Table 1.
Baseline characteristics of participants.
Fig 1.
Total pre-HIV IL-17-producing CD4+ T cells were correlated with CD4/CD8 ratio in HVTN 503.
PBMCs from study participants were stimulated ex vivo for 4 hours with PMA and ionomycin in the presence of Golgi Plug and Golgi Stop. The frequency of CD4+ T cells producing IL-17 was measured by flow cytometry. (A) Representative flow cytometry gating strategy for total IL-17-producing CD4+ T cells. (B, C) Correlation with mean CD4/CD8 ratio within the first 180 days post-infection and ≥180 days post-infection. (D, E) Correlation with CD4/CD8 ratio at last available measurement within the first 180 days post-infection and ≥180 days post-infection. CD4/CD8 ratios were calculated from absolute CD4 and CD8 counts obtained within the first 180 days (n = 35) or ≥180 days (n = 32) post-infection, as indicated. Measurements obtained after ART initiation or beyond 1-year post-infection were excluded. CD4/CD8 ratio at last available measurement refers to the last CD4/CD8 measurement within the first year of infection prior to ART initiation, if applicable. Correlations were assessed using Spearman rank correlation coefficient (rs), with linear regression lines shown for visualization purposes only. Two-tailed p-values are shown; statistical significance was defined as p < 0.05.
Fig 2.
Correlation between pre-HIV IL-17-producing CD4+ T cell subsets and mean CD4/CD8 ratio <180 days post-infection in HVTN 503.
PBMCs from study participants were stimulated ex vivo for 4 hours with PMA and ionomycin in the presence of Golgi Plug and Golgi Stop. The frequency of CD4+ T cells producing IL-17 was measured by flow cytometry. (A) IL-17+ TNF-α+. (B) IL-17+ TNF-α-. (C) IL-17+ IFN-γ+. (D) IL-17+ IFN-γ-. (E) IL-17+ GM-CSF+. (F) IL-17+ GM-CSF-. (G) IL-17+ IL-22+. (H) IL-17+ IL-22-. CD4/CD8 ratios were calculated from absolute CD4 and CD8 counts obtained within the first 180 days (n = 35) post-infection. Measurements obtained after ART initiation or beyond 1-year post-infection were excluded. Correlations were assessed using Spearman rank correlation coefficient (rs), with linear regression lines shown for visualization purposes only. Two-tailed p-values are shown; statistical significance was defined as p < 0.05.
Fig 3.
Correlation between pre-HIV IL-17-producing CD4+ T cell subsets and mean CD4/CD8 ratio ≥ 180 days post-infection in HVTN 503.
PBMCs from study participants were stimulated ex vivo for 4 hours with PMA and ionomycin in the presence of Golgi Plug and Golgi Stop. The frequency of CD4+ T cells producing IL-17 was measured by flow cytometry. (A) IL-17+ TNF-α+. (B) IL-17+ TNF-α-. (C) IL-17+ IFN-γ+. (D) IL-17+ IFN-γ-. (E) IL-17+ GM-CSF+. (F) IL-17+ GM-CSF-. (G) IL-17+ IL-22+. (H) IL-17+ IL-22-. CD4/CD8 ratios were calculated from absolute CD4 and CD8 counts obtained beyond the first 180 days (n = 32) post-infection. Measurements obtained after ART initiation or beyond 1-year post-infection were excluded. Correlations were assessed using Spearman rank correlation coefficient (rs), with linear regression lines shown for visualization purposes only. Two-tailed p-values are shown; statistical significance was defined as p < 0.05.
Fig 4.
Total pre-HIV IL-17-producing CD4+ T cells were associated with faster CD4+ T cell decline in HVTN 503 but not PP/COS.
PBMCs from study participants were stimulated ex vivo for 4 hours with PMA and ionomycin in the presence of Golgi Plug and Golgi Stop. The frequency of CD4+ T cells producing IL-17 was measured by flow cytometry. (A) Association in HVTN 503 (n = 32). (B) Association in PP/COS (n = 27). (C) Association in the combined cohort (n = 59). CD4 counts measured after ART initiation, after 1-year post-infection, or within the first 180 days post-infection were excluded. Hazard ratios (HR) were estimated using unadjusted Cox proportional hazards models. Multivariable models (adjusted HR[aHR]) were adjusted for sex, adenovirus type 5 (Ad5) titer, herpes simplex virus-2 (HSV-2) serostatus, age, and peak viral load for HVTN 503. For PP/COS and combined cohorts, models were adjusted for age, sex, HSV-2 serostatus, and peak viral load. Two-tailed p-values are shown; statistical significance was defined as p < 0.05.
Fig 5.
Stratified analyses of pre-HIV IL-17-producing CD4+ T cells and markers of HIV disease progression.
PBMCs from study participants were stimulated ex vivo for 4 hours with PMA and ionomycin in the presence of Golgi Plug and Golgi Stop. The frequency of CD4+ T cells producing IL-17 was measured by flow cytometry. (A) Forest plot showing associations between total pre-HIV IL-17-producing CD4+ T cells and CD4 decline below 500 cells/mm3, stratified by sex, age, and treatment arm across cohorts. (B, C) Sex-stratified correlations between total pre-HIV IL-17-producing CD4+ T cell frequencies and mean CD4/CD8 ratio in HVTN 503. (D, E) Treatment arm-stratified correlations between total IL-17-producing CD4+ T cell frequencies and mean CD4/CD8 ratio in HVTN 503. (F) Comparison of total pre-HIV IL-17-producing CD4+ T cell frequencies stratified by sex in HVTN 503. (G) Comparison of total pre-HIV IL-17-producing CD4+ T cell frequencies stratified by age group in the combined cohort. (H) Comparison of total pre-HIV IL-17-producing CD4+ T cell frequencies stratified by treatment arm in HVTN 503. Hazard ratios were estimated using unadjusted Cox proportional hazards models. Multivariable models (Adj) were adjusted for sex, adenovirus type 5 (Ad5) titer, herpes simplex virus-2 (HSV-2) serostatus, age, and peak viral load for HVTN 503. For PP/COS and combined cohorts, models were adjusted for age, sex, HSV-2 serostatus, and peak viral load. CD4/CD8 ratios were calculated from absolute CD4 and CD8 counts obtained within the first 180 days or ≥180 days post-infection, as indicated. Measurements obtained after ART initiation or beyond 1-year post-infection were excluded. Correlations were assessed using Spearman rank correlation coefficient (rs). Group comparisons were performed using Mann-Whitney test. Two-tailed p-values are shown; statistical significance was defined as p < 0.05.
Fig 6.
Pre-HIV IL-17-producing CD4+ T cells are not associated with viral load.
(A, B) Correlation between pre-HIV IL-17-prooducing CD4+ T cells and peak (n = 35) and set-point (n = 32) viral load in HVTN 503. (C, D) Correlation between pre-HIV IL-17-prooducing CD4+ T cells and peak (n = 32) and set-point (n = 18) viral loads in PP/COS. Peak viral load was defined as the highest viral load measured within the first 180 days post-infection, while set-point viral load was defined as the mean viral load after 180 days post-infection. Viral load measurements obtained after ART initiation or beyond 1-year post-infection were excluded. Correlations were assessed using Spearman rank correlation coefficient (rs). Two-tailed p-values are shown; statistical significance was defined as p < 0.05.