Abstract
BACKGROUND: Rash, liver dysfunction, and diarrhea are known major adverse events associated with erlotinib and gefitinib. However, clinical trials with gefitinib have reported different proportions of adverse events compared to trials with erlotinib. In an in vitro study, cytochrome P450 (CYP) 2D6 was shown to be involved in the metabolism of gefitinib but not erlotinib. It has been hypothesized that CYP2D6 phenotypes may be implicated in different adverse events associated with gefitinib and erlotinib therapies. METHODS: The frequency of each adverse event was evaluated during the period in which the patients received gefitinib or erlotinib therapy. CYP2D6 phenotypes were determined by analysis of CYP2D6 genotypes using real-time polymerase chain reaction techniques, which can detect single-nucleotide polymorphisms. The CYP2D6 phenotypes were categorized into 2 groups according to functional or reduced metabolic levels. In addition, we evaluated the odds ratio (OR) of the adverse events associated with each factor, including CYP2D6 activities and treatment types. RESULTS: A total of 232 patients received gefitinib therapy, and 86 received erlotinib therapy. Reduced function of CYP2D6 was associated with an increased risk of rash of grade 2 or more (OR, 0.44; 95% confidence interval [CI], 0.21-0.94; *p = 0.03), but not diarrhea ≥ grade 2 (OR, 0.49; 95% CI, 0.17-1.51; *p = 0.20) or liver dysfunction ≥ grade 2 (OR, 1.08; 95% CI, 0.52-2.34; *p = 0.84) in the gefitinib cohort. No associations were observed between any adverse events in the erlotinib cohort and CYP2D6 phenotypes (rash: OR, 1.77; 95% CI, 0.54-6.41; *p = 0.35/diarrhea: OR, 1.08; 95% CI, 0.21-7.43; *p = 0.93/liver dysfunction: OR, 0.93; 95% CI, 0.20-5.07; *p = 0.93). CONCLUSIONS: The frequency of rash was significantly higher in patients with reduced CYP2D6 activity who treated with gefitinib compared to patients with functional CYP2D6. CYP2D6 phenotypes are a risk factor for the development of rash in response to gefitinib therapy.