We evaluated pertussis-specific cell-mediated immunity (CMI) and humoral immunity in adolescents 3 years after they received an acellular pertussis booster immunization. Two hundred sixty-four adolescents were examined for immunoglobulin G antibodies, and 49 were examined for CMI against Bordetella pertussis antigens 40 months after receiving the booster. A control group of similarly aged adolescents who had received diphtheria and tetanus vaccination 3 years earlier was included for comparison. Pertussis-specific CMI persisted at greater than prebooster immunization levels. Although they had decreased by the 3-year follow-up, antibody levels remained significantly higher than prebooster immunization levels. Antibodies against pertussis antigens and CMI against filamentous hemagglutinin and pertactin were significantly higher in vaccinated adolescents than in control subjects. The acellular pertussis booster immunization provides long-term CMI and humoral immunity lasting for ⩾3 years. The significantly higher immunity observed in the diphtheria, tetanus, and acellular pertussis vaccine recipients, compared with that in control subjects, indicates that these responses are more likely to have resulted from the booster immunization than from the boosting effects of natural B. pertussis infection.
Pertussis has been traditionally considered a childhood illness. During the 1990s, there was a change in the age distribution of patients with pertussis, indicating an increasing incidence among adolescents and adults [1–3]. There are several possible explanations for this increase, such as waning of natural and vaccine-induced immunity [4, 5], improved clinical and laboratory recognition of pertussis in older individuals, and genetic changes in Bordetella pertussis bacteria [6]. Infected adolescents and adults may introduce pertussis to nonimmunized or partially immunized susceptible infants [7, 8], who are at greatest risk of developing severe disease and its complications [1, 9]. Therefore, better control of pertussis infection is highly necessary. The availability of safer and less-reactogenic acellular pertussis vaccines makes adolescent and adult booster immunization programs possible, and this may be an effective approach to prolonging vaccine-induced immunity.
Acellular pertussis vaccines are currently licensed for adolescent use in several countries [10, 11]. Because postimmunization antibody levels induced by acellular booster vaccination are well in excess of those measured in the infant studies [12, 13] that established vaccine efficacy after primary immunization, the clinical efficacy of acellular booster immunization is expected to be at least similar to that observed after primary immunization [13–15]. Because protection against pertussis is most likely multifactorial [16] and is believed to be mediated by both humoral immunity [17–19] and cell-mediated immunity (CMI) [20, 21], it is important to study both arms of immune responses when evaluating long-term immunity after vaccination.
The majority of studies of the mechanisms of vaccine-induced immunity have focused on assessing immune responses shortly after vaccination. These studies, which were based on investigations of primary immunizations in infants and young children, demonstrate a relatively rapid decrease in levels of pertussis antibodies to low levels just 1–2 years after immunization [22, 23]. In a large Italian study, almost one-half of the children primed with acellular pertussis vaccine had undetectable levels of pertussis toxin (PT) IgG at the age of 5–6 years, although protection appeared to occur for a longer duration [24]. In contrast, CMI seems to persist much longer than do antibody levels in children who have received primary immunization with acellular vaccine. Studies involving young children indicate that CMI remains 1–2 years after primary immunization, and even an increase in CMI has been observed in prebooster blood samples [22, 25]. A recent long-term follow-up study from Italy, however, indicates that only a minority of children have positive CMI responses to all of the vaccine components 5 years after the primary immunization [26].
Earlier, we showed that booster immunization of adolescents with an acellular vaccine containing reduced quantities of pertussis antigens, in addition to diphtheria and tetanus toxoids, induced good responses in both cellular and humoral arms of the immune system [27]. In this 3-year follow-up study, we report the persistence of immunity after the booster immunization. This work represents the first long-term study of CMI after adolescent acellular pertussis booster immunization.
Subjects and study design. The present study was conducted among children who had been previously recruited into an acellular pertussis booster immunization study that was performed in 1997 in southwestern Finland in the city of Turku (population, 174,000) [27]. At that time, a total of 510 healthy children (age, 10–13 years) were enrolled. A total of 450 of them received SB Biological's combined diphtheria, tetanus, and tricomponent acellular pertussis (dTpa) vaccine (Boostrix; GlaxoSmithKline [GSK]), and 60 children received Lederle's Td (tetanus and diptheria) vaccine and, 1 month later, SB Biological's acellular pertussis (pa) vaccine. The quantities of the 3 pertussis antigens (PT, 8 µg; filamentous hemagglutinin [FHA], 8 µg; and pertactin [PRN], 2.5 µg) in the dTpa and pa vaccines were one-third of those of the licensed Infanrix vaccine (GSK). Three years later, 485 of the 509 children who completed the previous study were still residing in the study area and were eligible for the follow-up. They were contacted by letter. Of the 447 subjects who responded, 305 agreed to be reevaluated. Serum samples were obtained from 294 of these 305 subjects. No other pertussis vaccines were given during the follow-up period. For each subject, the follow-up consisted of 1 study visit at which a physical examination was performed and a blood sample was taken. The study protocol was approved by the joint commission on ethics of the Turku University and the Turku University Central Hospital and conducted according to the Declaration of Helsinki and Good Clinical Practice. Written informed consent was obtained from each subject's parent before entry into the study.
Study population. Serological follow-up data were obtained for 294 adolescents (264 in the dTpa group and 30 in the dT-pa group), representing 58% of the vaccine recipients in the initial trial. Because the results of the antibody responses to pertussis antigens were similar between recipients of dTpa and dT-pa (data not shown), only the results for the dTpa group are reported. All subjects were white, and the overall ratio of male subjects to female subjects was 122 : 142. At the time of the follow-up sampling, the mean age of subjects was 14.6 years (range, 14.0–15.9 years). The mean time interval after the booster vaccination was 40 months (range, 39–41 months). CMI to PT, FHA, and PRN was assessed in the original study in every fourth subject [27]. Fifty-three of these subjects (29 male and 24 female subjects), representing 48% of those analyzed in the first phase, participated in the 3-year follow-up evaluation of CMI. The laboratory criteria for acceptable proliferation assay were fulfilled in 49 samples.
A control group from the same geographic region was included for comparison between adolescents boosted with dTpa and those boosted with a commercial diphtheria tetanus (dT) vaccine currently recommended as a part of routine immunizations. Two hundred fifty-five age- and sex-matched adolescents were contacted by letter; 121 (47%) of them answered, and, finally, 42 agreed to be evaluated. The results for 38 of the subjects (19 male and 19 female subjects) were eligible, according to our laboratory criteria. The age of the control subjects ranged from 14.4 to 16.3 years.
Laboratory assays. CMI was evaluated at the National Public Health Institute in Turku, as described elsewhere [27, 28]. The assay has been optimized previously [29]. In brief, fresh blood samples were obtained and processed within 3 h. Triplicate cultures (105 cells/well) of a suspension of PBMCs were incubated with 1 µg of heat-inactivated PT, 1 µg of FHA, or 2.5 µg of PRN (kindly provided by GSK). RMPI 1640 and pokeweed mitogen (PWM) were used as controls. After a 5-day incubation at 37°C in an atmosphere with 5% CO2, 3H-thymidine (0.5 µCi/well) was added for the last 16 h of incubation. The cells were harvested, and incorporated radioactivity was measured with a scintillation counter (Wallac Oy). The results were first expressed as mean counts per minute (cpm) for triplicate cultures. As a criterion for cell reactivity, the proliferative response to PWM had to exceed 1000 cpm. A stimulation index (SI) was used for final expression of the CMI results. SI was defined as specific antigen-induced proliferation divided by the spontaneous proliferation of the cells without antigen stimulation. A CMI response was considered to be positive when the antigen-induced proliferation was at least 4-fold higher than the spontaneous proliferation (SI, ⩾4) [27, 28].
IgG antibodies to pertussis antigens PT, FHA, and PRN and specific antibodies to diphtheria and tetanus toxoid were measured by ELISA techniques in the laboratory at GSK Biologicals (Rixensart, Belgium). The cutoff value for the test for pertussis antigens was 5 EIU/mL, and the cutoff value for diphtheria and tetanus toxoids was 0.1 IU/L. Serum samples with ELISA antidiphtheria antibody titers of <0.1 IU/L were retested using a Vero cell neutralization assay. The cutoff value for the Vero cell assay was 0.016 IU/mL.
Statistical analysis. Statistical analyses were performed using SPSS for Windows (SPSS). Antibody concentrations that were less than the lower limit of detection for an assay were arbitrarily assigned a value of one-half of the assay's cutoff value. Calculations of geometric mean values of antibodies and proliferations were performed on log10-transformed data, reposting the antilogarithm. Comparisons between geometric means were done using the Wilcoxon rank sum test. Analysis of correlations was performed using the Spearman rank correlation coefficient. Comparisons manifesting a 2-tailed P value of <.05 were considered to be statistically significant.
CMI. The results of the pertussis antigen—induced proliferations of PBMCs are presented in figure 1. Three years after booster immunization, the CMI responses had decreased from post—booster vaccination levels, but they were still significantly higher than the prebooster levels when measured with FHA and PRN antigens (P < .0001 for both). In most adolescents, the PT responses had decreased back to the same level as before booster immunization. The CMI responses to FHA and PRN were also significantly higher in dTpa-boosted adolescents than in the control group (P = .001 and P < .001, respectively), whereas no significant difference was observed in CMI to PT (figure 1). Table 1 shows the percentages of dTpa-vaccinated adolescents and control subjects with a positive CMI response. A positive CMI response to ⩾1 pertussis vaccine antigen was noted in 92% of the dTpa vaccine recipients, compared with 82% of subjects in the control group. The proportions of CMI-positive responses to all vaccine antigens were 37% in the dTpa-vaccinated group and 11% in the control group (P = .006). Three years after immunization, the CMI positivity rates for PT, FHA, and PRN were 57%, 88%, and 61%, respectively (table 1). Only 4 (8%) of 49 subjects had negative stimulation indices against all 3 antigens at the follow-up visits. Two subjects also had undetectable IgG antibody levels to PT.
Cell-mediated immunity responses, as indicated by stimulation indices in recipients of diphtheria, tetanus toxoids, acellular pertussis vaccine (dTpa) vaccine before dTpa vaccination, 1 month and 3 years after receipt of the dTpa booster vaccination, and, in control subjects, 3 years after receipt of diphtheria tetanus booster. Significant stimulation is defined as a stimulation index of ⩾4. Horizontal lines, mean value of the individual stimulation indices. FHA, filamentous hemagglutinin; PRN, pertactin; PT, pertussis toxin.
Three years after receiving booster immunization, a significant positive correlation between humoral immunity (IgG) and corresponding cellular response was demonstrated for FHA (r = 0.338; P < .05) and PRN (r = 0.412; P < .01), but not for PT (r = 0.271; P = .052). A positive correlation was observed between the 1-month postbooster and 3-year follow-up proliferative results after stimulation with PT (r = 0.282; P = .05), FHA (r = 0.545; P < .0001), and PRN (r = 0.438; P = .002), as assessed by SI.
At the follow-up visit, 20 (41%) of the 49 dTpa recipients yielded higher present CMI responses to ⩾1 pertussis antigen, compared with CMI responses determined 1 month after the booster immunization. None of these subjects had had a disease diagnosed as pertussis. Five subjects (10%) had an increase in CMI responses to all vaccine antigens; none of them, however, had an increase in any pertussis antibody levels or a history of confirmed pertussis during the follow-up period.
Humoral immunity. Figure 2 demonstrates the reverse cumulative distribution curves of present concentrations of IgG antibodies to PT, FHA, and PRN, in comparison with the reverse cumulative distribution curves before and 1 month after receipt of the booster vaccination, as well as in comparison with corresponding concentrations in control subjects. Geometric mean values of antibodies to pertussis antigens had decreased in 3 years from those measured 1 month after the booster immunization, but they were still significantly higher than the values in prebooster samples (table 2). Compared with prebooster levels, ⩾2-fold higher IgG antibody titers to PT, FHA, and PRN were observed in 36%, 57%, and 84% of participants, respectively. Antibody levels measured 1 month after booster vaccination correlated well with the corresponding follow-up levels: PT IgG (r = 0.768; P < .01), FHA IgG (r = 0.686; P < .01), and PRN IgG (r = 0.834; P < .01).
Reverse cumulative distribution curves of IgG antibodies against pertussis toxin (PT; top), filamentous hemagglutinin (FHA; center), and pertactin (PRN; bottom) among recipients of diphtheria, tetanus toxoids, acellular pertussis vaccine before the booster immunization, 1 month and 3 years after the booster immunization, and among controls. MLD, minimum level of detection (5 EIU/mL).
Rates of positive cell-mediated immunity (CMI) responses (i.e., stimulation index of ⩾4) in vaccine recipients before, 1 month after, and 3 years after diphtheria, tetanus toxoids, acellular pertussis booster vaccination, compared with a control group of similarly aged adolescents who had received the diphtheria tetanus booster (without the pertussis component) 3 years earlier.
The rates of detectable antibodies (DR; i.e., ⩾5 EIU/mL) and geometric mean values (GMVs) of IgG antibodies to pertussis antigens before receipt of booster immunization and during the follow-up, compared with values for the control group of similarly aged adolescents immunized with commercial diphtheria tetanus vaccine 3 years earlier.
Three years after booster immunization, all subjects had detectable FHA antibodies, and the rates for detectable PT IgG and PRN IgG were 82% and 99%, respectively (table 2). Detectable IgG antibody levels to PT, FHA, and PRN were measured in 56%, 94%, and 77% of the control subjects, respectively. Compared with control subjects, vaccine recipients had significantly higher levels of IgG antibodies to all 3 pertussis antigens, whereas the results for the control group were comparable to the prebooster immunization levels of the dTpa recipients (table 2 and figure 2).
At the time of follow-up, IgG antibodies to PT, FHA, and PRN were greater than the levels obtained 1 month after booster immunization in 3.1%, 4.2%, and 3.8% of the adolescents, respectively. Ten subjects (3.8%) had an increase in the level of antibodies to 2 or 3 pertussis antigens, and only 2 subjects had an increase in the level of antibodies to all 3 pertussis antigens. None of these subjects had a history of clinical pertussis disease. CMI was measured in only 1 of these 10 subjects, and there was no increase in follow-up CMI to any of the 3 pertussis antigens, compared with post—booster immunization values.
Three years after the dTpa booster immunization, antidiphtheria and antitetanus antibody concentrations had decreased from the 1-month postbooster levels, but they were still significantly higher than the prebooster levels. Ninety-two percent and 100% of subjects continued to have seroprotective concentrations of antidiphtheria (⩾0.1 IU/mL) and antitetanus (⩾0.1 IU/mL) antibodies, respectively. The blood samples for the subjects with an antidiphtheria antibody concentration level of <0.1 IU/mL (determined by ELISA) were retested with the in vitro Vero cell neutralization assay, and all of these subjects were shown to have seroprotective antidiphtheria antibody concentrations by that method.
The present study evaluated the long-term persistence of pertussis-specific cellular and humoral immunity after receipt of an adolescent acellular pertussis booster immunization. To the best of our knowledge, this is the first long-term follow-up study of cellular immunity in a vaccinated adolescent population. In the present study, we found that, 3 years after receiving a booster dose of vaccine with reduced quantities of pertussis antigens, 82%–100% of subjects had detectable levels of pertussis antibodies, and 92% had a positive CMI response to ⩾1 of the 3 pertussis antigens used in the vaccine. Because of a limited number of vaccinated adults, few previous studies have focused on the duration of CMI responses after adult booster vaccination. A follow-up study by Di Tommaso et al. [30], which involved 8 healthy adults who were vaccinated with an acellular vaccine containing detoxified pertussis toxin, showed that T cell responses persist up to 18–24 months after the vaccination. Similarly, we have previously shown that CMI responses can persist even up to 8 years after acellular pertussis booster immunization of adults who have been primed with whole-cell vaccine [31]. However, these adults were vaccinated with a booster vaccine with antigen content similar to that in vaccines used in the infant primary immunizations, whereas the quantities of pertussis antigens in the present booster vaccine were one-third of the quantity in the primary vaccine.
Similar to the findings after primary immunization in children, there are significant decreases in levels of antibodies to vaccine antigens 1 year following adult booster immunization, despite the fact that booster vaccines are capable of inducing an initially strong serological response [32–34]. However, the antibody levels obtained 1 year after immunization have remained higher than the prevaccination levels. The results of the present study indicate that, 3 years after booster immunization, both the antibody levels and the rate of positivity to all 3 pertussis antigens remain higher in dTpa recipients than in control subjects. The results further indicate that both humoral immunity and CMI responses against FHA and PRN persist longer than the PT responses. This might be one explanation for the multifactorial nature of pertussis immunity. Waning of immunity might be an antigen-dependent phenomenon, and, in the late phase after immunization, FHA and PRN might be more important antigens than PT.
Although PT IgG antibody levels were significantly higher in dTpa recipients than in control subjects, the PT CMI positivity rate in the vaccinated group was similar to that in the control subjects. In contrast to our study of adolescents, studies involving infants indicate that CMI appears to persist much longer than do detectable antibody levels [22, 25]. In our study, the PT CMI response rate was 45% before receipt of the booster dose, and it increased to 100% one month after the immunization and was again reduced to 57% three years after the immunization. The reason for this different behavior of infants and adolescents is not known. The adolescents had received 4 doses of whole-cell pertussis vaccine as a part of their routine childhood immunizations in Finland. This vaccine was produced in the National Public Health Institute (Helsinki, Finland) and is known to induce relatively weak PT antibody responses. The priming effect might thus be different than with infants primed with an acellular vaccine.
The reduction of the antibody levels and CMI after vaccination and after pertussis disease seems to be comparable. In their study, Esposito et al. [26] showed that, 5 years after acellular pertussis vaccination at 3.5 and 11 months of life, only a small proportion of children have significant concentrations of IgG to all B. pertussis antigens, and cellular immunity seems to persist only in a minority of subjects. However, although low, these residual immune responses were comparable with those observed 5 years after natural infection.
Ausiello et al. [35] have demonstrated an increase in CMI in children between the end of the primary vaccination schedule and before the booster dose. The authors suggested that the immunological basis for the increase in CMI was due to natural boosting with subclinical infections. At follow-up, we found that 41% of the dTpa recipients yielded higher present CMI responses to ⩾1 pertussis antigen, compared with those determined 1 month after immunization. Five subjects had an increase in CMI responses to all vaccine antigens; none of them, however, had an increase in any pertussis antibody levels or a history of clinical pertussis during follow-up. It is impossible to know whether this reflects a response to natural exposure to circulating B. pertussis, causing a silent booster effect. Therefore, the comparison with the control group is valuable. Table 1 shows that, at the age of 14 years, control subjects had positive CMI responses to PT and FHA more frequently than did adolescent vaccine recipients at the age of 11 years, before receiving booster immunization. This trend toward higher CMI positivity with increasing age does, however, not happen with PRN. These results are in line with our previous results [29]: we found that the currently circulating B. pertussis strains (most of which have PRN type 2 or 3) do not often induce PRN CMI when measured with PRN type 1 as an antigen in the proliferation assay [29]. These results might also indirectly imply that there has been some silent boosting with the circulating variants among the control subjects.
Recent epidemiological studies have shown a shift toward increasing prevalence of pertussis, especially in the adolescent population [3]. Although no conclusions regarding the degree of protection against pertussis can be drawn from the current observations, the results of our study show that, by providing long-term immunity, dTpa vaccination offers valuable means of achieving prolonged immunity in an adolescent population. Furthermore, at the time of the follow-up, all dTpa recipients showed protective levels against diphtheria and tetanus. Thus, replacing dT vaccine—currently recommended as a booster in this age group—with dTpa vaccine seems not to weaken the persistence of the immunogenicity profile of diphtheria and tetanus toxoids. Long-term follow-up studies are expected to provide information concerning the ideal timing of booster doses. Additional data are required on further longevity of booster-induced pertussis immunity before any recommendations on time interval between subsequent doses can be made.
In conclusion, our results show that the acellular pertussis booster vaccination provides long-term cellular and humoral immunity 40 months after vaccination. The significantly higher pertussis immunity observed in the dTpa-vaccinated group, compared with the control subjects, indicates that these responses are more likely results from the dTpa booster vaccination than from boosting effects of B. pertussis circulation on the population.
We are grateful to Birgitta Aittanen, Tuula Lehtonen, and Marianne Vahanne for their technical assistance. We also thank the participating study physician, Olli Honkinen, and the study nurse, Raija Lietzén, for their contributions.
Financial support. Turku University Foundation, the Academy of Finland, the Finnish Cultural Foundation, the Turku University Hospital (EVO), and GlaxoSmithKline Biologicals. GlaxoSmithKline Biologicals provided the vaccine and other materials necessary to conduct this study.
Conflict of interest. L.S. and J.W. are employees of GlaxoSmithKline Biologicals. All other authors: No conflict.
IDSA Members: For your free access to this journal, log in via the IDSA members area.
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy