Colorectal Cancer and Central Obesity

Menée à partir de données de la "UK Biobank" portant sur 458 543 personnes (äge médian : 57 ans ; 53 % de femmes), cette étude analyse la part des cancers colorectaux attribuable à un surpoids, qu'il soit évalué par l'indice de masse corporelle ou par le tour de taille et le rapport taille/hanches

Résumé en anglais

The proportion of colorectal cancer (CRC) cases attributable to excess weight, known as population attributable fraction (PAF), has been commonly based on measures of body mass index (BMI). Central obesity metrics, such as waist circumference (WC) and waist to hip ratio (WHR), are potentially better indicators of adiposity and have demonstrated stronger associations with CRC incidence.To examine PAFs of CRC cases that are attributable to high WC and WHR and compare them to those attributable to high BMI.This population-based UK Biobank cohort study included 458 543 individuals aged 40 to 69 years at recruitment (March 2006 to July 2010) living within a reasonable distance of the 22 assessment centers across the UK. The analyses were conducted between May and July 2024.Exposures were BMI, as a measure of general obesity, and WC and WHR, as indicators of central obesity.Hazard ratios (HRs) and corresponding 95% CIs were calculated for the associations of BMI, WC, and WHR with CRC incidence. The PAFs and 95% CIs of CRC cases attributable to high BMI, WC, and WHR were also calculated.A total of 458 543 participants (median [IQR] age, 57 [50-63] years; 244 351 [53.3%] female) were included in the study. During a median follow-up of 11.8 (IQR, 10.9-12.5) years, 5944 participants were diagnosed with CRC. The HRs for the association with CRC incidence were notably smaller for BMI (HR for the highest vs lowest BMI quartile, 1.23; 95% CI, 1.14-1.33) than for WC (HR for the highest vs lowest WC quartile, 1.37; 95% CI, 1.27-1.49) and WHR (HR for the highest vs lowest WHR quartile, 1.40; 95% CI, 1.29-1.51); these associations became comparable only after accounting for possible reverse causality by excluding the initial years of follow-up. Similarly, the PAF of CRC for high BMI was 9.9% (95% CI, 5.5%-14.4%), substantially lower than the PAFs for high WC and WHR, which were 17.3% (95% CI, 12.3%-22.1%) and 17.6% (95% CI, 12.9%-22.2%), respectively. After excluding the initial 7 years of follow-up, PAF estimates became analogous across all measures of obesity and were 15.7% (95% CI, 8.9%-22.4%) for BMI, 16.9% (95% CI, 9.8%-23.8%) for WC, and 18.0% (95% CI, 11.5%-24.6%) for WHR.In this cohort study of approximately half a million participants, the PAF of CRC attributable to excess weight, defined as high BMI, was considerably underestimated. The PAFs attributable to WC and WHR were consistent and much higher, underlining the importance of efforts to limit and overcome the obesity epidemic in CRC prevention.