CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

Results: DMPA users had significantly higher CVX CUM p24 than the no HC (p<.0001) and Cu-IUD (p=.0048) users. LNG-IUD users had a trend of higher CVX CUM p24 compared to no HC (p=.0764) users. DMPA and LNG-IUD users had significantly higher VAG CUM p24 than the no HC users (p=.0029 and p=.0025, respectively). No difference in CUM p24 was found between no HC and Cu-IUD users for CVX and VAG tissues. There was high intra-person variability in CUM p24 for both tissues with an ICC of ≤0.234. Conclusion: DMPA and to a lesser degree LNG-IUD use increased HIV-1 replication in mucosal tissue, which supports the epidemiological findings of increased HIV acquisition among DMPA users. Our data suggest that DMPA users should be considered for inclusion into clinical trials and controlled for in the analysis where the ex vivo challenge assay is used. HIV-1 replication in CVX and VAG tissues showed high p24 variability within the same woman suggesting baseline biopsies are not needed for the ex vivo challenge assay so long as a placebo group is included in the study. 945 ZIM-CHIC: IMPACT OF CONTRACEPTIVE INITIATION ON IMMUNE CELLS IN THE CERVIX AND PBMCS Background: Injectable contraceptives have been linked to increased susceptibility to HIV. Our objective was to prospectively evaluate the impact of three injectable contraceptives (depot medroxyprogesterone actetate [DMPA], norethisterone enanthate [Net-EN], medroxyprogesterone acetate and ethinyl estradiol [MPA-EE]), two implantable contraceptives (levonorgestrel [LNG], etonogestrel [ENG]) and the copper IUD on HIV-target cells including CD4 T-cells expressing CCR5 and CD69 and antigen presenting cells (APCs) expressing CD11c, 90-days after initiation. Methods: 193 women in Harare, Zimbabwe aged 18-34 who had not used hormonal or intrauterine contraception for >30 days or DMPA for >10 months were enrolled. Women were negative for sexually transmitted infections and HIV, and were confirmed by mass spectrometry to be free of exogenous hormones and in the follicular phase of the menstrual cycle. Cervical cytobrush samples and PBMCs were obtained at baseline and 90-days after contraceptive initiation. Immune cell populations were quantified by flow cytometry and gating was independently performed by three scientists masked to contraceptive group. Paired changes were evaluated using the Wilcoxon Signed Rank test. Results: Women initiating Net-EN had significant increases in both the number and percent (%) of local and systemic CD4 cells expressing CCR5 (Fig) as well as CD4 cells expressing the activation marker CD69 (p=.001 in cervical cytobrushes and p=.03 in PBMCs). Women initiating copper IUD use had an increase in the number of CD4 cells expressing CCR5 (Fig) and CD69 (p=.002) as well as the number and % of CD11c+ APCs in the cytobrush (p=.002 and .007 respectively). Women initiating DMPA, MPA-EE or implantable progestins had no significant change in the number or % of CD4 cells expressing CCR5 or CD11c cells in the cytobrush or PMBCs. Conclusion: Initiation of the injectable contraceptive Net-EN was associated with increases in CCR5 and CD69 on CD4 lymphocyte populations in the genital tract and systemically, a finding not observed following initiation of DMPA or MPA-EE. These data suggest that any increased susceptibility to HIV associated with MPA may occur through an alternative mechanism. The ECHO trial is comparing HIV incidence among women randomized to contraception with DMPA, LNG implants, or copper IUDs. Because initiation of copper IUDs was associated with local changes in immune cells, IUDs should not be considered as placebos when evaluating the impact of contraceptives on HIV risk. 946 HIV GENITAL-TRACT SHEDDING WHILE ON PROGESTIN CONTRACEPTION Julie A. Nelson 1 , Jeff Wiener 2 , Jennifer H. Tang 1 , Lameck Chinula 3 , Stacey Hurst 2 , Gerald Tegha 3 , Albans Msika 3 , Mina C. Hosseinipour 3 , Lisa B. Haddad 4 , Athena Kourtis 2 1 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA 2 CDC, Atlanta, GA, USA, 3 Univ of North Carolina Proj–Malawi, Lilongwe, Malawi, 4 Emory Univ, Atlanta, GA, USA Background: During a randomized trial of the Depot Medroxyprogesterone Acetate injectable (DMPA) and the Levonorgestrel (LNG)-implant that was designed to evaluate the effects of the two progestin contraceptives on HIV genital shedding in HIV-infected women, multiple genital specimens were collected up until 6 months after contraceptive initiation. Previous analysis indicated no difference in HIV RNA shedding between the two contraceptive arms using cervicovaginal lavage (CVL) among the subset of women who were taking antiretroviral therapy (ART). Methods: Both CVL and cervical TearFlo Strips (TFS) were collected at 2 time points before and 4 time points after contraceptive initiation. HIV RNA levels were measured (Abbott RealTime HIV-1 Assay) from CVL and TFS at all 6 time points. In addition, cell pellets from CVL were extracted for DNA and measured by Droplet Digital PCR for HIV DNA. Among women on ART, detectable genital HIV RNA was compared for the two specimen types between study arms using repeated measurements models fit by generalized estimating equations. Results: Among the 73 HIV-infected women who were randomized to either contraceptive, 68 (93%) were taking ART (35 using DMPA and 33 using LNG-implant) and included in this analysis. Overall, the TFS showed higher viral loads (max 300,000 cp/ml) than CVL (max 3700 cp/ml), and more frequently detectable HIV RNA overall (9.6% vs 3.7%). Despite more frequent detection of shedding with TFS, the risk of genital shedding of HIV was not different between the DMPA and LNG-implant arms (RR [95%CI] 1.19 [0.52-2.72] for TFS and 0.92 [0.22-3.83] for CVL adjusted for HIV viral load and CD4 count). There was also no increased risk for shedding after initiation of progestin contraception, either in TFS or CVL. Using TFS, the highest frequency of shedding was in the follicular phase prior to contraceptive initiation (14.9%), whereas with CVL, the highest frequency of shedding was 3 days after contraceptive initiation (4.7%). HIV DNA was detected in only one CVL cell pellet from a woman whose CVL and TFS HIV RNA measures were detectable. Conclusion: Among women on ART, there was no difference in the frequency of genital HIV RNA shedding between the DMPA and LNG-implant arms or before and after initiation of progestin contraception, regardless of the specimen type used. HIV RNA was detectable more often using TearFlo. HIV RNA in CVL fluid and HIV DNA in CVL cells are less frequently detected in the same women. 947 HORMONAL CONTRACEPTION DOES NOT AFFECT ART EFFECTIVENESS OR GENITAL HIV SHEDDING Rena Patel 1 , Jared Baeten 1 , Renee Heffron 1 , Ting Hong 1 , Nicole Davis 2 , Kavita Nanda 3 , Robert Coombs 1 , Elizabeth A. Bukusi 4 , Athena Kourtis 2 , Nelly Mugo 5 1 Univ of Washington, Seattle, WA, USA, 2 CDC, Atlanta, GA, USA, 3 FHI 360, Durham, NC, USA, 4 Kenya Med Rsr Inst, Nairobi, Kenya, 5 KEMRI, Nairobi, Kenya Background: To explore the association between hormonal contraceptive use and antiretroviral therapy (ART) effectiveness and genital HIV shedding among HIV-positive women on ART. Methods: We analyzed plasma viral load and genital viral RNA shedding from 1,079 HIV-positive women initiating ART who were followed prospectively in three HIV prevention studies in sub-Saharan Africa. Plasma and endocervical swab samples were collected every six months. Self-reported contraceptive use was categorized into injectable, implant, oral, or non-hormonal/no contraception. We used multivariate Cox regression to assess time to plasma viral suppression and logistic regression with generalized estimating equations to assess genital viral shedding for each contraceptive method. Results: At the time of ART initiation, there were 211 (20%) injectable, 69 (6%) implant, 50 (5%) oral, and 749 (69%) non-hormonal or no method users. Viral suppression was high (90% by 6 months) and hormonal contraceptives did not significantly diminish time to viral suppression as compared to non-hormonal or no methods [adjusted hazard ratios were: injectables 0.89 (95% CI 0.75-1.1), implants 0.91 (0.68-1.2), and oral methods 1.3 (1.1-1.7)]. Genital viral shedding at any time after ART initiation was uncommon (only 9% of samples had detectable viral shedding) and hormonal contraceptives were not significantly associated with changes in detection of genital viral shedding [adjusted odds ratios were: injectables 1.1 (0.73-1.7), implants 0.67 (0.30-1.5), and oral methods 0.55 (0.18-1.6)]. Conclusion: Hormonal contraceptive use was not significantly associated with reduced ART effectiveness or increased genital HIV shedding among HIV-positive women on ART. HIV-positive women should continue to be offered the full range of contraceptive options that best meet their needs. 948 ASYMPTOMATIC ANAL CHLAMYDIA AND GONORRHEA ARE ASSOCIATED WITH CD8+ T-CELL ACTIVATION Vinícius A. Vieira 1 , Natalia B. Cerqueira 1 , Dayane A. Costa 1 , Priscilla R. Costa 1 , Vivian I. Avelino-Silva 2 , Ricardo D. Vasconcelos 2 , Jose Valdez Madruga 2 , Valdilea Veloso 3 , Beatriz Grinsztejn 3 , Esper G. Kallas 1 Sharon L. Achilles 1 , Feliz Mhlanga 2 , Leslie A. Meyn 1 , Kevin A. Stoner 3 , Allen T. Matubu 2 , Zvavahera M. Chirenje 2 , Sharon L. Hillier 1 1 Univ of Pittsburgh, Pittsburgh, PA, USA, 2 Univ of Zimbabwe, Harare, Zimbabwe, 3 Magee–Womens Rsr Inst, Pittsburgh, PA, USA

Poster and Themed Discussion Abstracts

CROI 2017 410