As pertussis returns to pre-COVID19 endemicity, vaccination remains our best ally against an evolving Bordetella pertussis

Pertussis is a highly contagious respiratory disease caused by the bacterium Bordetella pertussis [1]. Despite being vaccine-preventable, pertussis remains endemic worldwide. The sizeable epidemics observed on all continents in the last two years have reminded us that pertussis remains one of the least well-controlled vaccine-preventable diseases [2]. Vaccination against pertussis is one of the cornerstones of national immunization programs (NIPs) in all countries. Yet, for infant pertussis vaccination as for subsequent booster vaccinations, the number of doses, the schedule, and the vaccine coverage rates of are extremely heterogeneous across countries.

Post-COVID-19 pertussis epidemiology showed a dramatic resurgence with large outbreaks recorded on every continent [2]. The first post-COVID-19 outbreaks were reported in the Central African Republic and in South Africa in 2022. The pertussis rise expanded geographically in 2023, with outbreaks reported in such diverse countries as Russia, Indonesia, Thailand, Malaysia, Bolivia, Peru, Denmark, and Israel. The expansion reached all continents in 2024, with outbreaks described across Europe, in France, Germany, Spain, and the UK, among many others, across the Western Pacific, in Australia, China, and the Philippines, and across the Americas, in the US, Mexico, Brazil, and Canada. The reported level of incidence and its distribution in the population varied greatly between the outbreaks, leading to different hypotheses on their causal factors. An important limitation in trying to explain outbreaks from notified cases of pertussis is that the magnitude and distribution of the incidence is highly dependent on the extent and sensitivity of the surveillance systems. Despite studies highlighting the burden of pertussis beyond childhood at least since the 1990s, pertussis continues to be perceived as mainly a childhood disease. As a result, under-diagnosis affects how well the burden of pertussis is characterized. Therefore, the heterogeneity of awareness and surveillance capacity across countries limits the interpretability of observed pertussis incidence.

In all countries, numerous cases were reported among age groups at least expected to be vaccinated if not verified to have received a prior vaccination [2]. This observation raised doubt about the effectiveness of the vaccines in protecting against pertussis; a valid question on face value, but one that would require a more thorough investigation of cases and outbreaks to analyze objectively. Lack of protection after vaccination can be categorized as primary or secondary vaccine failure [3]. Although rare, primary vaccine failure occurs when a vaccine does not elicit sufficient immunity, leaving recipients unprotected. No vaccine offers 100% efficacy, and while clinical trials have measured ≥ 85% efficacy for acellular pertussis vaccines, no data exist on the efficacy of the currently used whole-cell vaccines [1]. In the case of secondary vaccine failure, protection acquired from vaccination wanes over time, leaving individuals susceptible to disease several years after their last dose. Protection elicited by pertussis acellular and whole-cell vaccines lasts only a few years, and regular booster doses are needed, which have demonstrated effectiveness, albeit waning as well [4]. In primary and secondary pertussis vaccine failures, the individuals who are left unprotected may go mostly unnoticed in low incidence periods because their probability of encountering the pathogen and contracting pertussis is low. During surges of pertussis incidence which occur at 3- to 5-year intervals, the increased force of infection raises the probability of these individuals encountering the pathogen and becoming diagnosed cases of pertussis. The phenomenon is further reinforced with the improved awareness and testing capacity that may come with outbreaks and after widespread adoption of PCR. This explains why a proportion of cases detected during an outbreak were previously vaccinated, and the higher the overall vaccination coverage in the population, the greater this proportion of vaccinated cases. In fact, real-world studies of pertussis vaccine effectiveness have long documented this phenomenon, while still demonstrating high vaccine protective effect at population level.

Waning of vaccine-elicited protection and the impact of vaccines on infection and transmission are among the hypothesized contributing factors to pertussis resurgence [5]. Studies conducted in the context of pertussis epidemic surges pre- and post-COVID-19 have also evoked the possibility that genotypic and phenotypic changes in populations of circulating B. pertussis strains may lead to vaccine escape and explain, at least in part, the resurgence [6–8]. The most prominently studied and commented changes with comparable trends across countries concerned the increased prevalence of the ptxP3 allele of the pertussis toxin (PT) gene promoter, resulting in increased toxin expression, the loss of pertactin expression, and the resistance to macrolide antibiotics.

The ptxP3 allele of the pertussis toxin gene promoter, originating from a mutation in the 1970s, appears to have expanded in prevalence in the 1990s and is globally predominant today [6]. PT is both a toxin involved in pathogenesis and an immunomodulator with a crucial role in the pathogen survival in the respiratory tract and its transmission. Therefore, neutralization of PT is essential to the protection elicited by vaccines, and increasing PT production could present a selective advantage for the survival and dissemination of B. pertussis in highly vaccinated populations [6]. Despite limitations often due to small sample sizes and missing information, several studies have found that infection by ptxP3-harboring strains of B. pertussis tends to increase the risk of pertussis or its severity [6,9]. In these studies, age below vaccination age and absence of vaccination remained the strongest factors for predicting pertussis or its severity. Vaccines remain effective in preventing pertussis in a context of prevalent ptxP3-harboring strains so long as they induce a robust anti-PT antibody response [6]. However, it has been hypothesized that exposure to ptxP3-harboring strains would increase the likelihood of pertussis and pertussis complications among infants too young to be vaccinated or lacking maternal protective antibodies as well as among those whose immunity from the last vaccine dose may have waned.

Contrary to the PT promoter, deficiency in pertactin expression (PRN-) may occur in several genotypic forms ranging from single nucleotide mutations to insertion-sequence insertions. The PRN- phenotype increased in prevalence in the pre-COVID-19 period, but in the post-COVID-19 resurgence of pertussis, several European countries (France, Finland, the Netherlands and Belgium), reported a trend reversion with most isolates expressing PRN [10–13]. Previous studies had suggested that the presence of PRN in some acellular pertussis vaccines had created a selective pressure favouring PRN- strains, since PRN-specific antibodies have been shown to induce bacterial killing [14]. PRN- strains appear to have higher fitness to colonize and disseminate in highly vaccinated populations, including with repeated booster doses [15,16]. The effectiveness of acellular pertussis vaccines against disease did not appear diminished in the context of predominantly PRN- B. pertussis strains [9,17]. However, the impact of this phenotype on the risk and severity of disease remains to be fully demonstrated. Several studies explored this question and yielded contrasted results, but all faced significant methodological limitations that preclude a definitive conclusion [9,18,19].

In addition to virulence factors evolution, B. pertussis resistance to macrolide antibiotics has raised concerns in recent outbreaks. Macrolides, such as azithromycin, are the first line of antibiotics in the clinical management of pertussis. Macrolides are effective in treatment of pertussis when administered early in the course of infection, especially among young children; but their impact limited to clearing the bacteria when administered later means that they are also important in reducing transmission in the community [20]. Macrolide resistance resulting from different genome modifications has been sporadically identified in Europe, Asia and the Americas since the 1990s. However, in the last decade, macrolide resistance has become predominant among B. pertussis isolates collected in China, with increasing spread to countries in the region, and possibly beyond [8,10,21]. High prevalence of macrolide-resistant strains may be associated with local practices in antibiotics use [21]. The resistance of B. pertussis to macrolides does not affect the effectiveness of vaccines. However, it decreases the ability to control the spread of the pathogen, especially when vaccine boosters are not available beyond infancy and infection reservoirs are allowed to build up and persist over time.

B. pertussis has a dynamic genome; multiple mobile insertion-sequences and mutations enable rapid genomic modifications resulting in the emergence of new pathogenic strains. The expansion of pertussis vaccination may have played a role in selecting for strains of B. pertussis that can better circulate among highly vaccinated populations. The timing of the expansion and the distribution of these genotypic and phenotypic evolutions are clear indications that vaccines are effective in preventing disease and even more in hampering the pathogen dissemination in the community. The evidence available so far does not point to a diminished vaccine effectiveness against disease caused by these strains. However, strains of B. pertussis with better fitness to circulate in vaccinated and antibiotic-treated populations increase the force of infection and in turn the risk to unvaccinated individuals, or for individuals with primary or secondary vaccine failure, to become infected and diseased, and to transmit infection. School-age children and adolescents are the population segments where the highest circulation occurs. While they may be at lower risk of severe disease, they may also act as a source for transmission to infants too young to have received any vaccination [22]. Early infant protection, boosters in preschool children and adolescents, and vaccination in pregnancy have been well documented to reduce the risk of morbidity and mortality in young infants [23–25]. China’s recent schedule change of immunization program for DTP-containing vaccines illustrate the first two points. While B. pertussis is evolving to survive in a world where pertussis vaccines are increasingly adopted, optimizing the use of the current, effective vaccines is essential to continue improving pertussis control.

Disclosure statement

MZ reports no competing interest to declare. DM is an employee of Sanofi and holds shares of the Sanofi group of companies as part of his remuneration.