-0.99; and per 5g/day:0.84, 0.70-1.00, respectively) whereas no association was observed for participants with low genetically predicted production of propionate (p-heterogeneity=0.028 and 0.045, respectively). Conclusion: This study suggests that colorectal cancer risk varies by intake of different types and sources of carbohydrates; higher consumption of non-free sugars and fibre from wholegrains may lower the risk of colorectal cancer, and these associations may be modified by host genetically predicted SCFA production.