Control of Postprandial Plasma Glucose by an Oral Insulin Product (HIM2) in Patients With Type 2 Diabetes

Patients who participated in the study were men or women; were 36 to 65 years of age; were diagnosed with type 2 diabetes (1); had disease duration of at least 3 years; had fasting blood glucose levels of 125–252 mg/dl; had a HbA_1c level of 6.8–11.4%; had a BMI of 20.8–37.2 kg/m²; were following an appropriate regimen of diet and exercise; were on a stable regimen of oral hypoglycemic drugs but had no prior insulin exposure; and were without a major illness or disability. Daily fasting blood glucose levels were documented by each patient for 5 days preceding the first dose of study drug. Prior oral hypoglycemic drugs were discontinued for at least five half-lives before receiving study drug (metformin therapy was discontinued for at least 5 days).

The study was conducted in compliance with the Declaration of Helsinki and Good Clinical Practices. The study protocol and informed consent forms were approved by the Quorum Review IRB (Seattle, WA) and the Human Research Committee Kaleida Health (Buffalo, NY).

In this randomized, placebo- and regular insulin-controlled, three-way crossover, dose-escalation study, patients were randomized to the order in which they received each study drug (oral HIM2, subcutaneous regular insulin, and oral placebo). Patients were blinded to treatment with oral HIM2 and oral placebo; however, patients were not blinded when they received subcutaneous injections of regular insulin (Humulin R). A single dose of each of the following study drugs was administered to each patient on 3 separate days after an overnight fast of at least 8 h: oral HIM2 (at one of three dose levels: 0.375, 0.5, or 1.0 mg/kg), subcutaneous regular insulin (8 units Humulin R), and oral placebo. The 0.375-, 0.5-, and 1.0-mg/kg doses of HIM2 corresponded to ∼10.8, 14.4, and 28.7 IU/kg. The dose of 8 IU of subcutaneous regular insulin was estimated, on average, for an 80-kg person based on the insulin dosage algorithm described by Skyler et al. (14).

The time of dosing was assigned as time 0; all other time points on each dosing day are relative to that time. At 30 min after dosing, patients ingested a standardized test meal (16 oz/720 calories/90 g carbohydrate of Boost Plus). The entire amount of Boost Plus was ingested over a 10-min period. Blood samples were collected for determination of plasma glucose and insulin concentrations at 0, 10, 20, 30, 40, 50, 60, 75, 90, 105, 120, 135, 150, 165, 180, 200, 220, and 240 min after dosing.

A total of nine patients were enrolled at the Diabetes and Glandular Disease Research Clinic (San Antonio, TX), and nine patients were enrolled at the Diabetes-Endocrinology Center of Western New York (Buffalo, NY). A total of six patients (three patients at each study center) were enrolled in each HIM2 dosing group. At each study center, escalations from the 0.375 to the 0.5 and 1.0 mg/kg dose levels were permitted only if the safety and tolerability of HIM2 was confirmed in the previous dosing group.

Venous blood samples for determination of plasma glucose and insulin concentrations were drawn into tubes containing ethylenediamine tetraacetic acid at the following time points: 0 (immediately before dosing), 10, 20, 30 (immediately before the meal), 40, 50, 60, 75, 90, 105, 120, 135, 150, 165, 180, 200, 220, and 240 min. Samples were gently inverted 8–10 times and centrifuged at 1,300g for 10 min at 4°C. The plasma was separated, frozen within 60 min of collection, and stored at −7°C until assayed. Plasma glucose concentrations were determined using a glucose oxidase-based enzymatic assay (Roche Diagnostics/Boehringer Mannheim Hitachi 911 analyzer), with a lower limit of detection of 2 mg/dl. Plasma insulin concentrations were determined using a radioimmunoassay kit (Linco Research, St. Charles, MO), with a lower limit of detection of 0.07 ng/ml (2 μU/ml). Insulin concentration values less than the lower limit of detection were assigned a value of 0.07 ng/ml (2 μU/ml).

Blood glucose levels were determined on site immediately after collection using a Yellow Springs Instruments glucose analyzer (YSI, Yellow Springs, OH).

The following glucodynamic and pharmacokinetic parameters were calculated from the plasma glucose or insulin concentrations after dosing: 2-h postprandial plasma glucose concentration (2pp); maximum plasma glucose concentration (G_max); area under the plasma glucose concentration time curve from 0 to 240 min (glucose AUC_0–240); 2-h postprandial plasma glucose excursion (2pp_ex); maximum plasma glucose excursion (Gex_max); area under the plasma glucose excursion time curve from 30 to 240 min (glucose AUCex_30–240); maximum plasma insulin concentration (insulin C_max); time to maximum plasma insulin concentration (insulin T_max); and area under the insulin concentration time curve from 0 to 240 min (insulin AUC_0–240).

Plasma glucose excursions were calculated by subtracting the 30-min plasma glucose concentration (start of the meal) from each plasma glucose concentration. The areas under the plasma concentration time curves were calculated using the linear trapezoidal method.

Calculations were performed with SAS version 6.12 (SAS Institute, Cary, NC). Data are expressed as means ± SD in the text and tables; data are expressed as means ± SE in the figure. Wilcoxon’s signed-rank test was used to test for statistically significant differences between HIM2 and placebo treatments. The glucodynamic equivalence of HIM2 and subcutaneous insulin was tested by calculating a 95% CI on the mean ratios for the six glucodynamic parameters (2pp, G_max, glucose AUC_0–240, 2pp_ex, Gex_max, and glucose AUCex_0–240 [log-transformed]) between HIM2 and subcutaneous insulin for each subject. Equivalence testing was performed on pooled data from the 0.5 and 1.0 mg/kg HIM2 dosing groups. Equivalence to within a stringent 10% tolerance was demonstrated if the entire 95% CI fell within the range 0.9–1.1. This is algebraically equivalent to the “two one-tailed tests” method of demonstrating equivalence.

Table 1 presents a summary of the demographic and baseline characteristics of the 18 patients who participated in this study. Patients were either Caucasian (11 patients) or Hispanic (7 patients), with a mean age of 54.4 ± 7.3 years, a mean HbA_1c level of 8.6 ± 1.2%, and a mean C-peptide level of 3.8 ± 1.3 ng/ml. The most common oral hypoglycemic agents used by patients were glyburide (seven patients), glipizide (seven patients), rosiglitazone (five patients), and metformin (three patients).

Demographic and baseline characteristics were generally similar among dosing groups; however, differences among the three groups were noted in duration of diabetes (based on time since diagnosis), HbA_1c levels, and fasting plasma glucose levels.

There were no deaths, serious adverse events, or discontinuations due to adverse events in this study. There were no reports of symptomatic hypoglycemia or hyperglycemia. Adverse events were experienced by a total of 13 patients (72.2%). The most common adverse events were headache (six patients, 33.3%) and anemia (three patients, 16.7%). The anemia, evidenced by decreasing hematocrit and decreasing erythrocyte count in these three patients, was most likely due to the multiple blood draws during the course of the study, since these parameters decreased as the study progressed regardless of which study drug the patient was receiving at any given time. The majority of adverse events were determined by the investigator to be mild in intensity and unrelated to study drug. One patient in the 1.0 mg/kg HIM2 dose group experienced mild adverse events of headache, nervousness, and dizziness that were determined by the investigator to be possibly related to study drug; this patient was receiving oral placebo and subcutaneous regular insulin when these events occurred.

Glucodynamic and pharmacokinetic parameters are summarized for each HIM2 dose group in Table 2. Although there were no consistent differences in glucodynamic parameters between HIM2 and placebo at the lowest dose level (0.375 mg/kg), HIM2 was more effective than placebo at controlling the postprandial rise in plasma glucose concentrations at the 0.5- and 1.0-mg/kg dose levels. This was represented by consistently lower glucodynamic parameter values for HIM2 (0.5 and 1.0 mg/kg) compared with placebo. For example, mean glucose AUC_0–240 and AUCex_30–240 were both lower following administration of 0.5 mg/kg HIM2 compared with placebo (1,097.1 vs. 1,196.9 and 256.6 vs. 301.7 mg · h⁻¹ · dl⁻¹, respectively). Similarly, mean glucose AUC_0–240 and AUCex_30–240 were both lower following administration of 1.0 mg/kg HIM2 compared with placebo (801.1 vs. 992.1 and 207.6 vs. 262.0 mg · h⁻¹ · dl⁻¹, respectively). The difference in glucose AUC_0–240 between HIM2 and placebo was statistically significant at the 1.0-mg/kg HIM2 dose level (P < 0.05 using Wilcoxon’s signed-rank test). In addition, several other glucodynamic parameters were statistically significant between HIM2 (0.5 and 1.0 mg/kg) and placebo as indicated in Table 2.

The lowest dose level of HIM2 (0.375 mg/kg) was less effective than 8 units of subcutaneous regular insulin (HumulinR) at decreasing glucodynamic parameters. However, HIM2 at the 0.5- and 1.0- mg/kg dose levels was as effective as 8 units of subcutaneous insulin at controlling postprandial plasma glucose concentrations. This was represented by similar values for HIM2 (0.5 and 1.0 mg/kg) and subcutaneous insulin (8 units) for the majority of glucodynamic parameters. For example, mean glucose AUC_0–240 was similar following the administration of 0.5 mg/kg HIM2 and 8 units of subcutaneous insulin (1,097.1 vs. 1,048.0 mg · h⁻¹ · dl⁻¹). Likewise, mean glucose AUC_0–240 and AUCex_30–240 were similar following administration of 1.0 mg/kg HIM2 and 8 units of subcutaneous insulin (801.1 vs. 875.2 and 207.6 vs. 201.2 mg · h⁻¹ · dl⁻¹, respectively).

We next explored the relationship between plasma glucose and plasma insulin concentrations following administration of HIM2, placebo, and subcutaneous insulin. In relation to placebo, HIM2 (0.5 mg/kg) resulted in exposure to peripheral insulin concentrations that were similar to placebo (mean insulin AUC_0–240 193.1 vs. 196.1 μU · h⁻¹ · ml⁻¹); however, this dose level of HIM2 was associated with a greater decrease in postprandial glucose concentrations compared with placebo (mean glucose AUC_0–240 of 1,097.1 vs. 1,196.9 mg · h⁻¹ · dl⁻¹). In relation to subcutaneous insulin, HIM2 (0.5 and 1.0 mg/kg) resulted in lower peripheral insulin concentrations compared with 8 units subcutaneous regular insulin (mean insulin AUC_0–240 of 193.1 vs. 233.6 and 230.8 vs. 270.3 μU · h⁻¹ · ml⁻¹, respectively); however, these dose levels of HIM2 resulted in comparable control of postprandial plasma glucose concentrations (mean glucose AUC_0–240 of 1,097.1 vs. 1,048.0 and 801.1 vs. 875.2 mg · h⁻¹ · dl⁻¹, respectively).

Because the 0.5- and 1.0-mg/kg HIM2 dose levels showed the same general effects on postprandial plasma glucose concentrations, we pooled and analyzed the combined data for these two dose groups. The glucose excursion data for the 0.5- and 1.0-mg/kg pooled data, depicted graphically in Fig. 1, illustrate the similarities and differences among HIM2, subcutaneous insulin, and placebo. All three curves are indistinguishable from each other during the first hour postdose. The placebo curve then separates from the other two curves, displaying a significantly higher peak excursion. The HIM2 and subcutaneous insulin curves remain nearly indistinguishable for at least another hour. During the fourth hour postdose, the HIM2 curve clearly separates from the subcutaneous insulin curve, becoming nearly identical to the placebo curve, as the glucose excursion values in all three groups decline toward baseline levels.

The glucodynamic parameters for the 0.5- and 1.0-mg/kg pooled data are summarized in Table 2. The differences in glucodynamic parameters between HIM2 and placebo were statistically significant for the majority of parameters (P < 0.05 using Wilcoxon’s signed-rank test for 2pp, G_max, AUC_0–240, 2pp_ex, and Gex_max). The results of the test of glucodynamic equivalence, presented in the last column of Table 2, imply that HIM2 and subcutaneous insulin were equivalent (to within 10%) with regard to 2pp, G_max, and AUC_0–240. However, glucodynamic equivalence was not demonstrated for the excursion parameters (2pp_ex, Gex_max, and glucose AUCex_30–240).

The peripheral plasma insulin data for the 0.5- and 1.0-mg/kg pooled data, depicted graphically in Fig. 1, revealed an initial peak in peripheral plasma insulin concentrations following administration of HIM2. This initial insulin peak resembled first-phase insulin secretion, which is missing in patients with type 2 diabetes. However, on closer examination, it was observed that this initial insulin peak was due primarily to one patient who had a rapid peak of insulin of 507 μU/ml following administration of 1.0 mg/kg HIM2, which is considerably higher than the C_max for all other subjects receiving 0.5 or 1.0 mg/kg HIM2 (C_max range, 30–155 μU/ml). The median insulin C_max values, following administration of 1.0 mg/kg HIM2 or 8 units subcutaneous regular insulin, were nearly identical (77.0 vs. 77.5 μU/ml, respectively).

In the context of this study, single oral doses of HIM2 were safe and well tolerated in patients with type 2 diabetes. The reported adverse events were predominantly mild, and there were no reports of serious adverse events. Symptomatic events of hyperglycemia or hypoglycemia were not observed during the course of the study.

The goal of this exploratory study was to determine the effectiveness of a single, oral dose of HIM2 in controlling postprandial glycemia in patients with type 2 diabetes. In comparison with placebo treatment, administration of 0.5 and 1.0 mg/kg of HIM2 at 30 min before a meal resulted in lower postprandial plasma glucose levels during a 4-h postdose evaluation period. Statistically significant differences in postprandial plasma glucose parameters between HIM2 and placebo were evident when mean plasma glucose parameters were compared in the 0.5- and 1.0-mg/kg dosing groups and after combining data for these dosing groups.

In comparison with subcutaneous insulin treatment, administration of 0.5 and 1.0 mg/kg of HIM2 at 30 min before a meal resulted in similar postprandial glucose levels during the postdose evaluation period. Equivalence testing demonstrated that HIM2 (0.5- and 1.0-mg/kg pooled data) and subcutaneous insulin (8 units) are equivalent to within 10% with respect to the glucodynamic parameters 2pp, G_max, and glucose AUC_0–240. We were not surprised, however, to see that the excursion parameters, 2pp_ex, Gex_max, and glucose AUCex_30–240, did not meet the stringent equivalence criteria because the excursion parameters themselves are differences between two quantities, and the ratios of differences are unstable numerically, especially when the magnitudes of the differences are small relative to their variability. The glucose excursion data are consistent with the fact that HIM2 is a rapid-acting form of insulin. HIM2 appears to exert its effects primarily within the first 2–3 h postdose, which is consistent with its intended therapeutic use of controlling postprandial glucose excursion.

The potential benefits of oral delivery of insulin include control of plasma glucose levels without peripheral hyperinsulinemia and restoration of the physiological pathway of endogenous insulin in individuals without diabetes. Delivery of therapeutic levels of insulin via the portal route decreases hyperinsulinemia and may result in preservation of the counterregulatory responses to hypoglycemia, with concomitant reduction in hypoglycemic events (15–17). An important observation from the current study was that oral HIM2 at doses of 0.5 and 1.0 mg/kg, in comparison with 8 units of subcutaneous regular insulin, provided comparable postprandial glucose control at lower peripheral insulin concentrations. These data are consistent with the hypothesis that oral delivery of insulin may lead to the development of a portal-to-peripheral insulin gradient similar to the state observed in individuals not given subcutaneous insulin. These data also suggest that a reduced risk of hypoglycemia, compared with subcutaneous insulin therapy, can reasonably be expected in patients who receive oral HIM2 therapy.

The results of this study suggest that oral HIM2 may be useful in patients with type 2 diabetes who experience inadequate postprandial glycemic control.

This study was supported by a grant from Nobex Corporation, Research Triangle Park, NC.

Portions of this study were presented in a late-breaking abstract at the 61st annual meeting of the American Diabetes Association in Philadelphia, June 2001.

The authors thank Robin Tragus and Lisa Martin for managing the patients; David Sorscher, PhD, Caren Carver, MTS, and Christine Chin, PhD, for manuscript preparation; John C. Pezzullo, PhD, for statistical analysis; and Dan Regina, MS, and Ginger Reaves, BS, for technical contributions and data display.