Our preliminary data suggested that 1,4-bis(1-naphthyl)-2,3-dinitro-1,3-butadiene [Viale M, Ottone M, Chiavarina B, Mariggiò MA, Prevosto C, Dell'Erba C, et al. Preliminary evaluation in vitro of the inhibition of cell proliferation, cytotoxicity and induction of apoptosis by 1,4-bis(1-naphthyl)-2,3-dinitro-1,3-butadiene. Invest New Drug 2004;22:359-67] (Naph-DNB), possesses good characteristics in terms of inhibition of cell proliferation in two cell lines derived from colon and gastric cancers. On this basis and to confirm the specificity of our compound towards gastrointestinal malignancies, we have analyzed the inhibition of cell proliferation, the cytotoxicity and the induction of apoptosis by Naph-DNB in seven cell lines derived from human colon (DLD-1, Lovo, HCT-8 and Colo 741), stomach (HGC-27) and pancreas (Panc-1 and Hup-T4) tumours. For the sake of comparison, cells have also been exposed to four anticancer drugs utilized for the treatment of gastrointestinal malignancies (oxaliplatin, irinotecan, gemcitabine and 5-fluorouracil). Moreover, toxicological data have been obtained in order to define the lethal dose (LD) and maximal tolerated dose (MTD) values and the spectrum of tissue alterations caused by the intraperitoneal (i.p.) and intravenous (i.v.) administration of Naph-DNB. IC50 data obtained by the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay suggest that Naph-DNB is generally more active than two or more of the anticancer drugs above in most cell lines: it displayed the lowest activity only in HGC-27 cells, although data concerning the IC75 parameter enlighten a significantly better activity than irinotecan and 5-fluorouracil. Using the equitoxic concentrations IC50 and IC75, we have also evaluated the ability of Naph-DNB and of the other anticancer drugs to kill cells and to induce apoptosis. Our data show that at these concentrations Naph-DNB has a cytotoxic activity comparable or even better than that of some anticancer drugs. Similarly, Naph-DNB induces apoptosis better than the other anticancer drugs in HCT-8 and HGC-27 cells, while in Lovo and Panc-1 cells the induction is comparable. On the basis of toxicological data we defined the LD10, LD50, LD90 (i.p., 17.6, 36.1 and 54.1 mg kg(-1), respectively; i.v., 6.1, 14.1 and 22.0 mg kg(-1), respectively) and the MTD (i.p., 15 mg kg(-1); i.v., 5 mg kg(-1)) parameters. Histochemical analysis has shown that, in general, the administration of even toxic doses of Naph-DNB does not cause great structural injuries, although it can have some effects on the metabolism of glicogen and iron in organs as liver and spleen. In conclusion, our preclinical studies in vitro suggest that Naph-DNB may represent a good anticancer compound for the treatment of generally unresponsive tumours such as those of pancreas, stomach and colon. Moreover, the analysis of its toxic effects has allowed the definition of LD and MTD parameters, which will be used in further experiments in vivo for the definition of its antitumour activity.