AIDS RESEARCH AND HUMAN RETROVIRUSES

Volume 28, Number 10, 2012

ª Mary Ann Liebert, Inc.
DOI: 10.1089/aid.2011.0327

Metabolic Effects of Darunavir/Ritonavir Versus

Atazanavir/Ritonavir in Treatment-Naive,
HIV Type 1-Infected Subjects over 48 Weeks

Judith A. Aberg,1 Pablo Tebas,2 Edgar Turner Overton,3 Samir K. Gupta,4 Paul E. Sax,5
Alan Landay,6 Ron Falcon,7 Robert Ryan,8 and Guy De La Rosa 9

Abstract

We assessed metabolic changes for darunavir/ritonavir (DRV/r) once daily (qd) versus atazanavir/ritonavir
(ATV/r) qd with fixed-dose tenofovir/emtricitabine. This was a phase 4, multicenter, open-label, randomized
exploratory study. Treatment-naive, HIV-1-infected adults received DRV/r 800/100 mg qd or ATV/r 300/
100 mg qd, both with emtricitabine/tenofovir 200/300 mg qd. Primary end point: change in triglyceride levels
from baseline to week 12. Secondary end points: week 12 and week 48 changes in lipid parameters, insulin
sensitivity, inflammatory/coagulation/bacterial translocation biomarkers, viral load, CD4+ cell count, and week
48 changes in adipose tissue distribution and subjects’ perceptions of body changes. In the DRV/r arm, 32/34
and 29/34 subjects completed weeks 12 and 48, respectively; in the ATV/r arm, 30/31 and 25/31 subjects
completed weeks 12 and 48, respectively. Small changes in lipid parameters from baseline to weeks 12 and 48
were observed in both arms. Differences were noted between arms in mean changes in total cholesterol (DRV/r,
20.3 mg/dl; ATV/r, 4.6 mg/dl) and apolipoprotein A1 (DRV/r, 10.7 mg/dl; ATV/r, –0.7 mg/dl) at week 12. At
week 48, no clinically relevant differences between arms were noted for changes in any lipid parameter, fasting
glucose, or insulin sensitivity. Biomarkers generally decreased and efficacy parameters improved in both arms
over 48 weeks. Changes in adipose tissue were small and comparable between arms. Subjects’ perceptions of
body changes generally improved in both study arms. This first pilot comparison in HIV-1-infected subjects
suggests that DRV/r has a metabolic profile similar to ATV/r over 48 weeks of treatment. Further randomized
studies are warranted.

Introduction boosted PIs have been associated with a worsening of lipid

parameters and increases in inflammatory markers.4–7 Ad-

H IV-infected individuals have a heightened risk of se-

rious non-AIDS conditions, such as cardiovascular dis-
ease (CVD), compared with uninfected individuals, due to
ditionally, some PIs have been associated with insulin resis-
tance in HIV-negative healthy volunteers.8–11 However, this
association with insulin resistance seems to be ARV-specific,
HIV-induced activation of inflammation and coagulation as opposed to class-specific, as studies evaluating other PIs
pathways.1,2 Antiretroviral (ARV) agents, including protease have failed to demonstrate such an association.12,13 The met-
inhibitors (PIs), also contribute to this increased risk as a result abolic complications associated with PI-based therapy can
of their metabolic complications.3 Specifically, some ritonavir- develop within the first week after treatment initiation.6,14–18

1
Bellevue Hospital Center at New York University School of Medicine, New York, New York.
2
AIDS Clinical Trials Unit, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania.
3
Washington University School of Medicine, St. Louis, Missouri.
4
Indiana University School of Medicine, Indianapolis, Indiana.
5
Division of Infectious Diseases, Brigham and Women’s Hospital, and Harvard Medical School, Boston, Massachusetts.
6
Rush University Medical Center, Chicago, Illinois.
7
Janssen Therapeutics, Titusville, New Jersey.
8
Janssen R&D, Titusville, New Jersey.
9
Janssen Global Services, Titusville, New Jersey.

1184
METABOLIC EFFECTS OF DARUNAVIR 1185

Commonly prescribed PIs exert differential effects on lipid, HIV-1 RNA 1000 copies/ml or higher; there were no CD4 +
glucose, and insulin parameters, and comprehensive assess- count restrictions. Subjects were required to have demon-
ment of these metabolic effects will clarify the optimal treat- strated sensitivity to DRV, ATV, TDF, and FTC by resistance
ment choices. testing (DRV, ATV, and TDF susceptibility determined by
Thymidine analogs and, to a lesser degree, PIs and non- Antivirogram, Virco Lab, Inc., Raritan, NJ; FTC susceptibility
nucleoside reverse transcriptase inhibitors have also been determined by virco TYPE HIV-1, Virco Lab, Inc., Raritan,
associated with lipoatrophy and lipodystrophy.5,19–22 In the NJ). Exclusion criteria included body mass index greater
HIV-1-infected population, lipodystrophy is an umbrella term than 30 kg/m2; fasting glucose greater than 110 mg/dl; low-
that encompasses all changes in fat distribution, including density lipoprotein (LDL) greater than 130 mg/dl; triglycer-
lipohypertrophy, which can include accumulation of fat in the ides greater than 200 mg/dl; alanine aminotransferase greater
neck, chest, back, breasts, and/or abdomen, and lipoatrophy, than 2.5 times the upper limit of normal; creatinine clearance
which may include fat loss from the limbs, buttocks, and/or 50 ml/min/m2 or lower; evidence of significantly decreased
face.23–25 HIV-1-associated changes in body shape have the hepatic function or decompensation; presence of any Centers
potential to impact a subject’s quality of life and well-being, for Disease Control and Prevention active AIDS-defining
and several studies have linked quality of life to survival of illness (Category C conditions), except stable cutaneous
HIV-1-infected subjects.26–28 Interestingly, subjects’ percep- Kaposi’s sarcoma or wasting syndrome; acute or chronic
tions of the changes to their body during ARV therapy are not hepatitis A, B, or C; grade 3 or 4 laboratory abnormalities;
always in accordance with the objective changes measured in history of significant cardiac, vascular, pulmonary, gastroin-
a clinical setting.29 Despite this, few studies have specifically testinal, endocrine, neurologic, hematologic, rheumatologic,
investigated subjects’ perceptions of changes in their bodies psychiatric, or metabolic disturbances; use of any non-
due to ARV therapy.30,31 ARV investigational agents within 90 days of screening; re-
Darunavir (DRV; PREZISTA, Janssen Therapeutics, Divi- ceipt of anabolic steroids, atypical antipsychotics, or growth
sion of Janssen Products, LP, Titusville, NJ), a PI, combined hormones; use of disallowed concomitant therapy; and
with low-dose ritonavir (DRV/r), has a favorable lipid profile pregnancy or breastfeeding. Use of lipid-lowering medica-
in healthy and treatment-naive subjects.32,33 Another PI, ata- tions, either prescription (e.g., statins or fibrates) or over-
zanavir (ATV; REYATAZ, Bristol-Myers Squibb, Princeton, the-counter (e.g., fish oil), was prohibited from 28 days before
NJ), boosted with low-dose ritonavir (ATV/r), has also baseline through week 12 of the trial. The use of lipid-lowering
demonstrated a favorable metabolic profile in treatment-na- medications was allowed after week 12. All subjects provided
ive subjects.34 The metabolic effects of DRV/r have been written informed consent.
shown to be comparable with those of ATV/r in HIV-negative
subjects.33 Furthermore, the metabolic effects of these ARVs
Study evaluations
have not been directly compared in HIV-1-infected, treatment-
naive subjects. The primary end point was the change in triglyceride levels
Presented here are the week 12 primary end point analysis from baseline to week 12. Secondary end points included
and the week 48 results of METABOLIK (Metabolic Evalua- week 12 and week 48 changes in other lipid parameters that
tion in Treatment-naı̈ves Assessing the impact of two BOosted included total cholesterol (TC), high-density lipoprotein
protease inhibitors on LIpids and other marKers), an explor- (HDL), LDL, and apolipoproteins (apo) A1 and B. Lipid pa-
atory study evaluating metabolic outcomes of DRV/r-based rameters were tested for normality. Additional secondary end
therapy compared with those of ATV/r-based therapy in points assessed at week 12 and week 48 included changes in
treatment-naive, HIV-1-infected adult subjects. Additionally, glucose and insulin levels, insulin sensitivity (as measured by
we report changes in adipose tissue distribution and subjects’ the homeostasis model assessment of insulin resistance
perceptions of body changes over the course of the trial. [HOMA-IR] method),35 inflammatory biomarkers (interleu-
kin [IL]-1 beta, IL-6, tumor necrosis factor receptor II [TNF
RII], high sensitivity C-reactive protein [hs-CRP]), coagula-
Materials and Methods
tion biomarkers (fibrinogen, d-dimer), and the microbial
Study design and treatment translocation biomarker lipopolysaccharide (LPS). In addi-
tion, polychromatic flow cytometry was used to assess
METABOLIK was a 48-week, phase 4, multicenter, open-
changes in the percentage of CD3+/CD4+ and CD3+/CD8+ T
label, randomized study that assessed changes in fasting lip-
cells, along with their markers of immune activation (CD38+/
ids, glucose, insulin, insulin sensitivity, biomarkers, and the
HLADR+); senescence (CD28-/CD57+); proliferation (Ki67+);
safety and efficacy of DRV/r-based versus ATV/r-based
and naive/memory subsets (naive, CCR7+/CD45RA+; central
therapy in HIV-1-infected, treatment-naive adults. Adult
memory, CCR7+/CD45RA-; effector memory, CCR7-/
subjects were randomized in a 1:1 ratio, stratified by sex, to
CD45RA-; terminal effector, CCR7-/CD45RA+).
receive DRV/r 800/100 mg once daily (qd) or ATV/r 300/
For measuring insulin sensitivity, HOMA-IR was calcu-
100 mg qd, both with a fixed-dose combination of em-
lated using the following formula: {[fasting insulin (lU/
tricitabine/tenofovir disoproxil fumarate (FTC/TDF) 200/
ml)] · [fasting glucose (mmol/liter)]}/22.5.36 The lower limits
300 mg qd. Both DRV and ATV, with ritonavir, were admin-
of quantification (laboratory variability; normal reference
istered within 30 min of a meal to maximize drug exposure.
range) for the biomarker assays are as follows: IL-1 beta,
0.125 pg/ml (6.7%; < 0.201 pg/ml); IL-6, 0.2 pg/ml (4.3%;
Subject population
< 11.83 pg/ml); TNF RII, 1 pg/ml (2.6%–4.8%; 1003–3170 pg/
Eligible subjects were at least 18 years old and naive to ARV ml); hs-CRP, 0.15 mg/liter ( < 2%; < 5.00 mg/liter); fibrinogen,
therapy ( £ 10 days’ previous ARV therapy at any point) with 42 mg/dl ( < 4%; 211–372 mg/dl); and d-dimer, 109 ng/ml
1186 ABERG ET AL.

( < 4%; < 201 ng/ml; ICON Central Laboratories Inc., Farm- ues. All other end points use observed values for the intent-to-
ingdale, NY). For the determination of LPS concentration, the treat (ITT) population. The lipid and biomarker end points were
QCL-1000 assay (Lonza Biologics Inc., Walkersville, MD), estimated using mean values with 95% CIs. Response rates were
with a sensitivity range of 0.1–1.0 endotoxin units/ml (1 en- derived using the confirmed virologic response (CVR; HIV-1
dotoxin unit = 100 pg), and low intraassay variability, was RNA < 50 copies/ml, confirmed by two consecutive assess-
used. Change in log10 HIV-1 RNA from baseline, proportion ments at least 14 days apart) algorithm for the ITT population.
of subjects with virologic response (HIV-1 RNA < 50 copies/
ml), change in CD4+ cell count through week 48, and safety Results
were also assessed.
Subject population and baseline characteristics
For analysis of fat redistribution, computed tomography
(CT) scans were performed at the L4–L5 level and mid-thigh In total, 34 subjects (29 men) received DRV/r-based regi-
at baseline and week 48, and centrally analyzed for total mens and 31 (27 men) received ATV/r-based regimens. In the
(TAT), subcutaneous (SAT), visceral (VAT), and peripheral DRV/r arm, 32 subjects completed week 12 and 29 subjects
(PAT) adipose tissue by BioClinica Imaging Technologies, Inc. completed week 48; in the ATV/r arm, 30 and 25 subjects
(Newtown, PA). The 27-item, self-reported Assessment of completed week 12 and week 48, respectively. The lipid-
Body Change and Distress (ABCD) questionnaire37 was ad- evaluable populations included 28 and 27 subjects in the
ministered at baseline and at weeks 12 and 48 and was com- DRV/r and ATV/r arms, respectively. Of five (14.7%) subjects
posed of three parts. The categorical portion of the in the DRV/r arm who discontinued prior to week 48, two
questionnaire asked subjects to assign one of five answers, withdrew consent, one was noncompliant, one was lost to
ranging from ’’all of the time’’ to ’’none of the time,‘‘ to follow-up, and one relocated. Of six (19.4%) subjects in the
questions encompassing a wide range of disease-related is- ATV/r arm who discontinued early, two discontinued due to
sues. The yes/no portion of the questionnaire asked subjects AEs (one with grade 3 leukocytoclastic vasculitis and one
to indicate whether they had noticed changes in aspects of with grade 1 increased blood creatinine), one discontinued
their physical appearance. Finally, the questionnaire asked due to pregnancy, one discontinued because of investiga-
subjects to rate their overall satisfaction with their body. The tional product dispensing error, one was lost to follow-up,
questionnaire had a recall period of 4 weeks and measured and one withdrew consent. At baseline, DRV/r subjects had
subjects’ perceived body changes, physical and emotional higher mean log10 baseline viral loads, lower median CD4 +
distress, social concerns, and health behavioral changes. Re- counts, and lower TC and LDL levels compared with ATV/r
sponses to the ABCD questionnaire were not considered or subjects (Tables 1 and 2).
reported as adverse events (AEs). Over the course of the study, the median duration of ex-
Fasting (at least 8 h) blood samples for laboratory assess- posure to DRV or ATV was similar in the DRV/r (337 days)
ments, lipids, and efficacy parameters were taken at weeks 0,
4, 8, 12, 24, 36, and 48, and for biomarker tests at weeks 0, 4, 12,
24, and 48. Lipid parameters, glucose, and insulin assessments
Table 1. Baseline Demographics and Disease
were analyzed by ICON Central Laboratories (Farmingdale,
Characteristics
NY). Biomarkers were analyzed by Rush University Medical
Center (Chicago, IL) and ICON Central Laboratories. The DRV/r ATV/r
incidence and type of all AEs and serious AEs (SAEs) were Parameter n = 34 n = 31
recorded from the signing of the informed consent form
through the completion of the last study-related procedure. Male, n (%) 29 (85.3) 27 (87.1)
All procedures followed were in accordance with the ethical Age, median 36.5 (19.0–58.0) 35.0 (20.0–65.0)
(range), years
standards of the responsible committee on human experi-
mentation (institutional and national) and with the Helsinki Race, n (%)
Declaration of 1975, as revised in 2000. Asian 0 2 (6.5)
Black 13 (38.2) 17 (54.8)
White 21 (61.8) 12 (38.7)
Statistical analysis Ethnicity, n (%)
Hispanic/Latino 7 (20.6) 7 (22.6)
The primary end point for this study was change in tri-
glyceride levels from baseline to week 12. Assuming a BMI, mean (SD) 23.8 (3.1) 24.5 (3.6)
standard deviation (SD) of 75 mg/dl for the primary end Worst clinical stage of
point and a two-sided 95% confidence interval (CI) with a HIV infection, n (%)
precision of 42 mg/dl on each side of the estimated differ- A 30 (88.2) 26 (83.9)
ence, it would be required that at least 50 subjects complete B 4 (11.8) 4 (12.9)
C 0 1 (3.2)
the study (25 subjects per treatment arm). To allow for
CD4 + count, median 267 (10–532) 316 (39–813)
dropouts, an overall sample size of 60 subjects was planned. (range), cells/mm3
Descriptive statistics for the preplanned analyses of the pri-
Viral load, mean (SD), 5.0 (0.8) 4.6 (0.7)
mary and secondary end points at week 12 and week 48 are
log10 copies/ml
reported. The lipid-evaluable population used for the week
12 and week 48 lipid end points includes subjects completing Viral load, median 137,000 46,100
(range), copies/ml (642–2,450,000) (397–637,000)
week 12 who had a baseline value and at least one postdose
fasting lipid value and no relevant protocol deviations or DRV/r, darunavir/ritonavir; ATV/r, atazanavir/ritonavir; BMI,
violations; lipid analyses were completed using observed val- body mass index; SD, standard deviation.
 Table 2. Changes in Primary and Secondary Lipid End Points, Glucose, Insulin, Insulin Sensitivity,
and Inflammatory/Coagulation Biomarkers over 12 and 48 Weeks (Observed Values)

DRV/r ATV/r
n = 28 n = 27
Difference in week 12 mean
BL Change from BL at week 12 BL Change from BL at week 12 change between arms (95% CI)

Primary end point,a mean (SD)

TG, mg/dl 113.7 (57.4) 22.0 (62.7) 114.2 (84.1) 8.1 (81.2) 13.8 ( - 25.8, 53.4)

Change Change Change Change Difference in

from BL at from BL from BL from BL week 48 mean change
BL week 12 at week 48 BL at week 12 at week 48 between arms (95% CI)

Secondary lipid end points,a mean (SD)

TG, mg/dl 113.7 (57.4) 22.0 (62.7) 26.1 (69.0) 114.2 (84.1) 8.1 (81.2) 9.6 (73.7) 16.5 ( - 25.0, 58.0)
TC, mg/dl 141.8 (28.3) 20.3 (30.5) 22.3 (30.7) 165.1 (30.0) 4.6 (26.7) 11.8 (31.9) 10.5 ( - 7.7, 28.8)
LDL, mg/dl 84.6 (21.9) 13.6 (25.1) 14.7 (25.9) 100.2 (23.9) 9.6 (20.8) 13.9 (27.1) 0.8 ( - 14.6, 16.3)
HDL, mg/dl 37.9 (13.4) 6.6 (11.6) 6.0 (7.4) 45.0 (13.6) 2.2 (8.7) 3.7 (9.9) 2.3 ( - 2.8, 7.3)
TC/HDL ratio 4.1 (1.1) - 0.1 (0.9) 0.1 (1.06) 3.9 (1.0) - 0.1 (0.7) - 0.1 (0.75) 0.2 ( - 0.3, 0.8)
ApoA1, mg/dl 114.9 (25.7) 10.7 (21.3) 12 (16) 127.6 (21.9) - 0.7 (17.8) 3 (19) 9.7 ( - 0.5, 19.8)

1187
ApoB, mg/dl 74.5 (19.0) - 0.4 (20.0) 4 (21) 81.7 (18.5) - 4.9 (16.2) 2 (17) 2.0 ( - 9.3, 13.4)
ApoB/ApoA1 ratio 0.68 (0.20) - 0.06 (0.17) - 0.01 (0.20) 0.65 (0.16) - 0.04 (0.16) 0.01 (0.14) - 0.02 ( - 0.127, 0.079)
Glucose, insulin, and HOMA-IR,b mean (SD)
Glucose, mg/dl 88.5 (12.37) 1.5 (12.52) 2.8 (9.10) 89.7 (10.84) 5.8 (14.55) 6.4 (22.07) - 3.6 ( - 12.8, 5.6)
Insulin, lIU/ml 6.0 (5.57) - 1.1 (4.97) 1.0 (6.01) 8.6 (14.28) 0.7 (18.79) - 2.9 (16.73) 3.8 ( - 3.0, 10.6)
HOMA-IR 1.6 (1.70) - 0.5 (2.02) 0.04 (2.26) 2.9 (6.02) 0.1 (7.51) - 1.24 (8.01) 1.3 ( - 2.7, 5.2)
b
Biomarkers, mean (SD)
IL-1 beta, pg/ml 0.2 (0.32) 0.01 (0.22) 0.3 (1.45) 0.3 (0.33) - 0.01 (0.28) - 0.1 (0.28) 0.41 ( - 0.23, 1.04)
IL-6, pg/ml 1.9 (1.90) - 0.6 (2.88) 0.2 (7.30) 1.0 (1.32) 1.5 (6.34) 0.3 (0.94) - 0.08 ( - 3.24, 3.08)
hs-CRP, mg/liter 3.1 (5.16) - 0.6 (5.97) 1.2 (11.20) 2.2 (2.50) 0.7 (4.17) 0.6 (5.09) 0.65 ( - 4.55, 5.85)
d-dimer, ng/ml 373.0 (580.45) - 181.6 (580.91) - 192.1 (586.55) 189.0 (111.43) 51.0 (635.25) - 24.1 (143.98) - 168.0 ( - 432.2, 96.2)
Fibrinogen, g/liter 3.3 (1.05) - 0.5 (1.12) - 0.3 (1.08) 3.2 (0.70) - 0.1 (0.89) - 0.3 (0.90) 0.02 ( - 0.57, 0.61)
TNF RII, pg/ml 4207 (1701.7) - 1456 (1518.5) - 1384 (1722.3) 2957 (727.2) - 562 (529.6) - 442 (722.2) - 942.1 ( - 1735.3, - 149.0)
LPS, pg/ml 85.3 (29.2) - 2.5 (38.4) - 18.4 (34.9) 86.2 (30.0) - 6.9 (25.3) - 17.0 (50.6) - 1.4 ( - 25.6, 22.9)

Sample size varies by time point and parameter.

a
Lipid-evaluable population.
b
Intent-to-treat population.
DRV/r, darunavir/ritonavir; ATV/r, atazanavir/ritonavir; CI, confidence interval; BL, baseline; SD, standard deviation; TG, triglyceride; TC, total cholesterol; LDL, low-density lipoprotein; HDL,
high-density lipoprotein; Apo, apolipoprotein; HOMA-IR, homeostasis model assessment of insulin resistance; IL, interleukin; hs-CRP, high-sensitivity C-reactive protein; TNF RII, tumor necrosis factor
receptor II; LPS, lipopolysaccharide.
1188 ABERG ET AL.

and ATV/r (336 days) arms. No subjects received lipid- sulin sensitivity with DRV/r or ATV/r therapy (Table 2).
lowering medications after week 12. Biomarkers of inflammation, coagulation and microbial
translocation generally decreased from baseline to week 12 in
Lipid evaluations both treatment arms (Table 2). From baseline to week 48, fi-
brinogen, d-dimer, TNF RII, and LPS decreased in both
Primary end point: week 12 change in triglyceride
treatment arms, while hs-CRP and IL-6 demonstrated small
levels. From baseline to week 12, triglyceride levels in-
increases in both arms (Table 2). Small increases in IL-1 beta
creased by a mean of 22.0 mg/dl (SD: 62.7) in the DRV/r arm
were seen in the DRV/r arm, whereas small decreases were
and 8.1 mg/dl (SD: 81.2; Table 2) in the ATV/r arm. The
observed in the ATV/r arm, from baseline to week 48. At both
difference in week 12 mean change (95% CI) between the
time points, differences were noted between arms for changes
DRV/r and ATV/r arms was 13.8 (–25.8, 53.4). A sensitivity
in TNF RII, with the DRV/r arm experiencing greater de-
analysis was conducted on the changes in triglyceride levels
creases than the ATV/r arm.
using normalized (natural log) triglyceride data, yielding
similar results.
Adipose tissue distribution
Secondary end points. Small changes in other lipid pa-
At baseline, subjects in the ATV/r arm had higher mean
rameters were noted with DRV/r therapy from baseline to
values for all CT scan parameters compared with subjects in
week 12 (Table 2). Between arms, differences in lipid changes
the DRV/r arm (Table 3). Changes in TAT, VAT, SAT, and
were seen only in TC and apoA1, with the DRV/r arm ex-
the SAT/VAT ratio from baseline to week 48 were small and
periencing greater changes in both parameters than the ATV/
comparable between the DRV/r and ATV/r arms (Table 3).
r arm. The difference in week 12 mean change between arms
Although the mean change in PAT over 48 weeks was larger
(95% CI) was 15.7 mg/dl (0.0, 31.3) for TC (20.3 mg/dl in the
in the DRV/r arm than in the ATV/r arm (Table 3), none of
DRV/r group and 4.6 mg/dl in the ATV/r group) and
the changes in adipose tissue distribution over 48 weeks was
11.4 mg/dl (0.7, 22.1) for apoA1 (10.7 mg/dl in the DRV/r
considered clinically relevant.
group and –0.7 mg/dl in the ATV/r group) (Table 2). The
actual mean TC values at week 12 were 161.5 mg/dl and
169.7 mg/dl for DRV/r and ATV/r, respectively, while the Assessment of Body Change and Distress questionnaire
actual mean apoA1 values at week 12 were 125.6 mg/dl and
Subjects’ perceptions of body changes generally improved
126.8 mg/dl for DRV/r and ATV/r, respectively. Consistent
from baseline to week 48 in both study arms. Overall, higher
with the week 12 results, no clinically meaningful difference
proportions of subjects in both arms were ‘‘satisfied’’ or ‘‘very
was seen between arms for mean changes in triglyceride
satisfied’’ with their bodies after 48 weeks compared with
levels at week 48. In contrast to the week 12 results, changes
baseline (Fig. 1A). Within each study arm, responses to the
in TC and apoA1 were similar between arms by week 48
categorical portion of the ABCD questionnaire were similar at
(Table 2). Additionally, no clinically relevant differences be-
baseline and at week 48, with few subjects switching cate-
tween arms were noted for changes in the other fasting lipid
gories. Furthermore, these categorical responses were also
parameters, including the apoB/apoA1 ratio, at week 48
similar between the DRV/r and ATV/r arms at both time
(Table 2). Changes from baseline for the lipid parameters were
points.
generally normally distributed.
In the yes/no portion of the ABCD questionnaire (Fig.
1B), more subjects in both the DRV/r and ATV/r arms
Other laboratory evaluations and biomarkers
reported increases in waist size at week 48 compared with
No clinically relevant changes were seen from baseline to baseline (DRV/r, 46.4% vs. 9.1%; ATV/r, 32.0% vs. 12.9%).
week 12, or baseline to week 48, in glucose, insulin, and in- Fewer subjects in both the DRV/r and ATV/r arms

Table 3. Change in Adipose Tissue Distribution from Baseline to Week 48

DRV/r ATV/r
n = 34 n = 31
Difference in
Change from Change from mean change between
BL BL to week 48 BL BL to week 48 arms (95% CI)

Abdomen n529 n520 n530 n524

TAT, mean (SD), cm2 258.7 (89.38) 24.2 (98.40) 285.1 (120.80) 29.4 (38.32) - 5.20 ( - 49.19, 38.79)
VAT, mean (SD), cm2 97.6 (47.10) - 0.6 (49.20) 102.5 (44.60) 3.8 (24.48) - 4.38 ( - 27.43, 18.67)
SAT, mean (SD), cm2 161.1 (74.06) 24.8 (56.50) 182.7 (108.37) 25.6 (28.20) - 0.82 ( - 27.31, 25.67)
VAT/SAT ratio, 0.50 (0.281) 0.00 (0.135) 0.53 (0.289) - 0.04 (0.104) 0.04 ( - 0.03, 0.11)
mean (SD)
Mid-thigh n529 n521 n527 n521
PAT, mean (SD), cm2 55.4 (25.74) 7.0 (17.32) 61.0 (38.06) - 1.9 (9.53) 8.88 (0.16, 17.60)

DRV/r, darunavir/low-dose ritonavir; ATV/r, atazanavir/low-dose ritonavir; CI, confidence interval; BL, baseline; TAT, total
adipose tissue; SD, standard deviation; VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue; PAT, peripheral adipose
tissue.
METABOLIC EFFECTS OF DARUNAVIR 1189

reported that they had lost fat in their buttocks or in compared with baseline. Although responses to this portion
their arms and legs or that their cheeks had sunken in at of the questionnaire were generally similar between arms,
week 48, compared with baseline. Slightly more subjects in more subjects in the DRV/r arm reported increases in waist
the DRV/r arm and slightly fewer subjects in the ATV/r and chest size at week 48 compared with subjects in the
arm reported gaining fat at the back of their neck at week 48 ATV/r arm (Fig. 1B).

FIG. 1. Results of the Assessment

of Body Change and Distress
questionnaire: (A) categorical and
(B) yes versus no. *Data were
missing from one ATV/r subject at
week 48; DRV/r, darunavir/low-
dose ritonavir; ATV/r, atazanavir/
low-dose ritonavir; BL, baseline; Wk,
week.

(Figure continued /)
1190 ABERG ET AL.

FIG. 1. Continued.

Efficacy evaluations and senescence (CD28-/CD57+) generally decreased from

baseline to week 48 (Table 4). There was also a general increase
Viral load and CD4+ cell count improved over the course of
in the percent of naive CD4+ and CD8+ T cells in both treatment
the study. Mean viral load decreased and mean CD4+ count
arms.
increased from baseline to week 48 in both the DRV/r (change
in log10 viral load, –3.3 copies/ml; change in CD4+ count,
+ 217.4 cells/mm3) and ATV/r arms (change in log10 viral Safety evaluations
load, –2.9 copies/ml; change in CD4+ count, + 205.3 cells/
Over 48 weeks, rates of AEs were generally low and com-
mm3). At week 48, 76.5% of DRV/r and 71.0% of ATV/r
parable between the DRV/r and ATV/r arms (Table 5), except
subjects (CVR) achieved virologic response.
grade 2–4 hyperbilirubinemia (considered at least possibly
related to study drug), which was observed in more ATV/r
Cellular activation and senescence
subjects than DRV/r subjects. Increased total bilirubin re-
The proportion of CD4+ and CD8+ cells displaying markers ported as a grade 2–4 laboratory abnormality was observed
of cellular activation (HLADR+/CD38+), proliferation (Ki67+), far more frequently in the ATV/r arm compared with the

Table 4. Markers of Immune Activation and Senescence over 48 Weeks

(Intent-to-Treat Population, Observed Values)

DRV/r ATV/r

Parameter, mean % (SD) BL Week 48 BL Week 48

CD41
Naive (CCR7+/CD45RA+) 47.9 (21.14) 54.7 (19.55) 38.9 (21.10) 49.1 (20.91)
Central memory (CCR7+/CD45RA-) 11.9 (8.45) 14.2 (9.37) 19.1 (14.49) 15.0 (7.22)
Terminal effectors (CCR7-/CD45RA+) 10.9 (5.50) 9.4 (3.43) 12.1 (8.78) 11.4 (7.88)
Effector memory (CCR7-/CD45RA-) 29.3 (18.85) 21.7 (16.76) 30.0 (12.03) 24.5 (15.00)
CD38+/HLADR+ 13.6 (7.10) 8.5 (4.50) 13.0 (10.82) 10.9 (5.44)
Ki67+ 1.9 (1.52) 0.9 (0.86) 1.6 (1.55) 1.0 (0.97)
CD28-/CD57+ 7.7 (12.75) 1.7 (1.50) 9.1 (14.72) 3.5 (5.57)
CD81a
Naive (CCR7+/CD45RA+) 13.6 (10.09) 25.4 (15.63) 15.4 (9.89) 26.1 (12.33)
Central memory (CCR7+/CD45RA-) 12.37 (12.43) 8.7 (8.70) 16.2 (10.32) 8.5 (7.11)
Terminal effectors (CCR7-/CD45RA+) 26.4 (14.91) 27.2 (11.07) 26.0 (16.67) 33.9 (14.19)
Effector memory (CCR7-/CD45RA-) 47.7 (14.85) 38.7 (14.42) 42.4 (14.85) 31.4 (11.40)
CD38+/HLADR+ 43.8 (17.75) 19.4 (12.49) 40.5 (17.51) 22.2 (11.05)
Ki67+ 2.8 (1.75) 1.2 (0.75) 2.3 (2.06) 1.2 (0.78)
CD28-/CD57+ 32.1 (11.81) 19.6 (12.32) 31.8 (15.02) 22.8 (15.06)

Sample size varies by time point and parameter.

a
CD8 + cell subpopulations were based on CD8 + percentages and were not directly measured.
SD, standard deviation; DRV/r, darunavir/ritonavir; ATV/r, atazanavir/ritonavir; BL, baseline; HLADR, human leukocyte antigen-DR.
METABOLIC EFFECTS OF DARUNAVIR 1191

Table 5. Safety Parameters over 48 Weeks favorable lipid changes that support the return-to-health
phenomenon observed in subjects with lower CD4+ cell
DRV/r ATV/r counts initiating ARV therapy.38 The small changes in lipids
n = 34 n = 31
seen in this trial with DRV/r are in agreement with results
Adverse events, n (%) from the ARTEMIS (AntiRetroviral Therapy with TMC114
Subjects with ‡ 1 AE 31 (91.2) 29 (93.5) ExaMined In naı̈ve Subjects) trial of 689 treatment-naive
Subjects with ‡ 1 SAE 5 (14.7) 5 (16.1) subjects, which showed that once-daily DRV/r had a more
Subjects with ‡ 1 grade 3–4 AE 3 (8.8) 13 (41.9) favorable metabolic profile compared with that of lopinavir
Subjects with ‡ 1 AE at least 15 (44.1) 22 (71.0) (LPV)/r.39
possibly related to study drug Subjects infected with HIV-1 have increased levels of hs-
Grade 2–4 AEs at least possibly related to study drug,a CRP, TNF alpha (TNF-a), IL-6, d-dimer, and other biomarkers
n (%) compared with HIV-negative subjects.2,40–42 These elevations
Hyperbilirubinemia 0 3 (9.7) in biomarkers persist even after virologic suppression, likely
Ocular icterus 0 2 (6.5) due to HIV-induced activation of inflammation and coagu-
Pollakiuria 0 2 (6.5) lation pathways.2 Levels of TNF-a, IL-6, hs-CRP, and other
Neutropenia 0 1 (3.2)
proinflammatory cytokines are associated with HIV-1 viral
Diarrhea 1 (2.9) 0
Pancreatitis 1 (2.9) 0 load and may predict disease progression, as well as correlate
Dehydration 1 (2.9) 0 with a higher risk of CVD and all-cause mortality.42–45 Our
Hypercholesterolemia 1 (2.9) 0 findings corroborate another recent analysis, which reported
Joint swelling 0 1 (3.2) that only d-dimer, and not IL-6 or hs-CRP, is reduced over the
Acute renal failure 1 (2.9) 0 short term in those initiating ARV therapy.46 Additionally, the
Dermatitis 0 1 (3.2) reductions in LPS observed here are similar in magnitude to
Leukocytoclastic vasculitis 0 1 (3.2) those seen in a previous study47; the similar reductions in
Grade 2–4 laboratory abnormalities, n (%) the DRV/r and ATV/r arms suggest that these ARV agents
Glucose 4 (11.8) 3 (9.7) do not result in differential levels of LPS, a trigger of persis-
Total cholesterol 3 (8.8) 1 (3.2) tent immune activation in ARV-treated individuals. In con-
Neutrophils 2 (5.9) 4 (12.9) trast, no decreases in hs-CRP were seen in either arm of this
Low-density lipoprotein (direct) 2 (5.9) 3 (9.7) study over 48 weeks; this observation is in agreement with the
Aspartate aminotransferase 2 (5.9) 0
ACTG (AIDS Clinical Trials Group) A5095 study, which
Total bilirubin 1 (2.9) 27 (87.1)
Creatinine 1 (2.9) 1 (3.2) demonstrated similar results in subjects receiving efavirenz-
Sodium 1 (2.9) 0 based regimens over 96 weeks.48
Triglycerides 0 1 (3.2) During HIV infection, chronic viremia induces progressive
immune dysregulation characterized broadly by a decrease
a
Adverse events reported as laboratory abnormalities are not in CD4+ cells and an increase in CD8+ cells and specifically
included. by elevated expression of CD38 and HLADR. This persis-
DRV/r, darunavir/ritonavir; ATV/r, atazanavir/ritonavir; AE,
adverse event; SAE, serious adverse event. tent immune activation yields rapid and elevated lympho-
cyte turnover and a shift to an immunosenescent phenotype
(CD28-/CD57+ T cells).49,50 As expected, based on the results
of previous studies demonstrating the mitigating effect
DRV/r arm (Table 5). No other major differences in grade 2–4
of ARV therapy on these processes,51–53 the current study
safety parameters were noted between arms. A total of 10
found reductions in the proportion of activated T cells
subjects (Table 5) experienced 16 SAEs. Three events, chronic
(CD38-/HLADR+) in both the DRV/r and ATV/r arms.
obstructive pulmonary disease with respiratory distress
Furthermore, the decline in the proportion of cells with
(DRV/r arm), diabetes mellitus (DRV/r arm), and mitral
an immunosenescent phenotype (CD28-/CD57+) illustrates
valve incompetence (ATV/r arm), were grade 4 in severity; all
the relationship between immune activation and im-
three were considered not related to study medication. Of
munosenescence and a critical immunologic benefit of ARV
the 16 SAEs, only one event, grade 3 pancreatitis in a subject in
therapy.
the DRV/r arm, was considered possibly related to study
Individuals infected with HIV-1 have a higher risk of seri-
medication. No clinically relevant changes in creatinine
ous, non-AIDS conditions than do uninfected subjects.1 Given
clearance were seen from baseline to week 48 in the DRV/r
that HIV infection is now considered a manageable chronic
[mean (SD) change: –0.00 (0.288) ml/min] or ATV/r [mean
disease, there is a growing level of attention focused on the
(SD) change: –0.03 (0.244) ml/min] arm.
need for identification of metabolically favorable ARVs.
Atazanavir has generally been considered to have the most
Discussion
favorable metabolic profile among PIs34 and was, therefore,
These results illustrate that changes in triglyceride levels chosen as a comparator drug for this study. In the study
over 12 weeks of treatment were similar in subjects treated presented here, changes in metabolic parameters and bio-
with DRV/r-based and ATV/r-based regimens. Additionally, markers from baseline with DRV/r were comparable to
small changes in lipids and insulin sensitivity, decreases in changes observed with ATV/r. These results are in agreement
biomarkers, small changes in body fat, improvements in ef- with those seen in the TMC114-C159 trial, which investigated
ficacy parameters, and a low incidence of AEs were seen over metabolic changes in healthy subjects treated with DRV/r or
48 weeks in both treatment arms. In the DRV/r arm, the ATV/r over 28 days and observed similar mean changes in
increase in apoA1, the major component of HDL, indicates lipid and glucose parameters between treatment groups.33
1192 ABERG ET AL.

The TMC114-C159 trial did report significant differences in ABCD questionnaire are in line with those from a previous
the changes between treatment arms for insulin and the TC/ study, which demonstrated improvements in ABCD scores
HDL ratio; these results, however, were not observed in over 48 weeks of DRV/r-based therapy.30 In both treatment
METABOLIK, which noted differences between arms only in arms, the question ‘‘In the past 4 weeks has your belt or waist
changes in TNF RII over 48 weeks. size increased?’’ had the greatest increase in subjects an-
The effect of some boosted PIs on insulin sensitivity re- swering ‘‘yes’’ at 48 weeks compared with baseline; this result
mains controversial. In contrast to some other PIs,8,10,11 no is consistent with the increases in VAT and SAT seen during
clinically significant changes were seen in insulin sensitivity the trial. Although changes in adipose tissue distribution over
in either the DRV/r or ATV/r arm of this study. Treatment the study period were similar between arms, more DRV/r
with indinavir and treatment with LPV/r have both been subjects reported increases in waist and chest size compared
associated with the development of insulin resistance in with ATV/r subjects. This discrepancy may indicate subtle
healthy subjects9–11; however, other studies have demon- differences in lipodystrophy and lipoatrophy between the
strated that LPV/r does not affect insulin sensitivity in heal- two treatment regimens, or may have been partially due to the
thy subjects.12,13 Atazanavir has generally had little effect on differences in racial distribution between the two study arms.
insulin sensitivity in previous trials of HIV-negative sub- Both regimens had favorable safety profiles, with low in-
jects,10,12 and switching from other PI-based therapies to cidences of AEs and laboratory abnormalities; the increased
ATV/r has been shown to improve insulin sensitivity in HIV- incidence of grade 2–4 hyperbilirubinemia or grade 2–4 in-
1-infected subjects.54 The results from this trial suggest that creased total bilirubin in the ATV/r arm was expected, as they
DRV/r, likewise, has little impact on insulin sensitivity. It are known side effects of ATV/r-based therapy.34,58 Similar to
should be noted that unlike the other studies cited here, which the results obtained here, safety results from ARTEMIS sug-
used the euglycemic, hyperinsulinemic clamp technique, this gested low rates of grade 2–4 AEs, including gastrointestinal
study used HOMA-IR as a measure of insulin resistance. and renal AEs, in subjects receiving DRV/r over 96 weeks.39
However, several studies have demonstrated that results Subjects receiving DRV/r in ARTEMIS had a significantly
obtained using HOMA-IR correlate well with results using the lower rate of diarrhea compared with LPV/r subjects (4% vs.
euglycemic, hyperinsulinemic clamp technique.36,55 11%, respectively) and, in line with results from this trial, no
Slight reductions in creatinine clearance were seen from clinically relevant changes were seen in creatinine clearance in
baseline to week 12 in both treatment arms of this study; either treatment arm.
however, these reductions were no longer apparent after 48 Interpretation of the data reported in this pilot study may
weeks of treatment. A previous study suggested an associa- be limited by the small sample size, as well as the fact that the
tion between the use of ATV, TDF, or indinavir and creatinine trial was not powered to test for statistical significance, but
clearance, as indicated by a persistent reduction in glomerular was rather intended to be an exploratory analysis. The var-
filtration rate over time.56 The biological explanations for iation seen between arms in certain baseline characteristics,
these findings are unclear, but may include glomerular dys- despite the randomized study design, is likely related to the
function, high renal excretion rates, and/or crystalluria.56 small sample size. Somewhat larger changes in TC and
Though the current study observed modest changes over 48 apoA1 with DRV/r versus ATV/r, particularly at week 12,
weeks, the ARTEMIS study did not show any changes in are likely due to lower baseline values of these parameters in
creatinine clearance over 96 weeks,39 suggesting that long- the DRV/r arm. Likewise, the larger reduction in TNF RII is
term use of DRV/r has little effect on creatinine clearance. likely due to the higher baseline value in the DRV/r arm.
The small increases in TAT and SAT seen in this study are Additionally, it should be noted that subjects with abnormal
similar to those seen in previous trials of boosted PIs and lipid or glucose levels were excluded from the trial and may
nucleoside reverse transcriptase inhibitors (NRTIs) in treat- not, therefore, be reflective of the overall HIV-1-infected
ment-naive subjects.19–22,24 Increases in abdominal fat and population. Despite these limitations, the equally favorable
waist size have been associated with increased cardiovascular metabolic profile observed with DRV/r-based therapy when
risk factors in HIV-1-infected subjects, suggesting a need to compared with ATV/r-based therapy warrants further in-
monitor even small changes in fat distribution in this popu- vestigation. It is noteworthy that the ACTG is conducting a
lation. The increase in PAT seen with DRV/r treatment in this large study of approximately 1800 treatment-naive, HIV-1-
study was in contrast to data from other studies, which have infected subjects receiving DRV/r, ATV/r, or raltegravir (all
reported decreases in PAT in HIV-1-infected individuals re- receiving fixed-dose FTC/TDF in the background regimen);
ceiving other ARVs.19,20,22 Although some studies have end points from this study include changes in metabolic
demonstrated an association between lipoatrophy and use of parameters over 144 weeks (clinicaltrials.gov identifier:
specific PIs or NRTIs,19,22 others have shown no ARV-specific NCT00811954). This study is statistically powered to defin-
effects,20 and it has been observed that HIV-1-infected sub- itively evaluate whether DRV/r-based therapy and ATV/r-
jects, in general, have lower levels of PAT compared with based therapy are characterized by similarly favorable
uninfected control subjects.57 The increase in PAT in the metabolic profiles.
DRV/r arm of this study, which had more advanced disease
at baseline compared with the ATV/r arm, can be considered
Acknowledgments
as potentially favorable; however, the small sample size
and short duration of this trial limit clinical interpretation of The authors would like to thank the subjects and their
these data. families, the study sites, and the principal investigators
In this study, despite slight increases in VAT and SAT, for their participation in the trial. The authors would like
subjects’ perceptions of their body changes generally im- to acknowledge Gilead for supplying emtricitabine, ten-
proved or remained constant over time. The results from the ofovir, and emtricitabine/tenofovir. The authors would
METABOLIC EFFECTS OF DARUNAVIR 1193

additionally like to acknowledge internal study support G.D.L.R. is an employee of Janssen Global Services and is a
staff, as well as Cali Howitt, PhD, Medicus International Johnson & Johnson stockholder.
New York, for her editorial assistance. Funding for the study
and for editorial support was provided by Janssen Ther- References
apeutics.
1. Phillips AN, Neaton J, Lundgren JD: The role of HIV in
Original Presentations of These Data serious diseases other than AIDS. AIDS 2008;22:2409–2418.
2. Neuhaus J, Jacobs DR, Baker JV, et al.: Markers of inflam-
Aberg J, et al.: METABOLIK (Metabolic Evaluation in mation, coagulation, and renal function are elevated in
Treatment-naı̈ves Assessing the impact of two BOosted pro- adults with HIV infection. J Infect Dis 2010;201:1788–1795.
tease inhibitors on LIpids and other marKers): Comparison of 3. Friis-Moller N, Reiss P, Sabin CA, et al.: Class of anti-
the Metabolic Effects of Darunavir/Ritonavir versus Ataza- retroviral drugs and the risk of myocardial infarction. N
navir/Ritonavir over 12 Weeks. XVIII International AIDS Engl J Med 2007;356:1723–1735.
Conference (IAC), Vienna, Austria, July 18–23, 2010. Poster 4. Flint OP, Noor MA, Hruz PW, et al.: The role of protease
WEPE011. inhibitors in the pathogenesis of HIV-associated lipody-
Overton T, et al.: METABOLIK: Week 48 Comparison of strophy: Cellular mechanisms and clinical implications.
Metabolic Parameters and Biomarkers in Subjects Receiving Toxicol Pathol 2009;37:65–77.
Darunavir/Ritonavir or Atazanavir/Ritonavir. Presented at 5. Haubrich RH, Riddler SA, DiRienzo AG, et al.: Metabolic
outcomes in a randomized trial of nucleoside, nonnucleoside
the 10th International Congress on Drug Therapy and HIV
and protease inhibitor-sparing regimens for initial HIV
Infection (HIV10), Glasgow, UK, November 7–11, 2010. Poster
treatment. AIDS 2009;23:1109–1118.
P74
6. Stein JH, Komarow L, Cotter BR, et al.: Lipoprotein changes
Tebas P, et al.: METABOLIK (Metabolic Evaluation in in HIV-infected antiretroviral-naive individuals after start-
Treatment-naı̈ves Assessing the impact of two BOosted pro- ing antiretroviral therapy: ACTG Study A5152s Stein: Li-
tease inhibitors on LIpids and other marKers): Week 48 poprotein Changes on Antiretroviral Therapy. J Clin Lipidol
Comparison of Body Fat Changes in ARV-Naı̈ve Subjects 2008;2:464–471.
Receiving Darunavir/Ritonavir- or Atazanavir/Ritonavir- 7. Zhou H, Jarujaron S, Gurley EC, et al.: HIV protease inhib-
Based Therapy. Presented at the 12th International Workshop itors increase TNF-alpha and IL-6 expression in macro-
on Adverse Drug Reactions and Co-morbidities in HIV phages: Involvement of the RNA-binding protein HuR.
(ADRL), London, UK, November 4–6, 2010. Poster P07. Atherosclerosis 2007;195:e134–143.
8. Carper MJ, Cade WT, Cam M, et al.: HIV-protease inhibitors
Author Disclosure Statement induce expression of suppressor of cytokine signaling-1 in
insulin-sensitive tissues and promote insulin resistance and
J.A.A. has served as a scientific advisor to Abbott Labora- type 2 diabetes mellitus. Am J Physiol Endocrinol Metab
tories, Bristol-Myers Squibb, Gilead Sciences, Inc., Glaxo- 2008;294:E558–567.
SmithKline, Merck & Co, Inc., Pfizer Inc., Theratechnologies 9. Noor MA, Lo JC, Mulligan K, et al.: Metabolic effects of
Inc., Janssen Therapeutics, and ViiV Healthcare and has re- indinavir in healthy HIV-seronegative men. AIDS 2001;15:
ceived research support from Gilead Sciences, Inc., Glaxo- F11–18.
SmithKline, Merck & Co, Inc., Pfizer Inc., Schering-Plough 10. Noor MA, Parker RA, O’Mara E, et al.: The effects of HIV
Corp, Theratechnologies Inc., Janssen Therapeutics, Virco protease inhibitors atazanavir and lopinavir/ritonavir on
Lab, Inc., and Wyeth. P.T. has received grant support from insulin sensitivity in HIV-seronegative healthy adults. AIDS
Bristol-Myers Squibb, Gilead Sciences, Inc., GlaxoSmithKline, 2004;18:2137–2144.
Inovio Pharmaceuticals, Inc., Merck & Co, Inc., Janssen 11. Noor MA, Seneviratne T, Aweeka FT, et al.: Indinavir acutely
Therapeutics, and VIRxSYS. E.T.O. has served as a consultant, inhibits insulin-stimulated glucose disposal in humans: A
on a speakers bureau or on an advisory board for Gilead randomized, placebo-controlled study. AIDS 2002;16:F1–8.
Sciences, Inc., Bristol-Myers Squibb, GlaxoSmithKline, Janssen, 12. Dube MP, Shen C, Greenwald M, and Mather KJ: No im-
Merck & Co., Inc., Monogram Sciences, and Boehringer pairment of endothelial function or insulin sensitivity with 4
Ingelheim and has received research support from Abbott weeks of the HIV protease inhibitors atazanavir or lopina-
Laboratories, Gilead Sciences, Inc., Bavarian Nordic, Glaxo- vir-ritonavir in healthy subjects without HIV infection: A
SmithKline, Boehringer Ingelheim, and Janssen. S.K.G. has placebo-controlled trial. Clin Infect Dis 2008;47:567–574.
received grant/research support from Janssen Therapeutics, 13. Lee GA, Seneviratne T, Noor MA, et al.: The metabolic effects
of lopinavir/ritonavir in HIV-negative men. AIDS 2004;18:
Merck & Co., Inc., and Gilead Sciences, Inc. and is the site
641–649.
Principal Investigator on trials sponsored by TaiMed, Pfizer
14. Hui DY: Effects of HIV protease inhibitor therapy on lipid
Inc., and GlaxoSmithKline. P.E.S. has served as a consultant
metabolism. Prog Lipid Res 2003;42:81–92.
for Abbott Laboratories, Bristol-Myers Squibb, Gilead Sci- 15. Melroe NH, Kopaczewski J, Henry K, and Huebsch J: Lipid
ences, Inc., GlaxoSmithKline, Merck & Co, Inc., and Janssen abnormalities associated with protease inhibitors. J Assoc
Therapeutics and has received grant support from Gilead Nurses AIDS Care 1999;10:22–30.
Sciences, Inc., GlaxoSmithKline, Merck & Co, Inc., and Jans- 16. Periard D, Telenti A, Sudre P, et al.: Atherogenic dyslipide-
sen Therapeutics. A.L. has served as a scientific advisor to mia in HIV-infected individuals treated with protease in-
Merck, Pfizer, and Tobira and has received research support hibitors. The Swiss HIV Cohort Study. Circulation
from Merck & Co., Inc., Abbott Laboratories, and Janssen 1999;100:700–705.
Therapeutics. R.F. is an employee of Janssen Therapeutics and 17. Purnell JQ, Zambon A, Knopp RH, et al.: Effect of ritonavir
a Johnson & Johnson stockholder. R.R. is an employee of on lipids and post-heparin lipase activities in normal sub-
Janssen R&D and is a Johnson & Johnson stockholder. jects. AIDS 2000;14:51–57.
1194 ABERG ET AL.

18. Riddler SA, Li X, Otvos J, et al.: Antiretroviral therapy is 34. Molina JM, Andrade-Villanueva J, Echevarria J, et al.: Once-
associated with an atherogenic lipoprotein phenotype daily atazanavir/ritonavir compared with twice-daily lopi-
among HIV-1-infected men in the Multicenter AIDS Cohort navir/ritonavir, each in combination with tenofovir and
Study. J Acquir Immune Defic Syndr 2008;48:281–288. emtricitabine, for management of antiretroviral-naive HIV-
19. Hammond E, McKinnon E, and Nolan D: Human immuno- 1-infected patients: 96-week efficacy and safety results of the
deficiency virus treatment-induced adipose tissue pathology CASTLE study. J Acquir Immune Defic Syndr 2008;53:323–
and lipoatrophy: Prevalence and metabolic consequences. 332.
Clin Infect Dis 2010;51:591–599. 35. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Trea-
20. Martinez E, Mocroft A, Garcia-Viejo MA, et al.: Risk of lipo- cher DF, and Turner RC: Homeostasis model assessment:
dystrophy in HIV-1-infected patients treated with protease Insulin resistance and beta-cell function from fasting plasma
inhibitors: A prospective cohort study. Lancet 2001;357: glucose and insulin concentrations in man. Diabetologia
592–598. 1985;28:412–419.
21. Tsiodras S, Mantzoros C, Hammer S, and Samore M: Effects 36. Borai A, Livingstone C, and Ferns GA: The biochemical as-
of protease inhibitors on hyperglycemia, hyperlipidemia, sessment of insulin resistance. Ann Clin Biochem 2007;
and lipodystrophy: A 5-year cohort study. Arch Intern Med 44:324–342.
2000;160:2050–2056. 37. Neidig JL, Holmes W, Reynolds NR, et al.: ACTG 5089:
22. Walmsley S, Cheung AM, Fantus G, et al.: A prospective Development of assessment of body change distress (ABCD)
study of body fat redistribution, lipid, and glucose param- questionnaire for evaluation of fat redistribution. Presented
eters in HIV-infected patients initiating combination anti- at the XIV International AIDS Conference, Barcelona, Spain,
retroviral therapy. HIV Clin Trials 2008;9:314–323. July 7–12, 2002. Abstract B10609.
23. Innes S, Levin L, and Cotton M: Lipodystrophy syndrome in 38. Riddler SA, Smit E, Cole SR, et al.: Impact of HIV infection
HIV-infected children on HAART. South Afr J HIV Med and HAART on serum lipids in men. JAMA 2003;289:2978–
2009;10:76–80. 2982.
24. Safrin S and Grunfeld C: Fat distribution and metabolic chan- 39. Mills AM, Nelson M, Jayaweera D, et al.: Once-daily dar-
ges in patients with HIV infection. AIDS 1999;13:2493–2505. unavir/ritonavir vs. lopinavir/ritonavir in treatment-naive,
25. Tsiodras S, Perelas A, Wanke C, and Mantzoros CS: The HIV-1-infected patients: 96-week analysis. AIDS 2009;23:
HIV-1/HAART associated metabolic syndrome–novel adi- 1679–1688.
pokines, molecular associations and therapeutic implica- 40. Baker J, Ayenew W, Quick H, et al.: High-density lipoprotein
tions. J Infect 2010;61:101–113. particles and markers of inflammation and thrombotic ac-
26. Cunningham WE, Crystal S, Bozzette S, and Hays RD: The tivity in patients with untreated HIV infection. J Infect Dis
association of health-related quality of life with survival 2010;201:285–292.
among persons with HIV infection in the United States. J 41. Norris PJ, Pappalardo BL, Custer B, Spotts G, Hecht FM, and
Gen Intern Med 2005;20:21–27. Busch MP: Elevations in IL-10, TNF-alpha, and IFN-gamma
27. de Boer-van der Kolk IM, Sprangers MA, Prins JM, Smit C, from the earliest point of HIV Type 1 infection. AIDS Res
de Wolf F, and Nieuwkerk PT: Health-related quality of life Hum Retroviruses 2006;22:757–762.
and survival among HIV-infected patients receiving highly 42. Stacey AR, Norris PJ, Qin L, et al.: Induction of a striking
active antiretroviral therapy: A study of patients in the AIDS systemic cytokine cascade prior to peak viremia in acute
Therapy Evaluation in the Netherlands (ATHENA) Cohort. human immunodeficiency virus type 1 infection, in contrast
Clin Infect Dis 2010;50:255–263. to more modest and delayed responses in acute hepatitis B
28. Rajagopalan R, Laitinen D, and Dietz B: Impact of lipoa- and C virus infections. J Virol 2009;83:3719–3733.
trophy on quality of life in HIV patients receiving anti-ret- 43. Roberts L, Passmore JA, Williamson C, et al.: Plasma cyto-
roviral therapy. AIDS Care 2008;20:1197–1201. kine levels during acute HIV-1 infection predict HIV disease
29. Carter VM, Hoy JF, Bailey M, Colman PG, Nyulasi I, and progression. AIDS 2010;24:819–831.
Mijch AM: The prevalence of lipodystrophy in an ambulant 44. Sachdeva RK, Wanchu A, Bagga R, Malla N, and Sharma M:
HIV-infected population: It all depends on the definition. Effect of non-nucleoside reverse transcriptase inhibitors on
HIV Med 2001;2:174–180. cytokine, chemokine, and immunoglobulin profiles in serum
30. Currier JS, Martorell C, Osiyemi O, et al.: Effects of dar- and genital secretions of HIV-infected women. J Interferon
unavir/ritonavir-based therapy on metabolic and anthro- Cytokine Res 2009;30:13–24.
pometric parameters in women and men over 48 weeks. 45. Ford ES, Greenwald JH, Richterman AG, et al.: Traditional
AIDS Patient Care STDS 2011;25:333–340. risk factors and d-dimer predict incident cardiovascular dis-
31. Kelly JS, Langdon D, and Serpell L: The phenomenology of ease events in chronic HIV infection. AIDS 2010;24:1509–1517.
body image in men living with HIV. AIDS Care 2009;21: 46. Baker JV, Neuhaus J, Duprez D, et al.: Changes in inflam-
1560–1567. matory and coagulation biomarkers: A randomized com-
32. Baraldi E, Morales-Ramı́rez J, Schneider S, et al.: Effects of parison of immediate versus deferred antiretroviral therapy
once-daily darunavir/ritonavir versus lopinavir/ritonavir in patients with HIV infection. J Acquir Immune Defic Syndr
on lipid parameters and anthropometrics in treatment-naive, 2010;56:36–43.
HIV-1-infected ARTEMIS patients at Week 96. Presented at 47. Brenchley JM, Price DA, Schacker TW, et al.: Microbial
the 5th IAS Conference on HIV Pathogenesis, Treatment and translocation is a cause of systemic immune activation in
Prevention, Cape Town, South Africa, July 19–22, 2009. chronic HIV infection. Nat Med 2006;12:1365–1371.
Poster MOPEB034. 48. Shikuma CM, Ribaudo HS, and Zheng E: 96 week effects of
33. Tomaka F, Lefebvre E, Sekar V, et al.: Effects of ritonavir- suppressive efavirenz-containing antiretroviral therapy,
boosted darunavir vs. ritonavir-boosted atazanavir on lipid abacavir, and sex on hs-CRP in ACTG A5095. Presented at
and glucose parameters in HIV-negative, healthy volunteers. the 16th Conference on Retroviruses and Opportunistic In-
HIV Med 2009;10:318–327. fections, Montréal, Canada, February 8–11, 2009. Poster 736.
METABOLIC EFFECTS OF DARUNAVIR 1195

49. Kestens L, Vanham G, Gigase P, et al.: Expression of acti- 55. Noor M, Maa J, Witek J, and Falutz J: Predictive accuracy of
vation antigens, HLA-DR and CD38, on CD8 lymphocytes surrogate indices for predicting early insulin resistance after
during HIV-1 infection. AIDS 1992;6:793–797. treatment with HIV protease inhibitors (PI). Poster pre-
50. Kestens L, Vanham G, Vereecken C, et al.: Selective increase sented at the 44th Annual Meeting of the Infectious Diseases
of activation antigens HLA-DR and CD38 on CD4 + Society of America (IDSA), Toronto, Canada, October 12–15,
CD45RO + T lymphocytes during HIV-1 infection. Clin Exp 2006. Poster 986.
Immunol 1994;95:436–441. 56. Mocroft A, Kirk O, Reiss P, et al.: Estimated glomerular fil-
51. Angel JB, Kumar A, Parato K, et al.: Improvement in cell- tration rate, chronic kidney disease and antiretroviral drug
mediated immune function during potent anti-human im- use in HIV-positive patients. AIDS 2010;24:1667–1678.
munodeficiency virus therapy with ritonavir plus saquina- 57. Grunfeld C, Saag M, Cofrancesco J Jr, et al.: Regional adipose
vir. J Infect Dis 1998;177:898–904. tissue measured by MRI over 5 years in HIV-infected and
52. Ondoa P, Koblavi-Deme S, Borget MY, Nolan ML, Nkenga- control participants indicates persistence of HIV-associated
song JN, and Kestens L: Assessment of CD8 T cell immune lipoatrophy. AIDS 2010;24:1717–1726.
activation markers to monitor response to antiretroviral 58. REYATAZ (atazanavir): Full prescribing Information. Ac-
therapy among HIV-1-infected patients in Cote d’Ivoire. Clin cessed from http://packageinserts.bms.com/pi/pi_reyataz
Exp Immunol 2005;140:138–148. .pdf on August 17, 2010.
53. Steel A, John L, Shamji MH, et al.: CD38 expression on CD8
T cells has a weak association with CD4 T-cell recovery Address correspondence to:
and is a poor marker of viral replication in HIV-1-infected Judith A. Aberg
patients on antiretroviral therapy. HIV Med 2008;9: NYU School of Medicine
118–125. AIDS Clinical Trials Unit
54. Busti AJ, Bedimo R, Margolis DM, and Hardin DS: Im- 550 First Avenue, BCD 5, Room 558
provement in insulin sensitivity and dyslipidemia in prote-
New York, New York 10016
ase inhibitor-treated adult male patients after switch to
atazanavir/ritonavir. J Investig Med 2008;56:539–544. E-mail: judith.aberg@med.nyu.edu