A unilamellar liposomal formulation of amphotericin B (LAmB) known as AmBisome was safely administered intravenously to 20 rabbits at 0.5, 1.0, 2.5, 5, or 10 mg/kg of body weight, whereas of 12 rabbits given desoxycholate amphotericin B (DAmB) intravenously at 0.5, 1.0, or 1.5 mg/kg, 2 died of acute cardiac toxicity when DAmB was administered at the highest dose. Single-dose LAmB (1 mg/kg) achieved a maximum concentration in serum (Cmax) of 26 ± 2.4 ,ug/ml and an area under the curve to infinity (AUC O) of 60 ± 16 ,ug h/ml, while single-dose DAmB (1.0 mg/kg), by comparison, achieved a lower Cmax (4.7 ± 0.2 -ug/ml; P = 0.001) and a lower AUC -,, (30.6 ± 2.2 ,ug * h/ml; P = 0.07). Following administration of a single dose of LAmB (10 mg/kg), a disproportionately higher Cmax (287 ± 14 ,ug/ml) and AUC0_Q (2,223 ± 246 ,ug* h/ml) occurred, indicating saturable elimination. After chronic dosing (n = 4) with LAmB at 5.0 mg/kg/day for 28 days or DAmB at 1.0 mg/kg/day for 28 days, LAmB achieved daily peak levels of 122.8 ± 5.8 ,ug/ml and trough levels of 34.9 ± 1.8 ,ug/ml, while DAmB reached a peak of only 1.76 ± 0.11 ,ig/ml and a trough of 0.46 ± 0.04 ,ig/ml (P . 0.001). Significant accumulations of amphotericin B into reticuloendothelial organs were observed, with 239 ± 39 ,ug/g found in the liver after chronic LAmB dosing (5 mg/kg/day), which was seven times higher than the 33 ± 6 ,ug/g after DAmB dosing (1 mg/kg/day) (P = 0.002). Accumulation in kidneys, however, remained 14-fold lower (P = 0.04) following LAmB dosing (0.87 ± 0.61 j,g/g) than after DAmB dosing (12.7 + 4.6 ,ug/g). Nephrotoxicity occurred in only one of four LAmB-treated animals, while it occurred in all four chronically DAmB-treated animals; mild hepatotoxicity with transaminase elevations was seen in one MATERUILS AND METHODS Drugs. LAmB was provided as a lyophilized yellow powder (AmBisome) by Vestar, Inc. The drug's constituents are 53.3% hydrogenated soy phosphatidylcholine, 13.1% cholesterol, 21.1% distearoylphosphatidylglycerol, and 12.5% amphotericin B (by weight, or at a molar ratio of 2:1:0.8:0.4, respectively). The powder was initially reconstituted with sterile water to an initial concentration of 2 mg/ml, and the solution was then heated to 60°C for 10 min to ensure dissolution. The solution was then filtered through a 5-,um-pore-size filter to remove any aggregates that might remain and was diluted to a final concentration of 0.5 mg/ml with 5% glucose in water. DAmB was prepared from commercially available Fungizone (BristolMyers Squibb, Princeton, N.J.) by dissolving the initial powder with sterile water and then further diluting the solution with 5% glucose in water to a final concentration of 1 mg/ml.Rabbits. Female New Zealand White rabbits weighing between 2.5 and 3.0 kg were used for the study. Animals were housed individually and were provided food and water ad libitum. While the rabbits were under general anesthesia, a silastic central venous catheter was surgically placed in each rabbit for repeated, nontraumatic venous access as...

We investigated the safety and efficacy of amphotericin B colloidal dispersion (ABCD) for the treatment of invasive pulmonary aspergillosis in persistently granulocytopenic rabbits. Treatment groups included ABCD in dosages of 1, 5, and 10 mg/kg/day intravenously or conventional desoxycholate amphotericin B (DAmB) at 1 mg/kg/day intravenously. Antifungal activity was directly related to increasing dosage of ABCD as determined by the concentration ofAspergillusfumigatus organisms in lungs and the frequency of hemorrhagic pulmonary lesions. At 5 and 10 mg/kg/day, there was a significant reduction in the tissue burden of A. fumigatus as measured by percent culture-positive lobes and CFU per gram of tissue (P c 0.001), whereas at 1 mg/kg/day the tissue burden ofA.fimigatus was not significantly different from that in untreated controls. Microbiological clearance was significantly greater at 1 mg of DAmB per kg per day than at 1 mg of ABCD per kg per day (P c 0.001). There was no difference in microbiological clearance of bronchoalveolar lavage fluid among the treatment groups as measured by CFU per milliliter. As determined by survival, ABCD at 5.0 mg/kg/day was more effective than DAmB at 1.0 mg/kg/day and ABCD at 10 mg/kg/day. ABCD at 10 mg/kg/day was more nephrotoxic than the lower dosages of ABCD and resulted in higher mortality. Impairment of glomerular filtration developed as a direct function increasing the ABCD dosage (r = 0.77; P < 0.001). In summary, this study found dose-dependent antifungal activity and nephrotoxicity of ABCD against invasive pulmonary aspergillosis in persistently granulocytopenic rabbits and showed that the optimal dosage of ABCD for antifungal activity and safety was 5 mg/kg/day.

Cilofungin (LY-121019) is a fungicidal cell wall-active 1,3-0-glucan synthetase inhibitor with a short plasma half-life and saturable nonlinear plasma pharmacokinetics. To optimize the in vivo efficacy of this compound, we studied the effects of its linear and nonlinear pharmacokinetics during continuous versus intermittent intravenous infusion of cilofungin in the treatment of experimental disseminated candidiasis in persistently granulocytopenic rabbits. Six groups of rabbits were studied, untreated controls (n = 32) and five cilofungin dosage regimen groups consisting of the following: 25 mg/kg of body weight intravenously twice daily (VLoINT) (n = 9); 50 mg/kg twice daily (LoINT) (n = 9); 90 mg/kg twice daily (HiINT) (n = 11); 5 mg/kg/h for 18 h/day (LoCI) (n = 7); and 10 mg/kg/h for 18 h/day (HiCI) (n = 7). All regimens achieved plasma concentrations exceeding the MIC for Candida albicans (0.25 pg/ml). In vitro timed kill assays found that the fungicidal activity and rate of kill by cilofungin above the MIC for C. albicans was concentration dependent. At the lower dosage regimens (VLoINT, LoINT, and LoCI), cilofungin followed linear plasma pharmacokinetics, whereas at higher doses (HiCI and HiINT), nonlinear kinetics consistent with a saturated elimination pathway(s) were observed. Only HiCI and HiINT produced a 103-to 104-fold reduction in CFU per gram in candidiasis of the brain (P c 0.001). HiCI and HiINT also significantly reduced infection in the choroid (P c 0.05). All regimens, except VLoInt, significantly (P c 0.01) reduced tissue infections in lung, liver, spleen, and kidney. However, only the regimens with nonlinear saturation kinetics (HiCI and HiINT) produced a 106 reduction in the spleen and a > 105 reduction of C. albicans in the kidney and liver. A simple doubling of the dosage from LoCI to HiCI resulted in tissue concentrations that were 10 times higher and a 102-to 104-fold-greater antifungal effect. There was a direct correlation (r2 = 0.83) between tissue concentrations of cilofungin and antifungal activity. Thus, continuous and intermittent infusion dosage regimens that elicit nonlinear saturation plasma pharmacokinetics of cilofungin were associated with increased antifungal activity against experimental disseminated candidiasis.