The Effect of Atazanavir vs. Lopinavir/Ritonavir on

1
11th Conference on Retroviruses and Opportunistic
Infections, February 8-11, 2004 San Francisco, CA
The Effect of Atazanavir vs. Lopinavir/Ritonavir
on Insulin-Stimulated Glucose Disposal Rate in
Healthy Subjects MA Noor1, DM Grasela1, RA
Parker1, U Chaudhari1, D Tackett1, H Uderman1, A
Currie1, S Agarwala1, C Grunfeld2, MF Giordano3,
SL Hodder1, FT Fiedorek1, E OMara1 Bristol-Myers
Squibb Pharmaceutical Research Institute,
1Princeton NJ, and 3Wallingford CT, 2University
of California San Francisco, San Francisco, CA,
USA
4
Mustafa A. Noor, M.D. Phone [phone]/
897-5483 Fax [phone] E-mail
mustafa.noor_at_bms.com
Poster 702
R E S U L T S
I N T R O D U C T I O N
A C K N O W L E D G M E N T S
R E S U L T S (Contd)
D I S C U S S I O N
Thirty six subjects were enrolled and 30
completed the study. Five subjects did not
complete the study due to clamp procedure
technical difficulties one subject failed to
return for repeat studies. All subjects were
male 14 were Caucasian, 9 African-American and 7
Hispanic/Latino. Baseline characteristics are
listed in (Table 1). There were no adverse
clinical or laboratory events. The clamp
procedure was performed near peak plasma
concentrations for each drug (Fig 2a and Fig 2b).
A steady-state insulin level of ? 65 ?IU/ml was
achieved after 30 minutes (Fig 3a) and glucose
was clamped at ? 75 mg/dl (Fig 3b). The adjusted
mean values of insulin-stimulated glucose
disposal per unit of insulin (M/I) for LPV/r, ATV
are shown (Fig 4a). The difference in M/I between
the ATV and LPV/r groups was statistically
significant (2.28 with 95 CI 0.58, 3.97
p0.010). The difference in M/I between the LPV/r
and placebo groups was also statistically
significant (-2.35 with 95 CI -4.05, -0.66
p0.008). The difference in M/I between the ATV
and placebo group was not statistically
significant. The adjusted means of non-oxidative
component of glucose disposal (glycogen storage)
for individuals receiving LPV/r, ATV and placebo
are shown (Fig 4b). There were statistically
significant differences in glycogen storage rate
between the ATV and LPV/r groups (1.46 with 95
CI 0.55, 2.36 p0.003) and between the LPV/r
and placebo groups (-1.31 with 95 CI -2.22,
-0.40 p0.006). The difference in glycogen
storage rates between the ATV and placebo groups
was not statistically significant. Percentage
changes relative to placebo are shown in Figure
5. The adjusted means of fasting lipids for
LPV/r, ATV and placebo are shown in Fig 6.
There were statistically significant differences
in fasting triglycerides between the LPV/r and
ATV groups (-46 mg/dl, plt0.001) and between the
LPV/r and placebo groups (53 mg/dl, plt0.001).
The different in fasting triglycerides between
the ATV and placebo groups was not statistically
significant.
Treatment with some HIV protease inhibitors (PI)
has been associated with insulin resistance,
hyperglycemia, and development of diabetes
mellitus (1,2). Induction of insulin resistance
by a PI precedes any significant changes in
lipids, lipoproteins, or body composition (3) and
can be detected after a single dose of indinavir
(IDV) in healthy volunteers(4). In vitro studies
suggest blockade of the insulin-regulated glucose
transporter, GLUT-4, as a possible mechanism for
PI-associated insulin resistance (5).
Atazanavir (ATV) is a new PI that, unlike IDV,
lopinavir (LPV), and ritonavir (RTV), does not
block GLUT4 activity (6). This studies assesses
insulin-stimulated glucose disposal and glycogen
storage rates in healthy volunteers taking ATV
compared to those taking lopinavir/ritonavir
(LPV/r) or placebo.
This work was performed at the Clinical
Pharmacology Unit of Bristol-Myers Squibb
Company. The authors thank Bruce Oliver,
Marylou Bourgeois, R.N., and the CPU nursing
staff for technical assistance. Special thanks to
Melissa McManus, PharmD., for editorial
assistance.

• Consistent with in vitro observations, we found
  that ATV did not have an effect on glucose
  disposal while LPV/r administration resulted in
  insulin resistance when compared to both ATV and
  placebo. This finding confirms the favorable
  profile of ATV on glucose metabolism observed in
  clinical trials.
• Insulin-stimulated glucose disposal rates were
  consistently lower in individuals on LPV/r
  compared to both placebo and ATV. The magnitude
  of this effect was 24 and within range of the
  effect previously reported on IDV (4,5). The
  decrease is comparable in magnitude to total
  hepatic glucose production in fasted healthy
  individuals ( 2 mg/kg/min).
• Plasma concentrations of PI drugs are highly
  variable as are individual subjects sensitivity
  to insulin. We measured glucose disposal rates
  near peak of plasma concentration for lopinavir
  and atazanavir at comparable insulin levels.
• Our findings are in contrast to those by Lee et
  al who found little or no decrease in
  insulin-stimulated glucose disposal after 4 weeks
  of treatment with LPV/r in healthy volunteers
  (8). However, in that study insulin resistance
  was detected by oral glucose tolerance testing.
  In addition, Lee et al also observed significant
  increases in serum concentrations of free fatty
  acids and triglycerides, both known to be
  associated with insulin resistance.
• Although we have studied acute effects in healthy
  men, long-term results from clinical trials
  indicate that the extent and severity of
  metabolic complications are compounded by HIV and
  occur regardless of gender, suggesting clinically
  important adverse effects of ARV therapy on
  insulin sensitivity across diverse patient groups.

Figure 2a Pharmacokinetic Profile of ATV During
Euglycemic Clamp
Figure 2b Pharmacokinetic Profiles of LPV and
RTV During Euglycemic Clamp
R E F E R E N C E S
1. Dube MP et al. Lancet 1997 350713-4. 2.
Grinspoon S.. Clin Infect Dis. 2003 37 Suppl 2
S85-90.3. 3. Noor MA, et al AIDS. 2001 May 4
15(7) F11-8. 4. Noor MA, et al AIDS. 2002 Mar
29 16(5) F1-8. 5. Murata H et al. J Biol Chem.
2000 Jul 7 275(27) 20251-4. 6. Wang S, et al.
Antiviral Therapy 2003 8L36. 7. DeFronzo RA, et
al 1979 237E214-23. 8. G. A. Lee, et al 10th
Conference on Retroviruses and Opportunistic
Infections, Boston, MA 2003, Abstract No. 748.
LPV/r administered in fasted state on Day 6
ATV administered in fasted state on Day 6
M E T H O D S
Figure 3b Glucose Levels By Treatment During
Euglycemic Clamp
Figure 3a Insulin Levels by Treatment During
Euglycemic Clamp
After giving informed consent, healthy HIV-
seronegative volunteers enrolled into the study.
We excluded subjects with body mass index gt30
kg/m2, serum total cholesterol gt 240 mg/dl
triglycerides gt 200 mg/dl, fasting glucose gt 126
mg/dl , serum aspartate or alanine
aminotransferases gt50 U/l and creatinine gt1.4
mg/dl. Study Design This was a randomized,
double-blind, placebo-controlled, 3-treatment,
2-period cross-over study. All subjects were
admitted to the Bristol-Myers Squibb Clinical
Pharmacology Unit (CPU) as inpatients for 6
consecutive days. Subjects were randomly assigned
to receive atazanavir (Reyataz? Bristol-Myers
Squibb, Princeton, NJ) 400 mg daily,
lopinavir/ritonavir (Kaletra?, Abbot
Laboratories, Abbott Park, IL) 400/100 mg twice
daily, or matching placebo. On the morning of
day 6, a glucose clamp was performed. Treatment
was discontinued on day 7and the subjects were
discharged from the CPU. Two week later subjects
returned at which time they were randomly crossed
over to an alternate treatment arm for repeat
studies(Fig 1). Hyperinsulinemic Euglycemic
Clamp The clamp was performed as described by
DeFronzo et al (7). Insulin was administered as
a primed continuous intravenous infusion,
followed by a constant infusion at the rate of 40
?U/m2?min. A variable infusion of 20 dextrose
was used to maintain plasma glucose concentration
at 75 mg/dl. Steady state glucose disposal rate
was calculated between 60 to 180 minutes.
Statistical Analyses The effect of treatment
was compared by analysis of variance (ANOVA). The
factors in the analysis of variance were
sequence, subject within sequence, period and
treatment. Point estimates and 95 confidence
intervals were constructed for treatment
differences (ATV vs. LPV/r, ATV vs. placebo,
LPV/r vs. placebo). All data are adjusted mean?SE
unless stated otherwise.
Figure 4b Glucose Oxidation and Glycogen Storage
Rates
Figure 4a Mean Glucose Disposal Rate per Unit
of Insulin
C O N C L U S I O N

• Atazanavir (ATV) did not affect insulin
  sensitivity and had no effect on insulin-mediated
  glucose disposal rates as measured by the
  gold-standard glucose clamp technique.
• In contrast, lopinavir/ritonavir (LPV/r) induced
  significant insulin resistance when compared to
  both placebo-treated and ATV-treated subjects.
  LPV/r reduced the mean glucose disposal rate by
  23 relative to ATV and by 24 relative to
  placebo.
• ATV had no effect on the mean glycogen storage
  rate. In contrast, LPV/r significantly reduced
  the mean glycogen storage rate by 36 relative to
  ATV and 34 relative to placebo.
• ATV did not significantly increase fasting
  triglycerides. In contrast, LPV/r significantly
  increased the mean fasting triglycerides by 43
  compared to placebo and by 35 compared to ATV.
• Our findings are consistent with in vitro studies
  of glucose uptake and confirm the favorable
  clinical metabolic profile of ATV.

Figure 5 Effect of ATV vs. LPV/r on Insulin
Sensitivity
Figure 6 Lipid and Lipoproteins Prior to Clamp
(Day 6)
04-005