Previous experiments have shown that 64Cu transmutation was inefficient to cure mice which had been injected 6 days earlier with 5 x 10(5) Krebs ascitic cells. The experiments were repeated with 64CuCl2 administered only 1 day after the injection of 5 x 10(5) Krebs ascitic cells when no developing tumor existed. The results reported showed that, even in this case, only a delay in death could be observed. These negative results revealed that malignant cells injected 24 hours previously in a mouse (in vivo conditions) differ from a malignant cell suspension in a tube (in vitro conditions) where the lethal effect of 64Cu transmutation was clearly evidenced. We concluded that some kind of collaboration was established between the few accepted malignant cells and the host. Based on this collaboration we introduced a new concept, "the cancer mouse", in which in addition to the malignant cells some cells in the host, even though non - malignant, nonetheless feature a slightly modified functioning: "the cancer functioning". In this view, an efficient antitumor treatment must act at the same time and in a coordinated manner on the malignant cells and on the host cells which now feature "the cancer functioning". In the efficient treatment described (Anticancer Res 9: 947-954, 1989) some compounds (64Cu and thioproline) act against the malignant genomes, and other components (metal ions, amino acids, vitamin D2, thyroxine and chelating substances) act against the functioning of the host cells by taking into account certain characteristics of living systems. This treatment was efficient in curing mice injected 1 or 6 days previously with 5 x 10(5) Krebs ascitic cells.