How Important Is the Type of Acellular Pertussis Vaccine? Free

(See the Major Article by Carvalho et al on pages 200–7.)

Pertussis remains endemic, even in countries with robust childhood vaccination programs. The World Health Organization (WHO) acknowledges that the true pertussis disease burden is underrepresented, due to the lack of reliable surveillance data in many nations, but in 2008 estimated that pertussis-associated deaths may be as high as 89 000 per year [1]. In highly vaccinated populations, epidemic cycles of pertussis disease occur every 2 to 5 years, similar to the pre-vaccine era. The continued circulation of Bordetellapertussis occurs because immunity to pertussis, whether induced by a natural infection or vaccination, is not lifelong [2].

During the past 3 decades, a resurgence of pertussis in highly vaccinated nations occurred, with a shift in the distribution of cases towards adolescents. The reasons for this resurgence are multifactorial and include increased awareness; improved surveillance; more sensitive diagnostic tests, such as polymerase chain reaction; genetic variations in B. pertussis strains; and waning vaccine-induced immunity. The suspicion that waning immunity is associated with the transition from whole-cell pertussis vaccines (wP) to less reactogenic acellular pertussis vaccines (aP) in the primary infant immunization series has been supported by observations during the 2010 epidemic in California, where vaccine effectiveness declined more rapidly than expected, with the odds of acquiring pertussis increasing by 42% each year since the last dose of acellular vaccine [3, 4]. Case-control studies confirmed that children and adolescents who received 4 doses of wP vaccine were less likely to become infected than those who received 4 aP doses, that the risk was markedly increased among those receiving only aP vaccines, and that the receipt of 1 or more doses of wP within the series augmented the durability of the immune response [5, 6]. The order of wP dose within the series also appears to be important, with lower disease rates if the priming dose of the series is wP [7]. In 2015, the WHO concluded that true resurgences of pertussis had occurred in 4 of 19 countries, all of which had transitioned to a complete aP primary immunization series [1].

In this issue of the Journal, Carvalho and colleagues provide important further evidence that some aP vaccines demonstrate inferior immunogenicity to their wP counterparts [8]. The pertussis epidemic experienced in the United Kingdom in 2011–2012, combined with the system of centralized vaccine procurement and robust national surveillance data, facilitated this natural experiment. Their report is the first, to our knowledge, to simultaneously compare the performance of the 2 currently available aP vaccines, which differ in composition and number of pertussis antigens, using a wP vaccine as a reference. Children who received aP vaccines containing 3 antigens (aP3) for the primary vaccination series (2, 3, and 4 months of age) were more likely to be diagnosed with pertussis than those receiving wP. The risk of pertussis among children receiving aP vaccines containing 5 antigens (aP5) for the primary vaccination series was comparable to that of those receiving wP. While B. pertussis expresses a large number of virulence factors, including pertussis toxin (PT), filamentous hemagglutinin, pertactin, fimbriae (FIM) types 2 and 3, adenylate cyclate toxin, tracheal cytotoxin, and lipopolysaccharide, the roles of each factor in the pathogenesis of pertussis infections and disease are incompletely understood. Filamentous hemagglutinin, pertactin, and FIM facilitate attachments to ciliated respiratory epithelial cells, while PT, adenylate cyclate toxin, and tracheal cytotoxin allow evasion of the host immune response and are toxic to respiratory epithelial cells. It is biologically plausible that the inclusion of FIM in aP5 enhances protection against infections and disease. Studies conducted during the 1980s and funded by the National Institutes of Health demonstrated that aP5 had comparable FIM antibody titers to those generated by the wP vaccine included in those trials, while aP3 did not [2]. In addition, aP vaccines with ≥3 antigens had greater efficacy than those with ≤2 [9]. If the observations of Carvalho et al are confirmed by others, there are implications for the choice of aP vaccine type for the primary series and for future pertussis vaccine developments.

However, it is very important is to acknowledge the limitations of this report. A major caveat with this study is the shorter period of follow-up for those immunized with aP5 (administered starting 2004), compared with those who received aP3 (administered 1999–2003). The aP5 cohort was aged ≤8 years during the epidemic, compared to those immunized with aP3, who were 9–12 years, which is the age range when tetanus, diphtheria, and aP booster vaccines (Tdap) are recommended in many countries. Therefore, the increased risk of the aP3 cohort was arguably predictable. Since the waning of pertussis vaccine–induced immunity is a function of time [2–4], there may have been inadequate time for the differences between wP and aP5 to become apparent, and the effectiveness of aP5 and aP3 may be equivalent with similar periods of follow-up. Another possible limitation is that the PT antigen content of the vaccines differs and PT immunoglobulin G was the predominant diagnostic test for pertussis, rather than polymerase chain reaction, which provides sensitive, real-time diagnoses. This may have resulted in the decreased ascertainment of pertussis when comparing groups with each other, although the study results suggest that antibody responses did not vary significantly by vaccine received. Finally, the United Kingdom uses an accelerated schedule of 2, 3, and 4 months of age for the primary infant series, and these findings may not be applicable in countries using longer intervals between primary series doses, due to greater infant immunological maturity at the time of the series completion.

What are the implications of this study? Obviously, this investigation indicates an urgent need for additional studies (retrospective and prospective) to determine if these findings can be replicated. Such studies will require adequate sample sizes in each vaccine cohort, sufficient durations of follow-ups, and controls for sources of bias, such as differing primary immunization schedules and immune-boosting through circulation of natural pertussis. The further exploration of how aP vaccines, which are known to induce different T-helper (Th)1/Th2 and Th17 T-cell responses than wP, induce long-term humoral and cellular immunity is urgently needed.

This report adds additional evidence that aP vaccines are often inferior to their wP counterparts, and provides support for the WHO recommendation to continue wP vaccines in countries where pertussis-associated morbidity and mortality rates are high. In countries where the transition to aP has occurred, other approaches are needed. In an era where vaccine hesitancy is increasing, it appears unlikely that more reactogenic wP vaccines will replace aP in the primary series, even if only 1 priming dose were needed to overcome the less durable immune response. This supports the continued need for strategies like maternal Tdap vaccination to prevent infection in those with the highest risk of poor outcomes (young infants) and the boosting of immunity through vaccinations to control pertussis outbreaks until more immunogenic and durable pertussis vaccines are available. Limited data suggest that the differences seen between priming with wP versus aP vaccines may be partially mitigated by a booster dose of a Tdap vaccine [6]. Studies will need to address whether priming with aP or wP affects the immune response when Tdap is administered to pregnant women and if there are implications for the incidence of severe pertussis in young infants. Carvalho and colleagues, through this report, ensure that such studies will also need to be cognizant of the formulation of the aP vaccine received during the primary immunization.

Note

Potential conflicts of interest. C. M. H. received honoraria from the National Foundation of Infectious Diseases (as a speaker at a Clinical Vaccinology course). Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References