Salim S Abdool Karim

1
Impact of antiretroviral therapy on HIV
prevention

• Presented at
• 3rd IAS Conference on HIV Pathogenesis
• and Treatment, July 2005

Salim S Abdool Karim Pro Vice-Chancellor
(Research), University of KwaZulu-Natal Director
CAPRISA - Centre for the AIDS Programme of
Research in SA Professor in Clinical
Epidemiology, Columbia University Adjunct
Professor in Medicine, Cornell University
2
Overview

• Introduction why prevention in
• treatment is important
• Levels of ART impact on prevention
• Biological level effects
• Behavioral level effects
• Operational health systems effects
• Simulation models
• Conclusion

3
Introduction Scale of global HIV epidemic - 2004
(conservative estimate)
Source Joint UNAIDS and WHO AIDS epidemic update
December 2004
4
Impact of HIV/AIDS on mortality in South Africa
- 2000
Source Bradshaw et al. South African Medical
Journal 2005 95 496-503
5
The evolving HIV epidemic in Africa
Morocco
Algeria
Western Sahara
Libya
Egypt
Mauritania
Mali
Niger
Senegal
Benin
Chad
Ghana
Burkina Faso
Sudan
Guinea
Guinea Bissau
Nigeria
Sierra Leone
Somalia
Ethiopia
Central Afr Rep
Liberia
Togo
Cameroon
Cote DIvoire
Congo
Uganda
Kenya
Gabon
Zaire
Tanzania
Data unavailable
Angola
0 - 0.99
Malawi
Zambia
Mozambique
1 - 4.9
Zimbabwe
Namibia
Botswana
5 - 9.9
South Africa
Swaziland
10 - 19.9
Lesotho
20
1984
2003
1994
Source adapted from Abdool Karim SS. The African
Experience. In Kenneth Mayer and
HF Pizer (eds) The AIDS Pandemic Impact on
science and society. 2005
6
Impact of ART on mortality in the USA
45 40 35 30 25 20 15 10 5 0
Unintentional injury
Cancer
Deaths per 100,000 population
Heart Disease
Suicide
HIV infection
Homicide
Stroke
87 88 89 90 91 92 93 94 95 96
97 98 99 00
Year
Trends in annual rates of death due to leading
causes of death among persons 25-44 years, USA,
1987-2000
Source Adapted from lecture series by the
University of South Carolina School of Medicine
http//pathmicro.med.sc.edu/lecture
/hiv-USfigs.htm
7
HIV Infection scenarios of the impact of ART in
South Africa
40
Scenario B
30
Scenario A
HIV prevalence ()
20
Scenario C
10
0
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008



Year
Actual HIV prevalence rates in South
African pregnant women Projection
scenarios. Scenario A No ART Scenario B ART
without improvements in prevention, and Scenario
C ART accompanied by improved prevention
Source Abdool Karim SS, Abdool Karim Q, Baxter
C. Lancet 2003 362 1499.
8
The success and sustainability of AIDS treatment
will depend on the global impact of ART, not only
on AIDS morbidity and mortality, but also on its
impact on the prevention of new HIV infections
9
Levels of ART impact on HIV prevention

• Biological level effects
• Behavioral level effects
• Operational Health System effects

10
Biological level effects of ART on HIV prevention

• ART decreases viral load in blood, semen and
  female genital tract
• ART prevents mother-to-child transmission
• Post-exposure prophylaxis
• Pre-exposure prophylaxis

11
Biological effect Viral load and HIV transmission
Source Quinn T, et al, N Eng J Med 2000
12
Biological effect ART decreases blood viral load
Source O'Brien WA. MedGenMed 1, 1999. Available
at http//www.medscape.com/ viewarticle/408226
13
Biological effect ART decreases viral load in
semen
Blood plasma and seminal plasma viral load
reductions after the initiation of antiretroviral
therapy for patients 1 (a) and 2 (b). Viral load
at each time point is given as copies/ml with a
variation of 0.5 log10.
Source Taylor S, et al. AIDS 2001 15 424-6.
14
Biological effect ART decreases viral load in
vagina
The effect of HAART on plasma and cervicovaginal
HIV-1 RNA. Baseline PVL was 4.9 log10 and CVL
HIV-1 RNA was 4.7 log10. PVL was 3.1 log10 at 2
weeks and below detectable after 30 days. CVL
HIV-1 RNA was below detectable after 11 days of
therapy. Treatment included zidovudine,
lamivudine and indinavir.
Source Cu-Uvin S, et al. AIDS 2000 14415-21.
15
Biological effect ART decreases MTCT
60
50
40
30
Rates per 100 women
20
10
no therapy
0
ZDV
gt100K
HAART
40K-100K
gt3K-40K
400-3K
Undetectable (lt400)
Maternal Plasma HIV-1 RNA (copies/mL)
Source Adapted from Cooper et al, J Acquir
Immune Defic Syndr, 2002 and
Mofenson of the National Institute of Child
Health and Human Development.
16
Biological effect Post-exposure prophylaxis
(PEP) prevents HIV infection from needlestick
exposure

• Evidence from several observational studies eg.
• Case control study of needlestick exposure
• 33 cases of HIV infection
• 665 HIV negative controls
• Analysis for protection with ZDV prophylaxis
• Odds Ratio 0.19 (CI 0.06-0.52)
• 81 of HIV infections averted

Source Cardo et al. NEJM 2005 337(21) 1485-90.
17
Biological effect Post-exposure prophylaxis
(PEP) prevents HIV from sexual exposure in humans
Evidence from observational cohort studies
18
Biological effect Post-exposure prophylaxis
(PEP) prevents HIV from sexual exposure in animal
models

Source Otten et al Journal of Virol 2000
749771-5. Tsai et al Science 1995 2701197-9 T
19
Biological effect Pre - exposure prophylaxis
(PREP) and HIV prevention

• No available clinical efficacy data
• Critically important results awaited from
• studies of once daily Tenofovir for
• HIV prevention

20
Behavioral level effects of ART on HIV prevention

• AIDS care and treatment services provide an
  opportunity to promote safer sex in PLWHA
• ART prevents HIV by increasing VCT uptake
• May increase risky behaviors
• (behavioural dis-inhibition)

21
Behavioral effect integrating HIV prevention
activities focused on behavior change into AIDS
services

• New opportunities for integrating HIV prevention
• and care into Health Care systems

Source Global HIV Prevention working group 2004
22
Behavioral effect VCT decreases HIV incidence

• Multi-centre trial (n4,293) in Kenya, Tanzania,
  Trinidad
• Randomized to receive VCT or Health Information
  (HI)
• Those assigned to VCT arm reduced unprotected
  intercourse with
• non-primary partners

0
Source The HIV-1 voluntary counseling and
testing efficacy study. Lancet 2000 356 10312
23
Behavioral effect ART increases VCT uptake

• Khayelitsha, South Africa - VCT uptake gt12 fold
  increase over 4 years
• since introduction of ART.
• Haiti women seeking testing during pregnancy
  rose sharply
• from 15 - 20 in 1995/6 to over 90 two years
  later)
• Brazil Simultaneous increase in use of ART and
  VCT 1997-2003

Source Coetzee D et al. AIDS 2004 18887-895,
Koenig S et al. AIDS 2004 18(Supp 3)S21-5.
Gates Foundation Report 2004
24
Behavioral effect ART increases safer sex VCT
uptake
Source Parker et al. Commuter AIDS Information
Project. 2002
25
Behavioral effect ART may lead to behavioral
dis-inhibition
Distribution () of new HIV infections among MSM,
by time period
Source HIV Infections among MSM in Canada.
HIV/AIDS Epi Update - May 2004
26
Operational Health systems effects of ART on HIV
prevention

• Changing perceptions and improved morale in
  health services when AIDS treatment is available
• Improving health systems
• eg. improved STD treatment

27

• Operational health systems effect
• Pre-ART attitudes in health services

Selected studies reporting doctors unwillingness
to treat HIV/AIDS patients

• Physicians have right to refuse to care for
  AIDS patients
• - Raymond et al, 1991 (63, n480) Dubois
  et al, 1994 - (44, n70)
• Duyan et al,2001 - (42, n54)
• Refuse / hesitate to treat
• - Roumeliotou et al, 1992 - (16, n20)
  Adelekan et al, 1995 (43, n46)
• Never accept new patients who are HIV positive
  
• - Samuels et al,1995 - (14, n46)
• Not perform surgery despite adequate
  precautions
• - Adelekan et al,1995 - (33, n35).

28

• Operational health systems effect
• Post-ART attitude changes in health services

• Post-ART health service provider counselling
• reduced unprotected sex by 38 (plt0.001)
• (Richarson et al 2004)
• ART improves the morale of health care workers
• In North-West Province in South Africa ART has
  improved
• the morale of health care workers and patients
• (AfricaFocus Bulletin Jun 11, 2005)
• In Haiti, improving AIDS prevention and care led
  to
• dramatic improvement in general primary health
  care
• (Walton et al 2004)

29
Johnson/Dorrington simulation model Impact of
ART on prevention and mortality
Projected HIV infections and AIDS deaths averted,
2002 to 2015
Source Johnson and Dorrington, 2002
30
Solomon simulation model Impact of ART on new
HIV infections
HIV incidence in sub-Saharan Africa, 2003-2020
various intervention scenarios
Source Solomon et al. PLOS Medicine 2005 2(1)
50-6.
31
Blower simulation model Impact of ART on HIV
infection prevention
30 25 20 15 10 5 0 -5
of HIV infections prevented
10 20 30
40 50
of cases treated
Source Blower et al . CID Inf Diseases 2003
3335-53.
32
Conclusion

• Substantial evidence that ART is likely to
  prevent
• sexual transmission of HIV through
• reduced viral load in blood, semen and vagina
• increased VCT uptake and
• strengthening health care systems
• Need randomized control trials of ART impact on
• prevention
• Even in the absence of trial data, the evidence
  for
• integrated prevention and treatment is
  compelling
• For global control of the HIV epidemic, AIDS
  treatment
• must be accompanied by successful prevention

33
Useful Bibliography