Colorectal cancer is a significant cause of morbidity and mortality in indu
strialized societies and the second most frequent cause of cancer death in
the United States. Surrogate endpoint biomarkers are gaining wide acceptanc
e in early diagnosis and short-term cancer chemoprevention trials in place
of cancer endpoints. Molecular genetic biomarkers can be useful tools in id
entifying subjects at risk of developing cancer and screening for early can
cers amenable to complete cure. They may be useful both in predicting and a
ssessing response to a given therapy and in determining prognosis after an
initial diagnosis has been made. Ideally, biomarkers should fulfill some, i
f not all, of the following criteria: variability of expression between pha
ses of carcinogenesis, association with cancer risk, ability to undergo mod
ification in response to a chemopreventive agent, and finally, permit ease
of measurement. In consideration of colorectal cancer chemoprevention, seve
ral genetic biomarkers seem to meet many of these criteria, as they do exhi
bit distinct variability of expression at different phases of carcinogenesi
s, are often strongly associated with increased cancer risk (especially the
hereditary/familial syndromes), are generally able to be measured relative
ly easily through peripheral blood sampling (germline mutations) or by colo
nic mucosal sampling by endoscopic techniques (somatic mutations). In some
cases, genetic biomarkers have also been demonstrated to undergo modificati
on in response to a chemopreventive agent. With further understanding of th
e genetic and molecular changes involved in sporadic and familial colorecta
l carcinogenesis, genetic biomarkers appear to hold great potential for the
identification of subjects at high risk of developing colorectal cancer, a
s well as the development of novel chemopreventive approaches and form a pr
omising area for further research. (C) 2000 Wiley-Liss, Inc.