Age-related differences in humoral and cellular immune responses after primary immunisation: indications for stratified vaccination schedules

Abstract

Immunosenescence is characterised by reduced B and T cell responses. Evidence shows that booster vaccinations are less effective in elderly people, but data on the efficacy of primary immunisation are sparse. We conducted a monocentric, open label, phase IV trial to compare immune responses to primary vaccinations using the inactivated, adjuvanted Japanese Encephalitis vaccine by 30 elderly people (mean 69, range 61-78 years) and 30 younger people (mean 24, range 18-30 years). Humoral and cellular immune responses were analysed in relation to age and cytomegalovirus (CMV) seropositivity. Vaccine-specific antibody titres were significantly lower in elderly participants and 47% of them were non- or low responders after the two doses of the vaccine neo-antigen. The reduced humoral immune responses in elderly people correlated with reduced cytokine production, such as interferon gamma (IFN-γ) in vitro, as well as higher frequencies of late-differentiated effector and effector memory T cells and T regulatory cells. These cellular changes and lower antibody titres were particularly prominent in CMV-seropositive elderly participants. If primary vaccination before the age of 60 is not possible, elderly patients may require different vaccination strategies to ensure sufficient long-lasting immunity, such as adapted or accelerated schedules and the use of different adjuvants.

Figure 1

Vaccine-specific antibody titres. (a) Vaccine-specific JE virus (JEV-specific) antibody titres measured by neutralisation test (NT) in blood samples drawn on days 0, 35 and 70. Results are expressed as geometric mean titres (GMT) with 95% confidence intervals. (b) Vaccine-specific GMTs according to gender. (c) Proportions of vaccine responders (titre ≥ 1:20) and non/low responders (titre < 1:20) to JE vaccination. Statistical analysis by General Linear Model with log-transformed values. Individual comparisons by linear contrasts. *p < 0.05.

Figure 2

Vaccine-specific cytokine production. (a) IL-2, (b) IFN-γ and (c) IL-10 production of PBMC after 48 hour stimulation in vitro with JE virus antigen and TBE virus antigen (TBEV), respectively, or incubation with just medium. Cytokine levels were measured in supernatants. (d) JE-specific IFN-γ/IL-10 ratio after subtraction of medium values. Statistical analysis by General Linear Model with log-transformed values. Individual comparisons by linear contrasts. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 3

B cell subsets. (a) The percentage of B cells (CD3− CD19+) was determined after staining of PBMC, derived on days 0 and 35, with CD3, CD19, CD27 and IgD and gating on the live lymphocyte population in a SSC/FSC blot. (b) Further evaluation of naive (CD27− IgD+), (c) unswitched (CD27+IgD+) (d) and switched memory B (CD27+IgD−) cells were performed according to the expression of CD27 and IgD on gated B cells. Statistical analysis by General Linear Model with arcsine-transformed percentages. Individual comparisons by linear contrasts. *p < 0.05; **p < 0.01.

Figure 4

Naive and memory CD4+ cell subsets. (a) CD4+ T cells were identified as CD4+CD8− lymphocytes within the live lymphocyte population in a SSC/FSC blot of surface stained PBMC obtained on days 0 and 35. (b) Concomitant staining with CD45RA and CCR7 allowed differentiation of naive CD4+ T cells (CD45RA+CCR7+), (c) central memory (CM) CD4+ T cells (CD45RA− CCR7+), (d) effector memory (EM) CD4+ T cells (CD45RA− CCR7−) and (e) CD4+ T effector memory cells that re-express CD45RA (TEMRA) (CD45RA+ CCR7−) within the CD4+ T cells. Statistical analysis by General Linear Model with log-transformed values. Individual comparisons by linear contrasts. *p < 0.05; **p < 0.01; ***p < 0.001.

Figure 5

T and B cells with regulatory activity. (a) Regulatory T cells (CD25+ Foxp3+) were analysed by gating on the CD4+ live lymphocyte population. Additional staining with CD45RA allowed the sub-differentiation of regulatory T cells into (bI) naive (CD45RA+Foxp3^low), (bII) non-regulatory (CD45RA- Foxp3^low) and (bIII) effector (CD45RA- Foxp3^high) CD3+CD4 + T cells within the live lymphocyte population. (c) Immature transitional B cells with potential regulatory activity were characterized as CD24^highCD38^high gated B cells (CD3-CD19+) within the live lymphocyte population (cII). Sub-differentiation of T and B cells was performed in blood samples obtained on day 70. Statistical analysis by General Linear Model with arcsine-transformed percentages. Individual comparisons by linear contrasts. **p < 0.01; ***p < 0.001.

Figure 6

Antibody titre and distribution of different T cell subsets in CMV seropositive and seronegative participants. (a) CMV-specific antibody titres were measured by ELISA in sera obtained on day 0. Results of RU >16 were considered positive and individuals categorised as seropositive. (b) JEV-specific neutralising antibody titres (day 35), (c) CD4+ naive, (d) TEMRA, (e) late-differentiated effector cells and (f) regulatory T cells were analysed according to CMV serology results (all cell subsets were measured in samples obtained on day 0, for gating strategy see Figs 3 and 4). Statistical analysis by General Linear Model with arcsine-transformed percentages. Individual comparisons by linear contrasts. *p < 0.05; **p < 0.01; ***p < 0.001.