Single-Dose Lopinavir-Ritonavir Acutely Inhibits Insulin-Mediated Glucose Disposal in Healthy Volunteers

Abstract

Previously, we found that 4 weeks of treatment with lopinavir-ritonavir did not decrease insulin sensitivity but did increase adiponectin levels. In the present study, a single dose of lopinavir-ritonavir decreases insulin sensitivity but does not alter adiponectin levels. Insulin resistance from protease inhibitors may decrease with prolonged use; an increase in adiponectin levels may mediate this effect.

Protease inhibitor therapy has been associated with diabetes, hyperglycemia, and insulin resistance. Previously, we reported that 4 weeks of lopinavir-ritonavir therapy did not significantly decrease insulin-mediated glucose disposal [1]. Recently, another study found that insulin-mediated glucose disposal was 24% lower in healthy volunteers who were given lopinavir-ritonavir for 5 days than in a group of persons who received a placebo [2]. It is unclear why these 2 studies produced different results. Differences in duration of medication administration and study design make direct comparison of the studies difficult. In our 4-week study [1], each subject served as their own control, whereas in the 5-day study, comparable—but not identical—groups were used.

Here, we compare the effects of a single dose of lopinavir-ritonavir to placebo in a crossover design. We also measured adiponectin levels to determine whether late induction of adiponectin is a mechanism by which 4-week administration of lopinavir-ritonavir could ameliorate the acute induction of insulin resistance.

Methods

This is a double-blind, randomized, placebo-controlled crossover study. A total of 6 healthy men were recruited. Exclusion criteria included a body mass index >27 (calculated as body weight [kg] divided by height [m²]), a total cholesterol level in serum >6.2 mmol/L, a triglyceride level >3.8 mmol/L, a fasting glucose level >7.0 mmol/L, aspartate or alanine amino-transferase levels in serum >50 U/L, and a creatinine level >124 μM. The protocol was approved by the Committee on Human Research of University of California, San Francisco, and informed consent was obtained from each subject.

Subjects were instructed to eat a diet containing at least 150 g of carbohydrates for 3 days before the start of each arm of the study. Subjects kept a diet journal 3 days before each study began, which was reviewed by a dietitian to assess dietary adherence. Subjects were admitted to the General Clinical Research Center at the San Francisco General Hospital (San Francisco, CA) the morning before the study and began a 24-h urine collection. After an overnight (10-h) fast, patients received either a single dose of lopinavir-ritonavir (Abbott Laboratories; 533 mg/133 mg) or a placebo 2 h before the start of the euglycemic hyperinsulinemic clamp. On completion of the euglycemic hyperinsulinemic clamp, subjects returned within 7–28 days, and studies were repeated using the alternative treatment.

Lopinavir-ritonavir plasma concentrations are known to be variable, and the half-life is a mean ± SD of 5.3 ± 2.5 h [3]. Because lopinavir-ritonavir is normally given with food to increase absorption, we expected that the standard 400-mg/100-mg dose might not achieve therapeutic concentrations during both the previous study and the current study. We chose a 533-mg/133-mg dose to achieve therapeutic drug levels under the fasting conditions required for the performance of the euglycemic hyperinsulinemic clamp.

A 3-h euglycemic hyperinsulinemic clamp was performed as described by DeFronzo et al. [4]. At the start of the clamp (t = 0), insulin (Humulin R, Eli Lilly) was administered as a primed continuous intravenous infusion for 10 min, followed by a constant infusion at the rate of 40 mU/m² per min until the 180-min time point. Whole blood glucose concentration was measured every 5 min. Twenty percent dextrose was infused at a variable rate to maintain the plasma glucose concentration at 4.5 mmol/L, with a coefficient of variation <5%.

Oxidative and nonoxidative glucose disposal were calculated [5]. Oxygen consumption and carbon dioxide production were measured by indirect calorimetry with a metabolic monitor (DeltaTrac). Nonprotein respiratory quotient and substrate oxidation rates were calculated after correction for protein oxidation. The rate of nonoxidative glucose metabolism was calculated by subtracting the rate of carbohydrate oxidation from the rate of dextrose infusion during the clamp.

Fasting lipid, glucose, lactate, and lopinavir levels were measured at the start of the clamp. Lipid levels were measured by enzymatic colorimetric methods (Sigma Diagnostics and Wako Chemicals) [1, 6]. Whole blood and plasma glucose levels, as well as lactate levels, were measured using the 2300 STAT-Plus Glucose and Lactate Analyzer (YSI) [1, 6]. Serum insulin levels were determined by radioimmunoassay (Linco Research) with a 3.2% intra-assay coefficient of variation, a lower detection limit of 14.3 pmol/L. Adiponectin levels were measured by radioimmunoassay (Linco Research). Homeostasis model assessment insulin resistance index was calculated from fasting plasma glucose and fasting serum insulin levels [7]. Lopinavir levels were measured by liquid chromatography and tandem mass spectrometry at the Drug Research Unit at San Francisco General Hospital [1].

Paired t tests were used to compare data during treatment with administration of lopinavir-ritonavir and placebo using Sigma Stat software, version 3.0 (SPSS). Data were recorded as mean ± SEM. P values were 2-tailed.

Results

Subjects ranged in age from 25 to 68 years (mean, 42 ± 7 years); 4 were white, and 2 were African American. Baseline (before the administration of drug or placebo) body weight and body mass index did not differ in each study. Fasting serum insulin, plasma glucose, and lipid samples obtained immediately before the start of the clamp did not differ between the study arms (table 1). The level of lopinavir in plasma reached 7.1 ± 1.6 μM 60 min after dosing and remained >7.2 μM until the end of the study. These levels are comparable to those achieved in patients with HIV infection [3]. During the final hour of the euglycemic hyperinsulinemic clamp, similar steady-state insulin levels (648 ± 52 for the lopinavir-ritonavir arm vs. 637 ± 53 pmol/L for the placebo arm; P = .7) and glucose levels (4.5 ± 0.1 vs. 4.6 ± 0.1 mmol/L; P = .4) were achieved and maintained until the end of the study.

Table 1.

Metabolic parameters of the study participants.

Parameter	Administration of placebo	Treatment with lopinavir-ritonavir	P
Baseline weight, kga	79.4 ± 3.8	79.4 ± 3.8	NS
Baseline body mass indexa	24.6 ± 0.6	24.6 ± 0.7	NS
Preclamp fasting plasma glucose level, mmol/Lb	4.5 ± 0.1	4.6 ± 0.1	.10
Preclamp fasting serum insulin level, pmol/Lb	70.8 ± 5.2	75.6 ± 5.9	NS
Preclamp HOMA IRb	2.0 ± 0.2	2.2 ± 0.2	NS
Preclamp fasting lactate level, mmol/L	0.44 ± 0.04	0.42 ± 0.05	NS
Preclamp fasting cholesterol level, mmol/Lb
Total cholesterol	4.2 ± 0.1	4.1 ± 0.2	NS
LDL cholesterol	2.8 ± 0.04	2.7 ± 0.2	NS
HDL cholesterol	1.1 ± 0.1	1.1 ± 0.09	NS
Preclamp fasting triglyceride level, mmol/Lb	0.67 ± 0.1	0.65 ± 0.1	NS
Preclamp fasting free fatty acid level, mmol/Lb	0.47 ± 0.07	0.48 ± 0.09	NS
Insulin-mediated glucose disposal per unit of insulin at 120–180 min, mg/kg per min per μU/mL × 100	7.9 ± 1.1	6.9 ± 1.0	.048
Nonoxidative glucose disposal, mg/kg per min	5.3 ± 0.8	4.3 ± 1.0	.029
Oxidative glucose disposal, mg/kg per min	1.6 ± 0.3	1.8 ± 0.2	NS
Adiponectin level, μg/mL
Baselinea	13.2 ± 3.0	11.0 ± 1.9	NS
Preclampb	10.7 ± 2.5	11.2 ± 1.8	NS
End of clampc	11.5 ± 2.4	10.7 ± 1.6	NS
Free fatty acid level at 120–180 min time point, mmol/L	0.059 ± 0.005	0.065 ± 0.004	NS

NOTE. Data are mean ± SEM, unless otherwise indicated. Body mass index is calculated as body weight (kg) divided by height (m²). HDL, high-density lipoprotein; HOMA IR, homeostasis model assessment insulin resistance; LDL, low-density lipoprotein; NS, not significant.

^aBaseline measurements were obtained before administration of lopinavir-ritonavir, which occurred 2 h before the start of the euglycemic hyperinsulinemic clamp.

^bPre-clamp measurements were obtained 2 h after administration of lopinavir-ritonavir, immediately before the start of the euglycemic hyperinsulinemic clamp (at the 0-min time point).

^cEnd-of-clamp measurements were obtained at the 180-min time point.

The M/I (defined as the rate of insulin-mediated glucose disposal per unit of insulin) was lower in all 6 subjects in the study using lopinavir-ritonavir, compared with placebo (table 1). The average difference in M/I between the lopinavir-ritonavir and placebo arms was 13%. The nonoxidative component of total glucose disposal decreased by 18% (P = .03). Fasting free fatty acid levels were suppressed comparably with insulin administration in both arms. Lopinavir-ritonavir did not induce an increase in adiponectin levels.

Discussion

We found that a single dose of lopinavir-ritonavir acutely decreased insulin sensitivity, as measured by insulin-mediated glucose disposal during a euglycemic, hyperinsulinemic clamp. This reduction in glucose uptake occurred during therapeutic levels of lopinavir and reflected a reduction in the rate of nonoxidative glucose disposal, suggesting decreased glucose storage. The single dose of lopinavir-ritonavir was given only 2 h before the study began, which is the amount of time required for lopinavir-ritonavir to reach therapeutic levels in the body, demonstrating acute induction of peripheral insulin resistance by lopinavir-ritonavir.

This acute induction of peripheral insulin resistance contrasts with data from our previous study, in which we found no change in insulin sensitivity after 4 weeks of lopinavir-ritonavir treatment, despite induction of hypertriglyceridemia. A similar pattern was also observed in our single-dose and 4-week studies of indinavir, in which we showed that a single dose of indinavir decreased insulin-mediated glucose disposal by 34% in healthy volunteers, whereas 4 weeks of indinavir decreased insulin-mediated glucose disposal by only 17% [8, 9]. Taken together, these data suggest that chronic protease inhibitor administration ameliorates the acute induction of insulin resistance.

One possible explanation for this amelioration of insulin resistance is the increase in adiponectin levels seen with 4 weeks of protease inhibitor treatment. The adipocyte hormone, adiponectin, is associated with increased insulin sensitivity [10]. Previously, we found that adiponectin levels increased by 27% after 4 weeks of treatment with lopinavir-ritonavir [6]. However, in the current study, we found that a single dose of lopinavir-ritonavir did not increase adiponectin levels either at the beginning or by the end of the clamp. The increase in adiponectin levels after 4 weeks may ameliorate the small degree of induction of insulin resistance seen during the single-dose study. Likewise, increased adiponectin levels after 4 weeks of indinavir may partially explain the lesser degree of insulin resistance seen with 4 weeks of indinavir, compared with a single dose of indinavir [8, 9].

Although other factors, such as genetic predisposition towards developing diabetes mellitus, may have contributed to the differences between the 4-week and single-dose studies, 4 of the 6 subjects in this study also participated in the 4-week study. Dietary changes may have also occurred during the 4-week study, but the subjects’ weights remained constant. Finally, a very small reduction in insulin sensitivity may not have been detected in the 4-week lopinavir-ritonavir study; however, such a small reduction in insulin sensitivity may not be clinically significant.

There are several potential limitations to the current study. Because the sample size of the study was small, the overall magnitude of change in insulin sensitivity may have been affected. It should also be noted that we are comparing data from different studies with different, but similar, study populations. Although an increase in adiponectin levels may partially explain the amelioration of induction of insulin resistance, other mechanisms, such as chronic changes in dietary intake, may also compensate for the direct effects on insulin sensitivity. It is also possible that long-term, chronic use of HAART for patients with HIV infection may lead to a decrease in adiponectin levels because of body composition changes and a subsequent increase in insulin resistance. To definitely determine the effects of lopinavir-ritonavir on insulin sensitivity in clinical practice, longer studies are needed.

In summary, a single dose of lopinavir-ritonavir acutely decreases insulin-mediated glucose disposal but is not accompanied by an increase in adiponectin levels. These findings differ from the findings of our previous study, in which 4 weeks of lopinavir-ritonavir did not significantly change insulin sensitivity. An increase in adiponectin levels during chronic administration may contribute to the amelioration in acute induction of insulin resistance.