Table IV.
Joint effect of NAT2 phenotype and GST genotypes on risk of bladder cancer among non-Hispanic whites by level of exposure to bladder carcinogens, The Los Angeles Bladder Cancer Study
| NAT2 phenotype status by level of exposure to bladder carcinogensa | Summed GST genotype indexb |
||
| 0 | 1 | 2–3 | |
| All subjects | |||
| Rapid acetylators | |||
| Cases/controlsc | 72/99 | 120/114 | 51/44 |
| OR (95% CI)d | 1.00 (referent) | 1.44 (0.93–2.21) | 1.43 (0.83–2.48) |
| Slow acetylatorsc | |||
| Cases/controlsc | 91/113 | 155/134 | 66/37 |
| OR (95% CI)d | 0.88 (0.56–1.38) | 1.50 (0.99–2.27) | 2.27 (1.32–3.91) |
| Low exposure | |||
| Rapid acetylators | |||
| Cases/controlsc | 55/85 | 91/100 | 29/39 |
| OR (95% CI)d | 1.00 (referent) | 1.34 (0.83–2.14) | 1.12 (0.60–2.08) |
| Slow acetylators | |||
| Cases/controlsc | 54/96 | 103/123 | 44/34 |
| OR (95% CI)d | 0.73 (0.44–1.21) | 1.21 (0.77–1.92) | 1.77 (0.97–3.22) |
| High exposure | |||
| Rapid acetylators | |||
| Cases/controlsc | 17/14 | 29/14 | 22/5 |
| OR (95% CI)d | 1.00 (referent) | 2.00 (0.72–5.53) | 3.96 (1.12–14.03) |
| Slow acetylators | |||
| Cases/controlsc | 37/17 | 52/11 | 22/3 |
| OR (95% CI)d | 2.24 (0.84–5.94) | 5.12 (1.81–14.46) | 9.37 (2.15–40.86) |
See the Materials and Methods for the definition for the rapid/slow acetyplators. The high-exposure group consisted of smokers of ≥30 pack-years with high level of 4-ABP–Hb adducts (>30 pg/g Hb), whereas the low exposure group included all other subjects who never smoked or smoked <30 pack-years of cigarettes or had a relatively low level of 4-ABP–Hb adducts (≤30 pg/g Hb).
One was assigned to the GSTM1 null, GSTP1 BB or GSTT1 null genotypes, respectively, whereas zero was assigned to the GSTM1 non-null, GSTP1 AA or AB or GSTT1 non-null genotypes, respectively.
The number of bladder cancer patients/number of control subjects; a total of 555 bladder cancer patients and 541 control subjects who had complete data on all three GST genotypes, NAT2 phenotype and 4-ABP–Hb adducts were included in the analysis.
ORs were derived from unconditional logistic regression models that included the following covariates: age (years), gender, level of education (high school or below, 1–4 years of college, graduate school or above), currently smoking (no, yes), number of cigarettes smoked per day, number of years of smoking, total number of non-steroidal anti-inflammatory drug pills used over lifetime (non-user, <1441 and 1441+), ever use of permanent hair dyes (no, yes), ever held a high-risk job (no, yes) and 4-ABP–Hb adducts in tertile (for all subjects only).