Table 5.
VKORC1 haplotype frequency and influence on warfarin dose among European–American and African–American participants.
| Identification code |
Haplotype sequence |
Haplotype frequency (%) |
p-value | Mean maintenance dose (mg/day) (95% CI)* |
|
|---|---|---|---|---|---|
| Number of copies of VKORC1 haplotype | |||||
| 2 copies¶ | 1 or 2 copies# | ||||
| European-Americans‡ | |||||
| WHT1 | TGTCCGCA | 37.1 | Referent | 6.0 (5.4, 6.6) | 5.5 (5.2, 5.8) |
| WHT2 | TGTTCGCG | 23.9 | 0.99 | 5.2 (4.1, 6.2) | 5.6 (5.3, 6.0) |
| WHT3* | CAGCTCTG | 23.3 | <0.0001 | 3.1 (2.1, 4.1) | 4.5 (4.1, 4.9) |
| WHT4* | CATCTCTG | 11.3 | <0.0001 | 3.2 (1.8, 4.7) | 4.0 (3.5, 4.6) |
| WHT5 | TAGCTCTG | 1.1 | NA | NA | NA |
| WHT6 | CGTTCGCG | 0.7 | NA | NA | NA |
| African–Americans§ | |||||
| BHT1 | TTGCGCGCCAC | 24.4 | Referent | 6.3 (5.4, 7.3) | 6.1 (5.8, 6.6) |
| BHT2 | CCGCGCCCCGC | 16.2 | 0.96 | 6.5 (4.6, 8.4) | 6.1 (5.7, 6.6) |
| BHT3 | TCGCGCGCTAC | 15.3 | 0.73 | 5.7 (4.4, 7.1) | 6.1 (5.6, 6.6) |
| BHT4 | CTGCGCGTCGC | 9.6 | 0.018 | 5.8 (3.6, 8.1) | 5.4 (4.8, 6.0) |
| BHT5 | CTGCACGCCGC | 6.8 | 0.28 | NA | 5.9 (5.2, 6.6) |
| BHT6* | CTACGTCTCGC | 5.6 | 0.0028 | NA | 4.8 (4.2, 5.7) |
| BHT7 | TTGTGCGCCGC | 5.1 | 0.63 | NA | 6.0 (5.3, 6.7) |
| BHT8 | TTGCGCGCCGC | 4.9 | NA | NA | NA |
| BHT9 | CCGCGCGCTAC | 4.2 | NA | NA | NA |
| BHT10 | CTACGTCTCGT | 3.9 | NA | NA | NA |
| BHT11 | TTACGTCTCGT | 1.1 | NA | NA | NA |
| BHT12 | CTGCGCGCCAC | 1.1 | NA | NA | NA |
Five European–Americans and eight African–Americans individuals with more than 50% missing genotypes were excluded from the analysis.
For European–Americans each haplotype sequence SNPs are listed in the sequential order along the VKORC1 gene (rs7196161, rs9923231, rs2884737, rs17708472, rs9934438, rs8050894, rs2359612, rs7294). For African–Americans each haplotype sequence SNPs are listed in the sequential order along the VKORC1 gene (rs7196161, rs17878544, rs9923231, rs17708472, rs13336384, rs9934438, rs8050894, rs2359612, rs7200749, rs7294, rs17882368). Haplotypes with frequencies <0.5% are not presented.
The low prevalence did not allow multivariable evaluation.
Haplotypes independently predictive of warfarin dose are bolded.
Haplotypes independently predictive of warfarin dose after Tukey–Kramer adjustment for multiple comparisons.
For European–Americans multivariable model adjusted for significant (p < 0.2) clinical covariates (age, gender BMI, average vitamin K intake, use of amiodarone) and CYP2C9 (variant vs wild-type) and VKORC1 haplotype. CYP2C9 variant alleles assessed include *2 *3.
For African–Americans multivariable model adjusted for significant (p < 0.2) clinical covariates (age, gender BMI, average alcohol intake, presence of renal failure, presence of CHF, use of amiodarone) and CYP2C9 (variant vs wild-type) and VKORC1 haplotype. CYP2C9 variant alleles assessed include *2, *3, *5, *6 and *11.
Mean warfarin dose requirements among patients possessing two copies of VKORC1 haplotype.
Mean warfarin dose requirements among patients possessing one or two copies of VKORC1 haplotype.
Each haplotype is represented as a covariate.
Numbering of SNPs varies by race as we constructed race-specific haplotypes and haplotype groups.
CHF: Congestive heart failure; NA: Not analyzed as haplotype frequencies <5%.