A highly aggressive human CALLA+C mu+ pre-B acute lymphoblastic leukemia (ALL) cell line (NALM-6-UM1) causes disseminated and invariably fatal leukemia in CB.17 mice with severe combined immunodeficiency (SCID). We used this SCID mouse model of human pre-B ALL to evaluate and compare, in a total of 434 SCID mice, the antileukemic efficacy of B43 (anti-CD19)-pokeweed antiviral protein (PAP) immunotoxin and cyclophosphamide (CPA) as individual reagents and as combined immunochemotherapeutic regimens. B43-PAP plus CPA was superior to either the immunotoxin or drug alone, and combined immunochemotherapy markedly improved the event-free survival (EFS) of SCID mice challenged with NALM-6-UM1 pre-B ALL cells. Notably, 90% to 100% of SCID mice challenged with 1 x 10(6) leukemia cells and then treated with B43-PAP plus CPA combined immunochemotherapy regimens became long-term survivors, a result not achieved with B43-PAP alone or CPA alone. The advantage was particularly evident in mice inoculated with 5 x 10(6) leukemia cells. While neither 15 micrograms B43-PAP (median survival, 58 days) nor 1 mg CPA (median survival, 49 days) resulted in long-term EFS of SCID mice challenged with 5 x 10(6) NALM-6-UM1 pre-B ALL cells, the probability of EFS at 6 months was 50% +/- 16% for SCID mice treated with 15 micrograms B43-PAP plus 1 mg CPA (median survival, greater than 180 days) (P less than .0001). The probability of long-term EFS was only 14% +/- 7% for mice treated with 30 micrograms B43-PAP and 0% +/- 0% for mice treated with 1 mg CPA, but 40% +/- 16% for mice treated with 30 micrograms B43-PAP plus 1 mg CPA (P less than .0001). Similarly, the probability of EFS at 6 months was 40% +/- 16% for mice treated with 2 mg CPA alone, 70% +/- 15% for mice treated with 2 mg CPA plus 15 micrograms B43-PAP, and 70% +/- 15% for mice treated with 2 mg CPA plus 30 micrograms B43-PAP. Ten SCID mice in the B43-PAP plus CPA combined immunochemotherapy arms surviving long term after the inoculation of 5 x 10(6) NALM-6-UM1 pre-B ALL cells were electively killed at 174 to 181 days to assess their leukemia burden. We found no evidence of leukemia in any of the bone marrow specimens by two-color immunofluorescence and multiparameter flow cytometry.(ABSTRACT TRUNCATED AT 400 WORDS)