Results
A total of 86 patients treated with 110 mg DE twice daily were recruited for this study. All subjects were Chinese and had a mean (SD) age of 72.15±9.17 at the time of recruitment. Five SNPs were genotyped in these subjects, and all SNPs conformed to Hardy-Weinberg equilibrium (P > 0 05) (shown in Table 2). Eighty-three patients provided samples at the trough plasma level of dabigatran, and 58 patients provided samples at the peak plasma level of dabigatran. Peak and trough plasma concentrations of different genotypes are shown in Table 3.
There was a significant association between the CES1 SNP rs8192935 and trough plasma concentrations of dabigatran (P =0.013). For the trough levels, carriers of the G allele had increased trough plasma concentrations of dabigatran compared to the noncarriers. No significant association was observed between other SNPs and the plasma concentration of dabigatran (Table 3). Considering multiple-testing correction, we calculated the P value for the SNPs by false detection rate (FDR) for correction. However, rs8192935 remained not significantly associated after FDR correction (P =0.065).
In an exploratory analysis, ischemic and bleeding events related to patients treated with DE were compared among different genotypes of ABCB1 and CES1 (Table 4). Four of the SNPs were included in the final analysis. Clinical outcomes, including ischemic events and bleeding events, were recorded throughout the 1.5-year follow-up. All bleeding events were composed of major and minor bleeding events.
Overall, the incidence of ischemic events (stroke/SEE) in the included population was 10.47%. The incidence of ischemic events did not vary considerably among genotypes of ABCB1 or CES1. The incidence of any bleeding event in the included population was 19.77%. There was a significant association between all the SNPs and the incidence of any bleeding event (Table 4). Further analysis showed that there was a significant association between the ABCB1 SNP rs4148738 and major bleeding events in the addictive model (P =0.046, OR=3.29) and dominant model (P =0.040, OR=8.17). Additionally, the ABCB1 SNP rs1045642 was associated with the incidence of major bleeding events in the addictive model (P =0.043, OR=3.34) and dominant model (P =0.046, OR=7.77). However, no significant associations were found between all the SNPs and the incidence of minor bleeding events. Considering multiple-testing correction, we calculated the P value for the SNPs by false detection rate (FDR) for correction. However, no significant associations were found after FDR correction.