The telopeptide of type I collagen is thought to be responsible for causing immunogenic response when introduced into xenogenic hosts. To eliminate this problem, solubilized bovine skin collagen was filtered through a millipore membrane, digested repeatedly with pepsin to remove telopeptides, and used as a carrier for a water-soluble, partially-purified fraction of bone morphogenetic protein (BMP) in mice. Composite implants of this telopeptide-depleted collagen and the partially-purified BMP fraction consistently elicited ectopic bone formation in mice 3 weeks postimplantation. When implanted alone, collagen or BMP failed to show this response. Collagens, prepared by use of conventional methods (acid-solubilized collagen, or collagen-digested once with pepsin) were also assessed as carriers for BMP, but were found to be inferior in terms of consistency of bone formation and amount of induced bone mass. The results suggest that telopeptide-depleted collagen permitted a gradual release of purified BMP for induction of bone, with minimal immunogenic interference. Consequently, this collagen carrier represents an important development for future clinical application of BMP.

To improve the ability of an in vitro drug sensitivity test to predict in vivo effects, we applied a drug concentration that was pharmacokinetically equivalent to plasma levels and collagen gel droplet-embedded culture with a high cloning efficiency. We reported that the cell-killing effect of cell cycle phase-nonspecific drugs such as mitomycin C, cisplatin and Adriamycin depends on the area under the drug concentration-time curve (AUC). The plasma AUC values of these drugs were estimated after an injection into nude mice at the maximal tolerated doses (MTD). Tumor cells isolated from human tumor xenografts implanted into nude mice and cultured in collagen gel droplets were exposed to drugs under conditions that can reproduce the plasma AUC in vitro. The in vitro sensitivity to a drug was compared with the in vivo response of the same tumor treated with the MTD of the drug. When the criterion of sensitivity was taken as 50% or less of the growth inhibition (growth rate of treated group/that of control group, T/C), the correlation between the in vitro and in vivo growth inhibition of all 3 drugs tested was relatively high (86% of the true-positive rate, 82% of the true-negative rate and 83% of the correlation rate).

No abstract

No abstract

No abstract