Table 1.
Clinical cohort characteristics.
Fig 1.
Increased HSV-2 shedding rate, episode rate and episode duration in HIV infected men.
Frequency histograms of HSV-2 shedding in cohorts of HIV negative (orange), and HIV positive men with CD4+ T-cells >500/μL (magenta), 200-499/μL (green) and <200/μL (teal). (A) Quantitative shedding per swab; (B) Shedding episode classification scheme (adapted from reference 4; (C) Annualized episode rate; (D) Duration, (E) First positive viral load, (F) Peak positive viral load & (G) Last positive viral load per episode.
Table 2.
Rate outcomes were compared by arm between persons using generalized estimating equations with a log link. Continuous outcomes were compared using generalized estimating models with an identity link. Significant differences are noted in bold and are limited to differences in episode rate and episode duration. All other episode characteristics are equivalent across cohorts.
Table 3.
Parameter ranges used for model fitting.
Varied parameters were selected from uniform distributions. Fixed parameters were derived from prior model fitting or literature search (Methods).
Fig 2.
Mathematical model fit to quantitative shedding rate and episode rate.
Ten mathematical model simulations of HSV-2 shedding (thin colored lines) in reference to empirical shedding data (median marked with black dot and 95% CI with black horizontal bars). Quantitative shedding per swab for (A) HIV negative, and HIV positive men with (B) CD4+ T-cells >500/μL, (C) 200-499/μL and (D) <200/μL.
Fig 3.
Increased HSV-2 shedding rate, episode rate and episode duration in HIV infected men.
The red line indicates genital HSV-2 viral load over time. Model output for 300 days with parameters optimized for men with (A) HIV infection and CD4+ T-cell count<200/μL and (B) no HIV infection. Data from individual study participants with (C) HIV infection and CD4+ T-cell count<200/μL and (D) no HIV infection. Individuals in (C) and (D) performed daily swabs.
Fig 4.
Immunologic determinants of increased HSV-2 shedding in HIV-1 infected men.
Parameter sets represent the top 50 (5%) model fits for each group. Boxplots whiskers are inclusive of the full range of values. The box includes interquartile range and a median line. (A) HSV-2 release rate from ganglia to the genital tract, (B) Genital T-cell decay rate, (C) Genital HSV-2 DNA clearance rate and (D) Number of infected cells needed to stimulate half-maximal genital T-cell expansion. All comparisons are with Mann Whitney rank tests. Of note, genital T-cell levels in the model are intended as a surrogate measure of cytolytic immune potential.
Fig 5.
Individual determinants of increased HSV-2 shedding rate.
Critical model predictors of shedding rate in 10-year simulations with 1000 parameter sets. Curves in all panels indicate best fit with a polynomial model while dots are single simulations. (A) HSV-2 release rate from ganglia to the genital tract (R2 = 0.17), (B) Genital T-cell decay rate (R2 = 0.29), (C) HSV-2 DNA clearance rate (R2 = 0.30), predict shedding rate to various degrees. Genital T-cell levels in the model are intended as a surrogate measure of cytolytic immune potential.
Fig 6.
HSV-2 shedding episode characteristics in HIV-1 negative and positive men with CD4+ T-cells count <200/μL.
500 simulated episodes with parameters optimized for HIV negative (orange) and HIV positive men with CD4+ T-cells <200/μL (teal). Boxplots whiskers are inclusive of 90% of values and dots are simulations outside of this range. The box includes interquartile range and a median line. (A) Genital T-cell density in model region of episode initiation, (B) Peak HSV DNA copy number, (C) Number of infected regions per shedding episode, (D) Episode duration. Comparisons for each of the episode characteristics were not statistically significant with non-parametric Mann Whitney rank tests. However, simulations with CD4+ T-cells <200/μL had higher proportions of episodes involving more than 50 regions (C) and lasting more than 10 and 20 days (D).