Quantitative in vivo assay of human granulocyte colony-stimulating factor using cyclophosphamide-induced neutropenic mice.

Administration of human granulocyte colony-stimulating factor (hG-CSF) to mice with cyclophosphamide (CPA)-induced neutropenia for 4 consecutive days from the day after the CPA dosing (100 mg/kg) resulted in a dose-dependent increase in the peripheral blood neutrophil count 6 hours after the final hG-CSF injection. Within the hG-CSF dose range of 0.1 to 10 micrograms per mouse per day, there was a strong linear relationship (r greater than .9) between the logarithm of the dose and the peripheral blood neutrophil count in the treated mice. Using the same hG-CSF preparation, 38 experiments indicated that the regression lines are highly reproducible. Such an association never occurred with intact mice, and 100 mg/kg of CPA induced the highest response to hG-CSF. This linear relationship between the two variables allows us to determine the biologic potency of a test hG-CSF preparation relative to a reference standard using a parallel line assay, with a coefficient of precision of around .2. When assayed by this bioassay procedure, which we have termed CPA-mouse assay, natural hG-CSF and recombinant hG-CSF (produced by Chinese hamster ovary cells) were nearly equipotent in specific biologic activity. These results confirm the CPA-mouse assay as an especially useful assay method for quantifying the in vivo activity of hG-CSF.