Overexpression of cyclooxygenase-2 (COX-2) has been shown to play a major role in colorectal cancer pathogenesis. However, no human study has directly investigated whether biomarkers of COX-2 overexpression may predict colorectal cancer risk. We evaluated the association of urinary prostaglandin E₂ metabolite (PGE-M) levels and colorectal cancer risk

A nested case-control study was conducted within the Shanghai Women's Health Study, in which 74,942 Chinese women ages 40 to 70 years were recruited from 1997 to 2000. Urinary PGE-M in 150 cohort members who developed colorectal cancer during the follow-up were compared with 150 matched controls.

The baseline level of urinary PGE-M was more than 50% higher in cases than in controls. The relative risks (RRs) of developing colorectal cancer were elevated from 1.0 to 2.5 (95% CI, 1.1 to 5.8), 4.5 (95% CI, 1.9 to 10.9), and 5.6 (95% CI, 2.4 to 13.5) with increasing quartiles of urinary PGE-M levels (P for trend < .001). The positive association was observed for both colon cancer (RR = 4.9; 95% CI, 1.7 to 14.7 for the highest v lowest quartile; P for trend = .009) and rectal cancer (RR = 7.2; 95% CI, 1.7 to 30.7; P for trend = .048), and for colorectal cancer cases diagnosed in the first 30 months (RR = 7.6; 95% CI, 1.8 to 32.0; P for trend = .035) and subsequent months (RR = 4.4, 95% CI, 1.5 to 13.3; P for trend = .012) of follow-up.

Given its strong association with colorectal cancer risk, urinary PGE-M may be a promising biomarker for risk assessment of this common malignancy.