Background: Although CYP2D6 has been studied extensively in different population groups, relatively little is known for black Americans. Methods. CYP2D6 activity was assessed with dextromethorphan in 283 black American subjects and correlated with their genotype (2D6*2 to *12, 2D6*14, 2D6*15, 2D6*17, 2D6*18, and 2D6*29 and gene duplications). Volunteers provided information about ethnicity and concurrent medication, and they participated in either phenotyping (n = 225), genotyping (n = 251), or both (n = 193). Results. The median urinary dextromethorphan/dextrorphan metabolic ratio (MR) indicated significantly lower CYP2D6 activity in the black American group (0.016) than in a white control population (0.0044; P = .0001) studied previously. The reduced function allele 2D6*17 was more common (frequency [f] = 0.395) among intermediate metabolizers (0.03 less than or equal to MR less than or equal to 0.3) than extensive metabolizers (MR less than or equal to 0.03; f = 0.148; P = .0001). Consistent with reduced function toward dextromethorphan of COS cell-expressed 2D6.29 protein, 2D6*29 also was more frequent in intermediate metabolizers (f = 0.114) than in extensive metabolizers (f = 0.057; P = NS). Frequencies for 2D6*17 and 2D6*29 were f = 0.213 and 0.072, respectively. Of the 193 genotyped and phenotyped subjects, 14 were determined to be poor metabolizers, with dextromethorphan/dextrorphan ratios >0.3 (7.25%), but only 2 subjects (1.04%) carried 2 nonfunctional alleles (2D6*3/*4x2 and 2D6*4/*4). A new allelic variant, 2D6*40, was subsequently found in 2 discordant subjects (2D6*4/*40 and 2D6*6/*40), implying that the 18-base pair (bp) insertion found in 2D6*40 renders it nonfunctional. The frequency of 2D6*40 was 0.006. For genotypes that contain 2D6*2, median MR values were consistently higher in black Americans than in white subjects, indicating that other unidentified factors also contribute to lower CYP2D6 activity in black Americans. Conclusions. The lower CYP2D6 activity observed in a black American population is in part attributable to the presence of variant alleles that occur at a higher frequency in this population than in white subjects. Additional studies are required to ascertain the pharmacokinetic and pharmacodynamic consequences of these pharmacogenetic data in black Americans.
