Impaired immunity to recall antigens and neoantigens in severely immunocompromised children and adolescents during the first year of effective highly active antiretroviral therapy

Abstract

Background: We studied whether severely immunocompromised, human immunodeficiency virus (HIV)-infected children who were beginning highly active antiretroviral therapy (HAART) or changing HAART regimens could spontaneously respond to a recall antigen (tetanus toxoid [TT] vaccine) or respond to a recall antigen and neoantigen (hepatitis A virus [HAV] vaccine) after 3 vaccinations.

Methods: A total of 46 children who had CD4 cell percentages <15% and who demonstrated a >0.75-log reduction in plasma HIV RNA levels within 4 weeks of starting HAART were randomized to receive vaccinations with either TT or HAV vaccines during the first 6 months of HAART. Study subjects then received the alternate vaccine during the next 6 months of HAART.

Results: Despite the early decline in viremia and the later increase in the percentage of CD4 T cells, spontaneous recovery of cell-mediated immunity (CMI) was not seen for TT. Serologic responses to TT required 3 vaccinations and were comparable in both groups. Serologic responses to HAV were infrequent and of low titer, although the group that received HAV vaccine after receiving TT vaccine performed somewhat better. CMI to HAV was virtually absent.

Conclusions: Severely immunocompromised children who are receiving HAART develop CMI and antibody to a recall antigen independent of the timing of vaccination, but they require a primary series of vaccinations. Antibodies to a neoantigen, HAV, developed when vaccination was delayed after initiation of HAART. CMI to a neoantigen was difficult to establish.

Trial registration: Clinicaltrials.gov identifier: NCT00004735/PACTG P1006 .

Figure 1

CD4 cell population and viral load. Median and interquartile range (IQR) for CD4 cell counts (A), CD4 cell percentages (B), and log₁₀ HIV-1 RNA levels (C). V1–V6 indicate the time points when the 6 vaccinations were administered. The number of subjects who had data for each week is indicated immediately below the IQR bar.

Figure 2

Lymphoproliferative response to tetanus toxoid (TT). Group 1 subjects received a series of tetanus vaccinations followed by hepatitis A vaccinations; group 2 subjects received hepatitis A vaccinations followed by tetanus vaccinations. V1–V6 indicate the time points when the 6 vaccinations were administered. Data are medians and interquartile ranges for stimulation index (SI) values to TT for the 2 treatment groups, represented on the log₁₀ scale (A), and percentage of individuals with an SI≥3 (all CMI responders to TT) (B). The 2 groups were compared as described in the text, using the Wilcoxon rank sum test for the SI values and Fisher’s exact test for the percentage of CMI responders. *P < .05. The pairs of numbers at the bottom of each panel indicate the number of subjects who had data for each week.

Figure 3

Serologic response to tetanus toxoid (TT) vaccine. Group 1 subjects received a series of tetanus vaccinations followed by hepatitis A vaccinations; group 2 subjects received hepatitis A vaccinations followed by tetanus vaccinations. V1–V6 indicate the time points when the 6 vaccinations were administered. Data are median and interquartile range (IQR) TT titers for the 2 treatment groups (A) and the percentage of individuals with a protective level of antibody (positive titer, ≥ 0.1 IU/mL of anti-TT) (B). The 2 groups were compared as described in the text, by using the Wilcoxon rank sum test for median (IQR) TT titers and Fisher’s exact test for the percentage of individuals with a positive titer. *P < 0.05. The pairs of numbers at the bottom of each panel indicate the number of subjects who had data for each week.

Figure 4

Serologic Response to hepatitis A virus vaccine. Group 1 subjects received a series of tetanus vaccinations followed by hepatitis A vaccinations; group 2 subjects received hepatitis A vaccinations followed by tetanus vaccinations. V1–V6 indicate the time points when the 6 vaccinations were administered. Data are median and interquartile range (IQR) for hepatitis A titers for the 2 treatment groups (A) and the percentage of individuals with a positive titer of ≥20 mIU/mL (B). The 2 groups were compared as described in the text, by using the Wilcoxon rank sum test for median (IQR) titer values and Fisher’s exact test for the percentage of individuals with a positive titer. *P < .05. The pairs of numbers at the bottom of each panel indicate the number of subjects who had data for each week.