6 [1.4-9.3], Pc = 0.0242), but this correlation was abolished when hepatocellular type of injury was included (n = 42). There were no differences in the distribution of demographic characteristics, clinical findings, laboratory findings, and outcome among DILI patients classified MAPK inhibitor by the presence of a mutant allele of SOD2 or GPX1 and the wild-type

genotype. However, among the cholestatic/mixed cases, the mean age (61 years [range, 18-83 years]) was significantly higher in cases homozygous for the SOD2 Ala allele (P = 0.037). The total number of risk alleles (SOD2 C and GPX1 T alleles) was determined for each DILI patient and compared with those of the controls (Table 5). The presence of two or more risk alleles (n = 100) was significantly more frequent in DILI patients than in controls (OR = 2.1 [1.4-3.0], EPZ-6438 Pc = 0.0006), suggesting that these alleles constitute a cumulative effect on DILI susceptibility. The presence of a single risk allele (n = 52) was more frequently found in the controls (OR = 0.5 [0.4-0.8], Pc = 0.0097). Extending the risk allele analysis to also include the GSTM1 and GSTT1 null alleles determined in an earlier study17 showed a significant

risk of DILI development in the presence of four or more risk alleles (OR = 3.1 [1.7-5.6], Pc = 0.0004, n = 146). To further examine the role of risk alleles in DILI development, risk allele distribution was determined in DILI patients classified by time (days) to DILI onset. When divided into three distinct “time to onset” groups (≤30 days, 31-60 days, and >60 days), a clear trend was noted whereby DILI patients with higher number of risk alleles displayed a significantly shorter time to onset (P = 0.019) (Fig. 1). Among the patients with a time to onset of 30 or fewer days, 62% contained two or more

risk alleles, and 38% contained one or no risk allele. As the time to onset increased to 31 to 60 days or more than 60 days, the frequency of patients with two or more risk alleles decreased to 47% and 39%, respectively. The role of oxidative stress in DILI development is still relatively undefined. In this study, we have looked at functional polymorphisms in SOD2 and GPX1 that both lead to enhanced H2O2 generation, to seek potential associations between enhanced ROS net levels and risk of DILI development. Sclareol We found that carriers of the SOD2 Ala/Ala (CC) genotype or the GPX1 Leu/Leu (TT) genotype were both at higher risk of developing cholestatic and mixed type of DILI, but not hepatocellular type of injury. This result is not in agreement with Huang and coworkers,5 who found that the SOD2 Ala/Ala and Ala/Val genotypes increased the risk of developing hepatocellular DILI. However, most (55%) of the Taiwanese cohort were anti-tuberculosis drug–induced DILI cases, whereas only 3% of our DILI cohort corresponded to this category.