Preclinical and clinical development of biomarkers of the efficacy of curcumin in the prevention and treatment of colorectal cancer

This project explored the suitability of three biological indices as potential markers of the efficacy of curcumin in preventing or treating colorectal carcinoma: glutathione S-transferase (GST) activity; a deoxyguanosine adduct (M1G) associated with oxidative DNA damage and malondialdehyde (MDA) production via lipid peroxidation (LPO); and levels of cyclooxygenase-2 (COX-2) protein and its product prostaglandin E2 (PGE2). The viability of these biomarkers was studied at three levels: malignant and non-malignant human colon cells in vitro; a rat model of LPO and the multiple intestinal neoplasia (MIN) mouse model; and a pilot study of a standardised oral preparation of Curcuma extract in patients with advanced colorectal cancer.;The results in vitro demonstrated an association between COX-2 activity and intracellular MDA levels, and a correlation in malignant cells between MDA concentration and M1G adduct levels. A diet containing 2% curcumin decreased M1G levels in rat colon mucosa and increased GST liver activity, compared to controls; the same diet completely prevented chemically induced rises in M1G levels in colon mucosa. Elevated levels of M1G were discovered in intestinal adenomas in MIN mice, relative to normal mucosa, and 0.1% dietary curcumin significantly decreased those levels. Although addition of curcumin to human blood in vitro was shown to inhibit leukocyte COX-2 activity, dose-dependent effects definitely attributable to treatment were not observed in the three biomarkers measured in the clinical trial of Curcuma extract. Compatible with the tissue levels described in the preclinical studies, curcumin's systemic bioavailability following oral dosing appears to be low in humans.;Future development of curcumin as a cancer chemopreventive agent should study its effects on M1G levels, COX-2 activity and GST isoforms in colorectal adenomas.