CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

compared the Tfh responses from chronic HIV infection to several phase one and phase II clinical trials, which differed in the immunogen, adjuvant and route of delivery. Methods: We characterized the functional profiles of CD4 T cells using multicolor flow cytometry on cryporeserved PBMC isolated from chronically infected patients or vaccination study participants. The phase one and phase II clinical vaccination trials included in this analysis were RV138, RV172, RV114, RV132, RV135, RV158, and RV144. Results: The HIV-specific CD4 T cell response in chronic natural HIV infection and the response to different vaccine modalities differed markedly. While chronic HIV infection provoked a higher frequency of HIV-specific CD4 T cells than vaccination (p<0.001, n=6 for chronic HIV infection, n=97 for vaccination), the functional profiles between the responses induced in natural HIV infection versus after vaccination were also significantly different. Chronic natural HIV infection showed an HIV-specific CD4 response dominated by TNFα. In contrast, vaccination induced mostly CD40L-positive CD4 T cell responses. In addition, when we compared the effect of different vaccination modalities (route of delivery, prime/boost strategies) on the T cell functional profile, we found that vaccine delivery via antigen-loaded dendritic cells induced a stronger Th1 polarization than intramuscular injection. Additionally, the choice of prime and boost influenced the degree of polyfunctionality induced in T cells: The prime/ boost combination ALVAC-HIV/o-gp160 (RV132) led to the highest frequency of HIV-specific CD4 memory T cells with a polyfunctional profile (p=0.028 for the frequency of IFNγ, and p=0.035 for TNFα expressing cells compared to the other vaccine modalities). Conclusion: Overall, we describe the varying functional T cell response to different vaccination strategies and the differences in responses to chronic HIV-1 infection vs. vaccination. Future approaches to vaccine design will be informed by this and further studies with the goal to induce polyfunctional T cell responses, which support the production of protective antibodies. Katelyn J. Rittenhouse 1 , Humphrey Mwape 2 , John Mwale 2 , Gabriel Chipili 2 , Joan T. Price 2 , Bellington Vwalika 2 , Kristina De Paris 1 , Elizabeth M. Stringer 1 , Jeffrey S. Stringer 1 1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 University of North Carolina in Zambia, Lusaka, Zambia Background: Maternal HIV infection and its treatment are associated with increased risk of preterm birth (PTB). Conventional wisdom holds that an expansion of regulatory T cells (Tregs) during the 2nd trimester plays an integral role in maternal tolerance of the fetal allograft. However, recent studies (in HIV- uninfected populations) using updated immunophenotyping methods specific for viable, suppressive Tregs have not shown an expansion of these cells in the peripheral circulation during pregnancy. We sought to determine whether HIV infected (HIV+) pregnant women have decreased peripheral Treg frequencies compared to their HIV uninfected (HIV-) counterparts. Methods: Between May 2017 and January 2018, we immunophenotyped 64 1st trimester (HIV-: 53; HIV+: 11) and 270 2nd trimester (HIV-: 222; HIV+: 48) peripheral blood specimens collected fromwomen enrolled in the Zambian Preterm Birth Prevention Study (ZAPPS), a prospective cohort ongoing in Lusaka. We quantified Treg frequencies by flow cytometry (CD4+CD25+CD127low & CD4+CD25+CD127lowFoxP3+). T test was used to compare log-transformed Treg frequencies by HIV serostatus at study enrollment. ANOVA was used for sub-group analyses by preconceptional ART and viral load at enrollment. For patients with repeat 1st and 2nd trimester specimens (HIV-: 33; HIV+: 7), changes in Treg frequency were assessed by T test of log-transformed fold changes. Results: No significant differences in CD4+CD25+CD127low nor CD4+CD25+CD127lowFoxP3+ phenotypes were observed between HIV serostatus groups in either 1st or 2nd trimesters (figure). Additionally, individuals on preconceptional ART and with suppressed viral load were not found to differ significantly from their non-preconceptional ART and unsuppressed viral load counterparts, respectively. In individuals with repeat specimens, there were no statistically significant differences between groups for both CD4+CD25+CD127low (p = 0.67) and CD4+CD25+CD127lowFoxP3+ (p = 0.42) phenotypes. Conclusion: Exploratory data from this African cohort established specifically to study PTB do not demonstrate significant aberrations in peripheral Treg frequencies in HIV infected pregnant women. Although Tregs may play a role in

HIV-associated PTB at the maternal-fetal interface, this finding indicates that their role is unlikely to be systemic.

322 THE IMPACT OF HIV EXPOSURE ON PLACENTAL PATHOLOGY AND TREG CELLS Nadia M. Ikumi , Nonzwakazi Bangani, Michelle Barboure, Berenice Alinde, Bryan Gascon, Lizette Fick, Komala Pillay, Landon Myer, Thoko Malaba, Mohammed Lamorde, Saye Khoo, Heather Jaspan, Clive M. Gray University of Cape Town, Cape Town, South Africa Background: One of the main roles of the placenta is to maintain fetal- maternal (FM) tolerance. HIV and/or antiretroviral (ARV) exposure may interfere with this tolerance but data are sparse. We characterized placental decidua T regulatory cells (Treg) from HIV-infected women who initiated ART late in pregnancy compared to uninfected controls. Methods: Placentas of HIV-infected women were drawn from an ongoing study in which women commence antiretroviral therapy (ART) at or after 28 weeks’ gestation (n=14) and HIV-uninfected controls (n=6). The maternal decidua and villous tissue were dissected and enzymatically digested to obtain lymphocytes which were characterised using 15 colour multiparametric flow cytometry. Placenta tissue sections were formalin-fixed and wax embedded for Treg cell characterisation using immunofluorescence and pathology scoring based on the Amsterdam placental workshop group consensus statement. Results: A higher incidence of preterm deliveries was reported in the HIV infected mothers, 75%were very preterm (28+0-31+6 weeks), 18%moderate or late preterm (32+0-36+6 weeks) and 12% term (>37 weeks). The frequency of decidual CD4+ T cells was lower in placentas from HIV infected women when compared with HIV uninfected controls (p=0.005) and similarly, total CD8+ T cells were significantly higher in the HIV infected group (p=0.006). The variable expression of TIGIT (T cell Ig and ITIM domain) and CD45RA, expression on CD4+ T cells within decidual membranes was higher in HIV infected women vs uninfected. We identified a series of Treg subsets within the decidua that were all CD3+CD4+CD127-CD25HiFoxP3++ with variable expression of CD39, CTLA4 and TIGIT. Highly suppressive Treg cells, co-expressing all three markers more enriched in placentas from very preterm placentas in HIV infected women (Figure 1). Pathology indices including prolonged meconium exposure, cord vessel vasculitis and plasmacytic deciduitis were also reported in higher frequency in placentas from HIV infected women, specifically in very preterm deliveries in comparison to controls. Conclusion: The T cell phenotype in the maternal decidua appears to have a predominantly adult systemic footprint while the villous tissue mirrors foetal cells; an increased influx of naïve cells. There are unique and multiple Treg signatures in the placenta which appear to be associated with pre-term birth and may be influenced by HIV exposure and ARV that warrant further investigation.

Poster Abstracts

321 REGULATORY T CELLS IN HIV-INFECTED PREGNANT WOMEN

CROI 2019 117