Abstract
BACKGROUND: A comprehensive inflammatory lifestyle score (CILS) incorporating both dietary and non-dietary factors is lacking. This study aimed to develop and validate CILS and assess its association with colorectal cancer risk.</p>
METHODS: Eligible participants in the UK Biobank with data on baseline lifestyle factors and serum biomarkers measured 2 years later were randomly divided into training (N = 4,108) and validation (N = 1,760) datasets based on sex, age, and race. Reduced rank regression followed by stepwise linear regression identified significant lifestyle factors (i.e., components of CILS) correlated with high-sensitivity C-reactive protein and white blood cell count from 24 a priori-selected inflammation-related factors. CILS was validated using multiple linear regression. Cox models were applied to examine the associations of CILS with incidence and mortality risk of colorectal cancer and subtypes in the remaining 182,814 participants.</p>
RESULTS: CILS comprised three anti-inflammatory factors [low body weight (body mass index <18.5), higher intake of fish, and apples/berries] and five pro-inflammatory factors [obesity/overweight, prolonged sedentary time (the highest tertile in the population), current smoking, whole-fat dairy, and processed meat], with higher CILS indicating a more pro-inflammatory lifestyle. Validation analyses demonstrated a 22.2% and 2.8% increase in high-sensitivity C-reactive protein and white blood cell count for each SD increment in CILS, respectively. Higher CILS was linked to increased colorectal cancer incidence (HRper SD increase = 1.12; 95% confidence interval, 1.08-1.16) and mortality (HRper SD increase = 1.18; 95% confidence interval, 1.08-1.28) and increased risk of all colorectal cancer subtypes except incident rectal cancer.</p>
CONCLUSIONS: The validated hypothesis-driven and empirically derived CILS was positively associated with colorectal cancer risk in the UK Biobank.</p>
IMPACT: CILS may offer promising applications in cancer prevention.</p>