CONTEXT: Minimal residual disease (MRD) in patients treated for hairy cell (HC) leukemia as assessed by immunohistochemistry has not been included routinely in evaluation of treatment results. OBJECTIVE: To assess the presence of persistent HCs after treatment, as detected by immunohistochemistry, and to evaluate the correlation between the level of MRD and clinical outcome. DESIGN: Percentages of DBA.44-positive HCs were assessed on 116 biopsy specimens from 17 patients. The patients had a median follow-up of 55.4 months. RESULTS: Minimal residual disease was seen in 3 patterns. Group 1 (7 patients) had MRD levels ranging from "rare scattered suspicious HCs" to less than 1%. The MRD levels were stable throughout follow-up, and all patients remained in complete remission. Group 2 (6 patients) had MRD levels ranging from 1% to 5%, and 3 patients were in complete remission at 77.9, 63.8, and 108.0 months. Another patient showed evidence of disease activity (partial remission) at 47.6 months. Two other patients relapsed at 12.3 months and at 25.7 months, respectively, with greater than 1% HCs. Group 3 (4 patients) had MRD levels greater than 5%. Three patients relapsed at 11.3, 12.1, and 29.6 months, respectively, with greater than 5% HCs. The fourth patient had MRD levels of 5% at 14.6 months and 2% at 20.0 months but was subsequently lost to follow-up. CONCLUSIONS: Quantitative assessment of MRD may be of value in identifying patients at risk for relapse of hairy cell leukemia.