Immunological insights into why pertussis continues to be endemic

ABSTRACT

Pertussis (whooping cough) is a vaccine-preventable bacterial disease caused by Bordetella pertussis. Despite widespread vaccination, cases have increased globally since the 1990s, with notable outbreaks in 2012, 2016, and following the COVID-19 pandemic. Persistent endemicity and periodic epidemics are driven by limited prevention of nasal colonization, waning immunity after vaccination or natural infection, and pathogen adaptation. Differences between acellular (aP) and whole-cell (wP) vaccines reflect distinct immune profiles, with aP vaccines favoring Th2 responses and wP vaccines inducing Th1/Th17 responses and tissue-resident memory T cells. Circulating strains have evolved under vaccine pressure, including shifts in ptxP, ptxA and prn alleles, increasing pertussis toxin production, emergence of pertactin-deficient variants, and rising macrolide resistance linked to 23S rRNA mutations. Optimizing current vaccine strategies and developing next-generation vaccines that improve durable, Th1/Th17-polarized immunity and reduce transmission are essential for enhanced pertussis control.

Why pertussis persists

Introduction

Bordetella pertussis (B. pertussis) is a respiratory pathogen that can affect individuals of all ages, but disease is particularly severe, and can be life-threatening, in young infants.¹ Although pertussis is generally less severe in adolescents and adults, over half (52%) of reported pertussis cases across the EU/EEA (2023 data) were in adults.² Outcomes were known for 23,196 (89%) of all reported cases, including 10 deaths, five in the under‑one‑year age group and five in individuals over 75 y.² In addition, pertussis in adults is often under-diagnosed, serving as an important reservoir of infection for and transmission to younger, more vulnerable individuals.³

The primary protection against this public health burden is vaccination. Introduction of whole-cell pertussis (wP; inactivated whole‑cell pertussis) vaccines in the 1950s resulted in significant declines in the global incidence of pertussis and associated morbidity and mortality.⁴ However, concerns about the reactogenicity of wP vaccines and variation in effectiveness between formulations led to their replacement in most high-income countries with acellular pertussis (aP) vaccines.⁵ These contain up to five purified B. pertussis antigens, i.e., pertussis toxin (PT), filamentous hemagglutinin (FHA) and pertactin (PRN), with or without fimbrial protein types 2 and 3 (FIM 2/3) (Figure 1).⁵

Figure 1.

Mechanism of infection, antigenic components of vaccines against B. pertussis.

ACT, adenylate cyclase toxin; aP, acellular pertussis; FHA, filamentous hemagglutinin; FIM, fimbriae; PRN, pertactin; PT, pertussis toxin; TCT, tracheal cytotoxin; wP, whole-cell pertussis.

Despite global availability of pertussis vaccines (Table 1) and their inclusion in virtually all national immunization programs, pertussis remains endemic, with epidemics typically occurring every 3 to 5 y.⁶ Since the 1990s, increased numbers of pertussis cases have been seen in many countries, with notable outbreaks in 2012 − 2016, and after the COVID-19 pandemic.^7–10 Indeed, a global increase in pertussis cases was reported during 2023 − 2024.¹¹ Although the disease cannot be eradicated solely by vaccination, it can be effectively controlled if vaccination coverage reaches at least 90%.^3,4,6

Table 1.

Commonly used pertussis-containing vaccines (aP and wP).

Product	Vaccine type	Antigens included	Typical population	Regions typically used
Infanrix hexa® (GSK)	DTaP-IPV-Hib-HepB (aP, hexavalent)	D, T, aP, IPV, Hib, HepB	Infant primary series	Europe, Asia, LATAM
Vaxelis® (Sanofi / MSD)	DTaP-IPV-Hib-HepB (aP, hexavalent)	D, T, aP, IPV, Hib, HepB	Infant primary series	US, Europe
Hexaxim® / Hexyon® (Sanofi)	DTaP-IPV-Hib-HepB (aP, hexavalent)	D, T, aP, IPV, Hib, HepB	Infant primary series	Europe, emerging markets
Pentaxim® / Pentacel® (Sanofi)	DTaP-IPV-Hib (aP, pentavalent)	D, T, aP, IPV, Hib	Infant primary series	Global (Pentaxim) US (Pentacel)
Infanrix-IPV/Hib®	DTaP-IPV-Hib (aP)	D, T, aP, IPV, Hib	Infant primary series	Europe, global
Daptacel® (Sanofi) / Infanrix® (GSK)	DTaP (aP)	D, T, aP	Infant/child primary series	US Global
Boostrix® / Boostrix-IPV® (GSK)	Tdap ± IPV (aP, low dose)	d, T, aP (± IPV)	Adolescent/adult booster, maternal	Global
Adacel® / Adacel-Polio® (Sanofi)	Tdap ± IPV (aP, low dose)	d, T, aP (± IPV)	Adolescent/adult booster, maternal	Global
Repevax® (Sanofi)	Tdap-IPV (aP, low dose)	d, T, aP, IPV	Booster (child/adult)	Europe, Canada
DTwP-containing combinations (e.g., pentavalent DTwP-HepB-Hib) (Serum Institute of India, Bharat Biotech, Bio Farma, others)	DTwP ± HepB ± Hib (wP)	D, T, wP ± HepB, Hib	Infant primary series	Africa, Asia, LMICs
Standalone DTwP (Multiple manufacturers)	DTwP (wP)	D, T, wP	Infant primary series	LMICs

aP, acellular pertussis; D, diphtheria; HepB, hepatitis B; Hib, Haemophilus influenzae type b; IPV, inactivated polio vaccine; LATAM, Latin America; LMIC, low- and middle-income countries; T, tetanus; US, United States; wP, whole-cell pertussis.

Factors driving continued pertussis endemicity

Persistence of pertussis should not be interpreted as a failure of vaccination. Current vaccines remain central to pertussis control and remain critically important for preventing severe disease, particularly in young infants. Rather, ongoing endemicity reflects several factors: (1) extreme contagiousness of pertussis, which has a basic reproduction number (R₀) of 12–17 (so preventing epidemics requires extremely high vaccination coverage); (2) waning immunity; (3) pathogen adaptation; (4) limited mucosal protection; and (5) asymptomatic transmission.^12–15

Firstly, natural immunity after infection is highly variable. Although some studies suggest that protective immunity can last up to 20 y, robust data on its duration remain limited. In terms of humoral responses, although antibodies are not the only component involved in protection, levels of certain antibodies have been reported to decline more than eight‑fold within 3 y.¹⁶ Similarly, vaccine‑induced immunity depends on the vaccine type but generally lasts less than naturally acquired immunity, typically waning between 3 and 10 y after immunization, allowing susceptibility to re‑emerge in older children, adolescents, and adults, and causing recurrent increases in the number of infections.¹⁷ Also, waning immunity is closely linked to pathogen adaptation, as the evolution of B. pertussis toward more immune-evasive strains has shortened the period of effective protection, whether immunity is naturally acquired or vaccine‑induced.¹⁸

Secondly, although current vaccines are highly valuable for preventing disease, in comparison to infection, existing pertussis vaccinations only provide limited mucosal immunity when administered parenterally.^19–21 This may reduce their ability to prevent nasal colonization and onward transmission, even when protection against severe disease is maintained. Thirdly, as highlighted earlier, adolescents and adults infected with B. pertussis often have mild symptoms or may even be asymptomatic, allowing these populations to act as a reservoir to spread infections to individuals in more vulnerable groups.^3,14 This is intensified by suboptimal vaccination coverage rates.²²

A further emerging concern is the ongoing evolution of B. pertussis under immune and antimicrobial selection pressures, which result in two distinct outcomes: partial escape from vaccine‑induced immunity and the emergence of macrolide‑resistant strains. Circulating strains show antigenic variation in vaccine-targeted proteins and, in some regions, increasing macrolide resistance, both of which complicate disease control and emphasize the need for ongoing surveillance. Treatment with macrolides is the first-line option for pertussis, but macrolide-resistant strains of B. pertussis (MRBP) are being increasingly detected in different countries, most commonly associated with a single nucleotide polymorphism in the 23S rRNA gene.²³ In addition, circulation of B. pertussis strains with reduced susceptibility to vaccine-induced immunity has been recorded, with genes encoding for antigens included in pertussis vaccines having been found to be evolving at a significantly increased rate compared to those encoding for other antigens. Post−COVID-19 pandemic increases in pertussis may also have been influenced by reduced circulation of B. pertussis during the COVID-19 pandemic, theoretically leading to so-called “immunity debt”; that is, a reduction in natural immunity boosting due to lack of exposure and thus a lack of immune stimulation within the wider population. Other factors include declines in routine immunization and disruption of healthcare services seen during the COVID-19 pandemic.^11,17 Added to these, continuing improvements in surveillance efforts are also likely to be contributing to an increase in detected cases, with recent increased incidence rates in Europe having been previously attributed to more effective surveillance and case‑reporting frameworks.²⁴

Objective

To understand why pertussis continues to be endemic, this review is organized around the major drivers of its persistence. We outline the immune responses elicited by B. pertussis infection and vaccination, and highlight the challenges of current vaccine approaches. Finally, we discuss how these insights can inform or guide the development of next‑generation pertussis vaccines designed to reduce disease persistence.

The article content was derived from discussions at the most recent meeting of the Global Pertussis Initiative (GPI), which was held in April 2025 in Istanbul, Türkiye, and included the steering committee members plus several local and global pertussis experts. The GPI was established in 2001 to raise the profile of pertussis as an important vaccine-preventable disease, improve understanding of the increasing incidence of reported pertussis, develop and recommend immunization strategies, and advocate for improved disease surveillance.²⁵

Waning immunity after infection and vaccination

Immune response to B. pertussis infection

While vaccination is critical to reduce the likelihood of severe disease and curb persistence, it is important to understand the immune response to direct infection with B. pertussis to determine where susceptibility gaps continue to exist. Research over recent decades has shown that the immune response to B. pertussis infection involves a coordinated effort between innate and adaptive immunity; this evidence has mostly been derived from animal models (particularly murine), but more recently, increasing evidence from human studies has largely confirmed the animal model findings.

Innate immune response to B. pertussis infection

B. pertussis is transmitted from human to human. Infection begins most commonly through nasal acquisition, adhesion to surface cells, and mucosal disruption, followed by inspiration into the lungs, resulting in damage to the ciliated epithelium and compromising mucociliary clearance through the secretion of virulence factors such as tracheal cytotoxin (Figure 1).²⁶ Detection of the pathogen by the epithelium triggers communication within a cellular network, initiating a host innate immune response aimed at limiting the infection. Airway epithelial cells produce host‑defense peptides, in addition to cytokines and chemokines. Chemokines attract other immune cells to the infection site, and cytokines are involved in the activation of these cells, leading to an inflammatory response in the respiratory tract. This activation process includes recruitment of effector cells, such as macrophages, dendritic cells, neutrophils, and natural killer cells, to the site of infection, where they initiate phagocytosis, antigen processing, and antigen presentation.²⁷

Evidence demonstrates that macrophages utilize both nitric oxide-dependent and -independent mechanisms for B. pertussis clearance.^28,29 However, animal models of B. pertussis infection and in vitro studies using human monocytes indicate that B. pertussis has the capacity to enter and survive within macrophages, potentially prolonging infection.³⁰ While a substantial proportion of internalized bacteria are eliminated within acidic compartments shortly after phagocytosis, a fraction can evade destruction and replicate by residing in nonacidic compartments.³¹ Nevertheless, macrophages are recognized for their vital contribution to protective immunity. Macrophage-mediated killing of B. pertussis is augmented by T-cell–derived cytokines, specifically interferon-γ (IFN-γ) and interleukin-17 (IL-17).²⁷ This cytokine-mediated activation suggests the critical involvement of T-helper (Th) type 1 (Th1) and Th17 cells in recruiting neutrophils to the respiratory mucosa and promoting clearance of B. pertussis.^4,32

B. pertussis produces several virulence factors that suppress or subvert these early responses. PT inhibits the production of cytokines and chemokines by alveolar macrophages, thus reducing the recruitment of neutrophils.³³ Adenylate cyclase toxin hemolysin further disrupts innate immunity by impairing phagocyte function, promoting apoptosis of neutrophils and macrophages, and driving differentiation of monocytes into short-lived, less bactericidal phenotypes.³³ Together, these mechanisms allow B. pertussis to establish and maintain early colonization.

B-cell responses to B. pertussis infection

The adaptive immune system orchestrates a robust B-cell response following B. pertussis infection, characterized by the generation of predominantly immunoglobulin G1 (IgG1) antibodies that recognize multiple antigens, such as PT, PRN, FHA, and FIM 2/3 among others, and prevent severe disease.¹⁹ These antibodies include bactericidal antibodies capable of activating the complement system, as well as opsonising antibodies that promote recognition and uptake of B. pertussis by neutrophils.¹⁹ However, the kinetics of this response suggest a limited role for IgG in initial pathogen clearance. B‑cell numbers increase substantially in the lungs during infection, reaching their highest level only when the pathogen is almost cleared. Correspondingly, B. pertussis-specific IgG antibodies are only detectable at significant serum levels coinciding with near-clearance from the lungs.³⁴ This delayed IgG peak suggests that these antibodies may not be critical for clearing a primary infection, but instead play a more prominent role in adaptive immunity induced by prior infection or vaccination. Infection by B. pertussis also leads to the generation of potent anti-B. pertussis immunoglobulin A (IgA) in nasal secretions, which persists for several months after symptom onset and is important for mucosal immunity.^35,36 In contrast to IgG, IgA is detected earlier in the lungs, approximately 2 weeks post-challenge.³⁷

Despite the comprehensive antibody response, B. pertussis has evolved strategies for immune evasion.¹⁹ It produces complement-evasion factors, such as Bordetella resistance to killing A (BrkA) and virulence-associated gene 8 (Vag8), which binds to C1-inh, enabling escape from complement-mediated clearance. Similarly, B. pertussis has developed strategies for evading neutrophil-mediated opsonophagocytosis.¹⁹

B cells may also contribute to protective immunity through mechanisms beyond specific antibody production. Studies have highlighted the complex nature of this protection. Although Ig^−/− mice infected with B. pertussis developed chronic infection and failed to clear the pathogen, these mice also lack mature B cells and consequently failed to generate an effective T-cell response.³⁸ This observation suggests that B cells may contribute to the induction of memory CD4⁺ T cells, potentially by acting as antigen-presenting cells.^39,40

Regarding long-term immunity, antibody levels gradually decline over time following B. pertussis infection, with antigen-specific kinetics. Anti-PT IgG levels wane most rapidly, whereas anti-PRN IgG and anti-FHA IgG persist longer.¹⁹ Secretory IgA usually declines to low levels after 6 months.³⁶ Beyond quantity, qualitative antibody features, including avidity, neutralizing capability, and epitope specificity, also influence protective immunity.⁴¹ Memory B-cell levels are detectable following B. pertussis infection, peaking during the acute phase (within approximately 1.5 months) and decreasing thereafter, although they remain detectable at lower levels for at least 9 months.

T-cell responses to B. pertussis infection

Experimental evidence from animal models and observational studies in humans has shown that CD4⁺ T cells are essential for protective immunity against B. pertussis infection. IFN‑γ‑producing Th1 cells are induced during infection, and adoptive transfer of these cells confers protection in mice. Conversely, mice lacking the IFN‑γ receptor fail to clear the bacteria, which disseminates from the respiratory tract. Long‑lived B. pertussis‑specific CD4⁺ T cells have been identified in baboons and humans. Peripheral blood CD4⁺ T cells from convalescent children produce IFN‑γ upon antigen stimulation, and IL‑17‑producing CD4⁺ T cells are also induced in infected mice and baboons.²¹

A key limitation of many studies is their focus on circulating rather than tissue‑resident T cells. In mice, CD4⁺ T cells expressing CD44 and CD69 and secreting IL‑17 or IFN‑γ accumulate in the lungs and nasal mucosa as tissue‑resident memory T cells during infection and persist after bacterial clearance. These tissue‑resident memory T cells are first responders to re‑infection and mediate rapid bacterial clearance. IL‑17‑secreting γδ T cells also contribute to the protective response. Studies using IL‑17‑deficient mice or antibody blocking have demonstrated that IL‑17‑secreting CD4⁺ tissue‑resident memory T cells are essential for clearing nasal infection. In addition, B. pertussis infection induces IL‑10‑secreting regulatory T cells that suppress Th1 and Th17 responses, likely as a bacterial strategy to prolong infection.⁴²

Together, these data highlight that infection-induced immunity is heterogeneous and time-limited. Waning antibody and memory responses create recurrent susceptibility at the individual level and contribute to periodic increases of cases at the population level.

Vaccination-induced immunity

Vaccination vs. B. pertussis infection

Beyond the immune response to infection with B. pertussis, vaccination is the primary intervention designed to control the spread of disease. However, the immune response induced by vaccination differs qualitatively and quantitatively from that induced by natural infection.⁴³ The duration of protection provided by pertussis vaccination is generally shorter than that following B. pertussis infection and disease.⁴⁴ The immune response to vaccination also differs qualitatively.¹⁹ Parenteral vaccination induces more limited IgA and mucosal B-cell responses compared with infection, which robustly stimulates mucosal immunity.¹⁹ Differences in the duration and quality of immunity induced by aP and wP vaccines are thought to reflect differences in T-cell polarization, with aP vaccines favoring a Th2 (T-helper type 2) response and wP vaccines inducing a Th1/Th17 response with an increased CD4⁺ tissue-resident memory cell population.⁴⁴

Overall, despite differences in the immune responses they trigger, both wP and aP vaccines effectively reduce the risk of pertussis disease, severe illness, and death. Nevertheless, despite these benefits, B. pertussis transmission is not fully prevented, highlighting the need for improved next‑generation vaccines.

Impact of primary vaccination and natural boosting on immune responses later in life

The long‑term effects between wP and aP for primary vaccination, as well as the role of subsequent natural exposure to B. pertussis, can be observed when analyzing immune responses in adolescents and adults. Previous studies suggest that the nature of primary vaccination can influence subsequent immune responses to pertussis vaccination later in life, including responses to booster vaccination. Adults exhibit broad T-cell responses to aP antigens and to non-aP antigens, regardless of whether they received primary wP or aP vaccination in infancy. However, the functional profile of these responses appears to be influenced by the initial priming event. System-level analyses indicate that wP elicit stronger early innate immune activation and are associated with enhanced PT‑specific B-cell responses compared with aP vaccination.⁴⁵ Consistent with this, one experimental study showed that the response to a wP booster vaccination was determined by priming with either an initial wP or aP vaccine.⁴⁶ The results showed that B-cell responses and the resulting antibody generation were adjustable, with wP booster vaccines inducing a broader isotypic diversity of antibodies when an initial wP priming vaccine was used, rather than an aP priming vaccine.⁴⁶

Similarly, adolescents primed with wP vaccines in infancy tended to have higher memory B-cell levels before and after an aP booster vaccination, although the differences were not statistically significant and the sample sizes were small.⁴⁷ In contrast, aP-primed individuals generated higher proportions and concentrations of IgG4 antibodies following booster vaccination.⁴⁸ As IgG4 has limited ability to activate complement or engage Fcγ receptors, this may have implications for functional antibody activity.⁴⁸

Individuals primed with aP vaccine who have developed hybrid immunity due to subsequent exposure to B. pertussis (possibly subclinical) seem to respond better to aP booster vaccines than aP-primed individuals who showed no evidence of infection with B. pertussis.⁴⁹ Previous B. pertussis infection also appears to impact on IgA responses to vaccination. In a study evaluating the response to aP vaccination in children, adolescents, young adults, and older adults, concentrations of IgA to pertussis antigens increased with age.⁵⁰ These data indicate a form of hybrid immunity, whereby B. pertussis infections post-primary vaccination induce antigen-specific IgA memory B cells, which can then be boosted by parenteral aP vaccination later in life.

Impact of vaccination in pregnancy on subsequent immune responses to pertussis vaccination

Initial concerns about vaccination during pregnancy included the potential for blunting of infant responses to primary DTaP vaccination. While there is no evidence of increased pertussis prevalence among these children, a meta-analysis (n = 1,884) reported lower IgG responses to pertussis antigens, including PT, FHA, Prn, and Fim2/3, after primary DTaP vaccination in infants born to Tdap-vaccinated mothers compared with infants of unvaccinated women.^33,51–53 Evidence suggests that this effect may be more apparent in infants receiving wP rather than aP-containing primary series, although higher responses to non-aP antigens in wP-primed infants suggest a different immune profile rather than a uniformly weaker response.⁵⁴

Blunting has mainly been reported for post-primary antibody concentrations, with limited data on effects on memory B- or T-cell responses, and reduced antibody levels should not be assumed to translate directly into increased susceptibility. This distinction is important because protection before primary vaccination depends largely on maternally transferred antibodies, whereas protection after primary vaccination is likely to reflect both antibodies and immune memory. Findings also vary across settings, with studies in Vietnam⁵⁵ and South Africa⁵⁶ showing limited interference with infant responses to hexavalent aP-containing vaccines. Further studies are needed to determine the clinical significance of these findings as well as clarify the mechanisms and longer-term implications for immune memory and clinical protection. Across studies, despite some reductions in antibody levels and PT-specific memory B-cell responses, overall immune responses appear broadly preserved.^57–59

Nevertheless, the safety and effectiveness of pertussis vaccination in pregnancy are widely supported,⁶⁰ and the clear reduction in pertussis morbidity and mortality in early infancy following vaccination in pregnancy, when the disease burden is highest, far outweighs any uncertain clinical relevance of any potential immunological blunting effects.⁵⁷

Mucosal protection and nasal colonization

A key consideration in understanding pertussis persistence is the difference between protection from severe disease and the ability to prevent infection and onward transmission. Current vaccines are highly important for reducing clinical disease, particularly severe infant disease, but induce less mucosal IgA and tissue-resident memory responses than natural infection. Hence, vaccinated individuals may remain protected against severe outcomes while still harboring B. pertussis and potentially transmitting it.

T-cell responses to B. pertussis infection

Cell-mediated immunity plays a central role in protection against pertussis. The adaptive immune response to B. pertussis infection involves the recruitment of T cells to the respiratory tract, which is essential for mediating bacterial clearance. Studies in mice confirm that CD4⁺ T cells are key to clearance.³⁷ Mice lacking all T cells develop a persistent or lethal infection,²⁷ which is reversed by the adoptive transfer of immune T cells. Infection induces immune responses characterized by Th1 (IFN-γ) and Th17 (IL-17) cells, which are required for bacterial clearance and promoting neutrophil recruitment, respectively.²⁷ CD4⁺ T cells from infected mice produce IFN-γ and/or IL-17, confirming this Th1/Th17 bias.²⁷ Experimental evidence supports this: IFN-γ^−/− mice show impaired bacterial clearance,⁶¹ and IL-17^−/− mice have significantly greater bacterial loads, correlated with reduced neutrophil recruitment.⁶² Mucosal IL-17–producing Th17 cells are also important in providing sustained local protection against nasal colonization of B. pertussis.^63,64 These findings collectively demonstrate that both Th1 and Th17 cells contribute significantly to protective immunity against primary B. pertussis infection.

Furthermore, recent evidence has highlighted the critical role of tissue-resident memory CD4⁺ T (T_RM) cells in the respiratory tract. In murine models, B. pertussis infection generates a robust population of CD4⁺ T_RM cells in the lungs and nasal mucosa that secrete IFN-γ and IL-17. These cells are essential for rapid bacterial clearance and provide high and long-term protection against nasal colonization.^65,66 The importance of these cells is corroborated by findings in a baboon model, which closely mimics human infection and transmission. In baboons, natural infection induces a potent and durable Th1/Th17 response that not only protects against clinical disease but also prevents nasal colonization and subsequent transmission, a key immunological feature that differentiates infection-induced immunity from that provided by current aP vaccines.²¹ A complicating factor is that B. pertussis also induces regulatory T cells (Tregs), which can dampen the protective immune responses and may facilitate prolonged bacterial survival within the host.²⁷

Together, these findings provide an immunological explanation for why pertussis can persist despite high vaccine coverage and substantial protection against severe disease; they also support the rationale for mucosal immunity as a central target for future vaccine development.

Age-related responses and implications for persistence

Mild, atypical, or asymptomatic infection in adolescents and adults is a major contributor to pertussis persistence because these groups can transmit B. pertussis to infants and other vulnerable individuals while remaining undiagnosed. This reservoir effect is amplified by waning immunity and by age-related differences in immune responses.

Antibody responses to B. pertussis infection vary by age, with distinct IgG and IgA profiles observed across 11,386 US serum samples (2008–2010). While the overall proportion of individuals with elevated anti-PT IgG and IgA was similar across those aged ≤60y, younger individuals (aged <21y) predominantly showed IgG-only responses, whereas older individuals (aged ≥21y) more commonly exhibited combined IgG and IgA responses, particularly to FHA. Anti-PT IgG declined rapidly across all ages, while IgA responses (to PT, FHA, PRN, and FIM 2/3) waned more quickly in younger than older individuals. Peak anti-PT and anti-FHA IgG responses were higher in adults, whereas anti-PRN IgG responses were higher in younger populations. Anti-FIM 2/3 IgG responses were markedly lower in older adults. IgA responses were strongly age-dependent, with anti-PT and anti-FHA IgA levels being ~19-fold and ~76-fold higher, respectively, in older adults compared with infants aged <4y.

T-cell responses to B. pertussis also vary by age. A recent study compared cytokine activity, proliferative response, and phenotypical profiles of CD4⁺ T-cells in response to specific B. pertussis antigens across adolescent (11–15 y of age) and adult (25–56 y of age) groups in the Netherlands. The results showed that the adult group had lower IFN-γ levels and lower frequencies of anti-PT IFN-γ producing cells in the early phase following infection, reduced secretion of Th-type cytokines after infection, and a reduced proliferative response of B. pertussis-specific CD4⁺ T-cells than the adolescent group. Only minor phenotypical differences were observed in proliferated CD4⁺ T-cells between the two age groups, which the authors suggested were indicative of a reduced pool of B. pertussis-specific memory CD4⁺ T-cells in the adult group. However, similar frequencies of naïve and memory CD4⁺ T-cells, as well as Tregs, were observed between the two age groups.⁶⁷ The results of an earlier study have shown higher maintenance levels of B. pertussis-specific memory CD4⁺ T-cells in late childhood versus infancy, irrespective of booster vaccination status,⁶⁸ but studies directly comparing T-cell responses between infants, children, adults, and older adults remain limited.

Pathogen adaptation under immune and antimicrobial selection pressure

Evolution of B. pertussis strains

Among the proposed explanations for pertussis endemicity, the evolution of circulating B. pertussis strains, driven by vaccine-induced selection pressure, is a key consideration when evaluating current vaccine effectiveness and identifying areas for further research and development.⁶⁹ Comparative genomic analysis of isolates collected worldwide between 1920 and 2010 demonstrated that new strains rapidly spread globally and that their evolution coincided with the introduction of pertussis vaccines.⁷⁰ Since various mutations occur in the genomes of circulating B. pertussis strains, we focus here mainly on those encoding antigens included in aP vaccines and on those with significant mutations or changes.

Antigenic variation and immune selection

According to the 2022 European Center for Disease Prevention and Control (ECDC) Technical Report,⁷¹ 13 alleles of the ptxA gene have been identified (named ptxA1 to ptxA13). Excluding silent single-nucleotide polymorphisms, these alleles encode seven distinct PtxA protein variants, with ptxA1 being the most common, particularly in Europe. In the Netherlands, variations in the ptxA gene were observed in two regions of the bacterial genome, previously identified as T-cell epitopes.⁷² Similar findings were subsequently reported in other countries, including Finland,⁷³ Italy,⁷⁴ the United States,⁷⁵ and Australia.⁷⁶ Vaccine effectiveness studies to date in some of these countries have not indicated a diminished capacity of aP-induced immunity to protect against disease.

In 2009, B. pertussis strains circulating in the Netherlands were found to carry a novel allele of the PT promoter (ptxP3) region, associated with increased PT production.⁷⁷ Similar findings were subsequently reported in other countries, including Sweden,⁷⁸ Australia,⁷⁹ the United States,⁸⁰ and Finland.⁸¹ According to the 2022 ECDC technical report, 19 ptxP alleles have now been identified.⁷¹ In the pre-vaccination era, isolates with ptxP1 and ptxP2 alleles were most common, while ptxP3 now predominates in Europe.⁷¹ Although temporal associations between the emergence of ptxP3 lineages and increases in pertussis notifications have been described, current evidence does not indicate a substantial reduction in vaccine protection attributable to ptxP3 itself.^77–79 aP vaccines continue to provide good protection against severe disease in infants, although effectiveness against symptomatic disease and infection in older age groups wanes over time, irrespective of ptxP genotype.^{78,79,81–84}

Eighteen prn alleles have been identified (prn1 to prn18),⁷¹ with prn2 being the predominant subtype in Europe. Individuals infected with prn2 strains were previously shown to have very low antibody levels against the variable region of prn1, the strain included in pertussis vaccines.⁸⁵ Evidence indicates that the emergence and spread of PRN-deficient B. pertussis strains are associated with vaccine-driven selective pressure. This is supported by studies showing that the dominant circulating strains in different countries often reflect their respective vaccination strategies.^86,87 PRN-deficient strains have now been identified worldwide, apart from Africa.^87–95 The diversity of these alleles has continued to expand globally, with recent reports describing the emergence and spread of the prn150 allele in China.⁹⁶ Selective pressure driving PRN deficiency appears associated with aP vaccine use. This is evidenced by surveillance data from Argentina and Brazil, where wP-based primary schedules remain in place and PRN-deficient strains have rarely been detected.^97,98 Although the underlying mechanism conferring a selective advantage is not fully understood, experimental data indicate that aP vaccines generate bactericidal antibodies that are only directed against PRN,⁹⁹ and PRN deficiency is hypothesized to reduce vaccine-mediated opsonophagocytic clearance and alter the host immune response.⁹³ Nevertheless, despite evidence of vaccine-driven selection pressure, it is important to note that PRN-deficient strains have also appeared in countries using predominantly non-PRN‑containing aP vaccines.

Temporal trends in these genetic shifts indicate that they are vaccine‑driven.¹⁰⁰ For example, B. pertussis strains in many countries carrying non-vaccine ptxA and prn alleles emerged approximately 15–30 y after the introduction of wP vaccines.¹⁰⁰ The more targeted selective pressure from aP vaccines subsequently accelerated this timeline, with PRN-deficient strains becoming predominant within approximately 8 y after the introduction of exclusive aP vaccination schedules.⁸⁸

Emergence of antimicrobial resistance

Macrolides are the first-line antimicrobials used for the prophylaxis and treatment of pertussis.²³ The first MRBP was reported in the United States in 1994.¹⁰¹ Since then, additional cases have been reported in several countries. In China, circulating MRBP have now become the dominant phenotype, and resistant strains are spreading to other countries.²³ For example, MRBP emerged in several European countries during the pertussis outbreaks of 2024.^102,103 In August 2025, the Pan-American Health Organization issued a public health alert urging strengthened pertussis vaccination in response to the spread of antibiotic-resistant B. pertussis strains in the Americas.¹⁰⁴

Macrolide resistance in B. pertussis is primarily caused by the A2047G mutation in the 23S rRNA gene, which affects the macrolide binding site.²³ Rapid identification of this mutation is important to guide the appropriate use of antimicrobials, especially in countries where the prevalence of MRBP is increasing, but testing for macrolide resistance is not routine in all countries and settings. Wheezing has been shown to be a more common symptom in Chinese children infected with MRBP when compared with those with macrolide-sensitive B. pertussis.¹⁰⁵ While symptoms are otherwise expected to be the same, infection with MRBP may lead to illnesses that are more severe or of lengthened duration due to first-line treatment failure,¹⁰⁶ necessitating the use of alternative treatment options such as trimethoprim-sulfamethoxazole,¹⁰⁷ as well as increased utilization of healthcare resources. There is also a risk of increased spread of infection due to prescribed prophylactic treatment which is insufficient to control the disease in household or other contacts, where MRBP is not initially identified in the primary case.¹⁰⁶ The emergence and spread of MRBP further underline that vaccination remains the most effective strategy to prevent disease, including that caused by MRBP.

Overall, these data indicate that pertussis endemicity is significantly driven by the evolution of B. pertussis under immune selection pressure from vaccines and, in some regions, antibiotic use.⁸²

Implications for vaccine development and pertussis control

Research focus

Vaccination remains the cornerstone of public health protection, with a continued focus on the expansion of vaccination coverage. Notably, research into pertussis vaccination declined after 2020, likely reflecting a change in focus toward the COVID-19 pandemic and outbreaks of other diseases.¹⁰⁸ While current pertussis vaccines are highly effective and provide protection for several years, particularly against severe disease, additional strategies to address their limitations, extend protection, and reduce transmission may help further limit endemicity and outbreaks (Table 2). In this context, research priorities should target the main drivers of persistence identified earlier: waning immunity, pathogen adaptation limited mucosal protection, and asymptomatic transmission.

Table 2.

Strengths and limitations of current pertussis vaccines (aP and wP).

	Strengths	Limitations
Protection against disease	Highly effective at preventing pertussis-related severe disease, hospitalization, and death, particularly in infants	Protection against mild pertussis disease is less effective
Impact on public health	Major reductions in pertussis-related global morbidity and mortality since introduction of vaccination programs	Pertussis disease persists endemically with periodic outbreaks
Duration of protection	Provides several years of protection against pertussis (typically ~5–10 y) after primary series and boosters	Pertussis immunity wanes over time, leading to susceptibility in older children, adolescents, and adults
Safety and tolerability	aP vaccines have an excellent safety profile and are well tolerated, supporting high uptake	wP vaccines induce broader immunity but are more reactogenic, limiting their use in many settings
Immune response profile	Strong systemic antibody responses (IgG), effective at preventing severe pertussis clinical outcomes	aP vaccines induce limited mucosal immunity (IgA) and tissue-resident memory responses compared with natural pertussis infection or wP vaccines
Transmission and colonization	Contribute to reduced pertussis transmission through decreased symptomatic disease	Do not fully prevent pertussis infection or nasopharyngeal carriage, allowing ongoing transmission
T-cell responses	Induce protective immune responses against pertussis, particularly with boosting	aP vaccines tend to favor Th2 responses, with less robust Th1/Th17 responses compared with wP or natural infection
Use in pregnancy	Vaccination in pregnancy with aP vaccines is safe and highly effective in protecting young infants during the highest-risk period	Although clinical pertussis protection remains preserved, there may be some blunting of infant antibody responses to their primary series
Life-long protection	aP vaccines can be used across the life course (infant, booster, and pregnancy vaccination programs)	Requires booster doses to maintain protection against pertussis over time
Coverage against different strains	Continue to provide good protection against severe pertussis disease despite circulating strain variation	Antigenic divergence (e.g., PRN-deficient pertussis strains) and pathogen adaptation may affect some immune responses
Antimicrobial resistance context	Vaccination reduces pertussis disease burden and reliance on antibiotics	Does not directly address emergence of macrolide-resistant B. pertussis

aP, acellular pertussis; IgG, immunoglobulin G; Th, T helper cell; wP, whole-cell pertussis.

Future vaccine development should therefore aim to complement, rather than replace the public health achievements of current vaccines. The key objective is to retain the safety, tolerability, and disease-prevention benefits of existing aP vaccines while improving durability, mucosal protection, and breadth of coverage against evolving strains. Thus, the hypothetical ‘ideal’ pertussis vaccine would combine the safety and tolerability of existing aP vaccines with the following properties, to protect not only against severe disease but also reduce colonization and transmission of B. pertussis and increase the duration of vaccine-mediated immunity, thereby limiting the persistence of pertussis in the general population and minimizing the possibility of future outbreaks:^19,109

• Polarization of T-cell responses to a Th1/Th17 phenotype.

• Induction of IgA and tissue-resident memory T and B cells in the nasal mucosa.

• Generation of antibodies that block adhesion and bactericidal antibodies.

• Induction of sufficient PT-neutralizing antibodies (the high effectiveness of existing vaccines in infants suggests that they appear to do so).¹¹⁰

• A broader antigen repertoire.

The lessons learned from T-cell studies in mouse models and humans suggest that next-generation pertussis vaccines should be designed to induce CD4⁺ T cells with both a central memory cell and tissue-resident memory cell phenotype, as well as a mixed Th1/Th17-cell profile, closely resembling the robust response induced by B. pertussis infection. An important factor in both B. pertussis infection and wP vaccines that steers these protective adaptive immune responses is the presence of pathogen-associated molecular patterns (PAMPs).¹¹¹ These PAMPs activate innate cells via pattern‑recognition receptors and thereby promote dendritic‑cell maturation and the production of pro-inflammatory cytokines that direct the optimal induction of Th1 and Th17 cells.¹¹²

The following strategies have been explored in the development of novel pertussis vaccines with these properties:^113–115

• Enhancing existing aP vaccines by adding several additional immunogens (e.g., AC-Hly, Vag8, or outer membrane vesicles and/or new mucosal adjuvants.

• Using alternative vaccination routes, such as intranasal administration.

• Implementing prime-boost regimens combining intramuscular priming followed by an intranasal booster vaccine.

• Developing safer and more tolerable wP vaccine alternatives, including improved wP vaccines with better tolerability or outer membrane vesicle-based vaccines. Outer membrane vesicles represent a promising wP-like alternative, as they retain key PAMPs for potent innate immune activation while demonstrating an improved safety profile.

In particular, live-attenuated vaccines or aP vaccines formulated with novel adjuvants that induce respiratory IgA and tissue-resident memory cells, especially when delivered by the nasal route, are in development and have considerable promise as next-generation vaccines against pertussis.¹¹⁶

Research challenges

Several challenges complicate the development of novel pertussis vaccines. First, while the mouse model remains a valuable and accessible tool for studying immune responses to vaccination and B. pertussis challenge, it does not fully reproduce the symptomatic disease or the complete immune profile observed in humans.¹¹⁷ Second, although the baboon model provides excellent translational data, its high cost presents a significant barrier; making it more accessible will require the formation of specialized consortia to share resources and expertise. Third, the bacterial challenge strains used in human models need optimization. For instance, while the European strain Bp1917 is genetically representative of circulating populations, it is notoriously fastidious to culture and often yields heterogeneous growth, complicating its standardized use in challenge studies.

A fourth challenge is the lack of well-defined immune correlates of protection against B. pertussis infection.¹⁹ Arbitrary cutoffs for anti-PTx antibody values are often used,¹⁹ but protection is multifactorial and depends on both humoral and cellular immunity at mucosal surfaces and within the systemic compartment.⁴¹ It is also important to appreciate the distinction between prevention of infection and transmission and the prevention of severe disease only when selecting suitable correlates, as some proposed biomarkers may only indicate a history of infection whereas others demonstrate the strength of immune response. Previously proposed candidates for correlates of B. pertussis infection include IgG positivity,¹¹⁸ mucosal IgA positivity¹¹⁹ and T-cell responses.¹²⁰ A recently published study suggests that the use of a combination of correlate biomarkers may be more effective in determining long-term responses to pertussis vaccination, and this approach should be further explored.¹²¹

Another important challenge is the feasibility of clinical development and subsequent registration of pertussis vaccines that show promise in preclinical models. Despite substantial research activity in murine systems, only two candidates have entered clinical development.¹⁰⁹ Additional challenges, which also apply to other vaccines, include the complexity of manufacturing processes, optimization of assays to measure immune response, and the design and size of the study population required to demonstrate an acceptable benefit–risk ratio for recipients, parents, regulators, ethics committees, and policymakers.¹²²

Limitations

This study was a review of the current literature and a synthesis of information related to vaccine development against B. pertussis. As such, it was focussed on the evolution of B. pertussis strains, immune escape and challenges within vaccine development. It did not seek to be a systematic literature review of one of these areas, but rather offer expert insight into current trends.

Future considerations

Given the severity of pertussis in infants and the role of older age groups who have not received booster vaccinations as reservoirs of infection, continued action is needed not only to maintain protection against disease but also to reduce B. pertussis colonization and onward transmission. While pertussis vaccines have dramatically decreased the incidence, morbidity, and mortality of B. pertussis infection, they do not provide life-long protection against pertussis.

In the present and the immediate future, while next-generation vaccines continue to be developed, control of pertussis relies on ensuring timely and complete pediatric vaccinations, optimizing the availability of and compliance with vaccination in pregnancy, and establishing longer-term protection through regular booster vaccinations. Vaccination of infants and young children in a timely manner and with high coverage remain the most critical aspects for pertussis control; however, given evidence that immunity wanes over time, expanding booster vaccination programs into adolescence and adulthood is needed to fill the individual-level immunity gaps and reduce transmission.^123,124 Vaccination in pregnancy to protect neonates and infants before they can be immunized should be universally adopted. Central to each of these approaches is the continued education of healthcare workers on recognition, diagnosis, and vaccination schedules, as well as clear communication to the public on the importance of vaccination.^125–127

The sociocultural context of vaccination, including the presence of vaccine hesitancy and the spread of anti-vaccination views, is also an important consideration for policymakers when planning pertussis control strategies. Historical evidence suggests that such factors have, at times, been associated with disruptions to national immunization programs and subsequent increases in pertussis incidence.

A deeper understanding of the immune response to B. pertussis infection and how this differs from responses induced by existing vaccines is key to the development of less reactogenic and more immunogenic pertussis vaccines that provide longer-term protection; this is a goal that has been addressed by the PERtussIS Correlates Of Protection Europe (PERISCOPE) consortium.¹²⁸ This further enables efforts to develop innovative vaccines and vaccination strategies against pertussis, recognizing that different formulations, routes of administration, schedules, and vaccine combinations may be needed to confer better immunity. There is also a need for better long-term surveillance of B. pertussis strain evolution and how these changes impact vaccine effectiveness over a life-time. Altogether, this increased understanding of long-term immunity to B. pertussis can inform future healthcare policy decisions for the prevention of pertussis, with the aim of disrupting the endemicity of B. pertussis.

Conclusion

Global pertussis vaccination policies have helped reduce disease severity and transmission through vaccine-induced immune responses, yet long-term control of B. pertussis remains a persistent public health challenge. This is due, in part, to waning immunity, limited mucosal protection, and ongoing transmission from mild or asymptomatic infections. The evolution of B. pertussis under immune and antimicrobial selection pressure may contribute to immune and treatment escape. Taken together, these challenges highlight the importance of continued surveillance, tailoring immunization policies, and potential vaccine or antibiotic adaptation. While important scientific and translational challenges remain, we primarily recommend optimizing current vaccination strategies alongside developing next-generation vaccines that improve durable, Th1/Th17-polarized immunity to enhance control of pertussis in the future.

Biography

Mine Durusu Tanrıöver completed her internal medicine training in 2005 and served as the medical consultant for acute patients for several years. She has led numerous initiatives within medical societies and has undertaken leadership roles in national and international postgraduate medical education and professional development projects. She is currently the Director of the Vaccine Institute at Hacettepe University in Türkiye, where she leads postgraduate programs, multidisciplinary research projects, and policy-oriented initiatives in vaccinology. Her academic work focuses on adult vaccination, vaccine-preventable diseases, influenza and other respiratory infections, vaccine confidence, and health system quality improvement. She actively contributes to regional and global scientific collaborations on lifelong immunization policies and the prevention of respiratory‑pathogen‑related diseases, bridging clinical medicine, public health, and health systems strengthening. Prof. Tanrıöver is a strong advocate for planetary health, integrating climate‑change and health perspectives into medical education and health policy discussions. She is involved in projects that aim to enhance climate-resilient health systems, promote sustainable health infrastructure, and build capacity for climate and health training. Through her work, she seeks to align preventive medicine, immunization strategies, and environmental sustainability within a coherent planetary health framework.

Funding Statement

The Global Pertussis Initiative is supported by Sanofi, and Sanofi funded the expert meeting. The authors received no payment from Sanofi related to the development of this publication. Sanofi had the opportunity to review the publication; however, the authors remain responsible for all content, editorial decisions, and the decision to submit the manuscript.

Disclosure statement

The Global Pertussis Initiative, established in 2001, is supported by Sanofi and aims to assess ongoing global challenges related to pertussis and to recommend appropriate control strategies.

M.D.T., K.F., R.O.F., D.F.H., and A.O.L have previously received honoraria from Sanofi for participation in previous live meetings. Q.H. reports no other potential conflict of interest. D.D. reports funding to their institution from industry for conducting pertussis vaccine studies in adults. D.M. is a lecturer for Sanofi and Merck & Co., and a physician advisor for Sanofi, Pfizer, and Merck & Co. R.M. has received honoraria for educational activities from MSD and Sanofi. F.M. has served as a speaker for Sanofi and consultant for Novartis, GSK, and Novavax. She has also conducted clinical trials sponsored by Roche, GSK, and Gilead. T.Q.T. has received grants from Merck and Sanofi, personal fees from GSK Biologicals and Sanofi, and honoraria from Sanofi. U.H. has received honoraria for participation in previously associated live meetings from Sanofi, as well as receiving honoraria for educational activities from GSK, Pfizer, MSD, Sanofi, and InfectoPharm.

Abbreviations

AC-Hly

adenylate cyclase toxin

aP

acellular pertussis

B. pertussis

Bordetella pertussis

BrkA

Bordetella resistance to killing A

C1q

complement component 1q

COVID-19

Coronavirus disease 2019

ECDC

European Center for Disease Prevention and Control

EU/EEA

European Union/European Economic Area

FHA

filamentous hemagglutinin

FIM 2/3

fimbrial protein types 2 and 3

GPI

Global Pertussis Initiative

IFN-γ

interferon-γ

IgA

immunoglobulin A

IgG1

immunoglobulin G1

IL-17

interleukin-17

MRBP

macrolide-resistant strains of B. pertussis

PAMPS

pathogen-associated molecular patterns

PRN

pertactin

PT

pertussis toxin

Th1

T-helper type 1

Th2

T-helper type 2

Th17

T-helper type 17

Tregs

regulatory T cells

T_RM

tissue-resident memory

US

United States

Vag8

virulence-associated gene 8

wP

whole-cell pertussis

Data availability statement

This review of previously published literature does not involve the collection of any data.