Amrubicin, a completely synthetic 9-aminoanthracycline derivative, was previously shown to have potent antitumor activities against various human tumor xenografts. In this study, the in vitro activities of amrubicin and its major metabolite, amrubicinol, were examined using 17 human tumor cell lines. Amrubicinol was 5 to 54 times more potent than amrubicin, and as potent as doxorubicin, in inhibiting the growth of the cells following 3-day continuous drug exposure. Amrubicinol closely resembled doxorubicin in its profile of activities on the 17 human tumor cell lines. Cells were incubated with the drugs for 1 h, and the intracellular drug concentration and cell growth inhibition after 3 days were determined. Amrubicinol attained similar intracellular concentrations at lower medium concentrations compared to amrubicin, and the intracellular concentration of amrubicinol necessary to produce 50% cell growth inhibition was 3 to 8 times lower than that of amrubicin in 4 cell lines tested. Amrubicinol has a higher activity level inside the cells than does amrubicin. When cells were incubated with amrubicin for 5 h, a substantial amount of amrubicinol, more than 9% of that of amrubicin, was found in cells in 4 of the 8 cell lines tested. Amrubicinol may contribute to the in vitro growth-inhibitory effect of amrubicin on these cells. The results suggest that amrubicinol plays an important role in the in vivo antitumor effect of amrubicin as an active metabolite.Key words: Anthracycline -Amrubicin -SM-5887 -Metabolism Amrubicin hydrochloride, (+)-(7S,9S)-acetyl-9-amino-7-[(2-deoxy-β-D-erythro-pentopyranosyl)oxy]-7,8,9,10-tetrahydro-6,11-dihydroxy-5,12-naphthacenedione hydrochloride (SM-5887), is a completely synthetic 9-aminoanthracycline derivative.1) It has potent antitumor activities against various human tumor xenografts, being more potent than doxorubicin.2) In contrast to its potent in vivo antitumor activities, the in vitro growth-inhibitory activities of amrubicin were more than 5 times lower than those of doxorubicin in several human tumor cell lines.3)The dissociation between the in vitro and in vivo activities suggests that amrubicin is converted in vivo to metabolites which are more effective cell growth inhibitors than the parent compound. To test this possibility, the growthinhibitory activities of the metabolites of amrubicin were examined using various human tumor cell lines.A major pathway of anthracycline metabolism is known to be the reduction of the C-13 carbonyl group to a hydroxyl group by cytoplasmic carbonyl reductase, and another involves the reductive cleavage of the glycosidic bond between the amino sugar and the chromophore by microsomal glycosidases.4) The C-13 hydroxy derivatives and aglycones of doxorubicin, epirubicin, daunorubicin and idarubicin were found in the plasma of cancer patients, [5][6][7][8] as well as in the plasma and tissues of experimental animals treated with these drugs.9-11) These metabolites were also formed in vitro in human hepatocytes and human tumor cells. [12][13][...

The antitumor effects of SM‐5887, a totally synthetic 9‐aminoanthracycline derivative, were evaluated in six murinc experimental tumor systems (P38S, Ehrlich carcinoma, sarcoma 180, Lewis lung carcinoma, B16 melanoma and colon 38) and nine human tumor‐nude mouse systems (one breast cancer, two lung cancers and six gastric cancers). Characteristically SM‐5887 showed excellent antitumor activities, superior to adriamycin (ADR), against human tumor xenografts, although its activities against murine experimental tumors were almost equal to those of ADR. When the human tumors were implanted sc in female athymic mice (BALB/c, nu/nu) and their volume reached 100‐ 300 mm3, SM‐5887 and ADR were injected iv. All nine human tumors tested showed statistically significant responses to SM‐5887, and 7 of them were strongly suppressed in their growth by SM‐5887 so that minimum T/C values were less than 30% at the maximum tolerated dose (MTD, 25 mg/kg) with a single iv injection. Compared with ADR, SM‐5887 was statistically more effective in five tumors (one breast, one lung and three gastric), equal in two tumors (two gastric), and less potent in two tumors (one lung and one gastric). In addition, the 10‐day‐interval repeated iv treatments with SM‐5887 at the MTD (25 mg/kg) resulted in remarkably potent antitumor effects (including complete regression) against human gastric cancer, 4‐1ST, implanted in nude mice without enhancement of toxic effects, SM‐5887 was also effective against ip‐inoculated P388 by oral administration as well as iv injection.

The tissue distribution of a novel antitumor anthracycline antibiotic, amrubicin, was studied using seven human tumor xenografts implanted into nude mice, in order to identify the principal factors determining its therapeutic efficacy. We found a good correlation between the level of the metabolite amrubicinol in the tumor and the in vivo efficacy. High metabolic activity of amrubicin to amrubicinol was detected in tumor tissue homogenates, especially in cell lines highly sensitive to amrubicin in vivo. In contrast to amrubicin, the administration of amrubicinol showed less tumorselective toxicity in these human tumor xenograft models. These data indicate that the tumorselective metabolism of amrubicin to amrubicinol resulted in a tumor-selective disposition of amrubicinol, leading to good efficacy in in vivo experimental therapeutic models.

It has been reported that the 9-aminoanthracycline amrubicin shows good efficacy in human tumor xenograft models. We studied the disposition and metabolism of amrubicin in mice, in comparison with those of doxorubicin. Amrubicinol, a 13-hydroxy metabolite of amrubicin, which is 10 to 100 times more cytotoxic than amrubicin, was detected as a major metabolite in blood and tissues, and aglycones of amrubicin were also detected. A pharmacokinetic study revealed that amrubicin had a smaller distribution volume and a shorter half-life than doxorubicin. In several normal tissues, the levels of amrubicin and amrubicinol were lower than those of doxorubicin. In contrast, the tumor levels of amrubicinol in the mice treated with amrubicin were higher than those of doxorubicin in the mice treated with that drug, in tumors that are sensitive to amrubicin. These results suggest that the potent therapeutic activity of amrubicin is caused by the selective distribution of its highly active metabolite, amrubicinol, in tumors.

The toxicological characteristics of SM‐5887 were evaluated in mice after a bolus intravenous injection, and compared with those of adriamycin (ADR). The acute toxic signs observed after SM‐5887 administration were body weight decrease, ataxia, hair loss, and myelosuppression. They were qualitatively comparable to those induced by ADR. The 50% lethal dose values determined by 14‐day observation after drug administration were in the range of 32 to 50 mg/kg for SM‐5887 and 16 to more than 20 mg/kg for ADR in four strains of mice. The maximum tolerated doses (MTD) were estimated to be 25 mg/kg for SM‐5887 and 12.5 mg/kg for ADR (no death or body weight loss of more than 3 g occurred). When 14‐day survivors were further observed until 90 days after drug administration, ADR frequently and dose‐independently showed delayed‐type lethal toxicity at doses of more than 10 mg/kg, whereas SM‐5887 did not. The myelosuppression of SM‐5887 was more severe even at a half of the MTD than that of ADR at the MTD, but its recovery was more rapid than that after ADR, In addition, when the drugs were injected into the subplantar region of mouse hind paws, ADR induced a severe inflammatory reaction, whereas SM‐5887 yielded only a slight one. The data suggest that toxic effects of SM‐5887 are more reversible and more controllable than those of ADR.