Boston, Massachusetts
March 4–7, 2018


Conference Dates and Location: 
February 13–16, 2017 | Seattle, Washington
Abstract Number: 



Eileen Scully1, Monica Gandhi2, Rowena Johnston3, Rob Gorelick4, Jeffrey D. Lifson5, Sharon R. Lewin6, Jonathan Karn7, Nicolas Chomont8, Peter Bacchetti2, Steven G. Deeks2

1The Johns Hopkins Univ, Baltimore, MD, USA,2Univ of California San Francisco, San Francisco, CA, USA,3amfAR, New York, NY, USA,4Leidos Biomed Rsr, Inc, Frederick, MD, USA,5Frederick Natl Lab, Frederick, MD, USA,6Univ of Melbourne, Melbourne, Australia,7Case Western Reserve Univ, Cleveland, OH, USA,8Univ de Montréal, Montréal, QC, Canada

Abstract Body: 

Plasma HIV RNA levels are lower in women than men in the absence of antiretroviral therapy (ART), particularly in early infection. Less is known about sex differences in HIV during ART. We recently found that estrogen modulates HIV transcription ex vivo. Here, we sought to define sex differences in the size and activity of the HIV reservoir in a cohort of premenopausal women and matched men on ART.

Premenopausal women on ART with ≥1 year of viral suppression were prospectively enrolled (n=26) and matched with men (n=26) on age, duration of viral suppression, CD4 count/nadir and unusual clinical phenotypes. HIV persistence measurements include: integrated HIV DNA (iDNA) in CD4 T cells, residual plasma viremia by single copy assay (SCA) for HIV gag (HMMCgag), and cell associated (CA) multiply spliced (ms) and unspliced (us) HIV RNA in CD4 T cells. The frequency of CD4 T cells producing tat/rev RNA after activation was measured by TILDA in a subset of subjects. T cells were phenotyped by flow cytometry. Virologic data were analyzed by univariate and multivariate negative binomial regression and immune markers were compared with nonparameteric statistics (Mann Whitney).

HIV iDNA levels were comparable between men and women (p=0.47), but female sex was associated with a 76% lower level of residual viremia by SCA (p=0.011) in a multivariate model. CA msHIV RNA was also 6-fold lower in women (p=0.002) in a one-predictor model, and 4-fold lower when adjusted for nadir CD4 and controller phenotype (p=0.009). CA usRNA was ~35% lower in women by univariate and multivariate models (p>0.05). The frequency of inducible virus (TILDA: iDNA ratio) was lower in women (Mann-Whitney p=0.019), suggesting a lower inducible reservoir in women. Women demonstrated a lower proportion of HLADR/CD38+ CD4 and CD8 T cells than men (p<0.05) and lower PD-1 expression, most notably in central memory CD8 T cells (p<0.001).

In a well-matched cohort of ART-treated, virally suppressed women and men, multiple measures of virus activity and immune activation/exhaustion were lower in women despite comparable frequencies of CD4+ T cells harbouring HIV DNA. These data support sex differences in control of HIV latency. Biologic sex may impact the efficacy of curative interventions and manipulation of sex hormones may play a role in cure strategies.

Session Number: 
Session Title: 
Presenting Author: 
Eileen Scully
Presenter Institution: 
Johns Hopkins University School of Medicine