. Author manuscript; available in PMC: 2012 Nov 1.

Published in final edited form as: Cancer Epidemiol Biomarkers Prev. 2011 Sep 22;20(11):2438–2449. doi: 10.1158/1055-9965.EPI-11-0649

Table 4.

Selected haplotype^a main effects on SCCHN risk, additive genetic model

Gene (haplotype definition)	haplotype	Race^b (% prevalence)	OR (95% CI)^c
ALDH2 (rs4767939, rs2238151, rs7312055, rs2158029, rs16941667, rs16941669)	ACAGCT	AA (26%)	1.0 (ref)
	ATGGCT	AA (11%)	0.5 (0.3–0.8)

CYP2E1 (rs915908, rs7092584, rs743535, rs2249695)	GCCC	EA (65%)	1.0 (ref)
	GCCT	EA (10%)	0.7 (0.6–0.9)

GPX2 (rs11623705, rs2412065, rs2737844)	GGC	EA (70%)	1.0 (ref)
	GCT	EA (9%)	0.7 (0.5–0.9)

SOD1 (rs4998557, rs10432782, rs2070424, rs1041740)	GTAC	EA (58%)	1.0 (ref)
	AGGC	EA (6%)	1.4 (1.1–1.9)

	GTAC	AA (52%)	1.0 (ref)
	AGGC	AA (6%)	0.6 (0.4–0.9)

Criterion for selecting haplotypes for this table: ORs were statistically significant, or nearly so, after Bonferroni correction for multiple testing (13 for EA, 12 for AA)

AA=African-American (black), EA=European-American (Caucasian/white)

Unconditional logistic regression models for estimating main effect of each haplotype were adjusted for matching variables sex and age category and their 2-way interaction, and for continuous smoking duration rounded to whole years. The referent group for each OR was the most common haplotype.