Pertussis vaccines

I am honored to assist with this Special Focus Issue about pertussis. There is a clear need to improve the prevention of whooping cough. Detection methods have improved, awareness has risen, but undoubtedly we are facing a true increase in cases of whooping cough. Immunization has addressed part of the disease burden, but further actions and developments are needed. This special issue describes the problem of pertussis in the 21st century [1–4], the complexities of immune response to pertussis [5,6], issues around determining vaccine potency [7], the emergence of new strains potentially resistant to specific vaccine-induced antibody [8,9] and lessons to be learnt from other Bordetella species [10]. The limitations associated with the two classes of vaccines (whole cell vaccines [WCVs] and acellular vaccines [ACVs]) that are currently available are considered [11,12]. Potential improvements to vaccines and vaccine schedules are outlined [13–19] as well as the challenges inherent in diagnosing pertussis [20].

WCVs are used in the majority of childhood immunization programs worldwide, but can vary widely in terms of their immunogenicity, reactogenicity and effectiveness. This variability is a function of manufacturing methods which have changed little since the first WCVs were developed early in the 20th century and pre-release testing (the mouse intra-cerebral challenge) that is not always able to identify vaccines that are poorly effective in humans. However, based on long-term evidence of safety, efficacy and durability of the immune response, it is justifiable to continue with WCVs in regions where these vaccines are currently used.

ACVs, either co-purified mixtures of antigens or with purified and mixed antigens, have been proven to prevent pediatric pertussis disease, but the duration of protection is short-lived. Strikingly, ACV booster vaccination in ACV-primed individuals appears to induce an even shorter duration of protection than primary ACV vaccination. This is surprising, and original antigenic sin operating via chemically detoxified neo-epitopes on pertussis toxin (PT) (i.e., utilization of immune memory to the PT vaccine epitopes to produce antibodies that are ineffective against a wild-type strain) is one possible explanation. Upsurges of pertussis in adolescents, continued or possibly increased transmission, and a return of neonatal pertussis cases and associated deaths may be, at least partially, a consequence of the limited durability of protection induced by ACVs. This is suggested by studies showing more clinical pertussis cases in adolescents primed with ACV than WCV, and while subclinical cases in WCV recipients cannot be excluded, these observations point to a suboptimal priming event.

It seems likely that the limited antigenic composition of ACV has allowed the bacterium to escape vaccine-induced immunity. This appears to be the case for the emergence of pertactin (PRN)-deficient pertussis strains, reported from an increasing number of countries that use ACV. Multiple mechanisms for the non-expression of PRN have been identified, pointing to a true selective drive for such a mutation. Interestingly, the appearance of non-PRN-expressing pertussis strains in France after the introduction of ACV appears to be associated with the emergence of non-PRN-expressing Bordetella parapertussis.

How can pertussis be brought under control for the long term? Various means being explored include a return to WCV with concurrent improvements in pre-release testing and vaccine standardization, investigation to identify optimal primary and booster schedules, immunization of special populations (neonates, pregnant women) and development of new vaccines containing novel components or that use novel adjuvants or delivery systems to induce broad and durable protection. It is unlikely that countries that historically turned to ACV because of concerns with WCV reactogenicity will return to WCV. An urgent need is the protection of vulnerable infants who continue to die from pertussis. Preventing transmission using cocooning strategies has merits, but is difficult to implement effectively. Neonatal immunization has been attempted, but with mixed results and with questions remaining about the utility of this approach, including the impact on immune responses to later immunizations. Recently, attention has been drawn to maternal ACV immunization and implementation efforts are being undertaken in several countries. Good efficacy data from the UK have recently been published [21]. Prevention of neonatal pertussis would be a tremendous breakthrough and would represent a paradigm shift in the control of neonatal pertussis, opening the door for additional strategies to protect newborns through maternal immunization against pathogens such as respiratory syncytial virus and group B hemolytic streptococcus.

Bordetella pertussis is a very well antigenically conserved pathogen. Evasion of host immunity appears to be linked to major toxins such as PT and adenylate cyclase (ACT) that impact the host and protect the bacteria against phagocytosis. Another major pathogen, Staphylococcus aureus, shares this phenomenon of antigenic conservation and immune evasion of phagocytosis. B. pertussis ACT and S. aureus LukAB both bind to the neutrophil receptor C11b [22], which is in contrast to other pathogens such as the pneumococcus and meningococcus that appear to rely upon antigenic variability to avoid the host immune system.

Recently, the potential role of the Th1/Th17 immune response in enabling bacterial clearance has drawn attention. It is thought that Th1/Th17-associated cytokines may stimulate phagocytosis and increase neutrophil chemotaxis, thereby enhancing innate immune response to pertussis. It is not known if it is feasible, or even necessary, to induce Th1/Th17 immunity through vaccination. The means by which Th1/Th17 responses can be specifically enhanced is unclear, and there are potential risks associated with the induction of Th1/Th17 immunity. This is an area of ongoing research.

The use of PT as a major component in ACV relies upon toxin neutralization and indirectly enhancing bactericidal/opsonophagocytic innate immune mechanisms. The use of potent PT, avoiding chemical inactivation of protective epitopes is desirable. This may partly explain why a PT-only ACV (containing only mildly detoxified PT) appears to be reasonably successful in controlling whooping cough in Denmark. Too harsh a detoxification process could lead to original antigenic sin and immunodominance of non-protective neo-epitopes. It can be argued that inclusion of a properly detoxified ACT to further avoid immune evasion by the bacterium and/or the inclusion of other antigens that induce direct bactericidal and/or opsonophagocytic antibodies may be worthwhile, the latter are the key immune mechanisms that have led to the development of successful pneumococcal, Haemophilus influenzae and meningococcal vaccines. So far, these aspects of the immune response have received little attention in the pertussis vaccine field, although filamentous hemagglutinin and PRN have been described as inducing such antibodies. Similarly, no systematic research has been undertaken in developing robust bactericidal and opsonophagocytic assays for pertussis, or in deciphering the most suitable antigens enabling induction of such functional antibodies. Given the current upsurge in clinical pertussis cases, there is a clear incentive in this research area for progress and development.

Several approaches have been taken toward the development of new pertussis vaccines. A key problem is the identification of important antigens, toxins, opsonins and so on, and then, how to formulate these into a vaccine. One can argue that the emergence of PRN non-producing strains supports the construction of multivalent vaccines and potentially an improved ACV. A counter-argument to this approach is that more complex vaccines will have implications for manufacturing, cost and subsequent globalization of an improved ACV. Compromises may be needed that limit the composition to essential components that the bacterium cannot survive without (perhaps lipo-oligosaccharide or antigens that induce bactericidal antibodies). For instance, the emergence of non–PT-expressing B. pertussis strains is very limited and, in any case, the impact of failure to express PT is on virulence. A vaccine with composition limited to essential components that induce toxin neutralizing as well as bactericidal/opsonophagocytic immunity might be an achievable compromise, still enabling cost-effective manufacturing and global access.

Questions remain about vaccine formulation type and the potential role of novel adjuvants. But an optimized ACV that uses the widely accepted alum adjuvant warrants investigation. Such novel vaccines coupled with maternal ACV immunization using current vaccines could be part of an effective pertussis prevention strategy.

Continuing research in adjuvants, alternative delivery mechanisms and application sites (skin, oral, nasal delivery) deserve further research. In parallel, improved pertussis animal challenge models, such as the baboon intranasal challenge model, may facilitate vaccine testing.

Multiplicity of antigens could also be achieved with live-attenuated or mutated killed bacteria, reshaping the balance between potency and reactogenicity. Manufacturability and immune follow-up, identification of correlates of protection and issues of vaccine consistency are challenges to be addressed with such approaches.

Pertussis vaccines have made, and continue to make, significant contributions to global health. However, long-term disease control has not been achieved. As evident from the diverse papers on this issue by experts in the field, research continues at many levels to improve our understanding of the pathogenesis and immune response to pertussis and to develop improved vaccines.

Financial & competing interests disclosure

The author is an employee of Crucell. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing assistance was utilized in the production of this manuscript and was provided by Dr Joanne Wolter (independent writer on behalf of Crucell).