OBJECTIVES: There is some uncertainty regarding how to best dose and therapeutically monitor 6-mercaptopurine or azathioprine in patients with inflammatory bowel disease. The purpose of this study was to assess the relation between clinical response, 6-mercaptopurine metabolite levels, relative leukopenia, and drug dose. METHODS: 60 patients with inflammatory bowel disease who were on stable doses of 6-mercaptopurine or azathioprine for > or = 3 months and who had measurements of 6-mercaptopurine metabolite levels were evaluated. Patients were classified as complete responders (N = 24), partial responders (N = 7), or non-responders (N = 29). RESULTS: Drug dose was associated with clinical response when we analyzed adjusted doses based on molecular drug weight (P = 0.002). 6-Thioguanine levels also were associated with clinical response (P = 0.003) and the maximal difference between responders and non-responders was seen at 6-thioguanine levels greater than 260 pmol/8 x 10(8) RBC. Relative leukopenia, defined as white blood cell count less than either 5.0 or 4.0 K/uL, was not associated with clinical response (P = 0.13 and 0.77 respectively). CONCLUSIONS: 1. Drug dose and 6-thioguanine levels are related to clinical response in patients with inflammatory bowel disease on 6-mercaptopurine or azathioprine. 2. For 6-thioguanine levels, there is a fair amount of overlap, but maximal differentiation between responders and non-responders is seen at levels > 260 pmol/8 x 10(8) RBC. 3. Relative leukopenia does not correlate well with clinical response.