A novel method for evaluating natural and vaccine induced serological responses to Bordetella pertussis antigens
Introduction
The introduction of mass vaccination against pertussis or whooping cough in the 1940–1950s caused a drastic decrease in infant mortality and may have greatly reduced the amount of notified cases of morbidity and mortality of Bordetella pertussis. However, since 1993 (1996 in the Netherlands) an increase in symptomatic cases has been observed and pertussis has been recognized as a re-emerging infection [1], [2], [3].
Although the effects of improved diagnostic methods and increased knowledge among practitioners have contributed, waning immunity in combination with pathogen adaptation have been implied as possible main causes for the re-emergence of pertussis [4], [5], [6], [7], [8], [9]. The re-emergence of pertussis has led to a renewed interest in pertussis vaccines, and their protection against symptomatic disease [10]. Pertussis vaccines contain whole, killed bacteria (whole cell vaccines, mostly used in third world countries), or specific antigenic components (acellular vaccines). An important issue is comparison of the serological response among different vaccines, and between vaccinated and naturally infected subjects.
Studies of the immune response to vaccination usually report antibody levels before and after administration of the vaccine, recording (group) geometric mean levels [11], [12], [13], [14], or seroconversion above a specified threshold [15]. Antibody persistence is usually characterized as the decrease in (group) geometric mean antibody levels with time after vaccination [16], [17] or even frequency of detection (above an arbitrary threshold) [18].
In all these classical approaches the data are summarized prior to analysis or testing, neglecting the heterogeneity predominant in serological data [19], [20] and concentrating on group metrics. Between subject variation in the dynamics of antibody responses is fundamentally neglected [19], [21].
If the serological antibody response to infection or vaccination can be characterized by biologically meaningful variables like peak antibody level and antibody decay rate, instead of observing group level statistics of antibody levels at a range of times post infection or vaccination, comparison between vaccines is less hampered by design issues like times of sampling or numbers of samples per subject. As follow up is usually feasible only for few samples per individual subject, temporal information is sparse and little is observed about the decay rate or peak antibody levels in any individual subject. For that reason, a mixed model approach is appropriate [21], [22], explicitly modelling within and between subject variation.
We have previously studied IgG antibody responses to pertussis toxin (IgG-PT) in subjects naturally infected with B. pertussis across a wide range of ages using a dynamic model of the interaction between antigen and antibodies [21]. Here we use the same approach to model both natural and vaccine induced responses to various antigens, to derive model based characteristics of the serum antibody response, allowing comparison between the different antibodies in terms of peak antibody levels and antibody decay rates.
Section snippets
Data
Longitudinal serum samples of natural infections were obtained from a patient cohort from a paediatric practice, as described previously [21]. From this cohort, patients of ages >1 and ≤8 years were selected, to agree with the ages in the vaccinated cohorts while keeping the numbers included not too small (N = 44). The average number of sera per patient was 3.9 (minimum 1, maximum 9 blood samples per patient). The whole cohort included also an elderly group of pertussis patients and consisted in
Results
Fig. 1 shows observed IgG concentrations against 3 components of B. pertussis: pertussis toxin (PT), filamentous hemagglutinin (FHA) and pertactin (Prn) for natural infection, booster vaccination with the Dutch whole cell vaccine, and the acellular vaccines (ACV-SKB, ACV-WL and ACV-PM). Both concentrations and time (from symptom onset in natural infection or from vaccination) are transformed to a log-scale, to facilitate identification of a response pattern. The considerable amount of
Discussion
Our approach of comparing vaccines comprises two stages: (1) fitting a mathematical model for the kinetics of the antibody response (including rising and decaying phase of the response) to serological data from subjects naturally infected and subjects vaccinated with different vaccines, and (2) comparing key variables of the model for natural infection and different vaccines. We use predicted values of peak antibody concentrations and predicted halftime values (of antibody decay) to
Acknowledgement
The contribution of one of the authors (PT) to this study was funded by POLYMOD, a European Commission project funded within the Sixth Framework Programme, Contract Number: SSP22-CT-2004-502084.
References (41)
- et al.
Recent developments in pertussis
Lancet
(2006) - et al.
Antibody responses and persistence in the two years after immunization with two acellular vaccines and one whole-cell vaccine against pertussis
J Pediatr
(1998) - et al.
Pertussis antitoxin decay after vaccination with DTPa. Response to a first booster dose 3 1/2–6 1/2 years after the third vaccine dose
Vaccine
(2005) - et al.
Development and validation of a multiplex immunoassay for the simultaneous determination of serum antibodies to Bordetella pertussis, diphtheria and tetanus
J Immunol Methods
(2008) - et al.
Modeling the long-term antibody response of a human papillomavirus (HPV) virus-like particle (VLP) type 16 prophylactic vaccine
Vaccine
(2007) - et al.
Predicted long-term persistence of pertussis antibodies in adolescents after an adolescent and adult formulation combined tetanus, diphtheria, and 5-component acellular pertussis vaccine, based on mathematical modeling and 5-year observed data
Vaccine
(2008) - et al.
Levels of anti-pertussis antibodies related to protection after household exposure to Bordetella pertussis
Vaccine
(1998) - et al.
A search for serologic correlates of immunity to Bordetella pertussis cough illnesses
Vaccine
(1998) - et al.
Low levels of antipertussis antibodies plus lack of history of pertussis correlate with susceptibility after household exposure to Bordetella pertussis
Vaccine
(2003) - et al.
Pertussis of adults and infants
Lancet Infect Dis
(2002)