CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

Methods: We performed binding antibody multiplex assays to measure IgG binding against a cross-clade panel of 19 HIV-1 antigens in a cohort of 72 breastfeeding Kenyan mother-infant pairs enrolled during the pre-ART era. Infant plasma from the first week of life (before infection) and paired maternal plasma were screened for binding. IgG binding of the non-transmitting vs transmitting maternal samples (or uninfected vs infected infant samples) were compared by a logistic regression analysis adjusting for maternal viral load. Results: IgG binding to two antigens, a gp41 ectodomain protein and gp140 monomer, which contains the gp41 ectodomain, was significantly higher for transmitting mothers compared to non-transmitting mothers (ectodomain p=0.005; gp140 p=0.021), with a similar trend for maternal IgG binding to a gp41 full length protein (p=0.0501). Passively-acquired IgG binding for infected infants compared to uninfected infants also showed trends in the same direction for these antigens (ectodomain p=0.058; gp140 p=0.071; gp41=0.080). None of the gp120-only antigens or the envelope SOSIP trimer, where the gp41 ectodomain is present but partially buried, showed a difference in binding between the transmission groups for either maternal or passively- acquired IgG. Conclusion: These data suggest that in this cohort, IgG targeting the gp41 ectodomain is associated with increased odds of MTCT. This suggests that some antibody responses may be detrimental in terms of MTCT risk. Determining whether gp41 antibodies directly affect transmission or whether their presence reflects a redirecting of the responses toward epitopes exposed on non-native envelope may help better define protective antibody responses need to prevent MTCT. 1 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 2 University of Washington, Seattle, WA, USA, 3 University of Nairobi, Nairobi, Kenya Background: We previously reported that passively-acquired antibody- dependent cellular cytotoxicity (ADCC) in infants is associated with improved survival of infected infants and a trend towards protection against mother-to- child transmission (MTCT) of HIV-1. However, the epitopes of these beneficial ADCC-mediating antibodies have not been investigated. Because CD4-incudible (CD4i) epitopes are a common target of antibodies elicited by natural infection, we hypothesized that ADCC targeting these CD4i epitopes may contribute to improved infant survival. Methods: LALA variants to 3 CD4i antibodies, A32, C11, and 17b, were used as inhibitors in a competition rapid and fluorometric ADCC assay to measure CD4i epitope-specific ADCC of plasma samples from a cohort of 72 breastfeeding Kenyan mother-infant pairs enrolled during the pre-ART era. Infant plasma from the first week of life (pre-infection) and paired maternal plasma were tested. A32-like, C11-like, and 17b-like ADCC of the non-transmitting vs transmitting maternal plasma (or uninfected vs infected infant plasma) were compared using logistic regression adjusted for maternal viral load. The effect of ADCC targeting CD4i epitopes on infected infant survival was assessed by Cox-proportional hazards models. Results: A32-like and C11-like ADCC were common in this cohort but were not associated with MTCT (Table 1). A32-like ADCC was not associated with infected infant survival, but maternal C11-like ADCC was associated with a trend towards increased mortality of infected infants (HR=1.062; p=0.09; Table 1). Surprisingly, 17b-like ADCC was negative in the majority of samples, indicating that 17b-LALA mediated an enhancement of plasma ADCC. This enhancement was not associated with MTCT but was associated with increased infected infant mortality (Table 1). Enhancement with 17b-LALA was inversely correlated with total ADCC (Pearson R: -0.72), indicating that the negative association of enhancement with infant outcome may be due to lower total ADCC, consistent with our previous report. Conclusion: While CD4i-epitope-specific ADCC antibodies were elicited in this cohort, they likely are not responsible for improved infant outcome seen with higher passively-acquired ADCC. As suggested by the trend with maternal C11-like ADCC, C11-like ADCC may actually be associated with worse infant outcome, although further investigation is necessary. Nicole Naiman 1 , Jennifer Slyker 2 , Barbra A. Richardson 2 , Ruth Nduati 3 , Julie M. Overbaugh 1

1 University of Washington, Seattle, WA, USA, 2 Seattle Children’s Research Institute, Seattle, WA, USA, 3 Harvard University, Boston, MA, USA, 4 University of Zimbabwe, Harare, Zimbabwe, 5 Johns Hopkins University, Baltimore, MD, USA, 6 Makerere University–Johns Hopkins University Research Collaboration, Kampala, Uganda Background: HIV drug resistance (DR) can undermine antiretroviral treatment (ART). We examined risks of maternal DR on perinatal transmission (PT) in a case-control study, and infants’ acquisition of DR during breastfeeding (BF) in the PROMISE 1077BF trial. Methods: 85 HIV-infected infants and their transmitting mothers were compared to 254 HIV-infected, non-transmitting control mothers. PT was categorized by time of diagnosis: in utero (IU) at ≤2 weeks or during BF at >2 weeks of age. Controls for each PT category were matched for date of delivery and site. Plasma from infant’s date of HIV diagnosis, ART initiation, and last study visit, and mother’s plasma ≥40c/mL from the nearest date proximate to PT (or their matching case’s time of PT for controls) were genotyped by consensus sequencing (CS) of HIV pol. Infants and mothers were categorized as wild-type (WT) or DR based on major DR mutations (DRM) defined by the Stanford Database. Maternal viral loads (VL) and DR rates were independently compared using Mann-Whitney and Fisher’s Exact tests. Adjusted analyses used logistic regression. Results: Proximate to infant diagnosis, case mothers had higher median VL vs. controls (4.28 vs. 3.86 log10 c/mL, p<0.0001). DR was significantly higher in transmitting vs. control mothers (15.8% vs. 7.6%, p=0.048). DR was more prevalent in mothers who transmitted via BF compared to IU (29.7% vs. 4.4%; p=0.002). In a logistic regression adjusted for VL, antepartum (AP) treatment arm, and clinical site, DR was no longer associated with PT (p=0.618), while VL increased (p<0.0001) and AP triple ARV decreased (p=0.021) risk of PT. Of 75 infants with CS, 5/40 (12.5%) with IU vs. 19/35 (54.3%) with BF transmission had DRM at diagnosis (p<0.001). Of the 24 DR infants, 58.3% had 1 NNRTI DRM, 25% had ≥2 NNRTI DRM, 12.5% had 1 NRTI DRM, and 4.2% had dual-class DRM. Among 72 mother-infant pairs genotyped, 46 (64%) were concordant for WT, 7 (9.7%) concordant for DR and 19 (26.3%) were discordant (17 (89.5%) WT mothers had DR infants). Among 46 infants with longitudinal genotypic data, 8/13 (62%) WT infants at diagnosis in the IU cohort and 1/6 (17%) WT infants in the BF cohort acquired DRM resulting in 33/75 (44%) DR infants. Conclusion: DR was frequently detected among women with PT during BF. However, in this study, DR does not appear to be a driver of PT. DR was less prevalent in infants with IU vs. BF PT, and accumulated during early infancy, suggesting that exploration of additional prophylactic regimens is warranted.

771 EPITOPE TARGETS OF ADCC-MEDIATING ANTIBODIES AND THEIR RELATION TO MTCT OF HIV-1

Poster Abstracts

770 GP41 ECTODOMAIN-SPECIFIC IGG IS ASSOCIATED WITH INCREASED VERTICAL HIV-1 TRANSMISSION Nicole Naiman 1 , Jennifer Slyker 2 , Ruth Nduati 3 , Julie M. Overbaugh 1 1 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 2 University of Washington, Seattle, WA, USA, 3 University of Nairobi, Nairobi, Kenya Background: Studies of the epitope specificity of maternal antibodies in relation to reduced risk of mother-to-child transmission (MTCT) have identified correlates of protection in some studies. However, there is little consistency between results across studies. In addition, few studies have investigated pre-existing passively-acquired HIV-specific antibody responses in infants, which are most relevant because they are present prior to HIV exposure through breastfeeding. We hypothesized that pre-existing passively-acquired antibodies that target specific epitopes confer protection against MTCT of HIV-1.

CROI 2019 298