Abstract

A previous study suggested that drug release is the dominating factor controlling biological activity of liposomal mitoxantrone in tissues where the rate of liposome accumulation is rapid. The studies described here attempted to address the question: under conditions where the rate of liposome accumulation is slow, does drug release or liposome-mediated drug delivery become the dominant factor controlling therapeutic activity? Liposomal mitoxantrone formulations exhibiting different drug-release characteristics were injected i.v. in mice bearing human carcinoma xenografts: A431 human squamous cell carcinoma and LS180 human colon cell carcinoma in SCID/RAG 2 mice. When lipid and drug levels were measured in established (>100-mg) tumors, accumulation was more rapid in the LS180 tumors (C_max 4 h) than in the A431 tumors (C_max 48 h). Mean area under the curve values for mitoxantrone measured over a 96-h time course in A431 tumors were 505, 304, and 93 μg · g⁻¹ · h⁻¹ for 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC)/cholesterol (Chol), 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC)/Chol, and free mitoxantrone, respectively. When a similar analysis was completed in LS180 tumors, the area under the curve values were 999, 749, and 251 μg · g⁻¹ · h⁻¹ for DSPC/Chol, DMPC/Chol, and free mitoxantrone, respectively. Although drug delivery was less after administration of the DMPC/Chol liposomal mitoxantrone compared with the DSPC/Chol formulation, LS180 solid-tumor growth curves showed the treatment with the DMPC/Chol formulation produced greater delays in tumor growth compared with animals treated with the DSPC/Chol formulation. These data emphasize the importance of designing liposomal formulations that release drug after localization within a region of tumor growth.