Transcript Powerpoint - International AIDS Society

Initiating HAART: steering
a course
for long-term
treatment success
Sharon Walmsley MD
Professor of Medicine
University of Toronto
Senior Scientist
Toronto General Hospital
Research Institute
Considerations for initiating HAART
Suitability
for use in
special patient
groups
Efficacy/
durability
Metabolic profile/
CV risk
Genetic barrier
to resistance
Adherence/
pill burden
Short and long-term
tolerability
Initial HAART for naïve patients:
NNRTI-based vs PI-based regimens
• ACTG51421
• Efavirenz (EFV) + 2 NRTIs vs lopinavir-ritonavir (LPV/r) + 2
NRTIs vs LPV/r + EFV
• n=753
• Primary endpoints:
• Time to virological failure
• Time to regimen failure
• FIRST2
• NNRTI + 2 NRTI vs PI + 2 NRTI vs PI + NNRTI + 1 NRTI
• n=1397
• Primary endpoint:
• Composite of an AIDS-defining event, death or CD4 cell count
decline to <200 cells/mm3
1. Riddler SA et al. NEJM 2008;358:2095–2106; 2. MacArthur RD et al. Lancet 2006;368:2125–2135
NNRTI vs PI-based strategies:
virological response
Patients with HIV-RNA <50 copies/mL (%)
ACTG51421
FIRST2
100
100
90
90
80
70
70
60
50
EFV vs LPV/r at week 96:
p=0.003
40
30
EFV
LPV/r
20
% patients
% patients
80
60
50
Time to reduction varied
significantly among treatment
strategies: p<0.0001
40
30
NNRTI
PI
20
10
10
0
0
0 4 8 16 24 32 40 48 56 64 72 80 88 96
0
Weeks since randomisation
EFV 250 236
LPV/r 253 235
NRTI- 250 242
sparing
224
226
228
212
217
217
201
201
206
1
2
3
4
5
6
30
17
17
5
2
2
Years from randomisation
178
177
180
PI: 466
NNRTI: 458
PI+NNRTI: 459
179
113
120
118
74
77
89
60
47
56
42
31
Modified from: 1. Riddler SA et al. NEJM 2008;358:2095–2106 and 2. MacArthur RD et al. Lancet 2006;368:2125–2135
Initial HAART for naïve patients: NNRTI-based
vs PI-based regimens – main findings
• PI regimens inferior to NNRTI regimens in
terms of achieving and maintaining virological
suppression
• Rate of AIDS-defining events, death or severe
immunologic decline comparable
MacArthur RD et al. Lancet 2006;368:2125–2135; Riddler SA et al. NEJM 2008;358:2095–2106
ARTEN: NVP vs atazanavir/r in combination
with emtricitabine (FTC)/tenofovir (TDF)
Genotype report
Day -28
Week 2:
up-titration to 400 mg NVP/day
Screening
(n=187)
Nevirapine BID + FTC/TDF
Cont
or new
Screening
(n=187)
Nevirapine QD + FTC/TDF
Cont
or new
Screening
(n=187)
Atazanavir/r + FTC/TDF
Cont
or new
0
12
DSMB
200 pat
48
Primary
endpoint
Fully recruited; ongoing
144
or EOT
Post-trial
Factors to consider when choosing the NNRTI
Efficacy/
durability
Long-term safety/
tolerability
Metabolic profile/
CV risk
Suitability for use in
special patient groups
Factors to consider when choosing the NNRTI
Efficacy/
durability
Long-term safety/
tolerability
Metabolic profile/
CV risk
Suitability for use in
special patient groups
2NN: randomized controlled trial comparing
virological efficacy of nevirapine and efavirenz
Percentage of patients with viral load (VL)
<50 copies/mL through 48 weeks (ITT)
100
p=ns for all pairwise
comparisons
90
% patients
80
70
65.4
70
70
NVP QD (n=220)
EFV QD (n=400)
60
50
40
30
20
10
0
NVP BID (n=387)
van Leth F et al. Lancet 2004;636:1250–1253
2NN: virological efficacy of nevirapine and
efavirenz: long-term follow up
Percentage of patients with VL <400 copies/mL
through 144 weeks
100
90
87
91
87
% patients
80
70
60
50
40
30
20
10
0
NVP BID (n=224)
NVP QD (n=120)
Wit F et al. 4th IAS 2007, Sydney, Australia. Abstract WEPEB032
EFV QD (n=223)
Durability of NVP-based regimens in switch
patients: 3-year follow-up of NEFA
Proportion of non-failing patients*
Kaplan-Meier survival estimates
1.00
NVP
EFV
Abacavir
0.75
0.50
0.25
0.00
0
1
*Failure = death, progression to AIDS or virological failure
Martinez E et al. AIDS 2007;21(3):367–369
Years
2
3
Subanalysis of BICOMBO study:
NVP vs EFV in combination with once-daily
Kivexa® or Truvada®
Study population (n=335)
• Stable lamivudine-based regimen for ≥6 months (VL <200)
NRTI switch to Kivexa®
NRTI switch to Truvada®
Primary analysis, week 48
• Randomization stratified according to the third drug, either a PI (n=33) or a NNRTI (n=300)
• Patients treated with NVP (n=118) compared versus patients treated with EFV (n=182)
Martinez E et al. 17th IAC 2008. Abstract TUPE0110
Treatment and virological failure: subanalysis
of BICOMBO study
Primary endpoint: the proportion of patients with treatment failure
for any reason through week 48
100
90
Treatment failure
Virological failure
80
% patients
70
60
50
40
30
n=24
n=27
20
10
n=2
n=2
0
Nevirapine
p=0.271 for Therapeutic Failure (Fisher’s exact test)
p=0.647 for Virological Failure (Fisher’s exact test)
Martinez et al. 17th IAC 2008. Abstract TUPE0110
Efavirenz
Factors to consider when choosing the NNRTI
Efficacy/
durability
Long-term safety/
tolerability
Metabolic profile/
CV risk
Suitability for use in
special patient groups
Nevirapine: hepatic events are
related to gender and CD4 count in
treatment-naïve individuals
% patients with symptomatic
hepatic events
12
<150 cells/µL
150–250 cells/µL
250–400 cells/µL
8
4
0
Females
Males
Stern JO et al. XIVth International AIDS Conference 2002, Barcelona, Spain. Abstract LBOr15
Reanalysis of the 2NN study: hepatic events
in those meeting CD4+ criteria*
Asymptomatic hepatic event after NVP
Symptomatic hepatic event after NVP
Asymptomatic event after EFV
Symptomatic event after EFV
20
16
% hepatic events
p=ns
12
p=ns
8
4.2
4
2.8
3.9
3.2
3.9
3.6
0
NVP QD
119
n=
NVP BID
255
Storfer S et al. EACS 2005, Dublin, Ireland. Abstract PE9.6/2
EFV
225
*Excluding Thai centre
What about virologically suppressed patients
who switch to nevirapine?
• Meta-analysis of randomised controlled trials including:
• Virologically suppressed patients who switched to NVPcontaining HAART with 3 month follow up
• CD4 counts classified as high (HCD4; 400 cells/µL for males
and 250 cells/µL for females) or low (LCD4)
• Primary endpoint: hepatotoxicity within 3 months of switch
• Randomised control trials included:
•
•
•
•
NEFA (n=155, minimal follow-up 36 months)
GESIDA (n=101, minimal follow-up 3 months)
QDLluita (n=84, minimal follow-up 12 months)
Boehringer Ingelheim study 1100.138 (n=70, minimal
follow-up 24 months)
De Lazzari E et al. HIV Med 2008;9:221–226
What about virologically suppressed patients
who switch to nevirapine?
Development of hepatotoxicity or rash within
3 months of switch
% patients
40
35
Low CD4+ count (n=133)
30
High CD4+ count (n=277)
25
20
13
15
10
10
5
2
4
0
Hepatotoxicity
De Lazzari E et al. HIV Med 2008;9:221–226
Rash
What about virologically suppressed patients
who switch to nevirapine? Data from ATHENA
• 3752 patients classified in the following groups:
Naïve patients
Pretreated patients
Low CD4+ count
pre ART (PL)
Low CD4+
count at
start NVP
(NL; 77%)
High CD4+
count at
start NVP
(NH; 23%)
Low CD4+
count at
start NVP
(PLL; 49%)
High CD4+
count at
start NVP
(PLH; 30%)
High CD4+ count
pre ART (PH)
Low CD4+
count at
start NVP
(PHL; 4%)
• Rates of discontinuation due to rash and hepatotoxicity
calculated and compared between groups
Wit F et al. Clin Infect Dis 2008;46:933–940
High CD4+
count at
start NVP
(PHH; 17%)
ATHENA: risk of rash and hepatotoxicity in
patients with undetectable HIV RNA who
switch to nevirapine
Odds ratio (adjusted)
5
4
3
2
1
0
NL
NH
Naïve patients
PLL
PLH
Experienced patients
Wit F et al. 4th IAS 2007, Sydney, Australia. Abstract MOPEB008
PHH
Conclusions: impact of CD4+ count on rash
and hepatotoxicity in switch patients
• In meta-analyses, baseline CD4 count/gender
is not associated with increased risk of
rash/hepatotoxicity on switching to NVP in
virologically suppressed patients1
• Supported by data from ATHENA2 and
EUROSIDA3 suggesting risk similar in naïve
and treatment-experienced patients with
undetectable viral load
1. De Lazzari E et al. HIV Med 2008;9:221–226; 2. Wit F et al. Clin Infect Dis 2008;46:933–939;
3. Mocroft A et al. Antivir Ther 2007;12:325–333
Toxicities as a reason for treatment changes
• Reasons for treatment change assessed in
virologically-suppressed patients (n=508; 912
person-years follow-up)
• Reason for change available in 279 changes
• Most common reasons:
• Toxicity: 50.2%
• Patient choice: 17.6%
• Poor adherence: 3.6%
Lodwick R et al. AIDS 2008;22:1039–1046, Royal Free London, 2000-2005
Toxicities as a reason for treatment changes
Specific toxicities associated with
treatment changes
Lactic acidosis
Raised amylases
Myositis
Headache
Lipid abnormality
Diabetes or raised glucose
Diarrhoea
Allergic reaction
Rash
Intolerance
Abnormal LFTs
CNS effects or insomnia*
Peripheral neuropathy
Malaise or fatigue
Abdominal pain
Lipodystrophy
Anaemia
Nausea or vomiting
Renal problem
0
5
10
15
20
25
30
35
Frequency
*CNS effects comprised 23% of the total toxicities and all reports were in patients taking EFV
Lodwick R et al. AIDS 2008;22:1039–1046
Long-term CNS toxicity with efavirenz use
Cross-sectional study from one university hospital outpatient clinic – patients
receiving stable EFV-based regimen for at least 1 year
Neuropsychiatric adverse events with
efavirenz treatment (mean treatment
duration: 91.1 ± 39.5 weeks, n=60)
50
% of patients
40
30
20
10
0
Dizzy
Sad
Mood
changes
Irritable Impaired Abnormal
concentration dreams
Modified after Fumaz C et al. J Acquir Immune Defic Syndr 2005;38(5):560–565
Switching from efavirenz to nevirapine
leads to resolution/improvement of
neuropsychiatric side effects
18 patients with
sleep disturbances
20 patients with
neuropsychiatric
side effects
Switch from efavirenz
to nevirapine
14 pts
symptoms
resolved
completely
6 pts
reported
significant
improvements
Switch from efavirenz to nevirapine
14 pts
symptoms
resolved
completely
Ward D & Curtin J. AIDS Patient Care STDS 2006;20(8):542–548
3 pts
reported
improvement
1 pt
had no
change
Factors to consider when choosing the NNRTI
Efficacy/
durability
Long-term safety/
tolerability
Metabolic profile/
CV risk
Suitability for use in
special patient groups
D:A:D study: adjusted relative rates of MI
according to cumulative exposure to PIs
and NNRTIs
Protease inhibitors
Non-nucleotide reverse-transcriptase inhibitors
Adjusted relative rate
8.0
4.0
2.0
1.0
0.5
0
<1
1–2
2–3
3–4
Exposure (year)
Modified from Friis-Møller N et al. N Engl J Med 2007;356:1723–1735
4–5
5–6
>6
Interruption of ART and risk of cardiovascular
disease (CVD): findings from SMART
ARV therapy
(on at baseline)
HR for CVD events
(DC/VS*)
95% CI (p-value)
Any ARV
1.57
1.00–2.46 (0.05)
NNRTI based
1.37
0.85–2.21 (0.20)
NVP based
9.29
1.19–72.6 (0.05)
*DC, drug conservation (STI), VS, virological suppression (no STI)
Phillips A et al for the SMART Study Group. 14th CROI 2007, Los Angeles, USA. Abstract 41
Metabolic abnormalities and risk of MI
Unadjusted model
Multivariable Poisson model adjusted for age, sex, BMI, HIV risk, cohort,
calendar year, race, family hx CVD, smoking, previous CVD event, TC, HDL,
HTN, DM
Total cholesterol (per mmol/L incr)
RR: 1.31* (1.25–1.37)
Total cholesterol (per mmol/L incr)
RR: 1.26* (1.19–1.35)
Triglycerides (per 2-fold incr)
RR: 1.58* (1.43–1.75)
Hypertension
RR: 1.30† (0.99–1.72)
HDL (per mmol/L incr)
RR: 0.72‡ (0.52–0.99)
0.1
0.5
1
Relative rate of MI (95% CI)
*p<0.001; † p=0.06; ‡ p=0.05
Friis-Møller N et al. N Engl J Med 2007;356:1723–1735
5
10
NNRTI comparison of lipid changes: data
from the 2NN study
HDL-cholesterol
TC:HDL-C ratio
NVP
EFV
5.0
TC:HDL-C
HDL-C mmol/L
1.5
1.0
0.5
4.5
4.0
3.5
3.0
0 2 4 8 12
24
36
48
Weeks after start allocated treatment
Van Leth F et al. PLoS Med 2004;1:e73
0 2 4 8 12
24
36
48
Weeks after start allocated treatment
Switching from PI to nevirapine has a
favourable effect on lipid profile: LIPNEFA
study (24 months)
Percentage change (%)
(Median, 24 months from baseline)
Total
cholesterol
LDL
HDL
Triglycerides
cholesterol cholesterol
40
34
30
18
20
10
0
-10
5
-3
-7
-11
-20
-30
-40
2
0
NVP (n=29)
EFV (n=32)
ABC (n=29)
Fisac C et al. AIDS 2005;19:917–925
-8
-12
-14
-28
Switching from efavirenz or PI to nevirapine
has a beneficial effect on lipids
Median change in plasma lipid levels
from baseline to 36 months (mg/dL)
Prospective 3-year observational cohort study
Patients switched from EFV- (n=44), or PI-containing regimens (n=105)
20
10
0
-10
-20
-30
-40
-50
LDL-C
Goldbach R et al. 17th IAC 2008. Abstract TUPE0071
HDL-C
TG
Nevirapine intensive lipid evaluation
(NILE) study
• Nevirapine exerts favourable changes on the plasma
lipid profile, especially on HDL cholesterol
• The NILE study looked at the mechanisms of how
nevirapine influences lipids
• Patients with HIV-RNA ≤50 copies/mL while on
AZT/3TC/ABC for ≥6 months and no prior NNRTI
exposure (n=13) had NVP added
Sankatsing R et al. 4th IAS 2001, Sydney, Australia. Poster WEPEB032
NILE study: mechanism of lipid changes
with nevirapine
HDL
*p<0.05 for
% change from
week 0 to 24
*
*
Large HDL
particles
*
*
*p<0.05 for
% change from
week 0 to 6
*
*
Apo A-1
Cholesteryl ester
transfer protein
activity
-10
*p<0.05 for
% change from
week 6 to 24
*
0
10
Week 6
Week 24
*
20
30
% change from baseline
Sankatsing R et al. 4th IAS 2001, Sydney, Australia. Poster WEPEB032
40
50
Metabolic impact of ARV drugs/drug classes
Metabolic impact of drug classes
Less
Metabolic impact of drugs
Less
More
More
NNRTI
NRTI
PI
NVP
3TC/FTC
ABC
TDF
fAPV
EFV
ZDV
ATV/r
SQV/r
ddl
LPV/r
fAPV/r
DRV/r
d4T
IDV/r
TPV/r
RTV (full dose)
Lundgren J et al. HIV Med 2008;9:72–81. EACS guidelines
Association between HAART and
fibrinogen levels
• Fibrinogen is an inflammatory factor associated with
atherosclerosis
• Framingham study: each SD increase in fibrinogen level
(56 mg/dl) is associated with a 20% independent increment of
risk for cardiovascular disease
• Cross-sectional analysis of 1131 HIV-infected
participants and 281 controls with
fibrinogen measurements
• Median fibrinogen levels higher in HIV-infected individuals
compared with controls
• Associations between current use of each drug and
drug class and fibrinogen levels assessed by
multivariable linear regression
Madden E et al. AIDS 2008;22:707–715. CARDIA: substudy of Fat Redistribution And Metabolic change in HIV
infection (FRAM)
Fibrinogen levels with different antiretroviral agents
Median fibrinogen
(mg/dL) ±95% CI
450
400
p<0.0001
p<0.0001
† p<0.0001 vs no NNRTI
* p=0.0002 vs no NNRTI
*
†
350
300
0
No PI
All PI
(n=512) (n=616)
No
All
NNRTI NNRTI
(n=695) (n=433)
Modified from Madden E et al. AIDS 2008;22:707–715
NVP
(n=148)
EFV
(n=273)
Factors to consider when choosing the NNRTI
Efficacy/
durability
Long-term safety/
tolerability
Metabolic profile/
CV risk
Suitability for use in
special patient groups
Stage of liver fibrosis and levels
of antiretrovirals
Plasma levels in 279 HIV/HCV + pt on HAART
Liver fibrosis (Fibroscan); 37% F0–F1, 15% F2, 11% F3, 37% F4
Drug plasma levels (ug/ml)
•
•
12
10
8
p<0.01
6
4
2
0
n=
•
Cirrhotic patients
No
Yes
EFV
NVP
LPV
30 16
31 15
33 23
ATV 400 ATV 300
36 22
48 25
In compensated cirrhotics plasma levels of NNRTI, mainly EFV, may be increased,
plasma levels of PI remain similar to non-cirrhotics
Barreiro P et al. J Infect Dis 2007;195:973–979
Association between NNRTI and PI exposure
and liver fibrosis progression
• Objective: to analyse the effect of exposure to
NNRTI and PI on liver fibrosis progression in
HIV/HCV coinfected patients (n=201)
• Fibrosis stage score (F) was recorded from 0 (no
fibrosis) – 4 (definite cirrhosis)
• Fibrosis progression rate (FPR) was also calculated
by dividing F by the estimated duration of HCV
infection in years.
• Exposure to HAART reduced FPR and
development of bridging fibrosis and cirrhosis
Berenguer J et al. Clin Infect Dis 2008;46:137–143
Association between NNRTI and PI exposure
and liver fibrosis progression
AOR; P value
1.32; 0.020
1.24; 0.057
1.64; 0.003
1.32; 0.033
1.72; 0.008
F1 vs F3
HAART
PI
NNRTI
Efavirenz
Nevirapine
0.7 0.8 0.9 1
2
3
Harmful Beneficial
F2 vs F3
AOR; P value
1.20; 0.131
1.00; 0.990
1.51; 0.016
1.42; 0.091
1.58; 0.025
HAART
PI
NNRTI
Efavirenz
Nevirapine
0.7 0.8 0.9 1
Harmful Beneficial
Berenguer J et al. Clin Infect Dis 2008;46:137–143
2
3
AOR = adjusted odds ratio
Women of childbearing age (WCBA)
• The number of HIV positive women
is increasing
• With increased life expectancy, more
positive women are having children
• Family planning and open physician – patient
dialogue are necessities
• Treatment must be safe for both planned
and unplanned pregnancies
Nearly half of pregnancies are unintended
Finer L & Henshaw S. Perspect Sex Reprod Health 2006;38:90–96
FDA pregnancy categories for NNRTIs
 Upgrade of nevirapine to pregnancy category B
Category B Animal reproduction studies have failed to
demonstrate a risk to the fetus and there are
no adequate and well controlled studies in
pregnant women
Nevirapine
Category D There is positive evidence of human fetal risk
Efavirenz
based on adverse reaction data from
investigational or marketing experience or
studies in humans, but potential benefits may
warrant use of the drug in pregnant women
despite potential risks
Nevirapine AUC (µg*hr/mL)
Pregnancy does not alter NVP PK
160
140
120
100
80
60
40
20
0
2nd trimester
3rd trimester
Postpartum
Mean NVP AUC (n=26 patients)
– Antepartum: 56±13 µg*h/mL
– Postpartum: 61±15 µg*h/mL
Capparelli E et al. HIV Med 2008;9:214–220
What is the effect of pregnancy on the
development of liver enzyme elevations/rash?
• Objective:
• To assess predictors of developing liver enzyme elevations
and rash in pregnant and non-pregnant women initiating
HAART (total cohort n=612; total pregnancies n=146)
• Included 79 pregnant women and 61 non-pregnant women
on NVP
• Patients on NVP versus non-NVP regimens compared
• Analyses stratified by baseline and nadir CD4 count
and pregnancy status
• Conclusions
• Pregnancy was not an independent risk factor for developing
liver enzyme elevations or rash in patients receiving NVP
Aaron et al. 15th CROI 2008, Boston, USA. Poster 983
Factors to consider when choosing the NNRTI
Efficacy/
durability
Proven
Long-term safety/
tolerability
Metabolic profile/
CV risk
Favourable
Favourable
lipid profile
Suitability for use in
special patient groups
NNRTI of choice
in WCBA