Summary There is a pressing need to reduce the time and cost of developing new cytotoxic agents and to accurately identify clinically active agents at an early stage. In this study, the differential staining cytotoxicity (DiSC) assay was used to assess the efficacy of the novel antitumour cAMP analogue, 8-chloro-cAMP (8-Cl-cAMP) (and its metabolite 8-CI-adenosine) against 107 fresh specimens of human neoplastic and normal cells. Diagnoses included chronic and acute leukaemias, myeloma, non-Hodgkin's lymphoma (NHL) and miscellaneous solid tumours. The aim was to identify targets for subsequent phase 1, II and IlIl trials. 8-Cl-cAMP was tested at 4-985 gM, along with standard chemotherapeutic drugs. 8-Cl-cAMP and its metabolite caused no morphologically observable cell differentiation but induced dosedependent cytotoxicity. Compared with untreated patients, previously treated chronic lymphocytic leukaemia (CLL) patients showed no increase in ex vivo resistance to 8-Cl-cAMP (P = 0.878); minimal cross-resistance with other cytotoxic drugs was detected. Compared with normal cells (mean LC 9 = 1803 gM), 8-Cl-cAMP showed significant ex vivo activity against CLL (117.0 gM; P < 0.0001) and NHL (140.0 gM; P < 0.0001), of which eight were mantle cell NHL (84.7 gIM), and greatest activity against cells from patients with acute myeloid leukaemia (AML; mean LC 9 = 24.3 gm; in vitro therapeutic index 74-fold, P < 0.0001). Solid tumour specimens were comparatively resistant to 8-CI-cAMP. The results highlight the clinical potential of 8-Cl-cAMP, point to several new phase 1, II and IlIl trial possibilities and provide a rationale for the inclusion of ex vivo cytotoxic drug evaluation in the drug development process.Keywords: ex vivo phase 11 trial; Differential Staining Cytotoxicity assay; ex vivo cytotoxic drug evaluation; ex vivo therapeutic index Drug development is a lengthy and expensive process, costing in excess of £100 million per licensed drug. A crucial decision point in the development of any new compound is whether to take it to clinical trial, a process requiring substantial investment. Usually, indications for a new cytotoxic compound are gleaned from cell line and xenograft studies; toxicology tests are then undertaken before the drug is entered into clinical trials (Carmichael, 1994). The incorporation of a parallel series of ex vivo tests in different neoplasms at an early juncture ('ex vivo' is used to denote the determination of patient cellular response to drugs outside the body as a surrogate for treating the patient), using methodologies that predict well for subsequent patient response to known cytotoxic (Bosanquet, 1994), could increase the likelihood that drugs progressed to trials will be clinically active.Various ex vivo methods have been used to test new compounds (Nagoumey et al, 1993;Larsson et al, 1994;Martin et al, 1994; Hanauske et al, 1995). In this paper we outline a development of this approach to drug evaluation whereby new agents are tested against both fresh human neoplastic cells (from h...