A randomized clinical trial comparing revaccination with pneumococcal conjugate vaccine to polysaccharide vaccine among HIV-infected adults

Abstract

Background: The risk of pneumococcal disease persists, and antibody responses to revaccination with the 23-valent polysaccharide vaccine (PPV) are low among human immunodeficiency virus (HIV)-infected adults. We determined whether revaccination with the 7-valent pneumococcal conjugate vaccine (PCV) would enhance these responses.

Methods: In a randomized clinical trial, we compared the immunogenicity of revaccination with PCV ( n = 131) or PPV (n = 73) among HIV-infected adults (median CD4 cell count, 533 cells/mm(3)) who had been vaccinated with PPV 3-8 years earlier. HIV-uninfected adults (n = 25) without prior pneumococcal vaccination received 1 dose of PCV. A positive response was defined as a >or=2-fold increase (from baseline to day 60) in capsule-specific immunoglobulin G, with a postvaccination level >or=1000 ng/mL for at least 2 of the 4 serotypes.

Results: HIV-infected persons demonstrated a higher frequency of positive antibody responses to PCV than to PPV (57% vs 36%) (P = .004) and greater mean changes in the immunoglobulin G concentration from baseline to day 60 for serotypes 4, 9V, and 19F (P < .05, for all), but not for serotype 14. However, by day 180, both outcomes were similar. Responses to PCV were greater in frequency and magnitude for all serotypes in HIV-uninfected adults, compared with those in HIV-infected adults.

Conclusions: Among persons with HIV infection, revaccination with PCV was only transiently more immunogenic than PPV, and responses were inferior to those in HIV-uninfected subjects with primary vaccination. Pneumococcal vaccines with more robust and sustained immunogenicity are needed for HIV-infected adults. Clinical trial registration. ClinicalTrials.gov identifier NCT00622843.

Conflict of interest statement

Conflict of Interest: None

Figures

Figure 1

Geometric mean concentrations (GMC) in μg/mL with 95% confidence intervals for each of the four serotypes and each arm at baseline and days 14, 60, and 180 post-vaccination.

Figure 2

Results from models estimating the change in IgG concentration (log10 ng/mL) from baseline to days 14, 60 and 180. For each plot, the panel on the left shows the mean changes for Group A (HIV-uninfected PCV arm), Group B (HIV-infected PCV arm) and Group C (HIV-infected PPV arm); while the panel on the right shows the differences (with 95% confidence intervals) between Groups A and B and between Groups B and C for the change from baseline to each post-vaccination visit. All models were adjusted for age, ethnicity, prior pneumonia, and baseline serotype concentration; models comparing Groups B and C were also adjusted for CD4 cell count, HIV RNA level, and use of HAART at baseline.