Short Communication

Alternative measures of Brassica vegetable consumption (e.g., cabbage) may clarify the association between Brassica and cancer risk. Brassica isothiocyanates (ITCs) are excreted in urine and may provide a sensitive and food-specific dietary biomarker. However, the persistence of ITCs in the body may be brief and dependent on the activity of several Phase II enzymes, raising questions about the relationship between a single ITC measure and habitual dietary patterns. This study investigates the association between urinary ITC excretion and habitual Brassica consumption, estimated by a food frequency questionnaire, among healthy Chinese women enrolled in the Shanghai Breast Cancer Study. Participants (n 347) completed a validated food frequency questionnaire querying habitual dietary intake during the prior 5 years and provided a fasting first-morning urine specimen. Genetic deletion of glutathione S-transferases (GSTM1/ GSTT1), and single nucleotide substitutions in GSTP1 (A313G) and NAD(P)H:quinone oxidoreductase 1 (NQO1: C609T), were identified from blood DNA. Urinary ITC excretion levels were marginally higher with the GSTT1-null or GSTP1-G/G genotypes (P 0.07, P 0.05, respectively). Mean habitual Brassica intake was 98.3 g/day, primarily as bok choy, and Brassica intake significantly increased across quartile categories of ITC levels. The association between habitual Brassica intake and urinary ITC levels was stronger among women with GSTT1-null or GSTP1-A/A genotypes, or NQO1 T-allele, and the interaction was statistically significant across GSTP1 genotype. In conclusion, a single urinary ITC measure, in conjunction with markers of Phase II enzyme activity, provides a complementary measure of habitual Brassica intake among Shanghai women