Proper prospective pharmacokinetic studies of 6-mercaptopurine (6-MP) in inflammatory bowel disease (IBD) patients are lacking. As a result, conflicting recommendations have been made for metabolite monitoring in routine practice. The authors have evaluated 6-MP pharmacokinetics in IBD patients, including the genetic background for thiopurine methyltransferase (TPMT). Red blood cell (RBC) 6-thioguanine nucleotide (6-TGN) and 6-methylmercaptopurine ribonucleotide (6-MMPR) concentrations were measured in 30 IBD patients at 1, 2, 4, and 8 weeks after starting 6-MP, 50 mg once daily. Outcome measures included mean 6-TGN and 6-MMPR concentrations (± 95% confidence interval, CI95%) and their associations with TPMT genotype, 6-MP dose, and hematologic, hepatic, pancreatic, and efficacy parameters during the 8-week period. Steady-state concentrations were reached after 4 weeks, indicating a half-life of approximately 5 days for both 6-TGN and 6-MMPR; the concentrations were 368 (CI95% 284–452) and 2837 (CI95% 2101–3573) pmol/8 × 108 RBCs, respectively. Large interpatient variability occurred at all time points. TPMT genotype correlated with 6-TGN concentrations (0.576, P < 0.01), and patients with mutant alleles had a relative risk (RR) of 12.0 (CI95% 1.7–92.3) of developing leukopenia. A 6-MMPR/6-TGN ratio less than 11 was associated with therapeutic efficacy. Based on this pharmacokinetic analysis, therapeutic drug monitoring is essential for rational 6-MP dosing.