Ability of aberrant crypt foci characteristics to predict colonic tumor incidence in rats fed cholic acid.

Aberrant crypt foci (ACF) are putative preneoplastic lesions of colon cancer which are being utilized currently as a biological end point to evaluate the induction and modulation of colon carcinogenesis. In several previous short-term studies, the unexpected reduction of ACF by the reported colonic tumor promoter cholic acid (CHA) emphasized the need for a systematic evaluation of the growth of ACF in response to a tumor promoter. The present study was conducted to determine if any characteristic(s) of ACF at various early stages of carcinogenesis would predict resulting tumor incidence in rats fed CHA. Male Sprague-Dawley rats received two injections of azoxymethane (20 mg/kg) and were fed either the AIN-76 diet or AIN-76 plus 0.2% CHA. The number, crypt multiplicity (number of crypts/focus), and size (area) of ACF were measured after 2, 8, 14, and 18 weeks in 5 rats/group. The number of ACF was lower (P < 0.033) in animals fed CHA at all time points. Average crypt multiplicity of ACF was greater (P = 0.045) from CHA-fed animals after 8 weeks compared to animals fed the AIN-76 diet. The average size of ACF was smaller in CHA-fed animals after 2 weeks and then tended to be larger than the sizes of the ACF from animals fed the AIN-76 diet. All remaining animals were killed after 18 weeks. Tumor incidence was higher (P < 0.001) in the CHA-fed group (63.2%) compared to the control diet group (29.4%). CHA-fed rats also had a higher number of tumors/tumor-bearing rat compared to control diet rats (1.96 versus 1.13). The main finding of this study is that the number of ACF at early time points did not predict tumor incidence. Crypt multiplicity was a consistent predictor of tumor outcome and should be measured in future studies using ACF as a biological end point. The CHA diet appears to provide a unique tumor-modulating environment that selectively enhances the growth of a smaller number of ACF leading to an increased number of tumors compared to a control diet. The mechanism(s) by which CHA mediates this effect warrants further investigation.