摘要:Background : The aetiology of colorectal cancer has strong dietary links, and there may be an association between coffee and colorectal cancer risk. Objective : To study the effects of filtered (low levels of kahweol/cafestol) and unfiltered (high levels of kahweol/cafestol) coffee on tumour formation in multiple intestinal neoplasia ( Apc Min/+ ) mice. Design : Apc Min/+ mice ( n =11 per group) were fed for 9 weeks with 10% w/w of these two types of coffee. Coffee was served as a dietary ingredient mixed with a semi-synthetic AIN-93G-based diet. Plasma levels of caffeine and paraxanthine were used as compliance markers. At the end of the feeding period intestinal tumour number and size were determined. The levels of β-catenin and cyclin D1, two cell-signalling proteins important to the progression of neoplasia, were also analysed in the tumour tissue. Results : Plasma caffeine and paraxanthine concentrations were 3.2±1.4 and 1.7±0.4 µmol l -1 in the filtered coffee group and 3.6±2.3 and 1.6±0.6 µmol l -1 in the unfiltered coffee group. The level of plasma xanthines was below detection in the control group. The total number of tumours was equal between the dietary groups: 29 for the control, 30 ( p =0.767) for the filtered coffee and 29 ( p =0.430) for the unfiltered coffee groups. The levels of β-catenin and cyclin D1 in the nuclear fraction of the tumour tissue were also the same between the groups. Conclusions a : Filtered or unfiltered coffee (10% w/w) does not exert antitumorigenic activity in Apc Min/+ mice or change β-catenin and cyclin D1 signalling in the adenoma tissues. The results suggest that coffee does not change neoplasia progression in this animal model.