Increased hyperglycemia and diabetes were reported after the advent of HIV protease inhibitor use. For example, the Multicenter AIDS Cohort Study (MACS) found a 14% prevalence of diabetes .
Several protease inhibitors (indinavir and ritonavir) have been shown to induce insulin resistance [2–5], whereas others have little effect on insulin sensitivity [4,6,7]. Insulin resistance may contribute to diabetes, but the major pathophysiologic defect in the development of diabetes is impairment of insulin secretion [8,9]. The effects of protease inhibitors on insulin secretion are not well studied. The earliest manifestation of impaired insulin secretion is a decrease in first-phase insulin secretion, which occurs in the first 10 min after the administration of intravenous glucose .
In the MACS, ritonavir was the only protease inhibitor associated with diabetes . Ritonavir-boosted lopinavir is frequently used in the treatment of HIV . We have shown that 4 weeks of lopinavir/ritonavir does not induce insulin resistance , but in that study, lopinavir/ritonavir caused a small deterioration in glucose tolerance during oral glucose tolerance testing (OGTT) at 120 min. There was also a trend toward decreased insulin levels at the 30 min time point , which raises the possibility that lopinavir/ritonavir induces a decrease in first-phase insulin secretion, an effect that occurs before 30 min.
Few studies have addressed the issue of impairment of insulin secretion with protease inhibitor use. The only study using the hyperinsulinemic clamp to examine the effect of protease inhibitors on insulin secretion in humans reported that first-phase insulin secretion was lower after 12 weeks of protease inhibitor therapy in HIV-infected participants . However, these participants received several protease inhibitor regimens and a mixed nucleoside reverse transcriptase inhibitor background and their HIV RNA levels decreased by 80% with protease inhibitor therapy. Thus, it is difficult to determine whether the results are an effect of treatment of HIV infection. Two studies using a bolus glucose challenge (FSIVGTT) reported that indinavir or amprenavir had no effect on first-phase insulin secretion [2,7].
Therefore, we evaluated the effects of 4 weeks of administration of lopinavir/ritonavir on insulin secretion using the hyperglycemic clamp and OGTT. We studied healthy HIV-negative volunteers to eliminate other HIV-associated factors.
The protocol was approved by the Committee on Human Research at the University of California, San Francisco (UCSF). Informed consent was obtained. Study participants had no history of medical illness, had normal screening physical examination and hematology/chemistry results, and had negative HIV-1 antibody tests. Exclusion criteria included fasting glucose higher than 5.6 mmol/l, BMI higher than 30 kg/m2, triglycerides higher than 2.13 mmol/l, low-density lipoprotein (LDL) cholesterol higher than 4.87 mmol/l, taking lipid-lowering agents, blood pressure higher than 140/90 mmHg, taking antihypertensive medications, systolic blood pressure lower than 100 mmHg, creatinine higher than 140.8 μmol/l, liver function tests higher than normal, pregnancy, more than three alcoholic drinks per day for men or more than two per day for women, use of amphetamines, cocaine, heroin, or medications with known metabolic effects in the previous 30 days.
Preset post hoc exclusion criteria included missing more than 30% of the study drug, intercurrent illness, peak protease inhibitor level after morning dose during weekly monitoring more than 2 SDs below the mean published therapeutic levels, and peak protease inhibitor levels not within 2 SDs of published peak level during the clamp.
Ten healthy male volunteers were recruited. One was withdrawn for poor adherence (<70% medication compliance) and another for intercurrent infection.
Participants were admitted to the Clinical Research Center (CRC) at San Francisco General Hospital (SFGH) for 4 days and placed on a diet to maintain body weight and minimize influences on metabolism . After baseline studies, participants were discharged on 400 mg lopinavir/100 mg ritonavir (Kaletra; Abbott Laboratories, Abbott Park, Illinois, USA) twice daily. Participants were instructed to take lopinavir/ritonavir with food and maintain usual diet and physical activity. Adherence was monitored weekly by four methods: self-report/direct questioning, pill counting, electronic pill counts using a medication event monitoring system (MEMSCaps; Aprex Corp., Union City, California, USA), and measurement of peak plasma lopinavir and ritonavir levels drawn 1–3 h after last dose. After 4 weeks, participants were readmitted to the CRC for repeat studies. To achieve adequate drug levels, lopinavir/ritonavir was given 2 h before the OGTT and hyperglycemic clamp while fasting.
The hyperglycemic clamp was performed as described by DeFronzo  as modified by Gerich . Whole blood glucose was measured every 5 min using a glucose analyzer. A primed 20% glucose (150 mg/kg body weight) infusion was begun and adjusted every 5 min to clamp glucose at 10 mmol/l for 3 h. Serum insulin and C-peptide concentrations were measured as described [13,15]. Lopinavir and ritonavir levels were measured at baseline and 30, 60, 90, 120, 150, and 180 min. Urine was collected during the clamp to quantify glucose.
First-phase insulin secretion was calculated as sum C-peptide increments above baseline at 2.5, 5.0, 7.5, and 10 min . Second-phase insulin secretion was calculated as average C-peptide concentration during the last 60 min of the clamp . Insulin-mediated glucose disposal (M/I) was calculated using data from the last 60 min of the clamp . Two participants had missing urine samples (one baseline; one 4 weeks). However, the amount of glucose in the urine in other studies was negligible (<0.188 mg/min per kg), so it is unlikely that the missing urine samples would have had a significant effect on results.
Fasting plasma glucose and serum insulin were measured on 2 days; the average was used to calculate insulin resistance index by the homeostasis model assessment (HOMA) . Beta cell function was calculated as described . OGTT was performed as described . Resting energy expenditure (REE) was calculated from indirect calorimetry (DeltaTrac II metabolic monitor; Yorba Linda, California, USA).
Apolipoproteins A1 and B were measured using a BNII Nephelometer (Siemens Healthcare Diagnostics, Deerfield, Illinois, USA); insulin and C-peptide by radioimmunoassay (Millipore Bioscience, St. Charles, Missouri, USA); and glucose, lactate, cholesterol, high-density lipoprotein (HDL) cholesterol, direct LDL cholesterol and triglycerides as described .
Lopinavir and ritonavir levels were measured by liquid chromatography, tandem mass spectrometry within the Drug Research Unit, SFGH. Lopinavir has a lower quantitation limit of 50 ng/ml and inter-assay and intra-assay variation coefficient of 6.0–9.9 and 3.3–6.4%, respectively. Ritonavir has a lower quantitation limit of 25 ng/ml and inter-assay and intra-assay variation co-efficient of 4.2–10.6 and 1.1–9.1%, respectively.
Anthropometrics were measured as described . Total and regional body composition was measured by dual-energy X-ray absorptiometry (DEXA; Hologic scanner, Discovery Wi, software version 12.6) and analyzed as described .
Data are presented as mean ± SEM. Paired t-tests were used with two-tailed P values.
The eight male participants analyzed had mean age 51.5 ± 3.8 (range 31–64). Therapeutic adherence by MEMSCaps averaged 98 ± 1% (range 93–100%). All participants had protease inhibitor levels within 2 SDs of expected peak levels during monitoring (18.9 ± 1.4 μmol/l for lopinavir and 1.26 ± 0.15 μmol/l for boosting ritonavir). At the start of the hyperglycemic clamp, lopinavir levels were 15.4 ± 0.97 μmol/l at t0 min, decreasing during the study to 12.7 ± 1.1 μmol/l at t180 min (Fig. 1a), whereas ritonavir levels were 0.91 ± 0.12 μmol/l at t0 min, decreasing to 0.37 ± 0.03 μmol/l at t180 min (Fig. 1b). There were no significant changes in body composition or REE (Table 1).
Four weeks of lopinavir/ritonavir induced significant increases in triglycerides, total cholesterol, and apolipoprotein B-100 (Table 1). HDL cholesterol levels decreased slightly, but apolipoprotein A1 levels did not change. The difference in direct LDL cholesterol was small, not reaching significance.
There was also no significant difference in fasting glucose, insulin, or HOMA (Table 1). We found no trend toward deterioration in glucose tolerance or insulin levels during OGTT (Table 1). We found no significant change in insulin sensitivity (M/I) from lopinavir/ritonavir during the clamp (Table 1).
During the hyperglycemic clamp, there was no significant change in first-phase insulin secretion measured by C-peptide levels (<4% change, P = 0.60; Fig. 1c). There was also no change in second-phase C-peptide levels (Fig. 1d).
Given the report that protease inhibitor therapy of HIV-infected participants was associated with decreased insulin secretion  and the trend we observed previously toward decreased insulin levels at 30 min during OGTT after 4 weeks of lopinavir/ritonavir administration to healthy HIV-seronegative volunteers, we anticipated finding decreased insulin secretion after a similar 4-week administration . Surprisingly, lopinavir/ritonavir at therapeutic concentrations did not have significant effects on either first-phase or second-phase insulin secretion in healthy HIV-seronegative volunteers using the hyperglycemic insulin clamp. First-phase insulin secretion is best measured during the first 10 min after intravenous glucose using the hyperglycemic clamp or FSIVGTT.
In contrast, a 24% decrease in first-phase insulin secretion was found by hyperglycemic clamp after HIV-infected patients were treated with various protease inhibitor (including lopinavir/ritonavir) for 12 weeks . The possible explanations for these differences are of interest. Baseline first-phase insulin release in HIV-infected volunteers (5.9 ± 0.8 nmol/l) was somewhat higher than in normal volunteers (4.6 ± 0.6 nmol/l). Although first-phase insulin secretion decreased significantly to 4.5 ± 0.7 nmol/l after protease inhibitor therapy in these HIV-infected patients, it only reached the levels of healthy volunteers . Therefore, it is unclear whether the decrease in insulin secretion represents a pathological change that would lead to diabetes.
Furthermore, in that study, HIV RNA levels decreased by more than 80% after protease inhibitor therapy , making it difficult to separate the contributions of various HIV-related factors, including HIV infection, immune reconstitution, restoration to health, and body composition changes from possible direct effects of protease inhibitor drugs per se. Our study was designed to look at direct effects of protease inhibitors on insulin secretion by studying them in the absence of HIV-related effects through the use of healthy normal volunteers. As we found little difference in insulin secretion, our data suggest that HIV-related factors may affect insulin secretion rather than a direct effect of lopinavir/ritonavir.
Previously, we found a small deterioration in glucose tolerance and a trend toward decreased insulin levels at 30 min during OGTT after 4 weeks of lopinavir/ritonavir, suggesting a defect in insulin secretion . Noor et al.  found a similar deterioration in glucose tolerance in healthy volunteers given lopinavir/ritonavir for 10 days. However, here we found no change in glucose tolerance or insulin levels during OGTT. The OGTT is known to have significant intra-patient variability; one study found a 46% test–retest difference using the same volunteers . Therefore, the 28% deterioration in glucose tolerance found previously could reflect those limitations.
There are potential limitations to this study. We cannot rule out the effect of other antiretroviral drugs on insulin secretion. Of the 13 HIV-infected patients studied, only three received ritonavir, whereas nine received nelfinavir . Dube et al. [2,7] found no effect of unboosted indinavir or amprenavir on insulin release. However, it is the of the protease inhibitor ritonavir that has been linked to diabetes in HIV infection .
In summary, 4 weeks of lopinavir/ritonavir administration to healthy normal volunteers caused no changes in insulin secretion by hyperglycemic clamp, despite expected changes in lipids. Thus, direct effects of this combination therapy on insulin secretion cannot explain the increase in diabetes seen in HIV infection associated with ritonavir . Whether an increase in diabetes will continue to be seen with other antiretroviral drugs needs study.
V.P. and G.L. contributed to the manuscript in its acquisition, analysis, and interpretation of the data, drafting and critically revising the manuscript, and obtaining funding. S.T. contributed to the manuscript by acquisition, analysis, and interpretation of the data. F.A., J.-M.S., K.M., and M.S. contributed to the manuscript in its conception and design, analysis, and interpretation of the data, and critically revising the manuscript. C.G. contributed to the conception and design, acquisition, analysis and interpretation of the data, drafting and critically revising the manuscript, and obtaining funding.
The authors thank the CRC staff, Heather Southwell, Judy Shigenaga, Arthur Moser, Michael Wen, Sophie Patzek, Viva Tai, and Patricia Lizak for their assistance.
This study was supported by the National Institutes of Health (DK54615, DK63640, DK69185 and DK66999), Bristol-Myers-Squibb and with resources and the use of facilities of the VA Medical Center, San Francisco, California, USA. Individuals were studied in the CTSI Clinical Research Center at San Francisco General Hospital with support by the National Center for Research Resources, National Institutes of Health (RR00083) and NIH/NCRR UCSF-CTSI Grant UL1 RR024131. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH or Bristol-Myers-Squibb, who played no role in analyzing the data or writing the manuscript.
Clinicaltrials.gov ID: NCT00525239.
1. Brown TT, Cole SR, Li X, Kingsley LA, Palella FJ, Riddler SA, et al
. Antiretroviral therapy and the prevalence and incidence of diabetes mellitus in the multicenter AIDS cohort study. Arch Intern Med 2005; 165:1179–1184.
2. Dube MP, Edmondson-Melancon H, Qian D, Aqeel R, Johnson D, Buchanan TA. Prospective evaluation of the effect of initiating indinavir-based therapy on insulin sensitivity and b-cell function in HIV-infected patients. J Acquir Immune Defic Syndr 2001; 27:130–134.
3. Noor MA, Lo JC, Mulligan K, Schwarz JM, Halvorsen RA, Schambelan M, Grunfeld C. Metabolic effects of indinavir in healthy HIV-seronegative men. AIDS 2001; 15:F11–F18.
4. Lee GA, Rao MN, Mulligan K, Lo JC, Aweeka F, Schwarz JM, et al. The effects of ritonavir and amprenavir on insulin-mediated glucose disposal in healthy volunteers: two randomized, placebo-controlled, cross-over trials
5. Noor MA, Seneviratne T, Aweeka FT, Lo JC, Schwarz JM, Mulligan K, et al
. Indinavir acutely inhibits insulin-stimulated glucose disposal in humans: a randomized, placebo-controlled study. AIDS 2002; 16:F1–F8.
6. Noor MA, Parker RA, O'Mara E, Grasela DM, Currie A, Hodder SL, et al
. The effects of HIV protease inhibitors atazanavir and lopinavir/ritonavir on insulin sensitivity in HIV-seronegative healthy adults. AIDS 2004; 18:2137–2144.
7. Dube MP, Qian D, Edmondson-Melancon H, Sattler FR, Goodwin D, Martinez C, et al
. Prospective, intensive study of metabolic changes associated with 48 weeks of amprenavir-based antiretroviral therapy. Clin Infect Dis 2002; 35:475–481.
8. Defronzo RA. Banting Lecture. From the triumvirate to the ominous octet: a new paradigm for the treatment of type 2 diabetes mellitus. Diabetes 2009; 58:773–795.
9. DeFronzo RA. Pathogenesis of type 2 diabetes mellitus. Med Clin North Am 2004; 88:787–835, ix.
10. Hammer SM, Eron JJ Jr, Reiss P, Schooley RT, Thompson MA, Walmsley S, et al
. Antiretroviral treatment of adult HIV infection: 2008 recommendations of the International AIDS Society-USA panel. JAMA 2008; 300:555–570.
11. Lee G, Lo J, Aweeka F, Schwarz J, Mulligan K, Schambelan M, Grunfeld C. Single-dose lopinavir-ritonavir acutely inhibits insulin-mediated glucose disposal in healthy volunteers. Clin Infect Dis 2006; 43:658–660.
12. Lee GA, Seneviratne T, Noor MA, Lo JC, Schwarz JM, Aweeka FT, et al
. The metabolic effects of lopinavir/ritonavir in HIV-negative men. AIDS 2004; 18:641–649.
13. Woerle HJ, Mariuz PR, Meyer C, Reichman RC, Popa EM, Dostou JM, et al
. Mechanisms for the deterioration in glucose tolerance associated with HIV protease inhibitor regimens. Diabetes 2003; 52:918–925.
14. Schwarz J, Neese R, Turner S, Dare D, Hellerstein M. Short-term alterations in carbohydrate energy intake in humans. Striking effects on hepatic glucose production, de novo lipogenesis, lipolysis, and whole-body fuel selection. J Clin Invest 1995; 96:2735–2743.
15. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979; 237:E214–E223.
16. Matthews D, Hosker J, Rudenski A, Naylor B, Treacher D, Turner R. Homeostasis model assessment: insulin resistance and beta-cell function from fasting plasma glucose and insulin concentrations in man. Diabetologia 1985; 28:412–419.
17. Lohman TG, Roche AF, Martoll R, editors. Anthropometric standardization reference manual, abridged edition
. Champaign, IL: Human Kinetics Books; 1988.
18. Noor MA, Flint OP, Maa JF, Parker RA. Effects of atazanavir/ritonavir and lopinavir/ritonavir on glucose uptake and insulin sensitivity: demonstrable differences in vitro and clinically. AIDS 2006; 20:1813–1821.
19. Mooy JM, Grootenhuis PA, de Vries H, Kostense PJ, Popp-Snijders C, Bouter LM, Heine RJ. Intra-individual variation of glucose, specific insulin and proinsulin concentrations measured by two oral glucose tolerance tests in a general Caucasian population: the Hoorn Study. Diabetologia 1996; 39:298–305.
Keywords:© 2010 Lippincott Williams & Wilkins, Inc.
HIV; HIV protease inhibitors; insulin resistance; ritonavir; triglycerides