HIV and Inflammation: A Paradigm Shift

1
HIV and InflammationA Paradigm Shift

• Wafaa El-Sadr, MD, MPH
• Columbia University Harlem Hospital
• New York

2
Effect of Protease Inhibitor-Containing Regimens
on Mortality in Patients with lt100 CD4 cells
40
100
Deaths
80
30
60
Deaths per 100 Person-Years
Therapy with a Protease Inhibitor ( of
patient-days)
20
40
Antiretroviral Therapy
10
20
0
0
1994
1995
1996
1997
Adapted from Palella F, et al. N Engl J Med, 1998.
3
Survival from Seroconversion Compared to Pre 1996
Hazard Ratio of Death
1
0.63
0.24
0.14
0.08
0.03
Adapted from Ewings et al, 2008.
4
Change in Mortality over Time
All cause
AIDS
HAART
Mortality (per 1000 person-years)
Percent Receiving Therapy
Non-AIDS
Calendar Year
Adapted from Lau et al, JAIDS 2007
5
Causes of Death in HIV France 2005
AIDS
Cancer Hepatitis C
CVD Suicide Non-AIDS
infection Accident
Hepatitis B Liver disease OD /
drug abuse neurologic
renal pulmonary
digestive iatrogenic
metabolic psychiatric
other unknown
N 937 deaths
Adapted from Lewden et al, CROI 2007
Percent
6
(No Transcript)
7
Optimization of Use of Antiretroviral Therapy
Risks
Benefits
8
SMART Study
CD4 cell count gt350 cells/mm3 N 5,472
n 2,720
n 2,752
Drug Conservation (DC) Defer use of ART until
CD4 lt 250 episodic ART based on CD4 cell
count to increase counts to gt 350
Viral Suppression (VS) Continuous use of ART to
maintain viral load as low as possible
Primary Endpoint Opportunistic Disease or Death
9
Increased Risk Opportunistic Disease or Death
with DC versus VS Strategy
20
Logrank 31.1 p lt 0.0001
15
DC Group
Percent with Event
10
VS Group
5
0
0 4 8 12 16 20 24 28 32 36 40 44
Months from randomization
DC 2720 1170 589 322 VS
2752 1167 625 334
10
Drug Conservation (DC) Strategy Associated with
Increased Risk of Serious AIDS and Non-AIDS Events
Hazard Ratio (DC/VS) (95 CI)
No. of Patients with Events
Rate
Endpoint
DC
VS
3.6
Serious AIDS 59 1.3 0.4
1.6
Serious non-AIDS 186 3.2 2.0
1.9
Serious AIDS or 239 4.4 2.4non-AIDS
0.1
1
10
Favors VS ?
Favors DC
?

• Cardiovascular, renal, hepatic, non-AIDS
  malignancy, others
• Per 100 person-years

Adapted from Curr Opin HIV AIDS 20083112-17.
11
Unifying FrameworkHIV-Associated Immune
Activation

• HIV replication
• T cell apoptosis immunosuppression
• Coagulation cascade
• Inflammation
• Atherosclerosis - Liver disease
• Osteoporosis - Neurocognitive decline
• Renal disease

Michael Ross
Russell Ross, NEJM 1999
12
Inflammatory and Coagulation Markers in SMART

• Inflammatory
• hs C-reactive protein (hs-CRP)
• IL-6
• Serum amyloid A
• Serum amyloid P
• Coagulation
• D-dimer
• Prothrombin fragment 12 (F1.2)

13
Baseline Biomarker Levels Associated with All
Cause Mortality SMART Study
Biomarker Baseline Level DC arm OR (95 CI) P value DC arm OR (95 CI) P value VS Arm OR (95CI) P value VS Arm OR (95CI) P value
Hs CRP (Ug/ml) 2.3 (1.2-4.4) 0.01 2.7 (0.9-7.9) 0.08
IL-6 (Ug/ml) 3.8 (2.1-7.2) 0.0002 2.4 (1.1-5.2) 0.03
Amyloid A (mg/l) 1.6 (0.9-2.8) 0.11 1.5 (0.6-3.8) 0.40
Amyloid P (Ug/ml 0.8 (0.5-1.3) 0.40 0.7 (0.3-1.6) 0.46
D-dimer (ug/ml) 5.9 (1.9-18.7) 0.002 7.1 (0.8-63.2) 0.08
F1.2 (pmol/l) 0.8 (0.4-1.5) 0.47 0.7 (0.2-2.2) 0.55
Adapted from Kuller et al. Plos Medicine 2008.
14
Association of C Reactive Protein and HIV with
Myocardial Infarction
Marker CRP High vs Not High OR (95CI) P value CRP High vs Not High OR (95CI) P value HIV vs no HIV OR (95CI) P value HIV vs no HIV OR (95CI) P value
CRP 2.5 (2.4-2.8) lt0.0001
HIV 2.1 (1.3-3.1) 0.0009
CRP, HIV 2.5 (2.3-2.8) lt0.0001 1.74 (1.1-2.6) 0.01
CRP, HIV, age, sex, race, HPN, diabetes, dyslipidemia 2.1 (1.9-2.4) lt0.0001 1.9 (1.2-2.9) 0.0035
Adapted from Triant et al, J Acquir Immune Defiic
Syndr, 2009.
15
C-Reactive Protein Level is Associated with
AIDS-Free Survival
Proportion AIDS Free
Time from Baseline, years
Adapted from Lau et al, Arch Intern Med 2006.
16
C Reactive Protein Level is Associated with AIDS
Free Survival
Variable Relative Time (95 CI) P Value
CRP, mg/L lt1.2 1.3-2.3 gt2.3 1.0 0.86 (0.68-1.09) 0.63 (0.51-0.79) 0.21 lt0.001
CD4 cell count 1.12 (1.08-1.16) lt0.001
HIV RNA (log10) 0.34 (0.29-0.39) lt0.001
Hemoglobin (g/dL) 1.14 (1.06-1.23) lt0.001
Adapted from Lau et al, Arch Intern Med 2006.
17
C Reactive Protein Levels Increase over Time
prior to AIDS Diagnosis
AIDS
C reactive protein, geometric mean ug/L
Months from AIDS Diagnosis
Adapted from Lau et al, Arch Intern Med 2006.
18
The Natural History of HIV Infection Clinical
Latency?
Adapted from Pantaleo G, et al. N Engl J Med
1993.
19
Opportunistic Infections Occur at Higher CD4
Cell Count Strata
CMV / MAC / TOXO
PCP /EC
TB
Incidence per 1000 PYFU (95CI)
Latest CD4 count
N events 134 45 13 9 2
2 89 55 61 35
13 16 12 9 10 11
11 14
Adapted from Podlekareva et al. J Infect Dis
2006.
20
Non-AIDS-Related Deaths Occur at Higher CD4
Cell Counts
CASCADE
Rate per 100 person/yrs
DAD
CD4 Cell Count
Adapted from Phillips et al, AIDS 2008.
21
Deaths due to Non-AIDS Exceed AIDS Causes in
Patients enrolled with CD4 Count gt200
cell/mLPost 1999
0.8
AIDS
Non-AIDS
0.4
Cumulative mortality
Non-AIDS
Non-AIDS
AIDS
Non-AIDS
AIDS
AIDS
0
CD4lt200 CD4
201-350 CD4 351-500
CD4gt500
Adapted from Lau et al, JAIDS 2007.
22
A New Paradigm
Ongoing Morbidity from HIV
1000 800 600 400 200 0
Opportunistic Diseases
CD4 cells Count
1
2
3
4
5
6
7
8
9
10
11
12
13
14
Time in Years
Infection
23
Timing of Initiation of ART
Hazard Ratio for AIDS or Death
CD4 cell count threshold
Adapted from Sterne et al, Lancet 2009.
24
Earlier Initiation of ART and Risk of Death
Variable CD4 count 351-500 cells/ml CD4 count 351-500 cells/ml CD4 count gt500 cells/ml CD4 count gt500 cells/ml
Relative Risk P Value Relative Risk P Value
ART deferral 1.6 (1.2-2.2) 0.002 1.9 (1.2-2.9) 0.006
Female sex 1.9 (1.7-2.1) 0.04 1.4 (0.9-2.1) 0.20
Older age (10yr) 1.9 (1.7-2.1 lt0.001 1.8 (1.6-2.1) lt0.001
Baseline CD4 (100 cell increment) 0.7 (0.6-1.0) 0.06 1.0 (0.4-1.0) 0.45
Baseline HIV RNA (log 10 increment) 1.1 (1.0-1.3) 0.15 1.1 (1.0-1.3) 0.14
Adapted from Kitahata et al, New Eng J Med 2009 .
25
Effect of ART on C Reactive Protein
C Reactive Protein Level
Adapted from Henry et al, AIDS 2004.
26
Effect of ART Interruption on Biomarkers
Change from Baseline to Month 1 SMART Study
Marker DC Group DC Group DC Group VS Group VS Group VS Group
N Median M1-bl (IQR) N Median M1-bl (IQR) P-value1
IL-6 247 0.60(-0.17-1.87) 249 0.12(-0.88-0.97) lt.0001
D-dimer 248 0.05(-0.07-0.18) 248 0.00(-0.13-0.08) lt.0001
1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1 1 Wilcoxon 2-sided test comparing DC and VS from baseline to month 1
27
START Study
HIV-infected, ART-naïve
CD4 count gt 500 cells/mm3
Early ART Group Initiate ART immediately
Deferred ART Group Defer ART until CD4 count lt
350 cells/mm3 or AIDS
Primary Outcome Serious AIDS, Serious non-AIDS
Events or Death
Measurement of biomarkers
28
Effect of Rosuvastatin on CVD in General
Population with High CRP Low LDL-Jupiter Study
Cumulative Incidence
Years
Adapted from Ridker et al, N Engl J Med 2008.
29
A5275 Pilot Study of Effects of Atorvastatin
on Biomarkers in HIV
Arm A
Atorvastatin
Placebo

• HIV infected
• On boosted-PI regimen with HIV RNA lt50
  copies/ml
• LDL lt 130 mg/dl
• D-dimer gt0.34

WASHOUT
Week 0
20
48
28
Atorvastatin
Placebo
Arm B
Biomarkers of Inflammation, Coagulopathy,
Angiogenesis, and T-lymphocyte Activation
30
Mortality in HIV-infected Persons after
Seroconversion Compared to General Population
Age lt45 yrs at seroconversion
Age gt45 yrs at seroconversion
HIV Pre -1996
HIV Pre -1996
Cumulative Mortality, Proportion
Cumulative Mortality, Proportion
HIV 2004-2006
HIV 2004-2006
General 2004-2006
General 2004-2006
Time from Seroconversion, Years
Time from Seroconversion, Years
Adapted from Bhaskaran et al, Lancet 2008.
31
Dramatic Increase in Access to ARTLow Middle
Income Countries
32
Effect on HIV-related Deaths inResource-limited
Countries
PEPFAR Focus Countries (12)
Control Countries (29)
Deaths from HIV, Thousands
Deaths from HIV, Thousands
Year
Year
Adapted from Bendavid et al, Ann Int Med 2009.
33
High Mortality Pre-ART
Survival Probability
Days after Enrollment
Adapted from Lawn et al, AIDS 2005.
34
High Risk of Early Mortality after ART
InitiationResource Poor/Resource Settings
? HR unadjusted ? HR adjusted for cohort,
age, sex, baseline CD4, ART-regimen, disease
stage
Hazard Ratio (95 CI)
Months from Starting HAART
35
Summary

• Remarkable progress achieved with use of ART
• The spectrum of HIV-related complications evolved
  with a predominance of non-AIDS related events,
  particularly in patients with higher CD4 cell
  counts
• Inflammatory and coagulation markers associated
  with serious complications, AIDS and death
• A survival gap exists
• for those with HIV versus general population in
  resource-rich settings
• and an even more pronounced gap in outcomes in
  HIV infected individuals in resource rich versus
  limited settings

36
Conclusions

• A re-conceptualization of the pathogenesis of HIV
  disease is necessary-- clinical latency is a
  misperception
• Inflammation and coagulopathy are important
  causes of end-organ damage, disease progression
  and death
• Role of ART and of other interventions in
  averting and suppressing these processes and
  their consequences needs urgent definition.