The role of single drugs in the treatment of neuroblastoma is poorly defined. We, therefore, tested neuroblastoma cell survival after a 72 h exposure to one of 19 cytostatic drugs by monolayer proliferation assay. 6 cell lines (IMR-5, Kelly, SK-N-SH, GI-CA-N, CHP-100, CHP-134) were selected on the basis of MYCN amplification and PGY1 overexpression. ED50 drug concentrations were related to plasma levels achievable in patients during chemotherapy. More effective substances were mitoxantrone, doxorubicin, hydroxyurea, bleomycin, dactinomycin, cisplatinum, thiotepa, melphalan, carboplatinum, etoposide, vincristine, cytarabine, 6-thioguanine, cyclophosphamide, ifosfamide and zilascorb. Parental drugs (cyclophosphamide, cisplatinum) appeared more cytotoxic on a molar basis than derived drugs (ifosfamide, carboplatinum). Less effective drugs included 5-fluorouracil, 6-mercaptopurine, CCNU and procarbazine. Fractional application of a given dose was more efficient than a single dose of cyclophosphamide, ifosfamide and cisplatinum. The tested neuroblastoma cell lines showed distinct sensitivities to cytostatic drugs. Cell lines with MYCN amplification appeared more sensitive than PGY1 overexpressing cells. In conclusion, comparative in vitro testing of cytostatic drugs may provide a rationale for their clinical evaluation. Investigation of drug combinations and application of the monolayer proliferation assay to tumour biopsy material for preclinical chemosensitivity testing are clearly warranted.