Impact of Contraception on HIV Acquisition, Disease Progression, and Viral Shedding Elizabeth Stringer MD, MSc Center for Infectious Disease Research in Zambia, Lusaka, Zambia University of Alabama at Birmingham
Impact of Contraception on HIV Acquisition, Disease Progression, and Viral
Shedding
Elizabeth Stringer MD, MSc
Center for Infectious Disease Research in Zambia, Lusaka, Zambia
University of Alabama at Birmingham
The face of the HIV epidemic
Maternal Mortality sub-Saharan Africa
Shah et al Reproductive Health Matters 2007
Menstrual cycle
Contraceptives in common use worldwide
DMPA (Depo Provera) Depot MedroxyProgesterone Acetate
Injectable
Every 3 months
OCPs (Oral Contraceptive Pills)
Combined low dose estrogen and progesterone
IUD (Intrauterine Contraceptive Device)
Copper T
Progesterone releasing
Does hormonal contraception affect:
–HIV disease acquisition?
–HIV disease progression?
– HIV shedding?
Does hormonal contraception affect:
–HIV disease acquisition?
–HIV disease progression?
– HIV shedding?
HC-HIV Study• Prospective study • 6,109 HIV-negative women • Thailand, Uganda, Zimbabwe• Women were on method at entry• 3 monthly follow-up for 15-24 months• Outcome: HIV acquisition • 213 seroconversions (2.8 / 100 woman
years)
Morrison et al AIDS 2007
HIV acquisition
HC-HIV Study: Results Crude HR
(95% CI) pAdjusted HR
(95% CI)* p
No HC 1.0 1.0
OCP 1.0 (0.7-1.4) 0.9 1.0 (0.7-1.4) 0.9
DMPA 1.2 (0.9-1.7) 0.2 0.9 (0.6-1.8) 0.2
* Adjusting for site, living with partner, age, participant behavioral risk, primary partner risk, coital frequency, and condom use.
Morrison et al AIDS 2007
HIV acquisition
Mombasa Sex Worker Cohort
• Open cohort study
• HIV-1 negative sex workers
• Outcome: incident HIV
• Monthly follow-up
• 1272 women(> 4,700 woman-yrs follow-up)
• 248 seroconversions (8.5 per 100 woman-years)
Lavreys et al AIDS 2004
HIV acquisition
Predictors of HIV acquisition in Mombasa Cohort (n=1272)
Method AHR 95% CI pNone 1.0DMPA 1.8 1.4 - 2.4 <0.001OCPs 1.5 1.0 - 2.1 0.04Norplant 1.6 0.5 - 5.7 0.5IUD 1.1 0.4 - 3.0 0.9
Lavreys et al AIDS 2004
HIV acquisition
DMPA and HIV Acquisition
**About two-thirds DMPA, one-quarter Net-EN users
*About two-thirds Net-EN, one-third DMPA users
RR (95% CI)
0.1 1.0 10
Bulterys 1994
Ungchusak 1996
Kapiga 1998
Kilmarx 1998
Kiddugavu 2003
Lavreys 2004
Kleinschmidt 2005**
Myer 2006*
HIV acquisition
Slide Courtesy of C.Morrison
OCPs and HIV Acquisition
RR (95% CI)
0.1 1.0 10
Plummer 1991
Laga 1993
Saracco 1993
Plourde 1994
De Vincenzi 1994
Ungchusak 1996
Sinei 1996
Kilmarx 1998
Kapiga 1998
Kiddugavu 2003
Lavreys 2004
Morrison 2007
HIV acquisition
Slide Courtesy of C.Morrison
Does hormonal contraception affect:
–HIV disease acquisition?
–HIV disease progression?
– HIV shedding?
Mombasa Cohort and Disease Progression
• 248 seroconversions• 161 women had accurate date of seroconversion
– 91 / 161 HIV seroconversions occurred within one year of a clinic visit when stored samples were HIV-1 RNA and Ab negative
– 70 / 161 had HIV-1 RNA detected at visit immediately prior to seroconversion
Baeten et al AIDS 2005
Disease Progression
Viral load set point
• Viral load set point estimated to be established 16 weeks post acquisition
• Median set point: 4.46 log10 copies/ml (95% CI 4.32-4.60)
• DMPA users had higher set point (+0.33 log 10 copies/ml, p=0.03)
Disease Progression
Baeten et al JAIDS 2005
Multiple Viral Variants• 89/156 (57%) women had multiple viral
variants– More common in women with hormonal
contraceptive use (OR 2.7; p=0.003)
• Women with multiple viral genotypes– Had higher viral loads 4-24 months after
infection (4.84 vs. 4.64 log 10 copies/ml; p=0.04)
– Had lower CD4 counts (median 416 vs 617 cells/uL, p=0.01) and a faster decline
Disease Progression
Baeten et al JAIDS 2005
Viral diversity and disease progression
Sagar et al J Virol 2003
Disease Progression
Zambia IUD Study
• 599 recently-postpartum HIV+ women• Randomized trial
– Copper IUD– User-choice hormonal contraception (oral
contraceptive pills* or DMPA)– Between June 2002 and Oct 2003
• Outcomes: pregnancy, safety, and switching• Primary outcome: pregnancy
Stringer et al Am J Obstet Gynecol 2007
* “Mini pill” during early BF; combined preparation thereafter
Disease Progression
Enrollment and follow-up
• Follow-up every 6 months• CD4+ every 12 mo, but once ART
available, every 6 mo• Pregnancy testing and pelvic exam at
each visit• PID diagnosed by Hagar’s Criteria• All women followed to pregnancy event,
withdrawal, or until last enrolled participant had completed 24 months
Disease Progression
Hormonal (n=303)
IUD (n=296)
Age, mean (sd*) 25.9 (4.9) 26.3 (4.9)
Married, # (%) 272 (90) 264 (89)
Any prior contraceptive use, n (%) 186 (61) 181 (61)
Prior oral contraceptive use, n (%) 142 (47) 144 (49)
Prior injectable use, n (%) 47 (16) 49 (17)
Prior IUD use, n (%) 6 (2.0) 1 (0.3)
Enrollment CD4+ count, mean (sd) 493 (251) 514 (266)
BMI at enrollment, mean, kg/m2 (sd) 24.1 (4.0) 24.2 (3.9)
Baseline characteristics of cohort
Disease Progression
Hormonal IUD
Outcome #
Rate / 100 pt-yrs
# Rate / 100 pt-yrs HR
Pregnancy 27 4.62 14 2.17 2.2 (1.2, 4.2)
Discontinuation 38 6.81 146 28.6 0.23 (0.16, 0.33)
PID 0 0.00 1 0.16 ∞ (0.23, ∞)*
Analysis by Study Arm
*Fisher’s exact test
Disease Progression
Time to death by randomization arm Disease Progression
Time to CD4 count falling to below 200
cells/uL by randomization arm
Disease Progression
Time to either CD4 cell count falling to below
200 cells/uL or death by randomization arm
Disease Progression
Time-varying analysis Outcome
Type of Contraceptive method Initial Method Analysis* Time-varying analysis**
Death IUD
OCPs
DMPA
1.0
1.1 (0.4-3.0)
1.4 (0.6-3.0)
1.0
1.2 (0.4-3.6)
1.8 (0.8-4.1)
CD4 count dropping <200 cells/mm
IUD
OCPs
DMPA
1.0
1.3 (0.7-2.2)
1.4 (0.9-2.3)
1.0
1.2 (0.7-2.0)
1.31 (0.8-2.1)
CD4 count dropping below 200 cells/m3, or death
IUD
OCPs
DMPA
1.0
1.2 (0.7 - 2.0)
1.5 (1.0 - 2.2)
1.0
1.5(0.9 - 2.5)
1.7(1.2 - 2.6)
** extended Cox model treating contraception as a time-varying covariate.
Disease Progression
* Cox model with exposure based at start of f/u
Both models adjusted for initial CD4 count
Stringer unpublished data
Other studies HIV disease progression and HC
Study name Country Type Result
Cejtin et al, AIDS 2003
US Cross-sectional
analysis & longitudinal
No diff in change of VL or CD4
Richardson et al, AIDS 2007
Kenya Prospective cohort
No diff in VL or CD4 over timeShort follow-upSmall numbers
Allen S et al,J Women’s Health 2007
Rwanda Prospective cohort study
DMPA/OCPs protective on mortality
No CD4 counts or VL
Disease Progression
Does hormonal contraception affect:
–HIV disease acquisition?
–HIV disease progression?
– HIV shedding?
Contraception and HIV infectiousness
• Genital HIV-1 shedding used as marker for infectiousness(Pedraza, JAIDS 1999; Baeten, Curr HIV Res 2003)
• No association between use of contraception and female-to-male HIV-1 transmission (European Study Group on Heterosexual Transmission of HIV, BMJ 1992)
Slide courtesy of Ludo Lavreys
HIV shedding
Hormonal contraception and cervical DNA viral shedding
ANY HC DMPA OCPsWang AIDS (2004)
1.6 (1.03-2.6)
1.5 (0.8-2.9) 2.3 (0.9-6.7)
Mostad Lancet (1997)
2.9 (1.5-5.7) 3.8 (1.4-9.9)
Clemetson JAMA(1993)
11.6 (1.7-77.6)
HIV shedding
Detection of Cervical HIV-1 DNA Depending on Type of Hormonal Contraception Used
0
2
4
6
8
10
12
14
None DMP A L ow-dose O C High-dose O C
OR 2.9(1.5-5.7)
OR 3.8(1.4-9.9)
OR 12.3(1.5-101)
OR 1.0
318 HIV-infected Kenyan women Mostad,Lancet 1997
HIV shedding
What do animal models tell us?
Risk of SIV Acquisition of Female Monkeys with Progesterone Implants
Exposure Status
Number Exposed
Number Infected
Rate
RR 95% (CI)
Progesterone Implants
18 14 77.8 2.1 (0.9-4.9)
Placebo - Cycle Phase
11 4 36.4 7.8 (2.8-16.1)
Placebo - Follicular Phase
10 1 10.0
Source: Marx (1996); Duerr (1997)Slide courtesy of C. Morrison
Immune responses in macaques treated with SIV and DMPA
• No diff in α-SIV Ab response
• No IFN-γ secreting cells detectable in Depo-treated arm at 1-2 weeks
• Response of IFN-γ production delayed in Depo treatment arm
Trunova Virology 2006
Review
• Data on HC and HIV acquisition is varied• HC affected VL set point in new
acquisitions• In our RCT women on HC had faster
disease progression– Suggestion that DMPA worse than OCPs– Warrants more investigation
• HC shown to increase cervical DNA viral shedding
What could be happening here?
Changes in vaginal environment• Related to vaginal thinning
– Vaginal thinning in macaques treated with Progesterone
• Marx, Nature 1996
– Slight thinning of vaginal epithelial layer• Miller, AJOG 2000
• Changes in vaginal microenvironment– Progesterone associated with
decreased H202+ Lactobacillus • Miller, AJOG 2000
Other possible mechanisms
• Increasing HIV co-receptor expression – OCPs upregulates CCR5 receptors in
women• Prakash, J of Reproductive Immunology 2002
– Progesterone increases expression of CCR5 and CXCR4 expression
• Patterson et al, Int AIDS Conf, 1998
• Infection with multiple viral types• T cell activation
Summary
• Ensuring safe and effective contraception to HIV infected women is a priority
• Understanding mechanisms of potential interactions between HC and immune system is critical
• Evidence to date not compelling to change policy
END
• Special thanks to:– Lusaka District Health Board, Lusaka, Zambia– Charles Morrison, FHI– Ludo Lavreys, Tibotech– CIDRZ team– Mike Saag– Sten Vermund– EGPAF
Contraception on immune system function
• Estrogen and progesterone decreases immune response (Th1>Th2)
• Estrogen increases LTR induced viral replication
• Estrogen and progesterone induce upregulation of CCR5 receptor on CD4 T cells in cervical epithelium of non-HIV women- Prakash J of Reproductive Immunology 2002
Enomoto et al JAIDS 2007
Effect of progesterone on chemokine expression in vivo
• PBMCs incubated with 50 ng/ml progesterone for 6 days
• At 0, 3, and 6 days, RNA was extracted for CCR5, CXCR4, and CCR2b mRNA
• Flow cytometry performed to localize increases in CCR5 and CXCR4
Patterson, et al, AJP, 1998