The aim of this study was to develop an in vitro human peripheral lymphocyte micronucleus bioassay involving phytohemagglutinin stimulant for urines containing adriamycin (ADR) and cyclophosphamide (CP). In vitro studies with defined concentrations of ADR, CP, and fresh urine showed that mitotic indices and micronuclei counts/1,000 cells had to be log (X + 1) transformed to be able to use parametric statistics and that a specific micronucleus assay for ADR in the presence of CP and urine for 5-15 ng ADR/mL had been developed. Whereas CP alone could be detected between 196-522 micrograms/mL, this effect was abolished in the presence of 15 ng ADR/mL. Interdonor variabilities relative to ADR sensitivity and CP linear dynamic range were marked, but intradonor variability was small. The MN bioassay tolerated < 10% urine. Results for urines from nine patients receiving antineoplastic drugs (CP, all; ADR, 3; 5-fluorouracil, 3; methotrexate, 3; vincristine, 4; procarbazine, 1; and megestrol acetate, 1) showed that only 1/3 patients given ADR were detected, and two others not given ADR were positive. All frozen urines from the 12 control subjects and the nine patients exhibited depressed mitotic index, with, however, no control patient urines inducing increased micronuclei. Two patients had urines of undefined genotoxic potential since undepressed mitotic indices were not attainable by dilution. The effects of combination chemotherapy in addition to freezing and storage influences were complex. More research is required to be able to interpret the results.