Comparison of Insulin Aspart and Lispro: Pharmacokinetic and metabolic effects

We studied seven patients with type 1 diabetes (five women and two men). The subjects’ age, weight, height, and body composition are shown in Table 1. At the time of study, the diabetic patients were free of detectable diabetic complications. They had no other disease apart from diabetes and were not taking any medications other than insulin. Five patients used insulin pumps and two used multiple daily insulin injections. All patients were receiving lispro insulin before the study. The women were studied in the follicular phase of their menstrual cycle. Informed written consent was obtained from all subjects after explanation of the nature, purpose, and potential risks of the study. The study protocol was approved by the Institutional Review Board of Temple University Hospital.

All subjects were admitted to the General Clinical Research Center at Temple University Hospital the day before the studies. Their evening dose of insulin was withheld. At 6:00 p.m. they were fed a standard meal consisting of 53% carbohydrate, 15% protein, and 32% fat. After that, they fasted for the duration of the study but were allowed water ad libitum.

During the night, their blood glucose concentration was maintained between 5.5 and 6.7 mmol/l with an intravenous infusion of regular insulin. The insulin infusion was discontinued at ∼7:00 a.m. the following day. At ∼6:00 a.m. an infusion of [6,6²H₂]glucose was started and continued until the end of the study. At ∼8:00 a.m. they received a subcutaneous injection into the abdominal wall of a rapidly acting insulin analog (either aspart or lispro) at a dose equal to one-half of their normal daytime insulin dose. All subjects received both insulins, 1 month apart, in random order. The patients, the nurses who collected the blood, and the technician who analyzed the blood were all unaware as to which analog was given.

After injection of the insulin analogs, plasma glucose levels were kept at ∼5.5 mmol/l with a variable infusion of 20% dextrose in water (5). Eight hours after insulin injection, the study was terminated; the patients restarted on their normal insulin regimen, were fed a meal, and were discharged from the hospital.

Glucose turnover was determined using the stable isotope [6,6-²H₂]glucose as described (6). To assure isotope equilibration, the tracer infusion was started 120 min before initiation of the clamp starting with a bolus of 30 μmol, followed by a continuous infusion of 0.3 μmol · kg⁻¹ · min⁻¹. Blood was collected at 30- to 60-min intervals (−120, −30, 0, 30, 60, 120, 180, 240, 300, 360, 420, and 480 min) for determination of isotope enrichment. Rates of glucose appearance (G_Ra) and disappearance (G_Rd) were calculated from the isotope enrichments before (−30 to 0 min) and during the 8-h study using Steele’s equation for nonsteady state (7). Underestimation of G_Ra during hyperinsulinemia was avoided by adding [6,6-²H₂]glucose (6.9 mmol/100 ml) to the unlabeled glucose infused to maintain euglycemia.

Endogenous glucose production (EGP) was calculated as the difference between the isotopically determined G_Ra and the glucose infusion rates (GIR) needed to maintain stable blood glucose levels during insulin infusion (EGP = G_Ra − GIR).

Respiratory gas exchange rates were determined at 30- to 60-min intervals with a metabolic measurement cart as previously described (8). Rates of protein oxidation were estimated from the urinary nitrogen excretion after correction for changes in urea nitrogen pool size (9). Rates of protein oxidation were used to determine the nonprotein respiratory quotient. Rates of fat oxidation were determined with the tables of Lusk, which are based on anonprotein respiratory quotient of 0.707 for 100% oxidation and 1.00 for 100% carbohydrate oxidation.

Body composition was determined by bioimpedance analysis (10).

Plasma glucose was measured with a glucose analyzer with the glucose oxidase method every 15–20 min. Free insulin levels were determined in deproteinized serum by radioimmunoassay with a specific antibody that cross-reacts only minimally (<0.2%) with proinsulin (Linco, St. Charles, MO). This antiserum recognizes equally human insulin, insulin lispro, and aspart (Fig. 1). Plasma fatty acids were determined enzymatically in chilled plasma with a kit (Wako, Richmond, VA). C-peptide was determined by radioimmunoassay (Linco). Plasma β-hydroxybutyrate and acetoacetate were determined enzymatically.

All data are expressed as means ± SEM. Statistical significance was assessed using two-way repeated measures ANOVA and two-tailed Student’s t test. A P < 0.05 was considered significant.

Insulin concentrations rose from 39 ± 6 to 256 ± 63 pmol/l 30 min after injection of insulin aspart and from 43 ± 10 to 286 ± 99 pmol/l 30 min after injection of insulin lispro (P = 0.24). After that, insulin levels declined at similar rates in both groups reaching basal levels ∼4 h after injection (Fig. 2).

GIR reached 17.3 ± 5.2 and 15.3 ± 4.3 μmol · kg⁻¹ · min⁻¹ 120 min after injection of insulin aspart and lispro, respectively (P = 0.61) (Fig. 2). After that, GIR decreased in both groups, reaching levels not different from 0 at ∼300 min.

Preinjection glucose concentrations were 7.3 ± 0.6 and 7.2 ± 0.4 mmol/l (P = 0.68), respectively, in the insulin aspart and insulin lispro groups. After injection of the insulin analogs, blood glucose was prevented from falling for ∼300 min by infusion of exogenous glucose. After that, glucose levels rose, reaching 12.1 ± 2.8 and 10.7 ± 1.6 mmol/l (P = 0.67), respectively, in the insulin aspart and lispro groups at the end of the study (Fig. 2).

G_Rd rose from 16.6 ± 1.4 to 26.8 ± 6.6 and from 14.5 ± 1.1 to 27.6 ± 8.6 μmol · kg⁻¹ · min⁻¹, respectively, at ∼120 min in the insulin aspart and lispro groups. G_Rd then declined in both groups, reaching basal levels after ∼240 min. There were no statistically significant differences between the two groups at any time (P = 0.61) (Fig. 3).

In response to insulin injections, EGP decreased from 16.1 ± 1.7 to 9.6 ± 3.0 μmol · kg⁻¹ · min⁻¹ and from 15.2 ± 0.9 to 11.5 ± 2.2 μmol · kg⁻¹ · min⁻¹ at 240 min, respectively, in the insulin aspart and insulin lispro groups. The differences were not statistically significant (P = 0.86) (Fig. 3).

Basal carbohydrate oxidation rates were similar in both groups. They increased comparably after analog injection until ∼3 h and then declined. Carbohydrate oxidation rates in both groups were not statistically significant from each other at any time during the studies (P = 0.69). Similarly, free fatty acid (FFA) oxidation rates first decreased and then increased. There were no statistically significant differences between the two groups (P = 0.22) (Fig. 4).

Basal plasma FFA levels decreased from 413 ± 65 to 130 ± 19 and from 620 ± 109 to 207 ± 40 μmol/l, 180 and 120 min, respectively, after injection of insulin aspart and insulin lispro. After that, FFA levels rose in both groups reaching 1,071 ± 112 and 1,195 ± 185 μmol/l after 480 min. The differences between the groups were not statistically significant (P = 0.40). Ketone body concentrations changed in parallel with FFA concentrations. Again, there were no significant differences between the two groups at any time (P = 0.70) (Fig. 5).

Preinjection levels of C-peptide (0.07 ± 0.04 and 0.08 ± 0.04 nmol/l, P = 0.71) and glucagon (49 ± 5 and 46 ± 6 pg/ml, P = 0.82) for the aspart and lispro groups, respectively, did not change significantly throughout the studies.

After subcutaneous injection of an equal amount of either insulin lispro or aspart into the same patients, free insulin reached similar concentrations (∼250–300 pmol/l) at approximately the same time (30–60 min postinjection), and after that free insulin disappeared from serum at comparable rates, reaching preinjection levels at ∼240 min.

Insulin analog concentrations were measured with an anti-insulin antibody that bound insulin, insulin lispro, and aspart with equal affinity over the range of insulin levels observed in this study. Therefore, serum levels of these three insulins could be directly compared with each other.

A limitation of our study was that blood was sampled only every 30–60 min. This precluded precise definition of peak insulin levels, the time to reach peak levels, and the time of disappearance of injected insulin from the blood. Nevertheless, these limitations do not detract from the observation that both insulins produced virtually superimposable insulin concentration curves.

The results of the current study are similar to those of two other studies in which the pharmacokinetics of the two insulin analogs were compared (11,12). One study used a single blind, random crossover design in seven patients with type 1 diabetes and 5- to 10-min blood sampling during the first hour after injection. They reported that after subcutaneous injection of 10 units of either analog, plasma profiles resembled each other but lispro showed a slightly more rapid uptake and peaked and declined marginally faster. A second study, which appeared while our study was under review, showed complete equivalence of lispro and aspart with respect to pharmacokinetics profiles and effectiveness controlling postprandial glucose excursions (12).

To our knowledge, this is the first study comparing action of lispro and aspart on carbohydrate and fat metabolism in the same patients. We found no significant differences comparing the actions of both analogs on G_Rd (>80% of which occurs in skeletal muscle [13]), on EGP (>80% of which occurs in the liver [14]), on GIR (reflecting a combination of insulin action on muscle and liver), and on carbohydrate oxidation. All of these effects peaked at ∼2–3 h and had disappeared after ∼5–6 h. As shown in Fig. 3, the nonsignificant differences in G_Rd and EGP were in opposite directions (i.e., insulin aspart was more active promoting glucose uptake but less active suppressing EGP). This supports the notion that the bioactivity profiles of the two insulins are interchangeable.

Similarly, there were no significant differences comparing actions of the two analogs on plasma FFA and ketone body levels nor on FFA oxidation. (Changes in plasma FFA levels closely reflect insulin action on lipolysis and is a very sensitive indicator of insulin action [15].) Insulin effects peaked at ∼2 h and had disappeared after ∼4 h. These results are in accord with previously published data on effects of insulin lispro (16,17) and insulin aspart (18–20).

Thus, the actions of both insulin analogs on carbohydrate and on lipid metabolism appeared to be indistinguishable from each other. This was not surprising in view of the similarity of their blood levels. Because of the small number of patients studied, we cannot completely rule out the possibly of small differences in some of the action of these two insulin analogs. Nevertheless, our data suggested that these two rapidly acting insulin analogs should be equally effective in treating the metabolic abnormalities of patients with type 1 diabetes.

This work was supported by National Institutes of Health Grants R01-AG-15363 (to G.B.), R01-DK-58895 (to G.B.), M01-RR-00349 (General Clinical Research Center Branch of the National Center for Research Resources), a Mentor Based Training Grant from the American Diabetes Association (to G.B.), and a Grant-in-Aid from Novo Nordisk Pharmaceuticals, Princeton, NJ.

We thank the nurses of the General Clinical Research Center for help with the studies and for excellent patient care, Karen Kresge and Maria Mozzoli for outstanding technical assistance, and Constance Harris Crews for typing the manuscript.

Information from this article has previously appeared in abstract form and was presented as a poster at the American Diabetes Association Annual Meeting, New Orleans, Louisiana, 13–17 June 2003.