Summary The effect of the combined administration of doxorubicin (DX) and verapamil (VRP) on the induction of DX resistance of B16 melanoma cells, was investigated both in vitro and in vivo. Cells grown in the presence of increasing concentrations of DX and of I1M VRP, tested at several passages, were more resistant than cells grown with DX alone. The treatment of B16 melanoma bearing mice with the maximal tolerated dose of DX (12mgkg-1 i.v.) and of VRP (25mgkg-' i.p.) selected a line (B16-DX.VRP) completely resistant to DX after 17 transplants, while treatment with DX alone selected a DX resistant line after 27 transplants. Lung metastases were significantly lower in the B16-DX.VRP line compared to the original B16 melanoma. The results suggest that the association of VRP with DX increases the rate of resistance development to DX.The calcium channel blocker verapamil (VRP) has been shown to enhance anthracyclines, vincristine and vinblastine cytotoxicity in several resistant cell lines (Tsuruo et al., 1981(Tsuruo et al., , 1983Slater et al., 1982; Rogan et al., 1984;Simpson et al., 1984;Pradhan et al., 1984) probably by inhibiting their active efflux (Tsuruo et al., 1981(Tsuruo et al., ,1982(Tsuruo et al., ,1983 Rogan et al., 1984). Recently we have demonstrated that while the coadministration of VRP and doxorubicin (DX) is ineffective in mice bearing tumours against which DX is completely inactive, this combination may result in a significant potentiation of DX activity in mice bearing some sensitive tumours (Formelli et al., 1988). These in vivo results suggest a potential role of VRP not only after the failure of the secondary treatment, but also in the initial treatment with the aim of eliminating the highest number of tumour cells. Clinical trials combining VRP and DX are in progress both in DX resistant and responsive tumours (Presant et al., 1986;Ozols et al., 1987;Kerr et al., 1986) and the use of VRP with DX from the beginning of tumour treatment poses the question of the effect of this drug on the onset of DX resistance. To answer this question we investigated both in vitro and in vivo the sensitivity to DX of tumour cells exposed during several transplants to DX alone and to DX plus VRP. For this purpose we choose the B16 melanoma, an experimental model which may be representative of a tumour against which the use of DX plus VRP from the beginning can be justified. In fact, it has been shown that, in vivo, B 16 melanoma is a DX sensitive tumour whose sensitivity to DX is significantly increased by coadministration of VRP (Formelli et al., 1988) i.p. at the dose of 25mg kg-1, which also corresponds to the highest non-lethal dose. VRP was administered according to the schedule previously found effective in increasing DX activity in this tumour (Formelli et al., 1988). It was administered 3h before DX, since in vitro it potentiates DX activity in pretreated cells . It was administered repeatedly, i.e. 5 days per week, until the tumour was transplanted, since DX levels are maintained in this tumour lon...