Metastasis to the liver is a common event in clinicaloncology. Blood-borne tumor cells (TCs) arriving to the coined the term metastasis in 1829 in his treatise ''Reliver sinusoids run into a special vascular bed. The lining cherches du Cancer,'' the metastatic cascade has been of liver sinusoids is shared by Kupffer cells (KCs) and exhaustively studied.1,2 Metastases originate from a seendothelial cells. KCs, liver-fixed macrophages, are re-lective population of cells within the great biological sponsible for detection and removal of ''non-self'' parti-heterogeneity of the tumor.3 They must invade the vascles. To investigate their role in arresting blood-borne cular wall to be dislodged into the circulation and surTCs and controlling tumor growth, we injected a synge-vive in the blood stream until they arrest at the distant neic colon carcinoma cell line into a mesenteric vein of vascular bed of the target organ by adherence or metwo groups of rats; one group was without Kupffer cells chanical trapping. Then, they have to proliferate to iniand the other normal controls. We removed the liver of tiate a metastatic colony. these animals at different time intervals and performedLiver metastasis is a common event in cancer paimmunohistochemical analysis with monoclonal antitients. This is of particular relevance for neoplasias of bodies (MoAbs) against our tumor cell line, three macrophage subpopulations, natural killer cells, and B and T the gastrointestinal tract, because the liver is the first lymphocytes. Additionally, we showed in vitro spontane-vascular bed to be encountered by blood-borne tumor ous cytotoxicity of KCs against our tumor cell line. Re-cells (TCs) through the portal system. 4 The lumen of sults suggest that KCs play a relevant role in arresting liver sinusoids differ from the ordinary capillaries in circulating TCs at the liver sinusoid, although it is lim-that, besides endothelial cells, the stellate cells of von ited to a small number of malignant cells. They also seem Kupffer are lining the sinusoid wall.5 Kupffer cells to play a major role in clearing neoplastic cells from the (KCs) represent one of the largest populations of the liver parenchyma, in controlling tumor growth in the mononuclear phagocyte system in the body. Their privery early stages of metastatic development, and in mod-mary function is to discriminate between ''self'' and ulating the host immune response to cancer cells. (HEPA-''non-self'' particles, playing a prominent role as anti-

No abstract

We evaluated the activity of ruboxyl (Rbx), a nitroxyl analogue of daunorubicin (Dauno), in experimental models of hepatic metastases from colorectal carcinoma (CRC) and compared it with its parent compound and with 5-fluorouracil (5FU). In mice treated by intraperitoneal injections Rbx and 5FU proved more effective than Dauno: the Index of Inhibition of Metastases in comparison with controls was 43% for Dauno, 70% for 5FU and 84% for Rbx. In BDIX rats implanted with the syngeneic cell line DHD K12/TRb, both Rbx and 5FU, administered as a continuous intravenous infusion for 7 days, reduced the development of liver metastases from a median of 23.8 ± 2.16 for controls to 3.2 ± 1.3 for 5FU and 1.0 ± 1.4 for Rbx (p < 0.0001 versus controls for both treatments): the comparison of Rbx and 5FU showed a trend in favour of this new anthracycline. Median survival was prolonged from 40.6 ± 3.4 days in controls to 56.0 ± 5.8 days with Rbx and 58.0 ± 4.69 days with 5FU. Considering that in a phase I study Rbx showed only minor and manageable toxic side effect, its activity in the clinical treatment of CRC metastases may deserve further attention.

The mode of action of the combination treatment 5-fluorouracil (5-FU) and levamisole in colorectal cancer patients is unknown. It is postulated that the beneficial effect may be explained by an immunomodulatory effect on Kupffer cell (KC) cytotoxicity. We evaluated the effect of levamisole (200 micrograms/ml) and 5-FU (10 microM) on rat KC cytotoxicity against syngeneic CC531 tumor cells. Viability of KCs was unaffected by 5-FU and/or levamisole. The combination did not enhance growth inhibition of CC531 compared to 5-FU alone. A significant increase in KC cytotoxicity was observed after 24-hr incubation with 5-FU/levamisole especially at an effector/target ratio of 10 (P < 0.05). 5-FU alone had no effect on KC cytotoxicity, while levamisole induced only a slight increase. Our in vitro data suggest that the additive effect of the combination 5-FU/levamisole on KC cytotoxicity may attribute to the beneficial effect of the adjuvant treatment in colorectal cancer patients.