CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

further adjustment for KT ratio rendered mortality associations with all other biomarkers non-significant. The association between KT ratio and mortality persisted despite adjustment for all other biomarkers, with the possible exception of IL-6 at month 6 (HR 1.75, 95% CI 1.00, 3.01).

Conclusions: The immunologic pathways predicting mortality among HIV-infected Ugandans may be different than those described in resource-rich settings. IDO and T cell activation may have a greater impact during early ART in resource-limited settings where opportunistic infections remain an important cause of death. These findings also suggest that monocyte (sCD14) and T cell activation may predict mortality via the same causal pathways as IDO, while IL-6 may predict mortality via overlapping but potentially distinct pathways. 318 Immune Activation Impairs Yellow Fever Vaccine Efficacy in HIV-Infected Patients Vivian I. Avelino-Silva 1 ; KarinaT. Miyaji 1 ; Marisol Simoes 2 ; Marcos Freire 2 ; Ana Sartori 1 ; PeterW. Hunt 3 ; Karine Milani 1 ; Augusto Mathias 1 ; Ana Paula Batista 1 ; Esper G. Kallas 1 1 University of Sao Paulo Medical School, Sao Paulo, Brazil; 2 Fundaçao Oswaldo Cruz, Rio de Janeiro, Brazil; 3 University of California San Francisco, San Francisco, CA, US; 4 University of Sao Paulo Medical School, Sao Paulo, Brazil Background: While highly immunogenic in healthy vaccinees, Yellow Fever vaccine (YFV) induces weaker and less durable immune responses in HIV-infected individuals. Chronic immune activation may impair cellular and humoral responses to YFV in HIV-uninfected individuals, but little is known about the immunologic predictors of YFV response in HIV infection. We enrolled HIV-infected and –uninfected adults who received YFV to evaluate predictors of immunity, as assessed by specific neutralizing antibody (NAb) titers. Methods: 18-65 year-old volunteers who received a single YFV dose despite time since vaccination were eligible. All HIV-infected participants were ART-suppressed. Participants with diabetes, chronic liver, renal, or rheumatologic diseases, or prior cancer (except Kaposi Sarcoma) or immunosuppressive therapy were excluded. Results: Among 34 HIV-infected and 58 HIV-uninfected participants, median age was 46 and 38 years old, respectively, and more HIV-infected volunteers were men (79% vs. 29%). Median time since YFV was somewhat shorter in HIV-infected (42 months) than in HIV-uninfected participants (69 months). Few participants in either group recalled adverse events after YFV. Compared to HIV-uninfected controls, HIV-infected participants had lower median CD4 + T cell counts (790 vs.1120/mm 3 , p<0.0001) and CD4 + /CD8 + ratios (0.69 vs. 1.83, p<0.0001). Mean log 10 NAb titers were also significantly lower in HIV-infected than HIV-uninfected participants (3.3 vs. 3.6, p=0.044), a difference that remained significant after adjustment for age, gender, and time since vaccination (p=0.024). In HIV-infected participants, lower NAb titers were associated with longer time since YFV (rho: -0.38, p=0.027) and lower CD4 + /CD8 + ratio (rho: 0.42, P=0.014), but not CD4 + T cell count (p=0.519). None of these factors were associated with NAb titers in HIV-uninfected participant.

Poster Abstracts

Figure 1: Relationship between CD4/CD8 Ratio and YF Neutralizing Ab Titer in HIV-infected and –uninfected Participants. Conclusions: Treated HIV-infected individuals appear to have impaired and/or less durable neutralizing antibody responses to YFV than HIV-uninfected individuals, which were associated with lower CD4 + /CD8 + ratio. These results suggest that a low CD4 + /CD8 + ratio may be a better surrogate for functional immune defects than the CD4 + T cell count in treated HIV infection. Our findings are consistent with an emerging literature suggesting that CD4 + /CD8 + ratio is a stronger correlate of chronic immune activation and predictor of mortality in this setting.

256

CROI 2015