Investigation of the prevalence and number of aberrant crypt foci associated with human colorectal neoplasm.

BACKGROUND Aberrant crypt foci (ACF) are considered to be useful as surrogate biomarker for colorectal cancer (CRC), but the biological significance of ACF remains controversial. We attempted to investigate the relationship between the presence of ACF and human colorectal carcinogenesis using a relatively large sample size. METHODS We carried out high-magnification chromoscopic colonoscopy to identify ACFs in 861 subjects undergoing a diagnostic endoscopy at the Yokohama City University Hospital. The present study compared the prevalence and number of ACFs in three subject groups (normal subjects, adenoma cases, and CRC cases). The correlations between the demographic and behavioral characteristics of the subjects and the prevalence of ACFs were also assessed. RESULTS The prevalence of ACF was 64%, 88%, and 95%, and the mean number of ACF was 3.6, 6.2, and 10.1, in normal subjects, adenoma cases, and CRC cases, respectively. When differences in the prevalence and number of ACFs among age- and sex-stratified subject groups were examined, significant stepwise increments from normal subjects to adenoma cases to CRC cases were apparent (P < 0.001). Moreover, an age- and sex-adjusted multiple logistic regression analysis revealed that smoking and alcohol habits had a synergistic effect, increasing the prevalence of ACFs as well as the risk of CRC (P < 0.001). CONCLUSIONS These results suggested that ACF may serve as a reliable surrogate biomarker for human colorectal carcinogenesis. IMPACT The use of ACF as an endpoint may enable the size, duration, and cost of CRC chemoprevention studies to be reduced.