Immunogenicity of a Whole-Cell Pertussis Vaccine with Low Lipopolysaccharide Content in Infants

The lack of a clear correlation between the levels of antibodyto pertussis antigens and protection against disease lends credenceto the possibility that cell-mediated immunity provides primaryprotection against disease. This phase I comparative trial hadthe aim of comparing the in vitro cellular immune response andanti-pertussis toxin (anti-PT) immunoglobulin G (IgG) titersinduced by a cellular pertussis vaccine with low lipopolysaccharide(LPS) content (wP_low vaccine) with those induced by the conventionalwhole-cell pertussis (wP) vaccine. A total of 234 infants werevaccinated at 2, 4, and 6 months with the conventional wP vaccineor the wP_low vaccine. Proliferation of CD3⁺ T cells was evaluatedby flow cytometry after 6 days of peripheral blood mononuclearcell culture with stimulation with heat-killed Bordetella pertussisor phytohemagglutinin (PHA). CD3⁺, CD4⁺, CD8⁺, and T-cell receptor ${gamma}$ ${delta}$ -positive ( ${gamma}$ ${delta}$ ⁺) cells were identified in the gate of blast lymphocytes.Gamma interferon, tumor necrosis factor alpha, interleukin-4(IL-4), and IL-10 levels in supernatants and serum anti-PT IgGlevels were determined using enzyme-linked immunosorbent assay(ELISA). The net percentage of CD3⁺ blasts in cultures withB. pertussis in the group vaccinated with wP was higher thanthat in the group vaccinated with the wP_low vaccine (mediansof 6.2% for the wP vaccine and 3.9% for the wP_low vaccine; P= 0.029). The frequencies of proliferating CD4⁺, CD8⁺, and ${gamma}$ ${delta}$ ⁺cells, cytokine concentrations in supernatants, and the geometricmean titers of anti-PT IgG were similar for the two vaccinationgroups. There was a significant difference between the T-cellsubpopulations for B. pertussis and PHA cultures, with a higherpercentage of ${gamma}$ ${delta}$ ⁺ cells in the B. pertussis cultures (P < 0.001).The overall data did suggest that wP vaccination resulted inmodestly better specific CD3⁺ cell proliferation, and ${gamma}$ ${delta}$ ⁺ cellexpansions were similar with the two vaccines.

INTRODUCTION

Top

INTRODUCTION

Efforts to immunize against whooping cough began almost as soonas Bordet discovered the causative bacterium, Bordetella pertussis,in 1912 (8). A whole-cell pertussis (wP) vaccine has been availablesince 1940. However, the road to eliminate circulation of thepathogen has proven arduous. Throughout the world, pertussisremains a major cause of morbidity and mortality among infants(45). Some 20 million to 40 million cases of disease occur worldwideeach year, 90% of which are found in developing countries (44).

Although no causal link between wP vaccination and permanentbrain damage or death has been identified, concerns about systemicreactions after immunization with a wP vaccine have been a majorfactor in its reduced acceptance in developed countries (46).This has led to the development of more-purified and less locallyreactogenic acellular pertussis (aP) vaccines (21, 22, 43, 47).Similar immunogenicities are obtained with both types of vaccine,as the most efficacious vaccines of either category protect>80% of the recipients from clinical disease. Moreover, theconsiderably higher development costs of aP result in pricesper dose that are unlikely to be currently affordable for developingcountries (46).

Lipopolysaccharide (LPS) possesses endotoxic activity and alsohas powerful adjuvant activity. Both of these properties arebased upon the recognition of the LPS by the host Toll-likereceptor (TLR) complex TLR4/MD-2 and the subsequent activationof NF- ${kappa}$ B (35). The relatively high reactogenicity of wP vaccineshas been associated with proinflammatory cytokines (2, 28).Hence, a straightforward approach to reducing wP reactogenicitywould be the generation of a pertussis vaccine with a reducedquantity of LPS. Furthermore, the lack of a clear correlationbetween the levels of antibody (Ab) to pertussis antigens andprotection against disease, the persistence of protective immunitylong after the disappearance of pertussis antigen-specific Ab,and the longer duration of T-cell responses to pertussis antigenslend credence to the possibility that cell-mediated immunityprovides primary protection against disease (16, 34).

This phase I study was thus performed to obtain preliminaryimmunogenicity data on a new cellular pertussis vaccine withlow LPS content (wP_low vaccine), in comparison to the conventionalwP vaccine used in Brazil, both of which are formulated withdiphtheria and tetanus toxoids and given in three primary injectionsduring infancy.

MATERIALS AND METHODS

Top

MATERIALS AND METHODS

Subjects. Infants scheduled to receive the first dose of pertussis vaccinewere recruited in Campinas Public Health Centers, SãoPaulo, Brazil. A total of 247 infants were recruited to thestudy and randomized to receive three doses of wP_low or wP vaccineat 2, 4, and 6 months of age. The preterm (gestational age,<37 weeks) and low-birth-weight (weight, <2,500 g) newborns,infants whose mothers were younger than 18 years old or hadhepatitis B carrier status, and infants with a family historyof tuberculosis, syphilis, or human immunodeficiency virus infectionwere excluded. Exclusion criteria also included noncompliance,receipt of immunoglobulin (Ig) therapy or blood products, andany immunodeficiency, malignancy, or significant underlyingdisease or neurological impairment.

Vaccines. Vaccines were manufactured by the Butantan Institute, SãoPaulo, Brazil. Bordetella pertussis cells were treated withan organic solvent and washed in order to perform LPS extraction.The culture was then detoxified by the addition of 0.2% formalin,and the bacterial biomass was obtained by tangential flow filtrationto formulate the wP_low vaccine.

Each 0.5-ml dose of either vaccine contained four protectiveunits of pertussis toxin and two protective units of diphtheriaand tetanus toxoids. The antigens were adsorbed onto 1.25 mgof aluminum hydroxide, and 0.2 mg of thimerosal was used asa preservative. All infants received vaccine from the same batch.The vaccines were visually indistinguishable and identicallypackaged and were administered intramuscularly with the useof standard techniques.

Study design. This prospective, randomized, double-blind comparative trialwas conducted between August 2006 and July 2007 and followedthe principles outlined in the Declaration of Helsinki. Writteninformed consent was obtained from all participants' parentsor legal guardians before study procedures were initiated. Thestudy protocol was approved by the Committee for Ethics in Researchof the State University of Campinas, São Paulo, Brazil.

Antigens used in cell culture assays. A suspension of heat-killed B. pertussis (lot IB-CIIn/P14/06;Butantan Institute), without thimerosal, was used at a concentrationof 5 x 10⁶ cells/ml. Phytohemagglutinin (PHA; Sigma) was usedas a positive control at 7.5 µg/ml.

T-cell proliferation assay. Ten milliliters of heparinized peripheral blood was collectedand used for the evaluation of immune responses to pertussisvaccination at 7 months of age. The protocol for the proliferationassay was adapted from the method of Gaines and Biberfeld (20).Peripheral blood mononuclear cells (PBMC) were isolated by densitygradient centrifugation over Ficoll-Hypaque (Amersham Biosciences),washed, diluted to 1 x 10⁶ cells/ml in RPMI 1640 medium (Sigma)supplemented with 10% human AB serum (Sigma), 1% glutamine (Sigma),and 0.1% gentamicin, and stimulated for 6 days with heat-killedB. pertussis, PHA, or medium alone at 37°C with 5% CO₂ inround-bottom 96-well tissue culture plates (Nunc, Denmark).After being harvested with 20 mM EDTA, samples were incubatedwith human Ig and then stained with anti-human CD3, CD4, CD8,and T-cell receptor (TCR) pan- ${gamma}$ ${delta}$ fluorescent Abs (Beckman Coulter)before acquisition (Epics XL-MCL flow cytometer; Beckman-Coulter)and analysis (Expo software; Beckman Coulter). Isotype controlswere used to discriminate positive populations (Beckman Coulter).Only CD3⁺ T cells were used in analysis. Forward and side scatterswere used to gate resting and blast lymphocytes. Dead cellswere excluded from all analyses. CD4⁺, CD8⁺, and TCR ${gamma}$ ${delta}$ -positive( ${gamma}$ ${delta}$ ⁺) cells were identified in the gating of blast lymphocytes.Proliferation was measured by percentages of CD3⁺ blasts, inwhich basal proliferation was subtracted from B. pertussis-and PHA-stimulated cultures.

Cytokine quantification in supernatants. For the cytokine quantification assay, PBMC were diluted to2 x 10⁶ cells/ml in supplemented RPMI medium and incubated for48 h in round-bottom 96-well tissue culture plates with B. pertussis,PHA, or medium alone at 37°C. Supernatants were collectedand stored at –80°C. Gamma interferon (IFN- ${gamma}$ ), tumornecrosis factor alpha (TNF- ${alpha}$ ), interleukin-4 (IL-4), and IL-10levels were determined in duplicate by a two-monoclonal-Ab sandwichenzyme-linked immunosorbent assay (ELISA) (R & D Systems)in flat-bottom MultiSorp ELISA plates (Nunc, Denmark), accordingto the manufacturer's protocols. Recombinant cytokine was usedfor the standard curve. The limits of detection were 15.6 pg/mlfor IFN- ${gamma}$ and TNF- ${alpha}$ and 31.2 pg/ml for IL-4 and IL-10.

Quantitative determination of anti-PT IgG. Quantitative determination of anti-pertussis toxin (anti-PT)IgG was performed in a blind manner on serum samples collectedone month after the administration of the third dose of vaccineand stored at –80°C until tested. Using ELISA (32),anti-PT IgG levels were calculated in IU/ml by comparison witha U.S. reference human antiserum (lot 3; Food and Drug Administration).To evaluate the titers, the values were transformed into decimallogarithms, and then the means and the limits of the 95% confidenceinterval (CI) were determined. Thereafter, the antilogarithmand the corresponding 95% CI were calculated.

Quantitative determination of anti-diphtheria toxin and anti-tetanus toxin IgG. In vitro tests for measuring tetanus and diphtheria antitoxinlevels in sera from both vaccination groups were done by a standardizedmodified toxin-binding inhibition assay (41) at the QualityControl Service of the Butantan Institute, São Paulo,Brazil.

Adverse event monitoring. Parents were asked to notify the study staff immediately byphone of any unexpected or severe reactions. A standardizedquestionnaire using scripted questions and definitions was usedto collect information about any occurrence related to vaccination.After the first dose, parents' compliance to monitoring of adverseevents was assessed by phone interview. At the administrationsof the second and third doses and blood sampling, compliancewas assessed by the study nurses during visits to the publichealth center.

Statistical considerations. Analyses were carried out with SPSS for Windows (version 7.5.1).Mann-Whitney and Friedman nonparametric tests were used in proliferationand cytokine analyses. Comparisons manifesting a two-tailedP value of <0.05 were considered statistically significant.Prism software (version 4.0; GraphPad Software) was employedfor the figures.

To evaluate the anti-PT titers, calculation of the geometricmean titers (GMTs) of Abs was performed on log₁₀-transformeddata, and we report the antilogarithms. For each group, GMTsand 95% CIs were calculated. For comparison of the logarithmof the titers, Student's t test was applied for independentsamples. Comparisons manifesting a two-tailed P value of <0.05were considered statistically significant (25).

The differences in the proportions of subjects with seroprotectionagainst diphtheria and tetanus and 90% CI were calculated asrecommended for noninferiority studies (10, 36). Differencesor ratios equal to or lower than 10% were accepted as the limitfor defining noninferiority of the wP_low vaccine. The null hypothesisof noninferiority of the wP_low vaccine was accepted when thelower limit of the CI was not lower than –10%.

RESULTS

Top

RESULTS

Study participants. Out of 247 infants initially selected, 234 participated in theentire study and were distributed as follows: 115 infants inthe low-LPS-content vaccine group (57 males and 58 females)and 119 infants in the whole-cell vaccine group (65 males and54 females). Table 1 shows the mean ages (with standard deviations)of the infants at the administrations of the three doses ofvaccine and at blood sampling.

View this table:
[in this window]
[in a new window]

TABLE 1. Distributions of infants vaccinated with the wP_low or wP vaccine by age at the administrations of the three doses of vaccine and at blood sampling

B. pertussis-specific T-cell proliferation. T-cell proliferation was measured by flow cytometry (Fig. 1).CD3⁺, CD4⁺, and CD8⁺ immunophenotyping was carried out in thecultures of 205 infants, and ${gamma}$ ${delta}$ ⁺ blast cells in 55 children (28in the low-LPS-content vaccine group) were analyzed.

View larger version (22K):
[in this window]
[in a new window]

FIG. 1. Flow cytometry of unstimulated (control) and B. pertussis- and PHA-stimulated PBMC of a 7-month-old infant vaccinated with a cellular pertussis vaccine with low LPS content. After gating of CD3⁺ cells (A), events were analyzed for size and complexity (forward-scatter and side-scatter gates), from which resting and blast lymphocytes were separated (B). T-lymphocyte subsets (CD4⁺, CD8⁺, and ${delta}$ ${gamma}$ ⁺) were then verified in blast lymphocytes.

Median background CD3⁺ proliferation levels were 2.3% in thewP_low vaccine group and 2.6% in the wP vaccine group. The netpercentage of CD3⁺ blasts in cultures with B. pertussis in thegroup vaccinated with the wP vaccine was higher than that inthe group vaccinated with the wP_low vaccine (medians of 6.2%and 3.9% for those immunized with wP and wP_low vaccines, respectively;P = 0.029), but there was no difference between these valuesfor cultures stimulated with PHA (medians of 82.5% and 81.4%for those immunized with wP and wP_low vaccines, respectively;P = 0.166) (Fig. 2).

View larger version (19K):
[in this window]
[in a new window]

FIG. 2. Distribution of CD3⁺ blasts (percentages) determined by flow cytometry of B. pertussis (Bp; ${blacksquare}$ )- and PHA ( ${blacktriangleup}$ )-stimulated PBMC from wP_low vaccine- or wP vaccine-immunized infants. Medians (indicated by bars) of CD3⁺ blasts in B. pertussis-stimulated cultures were 3.9% and 6.2% for wP_low and wP vaccines, respectively. Median percentages of CD3⁺ blasts in PHA-stimulated cultures were 81.4% and 82.5%, respectively. n, number of individuals.

The frequencies of proliferating CD4⁺, CD8⁺, and ${gamma}$ ${delta}$ ⁺ cells inB. pertussis- and PHA-stimulated cultures were similar for thevaccination groups (Table 2). On the other hand, there was asignificant difference between the T-cell subpopulations forcontrol and B. pertussis- and PHA-stimulated cultures (Friedmantest, P < 0.001), with higher percentages of ${gamma}$ ${delta}$ ⁺ cells in theB. pertussis-stimulated cultures.

View this table:
[in this window]
[in a new window]

TABLE 2. Percentages of blast cells determined by flow cytometry in cultures of PBMC that were obtained from vaccinated infants and incubated with medium alone, B. pertussis, or PHA

B. pertussis-specific cytokine production. B. pertussis-specific cytokines in cell cultures of infantswere determined by ELISA after 48 h of stimulation. For thegroups of vaccinees, the amounts of cytokine secretion in supernatantswere compared. IL-4 was undetectable in stimulated and unstimulatedsamples (negative controls). Levels of IFN- ${gamma}$ , TNF- ${alpha}$ , and IL-10production (Fig. 3) were not different between B. pertussis-and PHA-stimulated cultures of infants vaccinated with the wP_lowor wP vaccine.

View larger version (12K):
[in this window]
[in a new window]

FIG. 3. IFN- ${gamma}$ (A), TNF- ${alpha}$ (B), and IL-10 (C) concentrations determined by ELISA (values are given in picograms per milliliter) by PBMC in cultures without stimuli (control), with B. pertussis (Bp), or with PHA (•) stimulation in infants immunized with wP_low or wP vaccine. Bars indicate medians, and the Mann-Whitney test was used to verify significances. n, number of individuals.

Anti-PT IgG titers. Anti-PT Ab GMTs following three doses of wP_low or wP vaccineindicated robust responses to both vaccines (Table 3), and nostatistically significant differences between the two groupswere observed (Student's t test, P = 0.464).

View this table:
[in this window]
[in a new window]

TABLE 3. GMTs and 95% CI for serum anti-PT IgG levels of infants vaccinated with three doses of wP_low or wP vaccine

Diphtheria and tetanus seroprotection. Table 4 shows the percentages (with 95% CIs) of subjects protectedagainst diphtheria and tetanus. As the lower limits of the CIswere not lower than –10% for the three studied parameters,the null hypothesis of noninferiority of the wP_low vaccine wasaccepted.

View this table:
[in this window]
[in a new window]

TABLE 4. Protection against tetanus toxin and diphtheria toxin in 7-month-old infants vaccinated with three doses of the wP_low or wP vaccine

Adverse events. Within the first 2 to 3 days after each injection, the parentsand legal guardians of the vaccinees reported local reactions,such as erythema ( $≥$ 20 mm; 3.6% for the wP_low vaccine group and1.4% for the wP vaccine group) and pain (5.3% for the wP_lowvaccine group and 2.3% for the wP vaccine group). Likewise,systemic reactions (body temperature, $≥$ 39°C) were observedin both groups (3.8% for the wP_low vaccine group and 2.0% forthe wP vaccine group). Irritability was reported after 4.4%of wP_low vaccine injections and 2.9% of wP vaccine injections,respectively. Severe systemic adverse effects after each diphtheria-tetanus-pertussis(DTP) vaccine dose were not reported. None of the participantswithdrew from the study because of vaccine-related adverse events.There were no significant differences between wP_low and wP vaccinegroups.

DISCUSSION

Top

DISCUSSION

Studies of immune responses to pertussis vaccines suggest thatboth B- and T-cell responses are elicited in humans (5, 7, 14)and mice (4, 33). By use of a murine model of infection, ithas been demonstrated that adoptive transfer of CD4⁺ T cellsfrom immune mice confers protection against B. pertussis duringchallenge in the absence of a detectable Ab response, but also,passive Ab transfer protected mice from B. pertussis infection(27). In an attempt to better understand the cell-mediated immuneresponses to pertussis vaccines in infants, we evaluated thein vitro proliferation of T cells specific to B. pertussis atone month after the third dose of vaccine was given.

We used a flow cytometry-based lymphocyte proliferation assaythat allowed the characterization of specific subpopulationexpansion in response to B. pertussis. We found that infantsvaccinated with the wP vaccine and those vaccinated with thewP_low vaccine exhibited similar frequencies of B. pertussis-specificproliferating CD4⁺, CD8⁺, and ${gamma}$ ${delta}$ ⁺ cells, although total CD3⁺ cellproliferation in the wP vaccine group was significantly higher.

We report here for the first time that, in humans, high percentagesof ${gamma}$ ${delta}$ ⁺ cells are found in B. pertussis-stimulated cultures (mediansof 20.3% for wP and 16.8% for wP_low), suggesting that thesecells may play an important role in the pertussis-specific response.T cells expressing the ${gamma}$ ${delta}$ receptor were identified in the mid-1980s(11, 12, 15, 39) and are disproportionately abundant withinepithelial surfaces, including those in the lung (26). Theirintraepithelial distribution and capacity for recognizing nonproteinantigens, sometimes in a non-major histocompatibility complex-restrictedfashion, have led to their consideration as part of the first-linedefense against pathogens (23). In humans, large expansion of ${gamma}$ ${delta}$ ⁺ T cells after Mycobacterium bovis bacillus Calmette-Guérinvaccination and during infection with Mycobacterium tuberculosis,Listeria monocytogenes, Brucella melitensis, and Ehrlichia chaffeensissuggests their importance in the host response (31, 42).

The activation of ${gamma}$ ${delta}$ ⁺ T lymphocytes can lead to IFN- ${gamma}$ production,which is instrumental in the upregulation of both macrophageand natural killer (NK) cell functions central to early antibacterialprotection prior to the ${alpha}$ β⁺ T-cell response (23). IFN- ${gamma}$ alsoinfluences the downstream acquisition of a Th1 phenotype by ${alpha}$ β⁺ T cells (42). ${gamma}$ ${delta}$ ⁺ T cells thus bridge the innate and acquiredimmune response by providing initial protection of epitheliafrom invasion and injury in instances where ${alpha}$ β⁺ T cellsare not yet operational and then downregulating the antigen-specificadaptive immune response after the danger has passed to minimizepotential immune-mediated injury (40).

Although little is known about the importance of ${gamma}$ ${delta}$ ⁺ T lymphocytesduring B. pertussis infection, a role for these cells in theresponse of infected children has previously been suggestedvia the migration of circulating ${gamma}$ ${delta}$ ⁺ T cells to the airways (9).Moreover, using a murine aerosol challenge model, Zachariadisand coworkers demonstrated that the absence of ${gamma}$ ${delta}$ ⁺ T cells couldinfluence the subsequent adaptive immune response to B. pertussisantigens, as evidenced by a shift from a Th1-type response toa Th2-type response against filamentous hemagglutinin in TCR ${gamma}$ ${delta}$ ^–/– mice (48).

Th1 cytokines are associated with protection in various B. pertussisinfection models, and in particular, in humans, protection againstpertussis after infection or vaccination is determined by thepresence of Th1 cytokines such as IL-12 and IFN- ${gamma}$ (37). Also,the clearance of B. pertussis or protection induced by wP vaccinesis dependent on the production of Th1 cytokines, since IFN- ${gamma}$ -or IFN- ${gamma}$ receptor-defective mice and mice depleted of NK cells,which infiltrate the lung and secrete IFN- ${gamma}$ early in infection,develop disseminating lethal infections (13).

In this study, ELISA measurements of cytokine levels in culturesupernatants demonstrated significant increases in IFN- ${gamma}$ andTNF- ${alpha}$ secretion by B. pertussis-stimulated PBMC. Since therewas no increase in IL-4 secretion, we speculate that the cytokineproduction in response to B. pertussis was more Th1-like forboth vaccines. However, our assay did not permit the identificationof the cells which were secreting these cytokines. Similar Th1cytokine secretion profiles for adults (3) and children (6,30, 37) in response to cellular pertussis vaccines have beenreported. In contrast, T cells from children immunized withaP vaccines can also secrete IL-5 following stimulation withB. pertussis antigens and generate a type 2 effector response(38).

Because of the bias against Th1-cell-polarizing cytokines, itwas initially thought that the neonatal immune system was generallyimpaired or depressed. However, mounting evidence suggests that,under some circumstances, human neonates seem able to developmature Th-cell responses, ranging from deficient or deviantto fully mature, depending on the conditions of antigen exposure(reviewed in reference 1). Our results demonstrate that infantsare able to mount Th1 responses to B. pertussis after wP orwP_low vaccine administration.

The induction of protective Th1 responses by immunization withthe wP vaccine or by previous infection with B. pertussis hasbeen associated with IL-12 production by macrophages or dendriticcells, and this has been linked with LPS present in the wP preparationsand in the live bacteria (29). LPS signaling through TLR4 ininnate immune cells plays a critical role in the generationof inflammatory cytokines IL-12, IL-23, and IL-1, which directthe induction of Th1 and Th17 cells in mice immunized with wPvaccine. Furthermore, the cytokines secreted by these T-cellsubtypes promote bacterial killing by macrophages, a responsethat is further enhanced at the effector level by TLR4-mediatedactivation of macrophages. Thus, TLR4 plays a critical rolein the induction and in the effector phase of the protectivecellular immune response to B. pertussis induced by vaccination(24), which could explain the lower level of proliferation ofCD3⁺ T cells observed in wP_low vaccine group.

Additionally, the null hypothesis of noninferiority of wP_lowvaccine was accepted because the wP_low and wP vaccines elicitedsimilar anti-PT IgG GMTs and did not interfere in tetanus ordiphtheria seroconversion. Theoretically, the wP_low vaccine,having 95% less LPS than the conventional vaccine, would haveweaker adjuvant activity and therefore produce a lower-levelAb response. In this sense, we can understand the significantlyhigher level of total CD3⁺ proliferation in the wP vaccine groupas a result of the LPS being linked to the host TLR complexTLR4/MD-2 and the subsequent activation of NF- ${kappa}$ B (35). Becausethere is no defined parameter for pertussis seroprotection (18),we could not estimate differences in the proportions of protectedinfants.

Our surveillance system for severe adverse events followingimmunization was able to detect an excellent primer safety profileof wP_low and wP vaccines. DTP vaccines differ from manufacturerto manufacturer because of the different Bordetella pertussisstrains used for production. Some manufacturers even use twodifferent strains. There is a rare although serious risk ofa severe adverse event related to the pertussis component ofthe DTP vaccine. The risks of encephalopathy and convulsionsvary according to the origin of the vaccine and the strain usedfor production. These risks have been estimated as 1/19,496for febrile convulsions, 1/76,133 for convulsion without fever,and 3/1,000,000 for encephalopathy (17, 19). The wP vaccineavailable in Brazil (licensed by the Butantan Institute) hasbeen administered free of charge to almost 100% of infants inthe past 15 years. More than 50 million doses have been used,without major adverse effects. According to the Brazilian Ministryof Health, pertussis incidence has declined from 10.8 per 100,000in 1990 to 0.55 per 100,000 in 2005. In the same period, diphtheriaincidence decreased from 0.45 per 100,000 to 0.01 per 100,000and tetanus incidence decreased from 1.0 per 100,000 to 0.25per 100,000.

Since there were similar patterns of responses in the two vaccinatedgroups, it is reasonable to assume that the low-LPS-contentcellular pertussis vaccine is capable of inducing a B. pertussis-specificresponse and, consequently, is immunogenic. Therefore, our dataendorse further investigations with the wP_low vaccine. On theother hand, it must be recognized that, although the low-LPS-contentcellular pertussis vaccine was similar to the conventional whole-cellone, based on Th1-polarized effector response and specific ${gamma}$ ${delta}$ ⁺T-cell expansion, the overall data did suggest that the wP vaccineperformed modestly better with regard to specific CD3⁺ T-cellproliferation.

ACKNOWLEDGMENTS

We are indebted to the participating parents and study nurses,whose intimate collaboration formed the backbone of the trial.

This study was supported by Fundação de Amparoà Pesquisa do Estado de São Paulo, SãoPaulo, Brazil, grant number 2005/03539-1, and by Financiadorade Estudos e Projetos, Brazil, grant number 01040957/0.

FOOTNOTES

* Corresponding author. Mailing address: Center for Investigation in Pediatrics and Pediatrics Department, State University of Campinas Medical School, Rua Tessália Vieira de Camargo 126, Campinas, São Paulo, Brazil CEP 13083-887. Phone: 55-19-35218963. Fax: 55-19-35218972. E-mail: marluce{at}fcm.unicamp.br

Published ahead of print on 4 March 2009.

REFERENCES

Top