Bioavailability and bioeffectiveness of subcutaneous human insulin and two of its analogs--LysB28ProB29-human insulin and AspB10LysB28ProB29-human insulin--assessed in a conscious pig model.

Abstract

In this study, human insulin was compared with two of its analogs--LysB28ProB29-human insulin and AspB10LysB28ProB29-human insulin-with respect to bioavailability and metabolic effectiveness. Absorption from the subcutaneous site was determined using kinetic parameters from the washout curve, following intravenous infusion of insulin or analog. Absorption was found to be more rapid for the two analogs, with 90% absorption by 100 min for the analogs and by 180 min for insulin. Total absorption was 97 +/- 10% for insulin, 99 +/- 7% for LysB28ProB29-human insulin, and 93 +/- 12% for AspB10LysB28ProB29-human insulin. Bioactivity was assessed from the glucose infusion and using tracer-determined metabolic clearance rates (MCRs) and glucose production rates. The fractional glucose requirements (relative to the total amount infused) increased more rapidly for the two analogs than for insulin, with 50% of the glucose infused by 105 min for both analogs vs. 145 min for insulin. The total amount of glucose required was, however, significantly less (19.7 +/- 1.5 mmol/kg) for AspB10LysB28ProB29-human insulin than for either LysB28ProB29-human insulin (25.9 +/- 3.0 mmol/kg) or human insulin (27.8 +/- 2.6 mmol/kg). The glucose requirements were reflected in a lower MCR for AspB10LysB28ProB29-human insulin but equivalent decreases in the rates of glucose production. Thus both analogs demonstrated more rapid rates of absorption, onset, and termination of action, but were not completely bioequivalent, with AspB10LysB28ProB29-human insulin demonstrating a 25% decrease in bioactivity.