| Summary: | <p><strong>Background:</strong> Oesophageal, stomach and colorectal cancers are three major gastrointestinal cancers that together account for approximately 1 million deaths in China annually. Certain dietary factors and adiposity traits have been associated with the risks of these cancers in Western populations, but prospective evidence is still limited in China, where dietary patterns, adiposity levels, and the burden and subtypes of these cancers differ from the West.</p>
<p><strong>Methods:</strong> During 2004-2008, the prospective China Kadoorie Biobank (CKB) study recruited 512,715 adults (41% men) aged 30-79 (mean 52) years from 10 diverse areas in China, including one high-risk area for oesophageal cancer (Huixian in Henan province). Self-reported dietary intake, objective measurements of body size and composition, and other socio-demographic and lifestyle factors were collected at baseline and two subsequent resurveys. During ~10 years of follow-up, 26,594 incident cancer cases were recorded (2350 oesophageal [1198 from Huixian]; 3350 stomach; 3061 colorectal) in participants without a prior history of cancer. Cox regression was used to estimate adjusted hazard ratios (HRs) for each cancer associated with the consumption of major food groups (including spicy food) and with different adiposity traits.</p>
<p><strong>Results:</strong> At baseline, 28%, 47%, 23% and 10% of participants reported regular intake (≥4 days/week) of fruits, red and processed meat, preserved vegetables and soy products, respectively, while 30% reported daily intake of spicy food. The mean (SD) baseline BMI and body fat percentage were 23.7 (3.4) kg/m² and 27.9 (8.4)%. The consumption of fruits, soy products and spicy food were inversely associated with oesophageal cancer risk (HR=0.85; 95%CI 0.78-0.93 per 50g/day higher usual fruit intake, 0.82; 0.73-0.92 per 25g/day higher usual soy intake, and 0.81; 0.71-0.93 for daily vs. no spicy food intake), while daily consumption of preserved vegetables was positively associated with risk (1.25; 1.08-1.46). For stomach cancer, only fruit intake showed a statistically-significant association (0.92; 0.86-0.97 per 50g/day), while for colorectal cancer none of the food groups examined showed clear significant associations. All general (BMI, fat-percentage, fat and fat-free mass) and central (waist circumference and waist-to-hip ratio) adiposity measures were inversely associated with oesophageal cancer risk, in an approximately log-linear manner (e.g. 0.73; 0.68-0.79 per 5kg/m² higher usual BMI). General and central adiposity (except fat-free mass) were also inversely associated with the risk of stomach cancer. The above inverse associations persisted after excluding participants with self-reported prior chronic diseases and the first three years of follow-up. Conversely, BMI at age 25 years was positively associated with these cancers. For colorectal cancer, all adiposity measures, except BMI at age 25, were positively associated with risk (e.g. 1.15; 1.09-1.22 per 5kg/m² higher usual BMI).</p>
<p><strong>Conclusion:</strong> The findings in this Chinese population were generally consistent with those reported in previous large cohorts, with several exceptions (e.g. meat and colorectal cancer). The same dietary factor and adiposity trait appeared to affect the risks of the three cancers differently.</p>
|
|---|