In order to determine if it was possible to reverse clinically evident chemotherapeutic drug-resistance, 51 evaluable patients received chemotherapy (in doses and schedules on which they had previously demonstrated tumor progression) together with amphotericin B (AMB). AMB was given in 1-, 2-, or 4-day courses. There was 1 complete response (2%), and 5 partial responses (10%). Response rates tended to be higher in the 4-day treatment program (23%) than in the 1- or 2-day AMB treatment schedules (8%). Toxicity was that expected with chemotherapy (myelosuppression), or AMB alone (fever, chills, and reversible mild azotemia). We conclude that AMB is only infrequently able to reverse clinical drug-resistance, but that this might have palliative effects in a small number of patients in whom other standard chemotherapeutic drugs lack clinical effectiveness.