A comprehensive analysis of serotype-specific invasive capacity, clinical presentations, and mortality trends of invasive pneumococcal disease

Abstract

Background:

Pneumococcal conjugate vaccines (PCV) reduced invasive disease, but the overall prevalence of pneumococcal nasopharyngeal colonization among children has not changed significantly. Our knowledge of which serotypes, once colonized, hold a higher likelihood to cause invasive disease is limited.

Methods:

Serotype-specific invasive capacity (IC) of Streptococcus pneumoniae was estimated using an enhanced population-based invasive pneumococcal disease (IPD) surveillance in children <7 years of age in Massachusetts and surveillance of nasopharyngeal (NP) colonization in selected Massachusetts communities in corresponding respiratory seasons. Serotype-specific IC was calculated by dividing the annual incidence of IPD by the carriage prevalence for each serotype. Serotype-specific relationship between NP carriage and IPD was evaluated by year, age group (<24 months vs 24–84 months), pre-PCV13 (2003/04, 2006/07, 2008/09) vs post-PCV13 (2010/11, 2013/14, 2015/2016) periods, clinical presentation, and outcome.

Results:

A total of 293 IPD and 1602 NP isolates were included in the analysis. Most common IPD serotypes were 19A (34.1 %), 7F (9.2 %), 15 BC (8.9 %), 3 (5.8 %), and 22F (4.8 %). Serotypes 18C, 38, 7F, 19A, 3, 22F, and 33F displayed a higher propensity to cause IPD once colonizing the nasopharynx compared to 11A, 35B, 6C, and 21. Serotype-specific IC was generally lower in children older than 24 months. During the study period, we observed shifts in the dominant serotypes in relation to IC as well as changes between pre- to post-PCV13 era. Except for serotypes 14, 6A, 7F, 11A, 23A, 20, 35F, 7C, 6C and 15F all serotypes presented primarily as bacteremia. Pneumonia was attributed to serotypes 14 and 20; serotypes 35B, 23B, and 11A were responsible for the highest percentage of deaths.

Conclusion:

This study highlights the need for continued serotype-specific surveillance to better understand the disease potential of emerging serotypes and to guide optimal vaccination strategies.

1. Introduction

Streptococcus pneumoniae (SPn) is major cause of morbidity and mortality among children globally [1,2]. In the United States, pneumococcus is commonly carried in the nasopharynx (NP) of school-aged children, with a prevalence of approximately 20–60 % among otherwise healthy individuals [3]. The epidemiology of pneumococcal colonization varies with age and geographical location with the highest rates among younger children [4]. Carriage is a prerequisite for invasive pneumococcal disease (IPD), yet only a small percentage of the colonization episodes result in invasive disease [5]. The propensity to progress to invasive disease following colonization can vary depending on multiple factors such as temporal opportunity due to repeated carriage or pathogen biological mechanisms allowing the ability to overcome host immune defense mechanisms or co-infection with respiratory viruses [5,6].

Pneumococcus utilizes different virulence factors including polysaccharide capsules, toxins such as pneumolysin, surface-associated proteins such as LPXTG anchored surface proteins and chlorine binding proteins, and enzymes such as autolysin to escape the host defenses [7–9]. Polysaccharide capsules are the most important virulence factor, playing roles in bacterial adherence and prevention of phagocytosis by host immune cells [8]. These capsules are the outermost layer of SPn and determine the specific serotype, which is the main target of vaccination [10].

Pneumococcal conjugate vaccines (PCV) provide serotype-specific protection against serotypes included in the vaccine with variable cross protectivity within serogroups [2]. Seven-valent pneumococcal conjugate vaccine (PCV7), approved in 2000 included serotypes 4, 6B, 9 V, 14, 18C, 19F, and 23F and was followed with other multivalency vaccines such as PCV13 including additional serotypes PCV7 plus 1, 3, 5, 6A, 7F, and 19A, PCV15 (PCV13 plus 22F and 33F), and PCV20 (PCV15 plus 8, 10A, 11A, 12F, and 15B) [11,12]. There are other PCVs, such as PCV21 (10 serotypes of PCV20 plus 9 N, 17F, 20, 15A, 15C, 16F, 23A, 23B, 24F, 31 and 35B), have recently been approved, and PCV24 (PCV20 plus 2, 9N, 17F, 20B) is currently in development [12,13]. Implementation of PCVs has been a very effective public health intervention and widespread use of PCVs was associated with significant reductions in the burden of vaccine-type IPD (VT-IPD) [6].

While carriage rates have stayed relatively similar to the pre-PCV13 era, the distribution of carriage serotypes has changed, with a significant increase in non-vaccine serotypes (NVST) [6]. A current hypothesis is that these NVSTs may have a decreased invasive capacity compared to the vaccine serotypes that have been widely eliminated through vaccination [6]. Further understanding of the serotype-specific invasive capacity and whether and how this differs across time, age groups, and clinical syndromes will be important to optimize future vaccine candidates and vaccination strategies.

There are other studies that have used this pneumococcal surveillance data, each contributing valuable insights into serotype prevalence, invasive capacity, and epidemiological patterns [5,6, 14–17]. Previous publications have detailed trends in pneumococcal carriage across different age groups, as well as shifts in serotype distribution following vaccine introduction [5,6,18]. However, this study uniquely focuses on differences in invasive capacity for emerging serotypes, particularly within the context of recent vaccine formulations (e.g., PCV21). Further analysis presents clinical outcomes, and the detailed clinical presentation of cases derived from this surveillance data, while also incorporating 2006 data to provide a balanced comparison of three years before and three years after the introduction of PCV13 for the first time. By examining serotype-specific invasive capacities and highlighting notable trends in children less than 2 and 2 years and older, this study adds a critical perspective on the evolving epidemiology of pneumococcal disease and the impact of immunization efforts.

2. Methods

2.1. Data and study population

We estimated the serotype-specific invasive capacity of SPn using the following data during October 1st through September 31st of the following year for the years 2003/04, 2006/07, 2008/09, 2010/11, 2013/14, and 2015/16: (1) The annual IPD incidence data of children less than 7 years of age in the entire state of Massachusetts (study population), expressed as cases per person-year [6,15,19]; (2) The NP carriage prevalence data of children less than 7 years of age from 16 Massachusetts communities (see section 2.2 for additional details) [14,18,20]. The annual birth cohort in Massachusetts has been stable since 2010 at approximately 70,000 children per year [21]. The National Immunization Survey data demonstrate that PCV coverage with 3 or more doses among those aged 0 to 35 months in Massachusetts approximates 94 % or higher, which is consistently higher than the national average [22].

2.2. Isolates

Cases of IPD were detected using a population-based surveillance program for SPn infection in children, which began in Massachusetts (MA) in October 2001 [5,16]. All clinical microbiology laboratories in MA submitted isolates of SPn from blood, cerebrospinal fluid, or other normally sterile body sites collected from MA residents less than 18 years of age to the Massachusetts Department of Public Health (MDPH). Epidemiologists at the MDPH subsequently interviewed parents/guardians and/or primary care providers to obtain demographic and clinical information about each case by using a standardized case report form. The isolates were sent to the Maxwell Finland Laboratory for Infectious Diseases at the Boston Medical Center for serotyping and for evaluation of antimicrobial susceptibility. Approval for this study was granted by the institutional review boards of both Boston University Medical Center and the MDPH.

SPn carriage isolates were identified from a series of NP colonization studies conducted in 16 MA communities during respiratory virus season between November and April 2003/04, and a subset of 8 of these 16 from October to April in 2006/07, 2008/09, 2010/11, 2013–2014, and 2015/16 seasons [5,14,18]. For the 2006/07, 2010/11, and 2013/14 respiratory seasons, we added a site (Boston Medical Center) to include children from urban Boston. Methods for sample collection and processing were previously described in detail [5,6]. After written consent was obtained from parents, NP swabs were collected from children aged 3 months to <7 years of age during well-child or sick visits at primary care practices. Parents provided survey responses detailing demographic variables and the purpose of their visit. Data concerning the presence of comorbidities was obtained through a review of medical records. This study was approved by the Harvard Pilgrim Health Care and Boston Medical Center Institutional Review Boards.

The presence of SPn in both IPD and NP specimens was confirmed by optochin sensitivity (≥14 mm inhibition) and bile solubility by using standard microbiologic methods according to guidelines from the Clinical and Laboratory Standards Institute [23] and was serotyped by Quellung reaction by using pneumococcal antisera (Statens Serum Institute, Copenhagen, Denmark). Vaccine serotypes were defined as serotypes included in PCV13 (4, 6B, 9 V, 14, 18C, 19F, 23F, 1, 3, 5, 7F, 6A, and 19A). Serotypes 15B and 15C were considered a single serotype as is done in other studies, 15BC, because of the phase shift that results in the expression of these capsules [24]. Isolates found to be unencapsulated were confirmed to be SPn by lytA gene amplification by using real-time polymerase chain reaction (PCR) [25].

2.3. Estimate of serotype-specific invasive potential

The annual IPD incidence data and NP carriage prevalence data for corresponding years were used to compute a serotype-specific invasive capacity (event to carrier ratio) by dividing the annual incidence of IPD due to serotype x by the carriage prevalence of that same serotype in children from the same age group using the following formula, which was previously described in detail [5,6].

The formula for calculating the invasive capacity (IC) is:

$${\mathit{IC}}_x=\frac{{\text{I}}_x}{q_x}$$

Here, I_x represents the annual incidence of IPD due to serotype x, while q_x denotes the prevalence of SPn carriage for the same serotype (x). The natural logarithm of the estimate of IC_x was approximately normally distributed with variance [26]. In this formula, N_x stands for newly diagnosed IPD cases of serotype x, and m represents the study period.

$$\text{Variance}=\frac{1}{{\text{N}}_{\text{x}}}+\frac{1-{\text{q}}_{\text{x}}}{{\text{q}}_{\mathrm{xm}}}$$

Ninety-five percent confidence intervals (95 % CIs) were calculated with normal approximation by multiplying the standard error by 1.96. The serotype-specific relationship between NP carriage and IPD was evaluated 1) overall, 2) by year, 3) by age groups (<24 months vs 24–84 months), 4) pre-PCV13 (2003/04, 2006/07, 2008/09) and post-PCV13 (2010/11, 2013/14, 2015/2016) periods, 5) by the outcome, and 6) by clinical presentation. In conducting these calculations, it was necessary to exclude certain serotypes from the IC calculations under specific conditions. This exclusion occurred when the number of IPD cases surpassed the count of NP isolates, or when IPD or NP isolates were nonexistent (refer to Supplemental Table 1 for full list).

3. Results

3.1. Serotype distribution

We identified 293 IPD cases in children less than 7 years old of which 53.2 % (156/293) were among children less than 2 years old (Table 1). There were 31 different serotypes, with 19 A, 7F, 15 BCE, 3, and 22F emerging as the prevailing serotypes, accounting for 34.1 %, 9.2 %, 8.9 %, 5.8 %, and 4.8 % of all IPD isolates, respectively. Among 6459 children enrolled in the NP colonization studies, 1602 (24.8 %) had pneumococcal colonization with 43 different serotypes. Serotypes 15 BCE, 19 A, 35B, 11 A, 23 A, and 23B emerged as the most prevalent colonizers, constituting 12.9 %, 10.4 %, 8.6 %, 8.1 %, 6.8 %, and 6.7 % of all carriage isolates, respectively. Throughout both the pre- and post-PCV13 periods, serotypes not covered by PCV13 accounted for the majority of all carriage episodes, representing 79.8 % overall—specifically, 72.3 % in the pre-PCV13 period and rising to 87.3 % in the post-PCV13 period. Additional serotypes introduced in PCV15 (22F and 33F) and PCV20 (10A, 11A, and 15 BC) composed 5.2 % and 23.8 % of all colonizing isolates, respectively.

Table 1.

Distribution for invasive pneumococcal disease and nasopharyngeal colonization samples by year, age group, underlying condition and race ^16,18,26.

Year	IPD cases					Pneumococcal colonization*
	Age groups					Age groups
	0 - <24 months			24 - <84 months		0 - <24 months			24 - <84 months
2004	35			31		99/395 (25.1 %)			131/592 (22.1 %)
2007	32			32		161/801 (20.1 %)			132/739 (17.9 %)
2009	48			20		151/485 (31.1 %)			139/526 (26.4 %)
2011	18			25		194/559 (34.7 %)			171/605 (28.3 %)
2014	13			16		165/555 (29.7 %)			161/602 (26.7 %)
2016	10			13		43/265 (16.2 %)			55/335 (16.4 %)
Total	156/293 (53.2 %)			137/293 (46.8 %)		813/3060 (26.6 %)			789/3399 (23.2 %)
Year	Underlying Condition		Race and Ethnicity: Non-Hispanic White	Race and Ethnicity: Non-Hispanic Black		Race and Ethnicity: Hispanic		Race and Ethnicity: Asian
	Age groups		Age groups	Age groups		Age groups		Age groups
	0 - <24 months	24 - <84 months	0 - <24 months	24 - <84 months	0 - <24 months	24 - <84 months	0 - <24 months	24 - <84 months	0 - <24 months	24 - <84 months
2004	7 (20 %)	11 (35 %)	17 (49 %)	12 (39 %)	6 (17 %)	6 (19 %)	6 (17 %)	4 (13 %)	2 (6 %)	1 (3 %)
2007	6 (19 %)	4 (13 %)	17 (53 %)	9 (28 %)	3 (9 %)	4 (13 %)	4 (13 %)	4 (13 %)	2 (6 %)	2 (6 %)
2009	10 (21 %)	5 (25 %)	16 (33 %)	8 (40 %)	7 (15 %)	0 (0 %)	8 (17 %)	4 (20 %)	1 (2 %)	2 (10 %)
2011	2 (11 %)	7 (18 %)	6 (33 %)	12 (48 %)	4 (22 %)	1 (4 %)	1 (6 %)	1 (4 %)	0 (0 %)	3 (12 %)
2014	2 (15 %)	5 (31 %)	5 (38 %)	7 (44 %)	2 (15 %)	3 (19 %)	0 (0 %)	0 (0 %)	1 (8 %)	3 (19 %)
2016	1 (10 %)	0 (0%)	7 (70 %)	10 (77 %)	1 (10 %)	0 (0%)	0 (0%)	0 (0%)	1 (10 %)	1 (8 %)

^*N colonized / N swabbed = 1602/6459 (24.8 %).

3.2. Serotype-specific invasive capacity and age-related variations

The propensity to cause invasive disease once colonization occurred (i.e., invasive capacity) exhibited notable variations among serotypes. Specifically, serotypes 18C, 38, 7F, 19A, 3, and 33F demonstrated significantly heightened invasive potential compared to serotypes 11A, 35B, 6C, and 21 (Fig. 1a). There was an approximately 50-fold difference in the point estimates between the serotypes with the highest and lowest IC. The point estimates of serotype-specific IC were generally higher in younger children (Fig. 1b). For example, in children younger vs. older than 2 years of age: (i) IC of 10A, a PCV20 serotype, was ~7 (99.3, 95 % CI 45.2–217.8 vs. 15.2, 95 % CI 3.6–64.3) times higher; (ii) IC of 33F, a PCV15 serotype, was 2 (129.4, 95 % CI 59.5–283.1 vs. 63.5, 95 % CI 20.5–196.8) times higher. Similar trend was particularly evident for serotypes 3, 34, 16F, and 35B. However, the confidence intervals are wide and overlap between some of the compared groups, indicating that these observed differences may not be statistically significant.

Fig. 1.

Serotype specific invasive capacity A) in children less than 7 years of age in Massachusetts, B) in children less than 2 years of age vs. 2 years and older (X-axis is displayed on log-scale).

3.3. Temporal trends by year and pre-PCV13 vs. post-PCV13 era

Fig. 2 illustrates the changes in IC across various years. In 2004, serotypes 18C and 15F exhibited the highest IC. In 2011, NVSTs surpassed all others in IC, with serotype 38 emerging as the leader among 27 strains, while serotype 7F, 7C, and non-typeable (NT) strains took precedence. A reduction in IC among NVSTs was observed in 2007, 2009, 2014, and 2016. Meanwhile, serotype 22F, a serotype included in PCV15, demonstrated a consistent IC over five years (2004, 2007, 2009, 2011, and 2014), with a slight decrease noted in 2016. Apart from serotypes 19A, 19F, 3, 15A, and 23B, the IC of vaccine-type serotypes either remained stable or declined from 2011 to 2016 with the introduction of PCV13.

Fig. 2.

Serotype specific invasive capacity in children less than 7 years of age in Massachusetts by year and vaccine group.

During the pre-PCV13 era, serotypes 33F, 19A and 7F exhibited significantly higher IC compared to approximately half of the serotypes (10A, 6A, 15BC, 19F, 15A, 35F, 23A, 11A, 23B, 6C and 35B) (Fig. 3). However, in the post-PCV13 era, serotype 38 displayed significantly higher IC compared to serotypes 23B, 15A, 15BC, 19A, 23A, 35F, 35B, 11A, and 6C. Additionally, we calculated IC with 95 % confidence interval for post-PCV13 period as follows: for serotypes in PCV15 14.6 (10.1, 21.0), serotypes in PCV20 9.9 (7.5,13.1), serotypes in PCV21 7.7 (6.1,9.8) and for non-vaccine serotypes as 6.4 (3.7,10.8).

Fig. 3.

Serotype specific invasive capacity in children less than 7 years of age in Massachusetts before and after PCV13 implementation (Y-axis is displayed on log-scale).

3.4. Serotype-specific clinical presentation and mortality trends in invasive pneumococcal disease

Notably, except for serotypes 14, 6A, 7F, 11A, 23A, and 20, all VT serotypes predominantly manifested as bacteremia (Fig. 4a). Conversely, all cases of two VT serotypes (14 in PCV13; and 20 in 23 valent pneumococcal polysaccharide vaccine and PCV21) and two NVSTs (7C and 6C) were attributed to pneumonia. In the single case of IPD caused by serotype 15F, meningitis was present.

Fig. 4a.

Clinical presentation in invasive pneumococcal disease and distribution of serotypes by vaccine groups.

The highest percentage of deaths observed among IPD cases was attributed to serotypes 35F (NVST), 35B (included in PCV21), 23B (included in PCV21), and 11A (included in PCV20) (Fig. 4b). Complete recovery occurred in IPD cases caused by 2 PCV7 serotypes (18C and 14), 1 PCV13 serotype (6A), and 3 PCV21 serotypes (16F, 31, 20).

Fig. 4b.

Survival in invasive pneumococcal disease and distribution of serotypes by vaccine groups.

Notes: 1) Serotypes with small case numbers may show a high predisposition due to limited data, and results should be interpreted with caution. 2) In Fig. 4a, N may include unknown clinical syndrome. 3) Serotypes 4, 8, and 10F were excluded from the graph as the study focuses on analyzing serotypes with sufficient data for invasive capacity evaluation. However, we observed one case of IPD for each of these serotypes, all of which resulted in full recovery (100 %). Specifically, the cases associated with serotypes 4 and 8 presented as pneumonia, while the case with serotype 10F was classified as bacteremia.

For the results illustrated in Figs. 4a and 4b, it is important to note that these findings are based on a limited number of cases for certain serotypes (e.g., N = 1 to 3), which increases variability and results in wider 95 % confidence intervals for these estimates.

4. Discussion

The findings from this study provide insights into the serotype-specific epidemiology and dynamics of pneumococcal disease in children less than 7 years old in Massachusetts. We evaluated serotype distribution, invasive capacity, clinical presentations, and mortality rates during the study period. First, our results show that over half of the IPD cases occurred in children less than two years of age, highlighting the vulnerability of young children to pneumococcal infections, and emphasizing the importance of vaccination strategies targeting infants. In children older than 2 years of age, serotype-specific invasive capacity was higher for 7 out of 18 serotypes (approximately 40 %). Differences in serotype-specific IC are highest with 10A (included in PCV20) and 23A (included in PCV21). In children younger vs. older than 2 years of age: (i) the IC of 10A, a PCV20 serotype, was 6.6 times higher among children less than 2 years (99 vs. 15); (ii) the IC of 23A, a PCV21 serotype, was 5.7 times higher among children 2 years and older (23 vs. 4). On the other hand, previous studies showed that older children are more vulnerable to PCV13 vaccine failure [17]. Age-related variations in IC suggest differential susceptibility to invasive disease across different age groups, highlighting the complex interplay between host factors and serotype characteristics.

Second, our findings indicate significant variations in IC among serotypes, with certain strains demonstrating increased invasive potential compared to others. During the study period, serotypes 38 and 15F (NVST), 20 (covered by PCV21), 33F and 22F (covered by PCV15), and 10A (covered by PCV20) exhibited the highest invasive capacity among non-PCV13 serotypes. This contrasts with the global prevalence of replacement serotypes found in NP isolates, where serotypes 15BC, 10A, 21, 16F, 9 N, 13, and 11A are among the most common [27]. Among these serotypes, the point estimates for 19F (included in PCV7), 10A and 11A (included in PCV20), 23A, 23B, and 35B (included in PCV21), 35F and 6C (NVST), and NT increased from the pre-PCV13 to post-PCV13 period, while the point estimate for serotype 7F, 3 and 19A (covered by PCV13), 33F (covered by PCV15), and 15 BCE (covered by PCV20) decreased. However, all estimates are accompanied by wide confidence intervals, making it challenging to support statistically significant differences in these findings except for 19A (included in PCV13). Serotypes 22F and 15A remained relatively stable across both periods. Due to limited data, we were unable to assess changes for serotypes 18C, 7C, 34, 14, 15F, 16F, 21 and 9 N.

Temporal trends analysis revealed shifts in serotype-specific IC over the study period, with fluctuations observed in both VT and NVST. 18C and 3 being predominant in 2004 whereas, 19A (included in PCV13) and 15A (included in PCV21) being dominant in 2016. Specifically, serotype 19A, although highly dominant in 2016, was even more prevalent before the PCV13 recommendations and was excluded in 2004. These findings resonate with prior research [14], reinforcing the temporal evolution of pneumococcal serotypes and highlighting the influence of vaccination strategies on serotype prevalence [14]. Of particular interest is the emergence of NVSTs as prevalent causes of invasive disease, indicating potential serotype replacement following the introduction of PCVs. Serotype 38 (NVST) stands out in the post-PCV13 era with the highest IC. In the post-PCV13 era, serotypes 11A and 15BC, 35B, and 6C (NVST) dominated the nasopharyngeal colonization, as indicated by their high prevalence (Supplemental Table 1). However, these serotypes exhibited the lowest invasive capacity in our analysis consistent with previous reports [6].

Furthermore, our study elucidated the clinical presentations and mortality trends associated with different pneumococcal serotypes. We found distinct patterns of disease manifestation, with certain serotypes predominantly causing bacteremia, consistent with findings reported in the literature [28]. Mortality trends among IPD cases underscore the impact of serotype-specific dynamics on disease outcomes. While certain nonvaccine serotypes such as 23B, 35B, 19F, and 11A (included in PCV20) were associated with higher case mortality rates, complete recovery was observed with others. This is consistent with previous studies reporting serotype 11A, 31, and 19F as serotypes frequently causing fatalities [29]. The effectiveness of pneumococcal vaccines in preventing severe disease outcomes, as evidenced by the complete recovery observed in cases caused by vaccine-targeted serotypes, highlights the importance of vaccination in reducing disease burden [30]. Our analysis shows that the top 3 serotypes currently linked to increased mortality rates are serotypes included in PCV20 and PCV21 (serotypes 23B, 35B, and 11A). However, these results should be interpreted with caution due to the limited number of cases for certain serotypes, which results in increased variability and wide confidence intervals. These limited data points may introduce variability that affects the stability and reliability of the observed trends. While these serotypes show an apparent predisposition toward higher mortality, the small sample sizes limit the generalizability of these findings. The limited number of IPD cases and low mortality rates constrain our ability to further stratify results by underlying high-risk conditions, highlighting the importance of future studies incorporating this data to improve interpretability.

Our study is subject to several limitations. Firstly, our analysis is based on data from a single state, Massachusetts, which may not be fully representative of the broader epidemiology of pneumococcal disease in other states. The generalizability of our findings is therefore limited, and further studies involving diverse geographic areas are needed to validate our results. Secondly, our study did not include multilocus sequence typing (MLST) data, which would have provided a more comprehensive understanding of the genetic diversity and evolutionary dynamics of the pneumococcal strains. Without MLST, our ability to differentiate between the contributions of different factors is constrained. Additionally, there are inherent limitations in our analytical approach and assumptions, such as potential biases in case reporting, variations in diagnostic practices over time, and the challenges associated with accurately estimating invasive capacity from observational data. The nasopharyngeal (NP) samples in this study were collected from multiple sites but do not represent the entire state, potentially limiting the generalizability of these findings to a broader population. Furthermore, certain serotypes in both IPD and NP samples were based on small case numbers, resulting in wide confidence intervals and greater variability in key estimates, including invasive capacity (IC), mortality, and clinical presentation. These limitations should be considered when interpreting the results, as small sample sizes may impact the robustness and stability of observed trends. Future studies with larger, more representative samples would be valuable in validating and refining these findings.

Despite these limitations, our findings reaffirm previous observations regarding the varying propensity of serotypes to cause invasive disease. Additionally, they offer insight into the ongoing reduction in IPD rates despite the absence of direct protection against currently circulating colonizing serotypes. Further investigations are warranted to gain a deeper understanding of the invasive disease potential of emerging non-PCV13 vaccine colonizers. With the introduction of new pneumococcal conjugate vaccines, further changes in the distribution of carriage serotypes are expected to be observed [31]. The emergence of highly invasive serotypes is a cause for concern, as it may lead to an increase in IPD cases [32]. Continuous surveillance of IPD is imperative to monitor whether serotypes with heightened invasive potential establish themselves as successful colonizers.

5. Conclusion

Overall, this study emphasizes the complex interplay between serotype-specific characteristics, clinical presentation, and disease outcomes in pneumococcal disease. The high invasive capacity and high mortality of some non PCV13/15 serotypes refutes the concept that such strains are of low virulence but rather that prior to the introduction of PCV7 these serotypes were outcompeted in the nasopharynx. Now that they have become more prevalent colonizers their potential to cause disease is revealed. Continued surveillance and research efforts are essential for monitoring trends in serotype distribution, assessing vaccine effectiveness, and informing strategies to mitigate the burden of pneumococcal disease in the pediatric population.

Abbreviations:

IC

invasive capacity

IPD

invasive pneumococcal disease

MA

Massachusetts

MDPH

Massachusetts Department of Public Health

NP

nasopharynx

NVST

non-vaccine serotypes

PCV

pneumococcal conjugate vaccine

PCV-7

pneumococcal 7-valent conjugate vaccine

PCV-13

pneumococcal 13-valent conjugate vaccine

PCV-15

pneumococcal 15-valent conjugate vaccine

PCV-20

pneumococcal 20-valent conjugate vaccine

PCV-21

pneumococcal 21-valent conjugate vaccine

PCV-24

pneumococcal 24-valent conjugate vaccine

SPn

Streptococcus pneumoniae

VT

vaccine-type serotypes

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.vaccine.2024.126692.

Data availability

The authors do not have permission to share data.