Female HIV acquisition per sex act is elevated in late pregnancy and postpartum

Transcription

1 Female HIV acquisition per sex act is elevated in late pregnancy and postpartum Kerry A. Thomson, 1 James Hughes, 1 Jared M. Baeten, 1 Grace John-Stewart, 1 Connie Celum, 1 Craig R. Cohen, 2 Nelly Mugo, 1 Kenneth Ngure, 1 James Kiarie, 1,3 and Renee Heffron 1 for the Partners in Prevention HSV/HIV Transmission Study and Partners PrEP Study Teams 1 University of Washington, Seattle, WA, USA 2 University of California, San Francisco, CA, USA 3 World Health Organization, Reproductive Health and Research Human Reproduction Programme, Geneva, Switzerland

2 HIV and women Disproportionate HIV risk Leading causes of death for women years worldwide HIV: 12% Tuberculosis: 5% Breast and cervical cancer: 6% Road injuries / self harm: 10% High HIV incidence during pregnancy and postpartum HIV incidence in Pregnancy 4.7 per 100 person-years HIV incidence in Postpartum 2.9 per 100 person-years UNAIDS 2016, IHME Global Burden of Disease, 2016, Drake et al. Plos Med

3 Studies of pregnancy as a risk factor for HIV acquisition Author Country N Hazard Ratio (95% CI) Mugo Africa (multiple) (1.3, 4.1) Wand South Africa (1.6, 2.7) Reid Africa (multiple) (0.5, 3.0) Braunstein Rwanda (0.2, 1.7) Morrison Zimbabwe (0.0, 1.0) Subtotal (I-squared = 79.9%; p=0.001) 1.3 (0.5, 2.1) Drake et al. Plos Med

4 Duration of risk Female average life expectancy: 63 years Total fertility rate per woman: 3.9 children Years pregnant/lactating per pregnancy: 1.75 years Total years pregnant/lactating: ~7 years 10% of lifetime 20% of reproductive years woman year pregnant/lactating Source: Kenya 2014 DHS, slide modified from Drake, IAS 9 th Annual Pediatric HIV Workshop,

5 Research objective To estimate the probability of HIV acquisition per sex act during periods when women were pregnant and postpartum and compare these probabilities to time periods unrelated to pregnancy 5

6 Data sources Two longitudinal HIV prevention studies HIV serodiscordant couples, ART naïve Frequent testing of HIV and pregnancy Monthly reports of sexual behavior (sex acts and condom use) Genetic linkage of incident infections The Partners in Prevention HSV/HIV Transmission Study Botswana, Kenya, Rwanda, South Africa, Tanzania, Uganda, Zambia 3,408 couples for 24 months No effect of acyclovir on HIV incidence The Partners PrEP Study Kenya and Uganda 4,747 couples up to 36 months Significant reduction in HIV incidence from daily oral PrEP 6

7 Analytic approach HIV uninfected women ages Censoring Study visits after male partner reported ART initiation Seroconversions determined to be unlinked HIV infections (from partners with unknown sexual activity) Outcome: First evidence of an HIV infection linked to the male study partner Earliest visit with positive HIV EIA or RNA 7

8 Defining reproductive stages Exposures 1. Pregnant/non-pregnant stage (time dependent) Pregnancy testing: Monthly (Partners PrEP) or as clinically indicated (HSV-2) Pregnancy start and stop dates defined using date of last menstrual period and pregnancy outcome* Early Pregnancy 0-13 wks gestation Late Pregnancy 14 wks delivery/loss Postpartum Delivery 6 months (less for losses) Unrelated Not pregnant or postpartum 2. Number of sex acts within partnership (reported monthly) *Postpartum definition for pregnancy loss 20 weeks or newborn death: 6 weeks; for pregnancy loss 6-19 weeks: 4 weeks; for pregnancy loss <6 weeks: 0 weeks. 8

9 Statistical model Complementary log-log model Adjustment HIV plasma viral load of male partner Active PrEP use Condom use Age Additional adjustment for STI/HSV-2 infection and parent study did not yield substantial changes Reference group for the absolute probability of HIV acquisition per condomless act: 25 year old woman not pregnant not using PrEP male partner viral load=10,000 copies/ml 9

10 Participant characteristics 2,751 HIV-Negative Women, Median (IQR) or N (%) HIV Negative Women Characteristics at enrollment Age (years) 32.0 ( ) Number sex acts with study partner, past month 4.0 ( ) Any condomless sex with study partner, past month 670 (24.4) Pregnancies during follow-up Ever pregnant during follow-up 615 (22.4) Total pregnancies 686 Live birth 426 (62.1) Loss 169 (24.6) Ongoing at study exit 91 (13.3) Thomson et al. Journal of Infectious Diseases

11 Sexual activity by reproductive stage Non-pregnant/ postpartum Early pregnancy Late pregnancy Postpartum All sex acts within study partnership Crude rate of sex acts per person-month (95% CI) Adjusted Relative Risk (95% CI, p-value) 4.62 (4.60, 4.64) (4.61, 4.82) 1.04 (0.95, 1.14) p= (3.17, 3.32) 0.66 (0.59, 0.73) p< (2.24, 2.40) 0.45 (0.39, 0.51) p< Condomless sex acts within study partnership Crude rate of sex acts per person-month (95% CI) Adjusted Relative Risk (95% CI, p-value) 0.47 (0.46, 0.47) (0.91, 1.00) 2.43 (1.91, 3.10) p< (0.40, 0.45) 0.91 (0.66, 1.25) p= (0.19, 0.24) 0.51 (0.27, 0.97) p=0.04 Poisson regression models with an independent correlation matrix and robust standard errors adjusted for female age and duration of partnership with HIV-positive male study partner. Thomson et al. Journal of Infectious Diseases

12 HIV incidence Analysis includes 78 new HIV infections Reproductive Stage HIV incidence per 100 person years (95% CI) Overall 1.62 (1.29, 2.01) During non-pregnant/postpartum time 1.25 (0.95, 1.62) During early pregnancy through postpartum 5.37 (3.44, 7.99) During early pregnancy 3.75 (1.22, 8.75) During late pregnancy 7.02 (3.74, 12.01) During postpartum 4.68 (1.72, 10.18) 12

13 HIV infectivity per 1,000 sex acts % CI: 0.55, 1.87) % CI: 0.38, % CI: 0.66, % CI: 0.72, Unrelated to pregnancy Early Pregnancy Late Pregnancy Postpartum Calculated using a reference case of a 25-year old woman not pregnant, not using PrEP, with a partner with viral load of 10,000 copies/ml Thomson et al. Journal of Infectious Diseases

14 Relative risk of HIV acquisition Reproductive Stage Non-pregnant/ postpartum Early pregnancy through postpartum RR (95% CI) Base Model * Adjusted Model ** p-value RR (95% CI) p-value (2.95, 8.38) < (1.58, 4.81) <0.001 * Adjusted for condom use, reproductive stage ** Adjusted for condom use, reproductive stage female age, active PrEP use, HIV RNA of male partner Thomson et al. Journal of Infectious Diseases

15 Relative risk of HIV acquisition Reproductive Stage Non-pregnant / postpartum Early pregnancy through postpartum Early pregnancy Late pregnancy Postpartum RR (95% CI) Base Model * Adjusted Model ** p-value RR (95% CI) p-value (2.95, 8.38) 3.20 (1.24, 8.25) 5.54 (2.62, 11.69) 7.80 (3.04, 20.02) < <0.001 < (1.58, 4.81) 2.07 (0.78, 5.49) 2.82 (1.29, 6.15) 3.97 (1.50, 10.51) < * Adjusted for condom use, reproductive stage ** Adjusted for condom use, reproductive stage female age, active PrEP use, HIV RNA of male partner 15

16 Additional analyses Similar results in sensitivity analyses Used estimated date of HIV infection Excluded women randomized to active PrEP arms Excluded women who were never pregnant during follow-up Included a longer postpartum period from 6-12 months Thomson et al. Journal of Infectious Diseases

17 Results summary Increased risk of HIV acquisition per sex act 3-fold increase in late pregnancy 4-fold increase in postpartum Similar results seen across multiple sensitivity analyses Results accounted for decreases in sexual frequency and condom use as pregnancy progressed Results of this per-coital act analysis suggest that biological changes associated with pregnancy and postpartum, contribute to increased HIV acquisition. However, we did not directly assess any biological mechanisms for increased HIV susceptibility Thomson et al. Journal of Infectious Diseases

18 Implications Antenatal and postnatal care presents tremendous opportunities to promote HIV prevention and care Counseling on increased HIV risk during pregnancy and postpartum Promoting repeat HIV testing during maternal health visits Identifying HIV infected male partners and linking to HIV care and ART initiation Promoting oral PrEP during pregnancy and postpartum 18

19 Counseling on increased HIV risk of HIV during pregnancy and postpartum WHO now recommends at least 8 contacts between pregnant woman and provider prior to delivery Many opportunities for HIV counseling and initial HIV testing Counseling on HIV risk can include descriptions that women s risk for HIV increase during pregnancy and postpartum due to changes in their sexual behavior and also likely due to the changes their bodies undergo while being pregnant and postpartum Source: WHO

20 Estimated HIV testing and counseling coverage among pregnant women, 2005, Source: WHO Global Update on the Health Sector Response to HIV,

21 New HIV infections among children (0-14 years) and coverage of ART for PMTCT, Global New HIV infections among children represent missed opportunities for HIV diagnosis and/or ART initiation during pregnancy/postpart 21

22 Repeat HIV testing during pregnancy and postpartum For high HIV prevalence settings, WHO recommends: Provider-initiated testing and counseling for women as a routine component of the package of care in all antenatal, childbirth, postpartum and paediatric care settings. Retesting in the third trimester, or during labour or shortly after delivery Periodic retesting of lactating mothers who are HIV-negative throughout the period of breastfeeding For low HIV prevalence settings, WHO recommends PITC for pregnant women HIV testing for pregnant women from key populations or who have partners with HIV or from key populations Source: Consolidated Guidelines on HIV Testing Services July keypopulations/ 22

23 Repeat HIV testing during pregnancy, Kenya MOH data, In this study, only 9.6% of women who could have had a 2 nd HIV test during the 3 rd trimester actually did Gaps include women coming for their first test too late, not returning for ANC visits, returning for ANC visits too soon after the 1 st HIV test, and missing a test at an eligible ANC visit Source: Rogers AJ et al. JIAS

24 Male engagement in HIV testing with onward linkage to HIV care or prevention Factors affecting men s uptake of HIV testing (from the SEARCH study in Kenya and Uganda) Men s employment schedules Gender norms health care seeking is for women Testing by proxy assuming their HIV status is the same as their partner s Camlin CS et al. AIDS Care

25 Male engagement in HIV testing with onward linkage to HIV care or prevention We will need innovative approaches to engage men more Home-based couples testing Self-testing, including secondary distribution of self-test kits from pregnant women to their partners Sources: Osoti AO et al. AIDS 2014; Mark J et al. STD 2017; Krakowiak D et al. JAIDS 2016; Thirumurthy H et al. Lancet HIV 2016; Masters SH et al. Plos Med

26 PrEP use during pregnancy WHO guidelines The existing safety data support the use of PrEP in pregnant and breastfeeding women who are at continuing substantial risk of HIV infection AND active surveillance of mother and infant outcomes during PrEP use in pregnancy and breastfeeding should be part of a PrEP programme 1. Maternal adverse outcomes 2. Adverse birth outcomes 3. Adverse infant and child outcomes Source: WHO. Preventing HIV during pregnancy and breastfeeding in the context of PrEP. July

27 National guidelines Countries with generalized HIV epidemics Few have pregnancy-specific guidance Among the 4 that do, (South Africa, Kenya, Swaziland, and Uganda), 3 recommend PrEP to be used during pregnancy South Africa: TDF/FTC is contra-indicated for use as PrEP in pregnant or breastfeeding women. However, as the risk of seroconversion during pregnancy is high, the risks and benefits of PrEP should be discussed with potential PrEP users, allowing these women at high risk of HIV acquisition to make an informed decision regarding PrEP use. Countries with concentrated HIV epidemics PrEP can be used during pregnancy, consideration should be given to whether there is an added benefit of PrEP in the context of ART use and viral suppression. Davies and Heffron, under review 27

28 Implications summary There are great opportunities to engage pregnant women in HIV prevention, beginning with HIV testing Structural and individual barriers present challenges with HIV testing and innovation is needed to overcome these barriers for women and men Oral PrEP is recommended for use during pregnancy; countries with PrEP guidelines are often permissive of PrEP use in pregnancy with options for women to weigh personal preferences BUT not all countries with large HIV epidemics among women have adopted WHO recommendations 28

29 Acknowledgments Thank you to all the participants from these studies Partners in Prevention HSV/HIV Transmission Study Team University of Washington Coordinating Center: Connie Celum (PI and Chair), Anna Wald (Co-Chair), Jairam Lingappa (Medical Director), Jared M. Baeten, Mary Campbell, Lawrence Corey, Robert W. Coombs, James P. Hughes, Amalia Magaret, M. Juliana McElrath, Rhoda Morrow, James I. Mullins Sites and collaborating partners: Cape Town, South Africa (U Cape Town): David Coetzee Eldoret, Kenya (Moi U, Indiana U): Kenneth Fife, Edwin Were Gaborone, Botswana (Botswana Harvard Partnership): Max Essex, Joseph Makhema Kampala, Uganda (Makerere U): Elly Katabira, Allan Ronald Kigali, Rwanda, Kitwe & Lusaka, Zambia (Rwanda Zambia HIV Research Group, and Emory U): Susan Allen, Kayitesi Kayitenkore, Etienne Karita, William Kanweka, Bellington Vwalika, Mubiana Inambao Kisumu, Kenya (KEMRI, UCSF): Elizabeth Bukusi, Craig Cohen Moshi, Tanzania (Kilimanjaro Christian Medical College, Harvard U): Saidi Kapiga, Rachel Manongi Nairobi, Kenya (U of Nairobi, U of Washington): Carey Farquhar, Grace John-Stewart, James Kiarie Orange Farm, South Africa (RHRU, U Witwatersrand): Sinead Delany-Moretlwe, Helen Rees Soweto, South Africa (PHRU, U Witwatersrand): Guy de Bruyn, Glenda Gray, James McIntyre Thika, Kenya (U Nairobi, U Washington): Nelly Mugo Partners PrEP Study Team University of Washington Coordinating Center: Connie Celum (PI and Co-Chair), Jared Baeten (Co-Chair and Medical Director), Deborah Donnell (Statistician), Robert Coombs, Lisa Frenkel, Jim Hughes, Jai Lingappa, Julie McElrath Sites and collaborating partners: Eldoret, Kenya (Moi U, Indiana U): Edwin Were, Ken Fife Jinja, Uganda (Makerere U, UW): Patrick Ndase, Elly Katabira Kabwohe, Uganda (KCRC): Elioda Tumwesigye Kampala, Uganda (Makerere U): Elly Katabira, Allan Ronald Kisumu, Kenya (KEMRI, UCSF): Elizabeth Bukusi, Craig Cohen Mbale, Uganda (TASO, CDC): Jonathan Wangisi Nairobi, Kenya (KNH/U Nairobi, UW): James Kiarie, Carey Farquhar, Grace John-Stewart Thika, Kenya (KNH/U Nairobi, UW): Nelly Mugo Tororo, Uganda (CDC, TASO): Jim Campbell, Jordan Tappero Funding Bill and Melinda Gates Foundation This project was supported by the National Center For Advancing Translational Sciences of the National Institutes of Health under Award Number TL1 TR The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health 29

30 Thank you! Questions? Dr. Kerry Thomson Dr. Renee Heffron 30

31 Comparison to other per-act studies Cohort Meta-analysis 5 low income countries 1 Male to Female Infectivity per 1000 acts Comparison Study Details 1.9 Mix of partners studies HSV2 cohort 1.9 HIV serodiscordant couples only 2 Did not account for pregnancy/pp Preg PP Unrelated Rakai, Uganda arr: 1.42 ( ) Mama Salama 4 Western Kenya HIV serodiscordant couples (not linked) Annual HIV testing (assumed midpt) Self-report pregnancy dates Reported average sexual behavior Extrapolated reported sex acts NA p=0.08 Already pregnant at enrollment No non-pregnant women/time Follow-up 20wks gest-14wks pp No male partner data / adjustment / unknown status 1. Bailey, Baggaley, et al. Lancet, 2009; 2. Hughes, Baeten, et al. JID, Gray, Li, et al. Lancet, 2005; 4. Kinuthia, Drake, et. al. AIDS, 2015

32 Sensitivity analyses Primary Model Sensitivity Analyses Reproductive Stage First Evidence of HIV All Female Report of Sex Acts Exclude any Imputation Exclude Active PrEP Arm * Exclude Women Never Pregnant Infections Early pregnancy through postpartum (combined) Non-pregnant/ postpartum time RR (95% CI) 2.76 (1.58, 4.81) p- value <0.001 RR (95% CI) p-value 2.37 (1.34,4.20) RR (95% CI) 3.11 (1.72, 5.62) p- value RR (95% CI) p-value 2.92 (1.59, 5.35) RR (95% CI) 3.37 (1.59, 7.16) p- value Adjusted for condom use, reproductive stage female age, active PrEP use, HIV-1 RNA of male partner HIV-1 infected study partner * Adjusted for condom use, reproductive stage female age, HIV-1 RNA of male partner HIV-1 infected study partner 3