Quantifying ongoing HIV-1 exposure in HIV-1-serodiscordant couples to identify individuals with potential host resistance to HIV-1

Abstract

Background: Immunogenetic correlates of resistance to HIV-1 in HIV-1-exposed seronegative (HESN) individuals with consistently high exposure may inform HIV-1 prevention strategies. We developed a novel approach for quantifying HIV-1 exposure to identify individuals remaining HIV-1 uninfected despite persistent high exposure.

Methods: We used longitudinal predictors of HIV-1 transmission in HIV-1 serodiscordant couples to score HIV-1 exposure and define HESN clusters with persistently high, low, and decreasing risk trajectories. The model was validated in an independent cohort of serodiscordant couples. We describe a statistical tool that can be applied to other HESN cohorts to identify individuals with high exposure to HIV-1.

Results: HIV-1 exposure was best quantified by frequency of unprotected sex with, plasma HIV-1 RNA levels among, and presence of genital ulcer disease among HIV-1-infected partners and by age, pregnancy status, herpes simplex virus 2 serostatus, and male circumcision status among HESN participants. Overall, 14% of HESN individuals persistently had high HIV-1 exposure and exhibited a declining incidence of HIV-1 infection over time.

Conclusions: A minority of HESN individuals from HIV-1-discordant couples had persistent high HIV-1 exposure over time. Decreasing incidence of infection in this group suggests these individuals were selected for resistance to HIV-1 and may be most appropriate for identifying biological correlates of natural host resistance to HIV-1 infection.

Figures

Figure 1.

Smoothed density curves representing human immunodeficiency virus type 1 (HIV-1) exposure score distributions for all HIV-1 seroconverters and HIV-1–exposed seronegative participants from the highest, lower, and decreasing exposure risk groups. Longitudinal HIV-1 exposure scores were quantified using time-dependent predictors (unprotected sex, plasma HIV-1 RNA levels, and symptomatic genital ulcer disease in HIV-1–infected partners and age, pregnancy, herpes simplex virus 2 serostatus, and male circumcision in HIV-1–exposed seronegative participants), with a 1-unit increase representing a exp(1)=2.7-fold increased risk of HIV-1 acquisition. HIV-1 exposure risk groups were based on individual exposure score trajectories over time and were created using longitudinal K-means cluster analysis. Area under the kernel density curves between 2 HIV-1 exposure risk scores represents the probability that exposure risk scores for individuals in the respective participant subgroup fell between those 2 values of the exposure score.

Figure 2.

Empirical hazard functions for human immunodeficiency virus type 1 (HIV-1) acquisition among HIV-1 exposure score risk groups, determined by clustering initially HIV-1-exposed seronegative individuals into homogenous groups on the basis of their longitudinal HIV-1 exposure trajectories. Hazard rates represent the instantaneous risk of HIV-1 acquisition at time t conditional on survival until time t or later.

Figure 3.

Receiver operating characteristic (ROC) curves comparing the ability of an individual's average human immunodeficiency virus type 1 (HIV-1) exposure score across all study visits to discriminate HIV-1 acquisition risk for participants in the primary and secondary cohorts. Models for HIV-1 acquisition were developed with primary cohort of 3408 initially seronegative partners from HIV-1–discordant couples in the Partners in Prevention HSV/HIV Transmission Study. The final best-fitting Cox proportional hazards model for HIV-1 acquisition included unprotected sex with, HIV-1 RNA level of, and genital ulcer disease for the infected partner and herpes simplex virus 2 (HSV-2) serostatus, pregnancy, sex, age, and male circumcision for the uninfected partner. The reduced model included the same variables as the primary model but used baseline rather than longitudinal plasma HIV-1 RNA levels. The secondary cohort included 485 seronegative partners from HIV-1 discordant couples in the Couples Observational Study. Only baseline HIV-1 RNA levels for infected partners were available for this cohort, so ROC curves were generated for a reduced model that included baseline but not time-varying HIV-1 RNA concentrations. Abbreviation: AUC, area under the ROC curve.

Figure 4.

Simulations demonstrating potential biases when evaluating potential correlates of human immunodeficiency virus type 1 (HIV-1) resistance if HIV-1 exposure is not considered. Scenario 1 assumes that a 1-unit increase in HIV-1 exposure score is associated with a 1-unit increase in a continuous hypothetical host factor and that the host factor is not associated with HIV-1 acquisition. A, True relationship of exposure score and HIV-1 exposure score with no difference in host factor for seroconverters and HIV-1–exposed seronegative (HESN) individuals. B, False-positive association of HESN with elevation in the hypothetical host factor due to random selection of controls without regard to HIV-1 exposure levels. C, True-negative association revealed by selecting controls with similar exposure scores as seroconverters. Scenario 2 assumes that a 1-unit increase in HIV-1 exposure score is associated with a 1-unit increase in a continuous hypothetical host factor and that the average level of the continuous host factor was 2 units lower among seroconverters than among HESN individuals of the same exposure level. D, True relationship of exposure score and HIV-1 exposure score with a 2-unit difference in host factor for seroconverters and HESN individuals. E, False-negative association of HESN with the hypothetical host factor due to random selection of controls. F, True-positive association revealed by selecting controls with similar exposure scores as seroconverters.