Abstract
Prematurity is a major risk factor for invasive Group B Streptococcus (iGBS) disease in infancy. Whether gestational age (GA) modifies the association between GBS genomic characteristics and postnatal age at disease onset remains unclear. We analyze isolates from infants with iGBS disease in Denmark between 2000 and 2023 using whole-genome sequencing to identify clonal complexes and genes encoding alpha-like surface proteins and capsular serotypes. Isolates are classified into clones based on shared genomic characteristics.Among 510 infants, 356 (70%) are term, 66 (13%) moderate preterm, and 88 (17%) very preterm. Approximately half of term and moderate preterm infants develop iGBS during days 1–3 of life, compared with 9% of very preterm infants. GBS clone distributions differ by GA and postnatal age, particularly for disease onset at 4–89 days. Term and moderate preterm infants are predominantly infected with CC17/Rib/III (71%), whereas very preterm infants show greater clonal diversity, with a lower prevalence of CC17/Rib/III (31%) and higher prevalences of CC23/Alp1/Ia (27%) and CC12/AlphaC/Ib-II (19%). These findings show that GA modifies associations between GBS genomic characteristics and postnatal age at disease onset, highlighting the importance of GA- and postnatal age-stratified GBS surveillance to inform preventive strategies and maternal vaccine implementation.
Subject terms: Paediatric research, Epidemiology, Bacterial genetics
This Danish study shows that gestational age modifies the association between Group B Streptococcus genomic characteristics and postnatal age at disease onset in infants. The findings may help inform surveillance and maternal vaccination strategies.
Introduction
Streptococcus agalactiae (Group B Streptococcus, GBS) remains a leading cause of severe infection in neonates and infants, most commonly presenting as sepsis or meningitis1,2. The global burden of invasive GBS (iGBS) disease is substantial and is associated with being born preterm, stillbirth, early-life mortality, and long-term morbidity3–7. Transmission occurs either vertically from colonized women at the time of birth or horizontally from mothers or caregivers during early infancy8. Vertically transmitted iGBS typically presents immediately after delivery or within the first few days of life and is associated with more severe clinical presentation and higher mortality than later disease onset. Horizontal transmission of iGBS presents after some days, with incidence declining rapidly with increasing postnatal age, particularly among term infants8–10.
Prematurity is a well-established major risk factor for iGBS disease11. Transplacental transfer of IgG occurs predominantly in the third trimester and increases substantially after gestational week 3612. Consequently, infants born preterm have lower protective antibodies at birth and increased susceptibility to iGBS disease13.
Despite advances in obstetric and neonatal care, morbidity and mortality associated with iGBS in the preterm infants appear not to have declined over recent decades9.
Current prevention of infant iGBS disease relies on intrapartum antibiotic prophylaxis (IAP), which provides only short-term protection because antibiotic concentrations in the newborn decline rapidly after birth14.
Maternal GBS vaccines represent a promising preventive strategy against infant iGBS disease, with the potential to provide longer-lasting protection2,15–18. Two vaccines have advanced in development; one targeting the alpha-like surface proteins (Alps; Rib, AlphaC, Alp1, and Alp2/3)18, and another targeting six (Ia-V) of ten known capsular polysaccharide (CPS) serotypes (Ia-IX)19. However, whether infants born preterm are potentially covered is not entirely clear.
Whole-genome sequencing (WGS) enables detailed genomic characterization of GBS isolates, important for surveillance, understanding pathogenesis, and informing preventive strategies. Previous studies have described associations between genomic characteristics of infant iGBS isolates, such as multi-locus sequence types (MLSTs), clonal complexes (CCs), Alps, and CPS serotypes, and postnatal age at disease onset. In particular, the hypervirulent CC17/Rib/III clone is strongly associated with iGBS disease presenting after the first days of life and with meningitis20–23. However, whether differences in the genomic characteristics of GBS exist between preterm and term infants have not been comprehensively investigated. Understanding whether gestational age (GA) influences the genomic characteristics of iGBS isolates is essential to clarify GBS pathogenesis in preterm infants and to inform GA-specific surveillance and preventive strategies in this high-risk population.
We aimed to assess whether GA modifies the association between the genomic characteristics of infant iGBS and postnatal age at GBS disease onset.
Results
Characteristics of infants with GBS
A total of 664 infants with a positive GBS culture from blood or cerebrospinal fluid (CSF) were identified between 2000 and 2023 using national microbiological data. Of these, 529 (80%) had GBS isolates available for WGS, whereas 135 (20%) did not, primarily because no viable isolates were stored at the relevant department of clinical microbiology (Table 1).
Table 1.
Characteristics of 664 infants younger than one year of age with GBS cultured in blood or CSF samples in Denmark between 2000 and 2023
| Infant characteristics | Infant with available | Infants without available |
|---|---|---|
| (n, %) | GBS isolate (n = 529) | GBS isolate (n = 135) |
| Sex | ||
| Female | 235 (44%) | 69 (51%) |
| Male | 294 (56%) | 66 (49%) |
| Gestational age | ||
| <32 weeks | 88 (17%) | 18 (15%) |
| 32 to 36 weeks | 66 (13%) | 18 (15%) |
| ≥ 37 weeks | 356 (70%) | 87 (70%) |
| GA missing (n) | 19 | 12 |
| Period (years) | ||
| 2000-2005 | 126 (24%) | 44 (33%) |
| 2006-2011 | 138 (26%) | 36 (27%) |
| 2012-2017 | 119 (23%) | 40 (30%) |
| 2018-2023 | 146 (28%) | 15 (11%) |
| Postnatal age at GBS disease onset | ||
| Age at onset: 1-3 days | 263 (50%) | 63 (46%) |
| Age at onset: 4-89 days | 239 (45%) | 67 (50%) |
| Age at onset: 90-365 days | 27 (5%) | 5 (4%) |
| Specimen type | ||
| Blood | 437 (83%) | 107 (79%) |
| CSF | 92 (17%) | 28 (21%) |
CSF Cerebrospinal fluid, GA Gestational age, GBS Group B Streptococcus.
Infants with and without available GBS isolates were similar with respect to sex, GA, postnatal age at disease onset, and specimen type. However, a higher proportion of infants without available isolates were identified in the early study period (2000–2005), whereas a larger proportion of infants with available GBS isolates were identified in more recent years (2018–2023; Table 1).
Among infants with available GBS isolates (n = 529), 510 had data on GA and 356 (70%) were term, 66 (13%) were moderate preterm, while 88 (17%) were very preterm (Table 1).
Among term infants, iGBS disease most frequently presented during the first 1–3 days of life (60%), whereas 37% presented during days 4–89 of life. Among moderate preterm infants, 48% presented with iGBS during day 1-3 and 47% during day 4–89 of life. In contrast, only 9% of the very preterm infants presented with iGBS disease during day 1–3 days of life, whereas the majority (80%) presented between days 4 and 89 (Table 2).
Table 2.
Association between gestational age at birth and presentation of invasive Group B Streptococcal disease in 529 infants younger than one year of age in Denmark, 2000 – 2023
| Categories of gestational age at birth | ||||||
|---|---|---|---|---|---|---|
| GA < 32 weeks (n = 88) |
GA 32-36 weeks (n = 66) |
GA ≥ 37 weeks (n = 356) |
p-value | GA missing (n = 19) |
||
| Sex | ||||||
| Female | 38 (43%) | 29 (44%) | 160 (45%) | 0.77 | 11 (58%) | |
| Male | 50 (57%) | 37 (56%) | 196 (55%) | 8 (42%) | ||
| Postnatal age at onset | ||||||
| 1-3 days | 8 (9%) | 32 (48%) | 213 (60%) | <0.001 | 10 (53%) | |
| 4-89 days | 70 (80%) | 31 (47%) | 130 (37%) | 9 (47%)* | ||
| 90-365 days | 10 (11%) | 3 (5%) | 13 (3%) | |||
| Site of infection | ||||||
| Blood (sepsis) | 79 (90%) | 54 (82%) | 292 (82%) | 0.20 | 12 (63%) | |
| CSF (meningitis) | 9 (10%) | 12 (18%) | 64 (18%) | 7 (37%) | ||
| GBS clone | ||||||
| CC17/Rib/III | 25 (28%) | 33 (50%) | 179 (50%) | <0.001 | 7 (37%) | |
| CC23/Alp1/Ia | 23 (26%) | 6 (9%) | 56 (16%) | 3 (16%) | ||
| CC12/AlphaC/Ib-II | 13 (15%) | 5 (8%) | 35 (10%) | 0 (0%) | ||
| CC19/Rib/III | 11 (13%) | 11 (17%) | 22 (6%) | 3 (16%)** | ||
| CC1/Alp2/3/V | 5 (6%) | 3 (5%) | 10 (3%) | |||
| Other GBS clones | 11 (13%) | 8 (12%) | 54 (15%) | 6 (32%) | ||
CSF Cerebrospinal fluid, GA Gestational age, GBS Group B Streptococcus.
* Combined for 4-89 days and 90-365 days; **Combined for CC19/Rib/III and CC1/Alp2/3/V.
Among term and moderate preterm infants, 18% had GBS isolates from CSF compared with 10% among very preterm infants; however, this difference was not statistically significant (Table 2). Because the distributions of CSF isolates across GA categories did not differ significantly, and only nine very preterm infants had CSF isolates, GBS clone distributions were not further analyzed separately for CSF isolates across GA categories.
Additional comparisons of blood and CSF isolates according to postnatal age at onset, gestational age, and major GBS clones are provided in Supplementary Table 4.
Genomic characteristics of the infant iGBS isolates
WGS of the 529 GBS isolates identified 37 distinct STs clustering into eight CCs. Maximum-likelihood phylogenetic analysis revealed five major clades comprising the five largest CCs (Fig. 1).
Fig. 1. Core SNP-based maximum-likelihood phylogeny of 529 invasive GBS isolates collected from infants ≤ 1 year of age in Denmark, 2000–2023.
From the inner to the outer ring, the annotation strips indicate clonal complex (CC), alpha-like surface protein (Alp), capsular serotype, and the hypervirulent GBS adhesin hvgA.GBS-ref-BIII (M 781) (ENA accession no. PRJEB43628) was used as reference. GBS: Group B Streptococcus.
The largest clade comprised CC17 isolates (46%), while the other clades mainly comprised CC23 (18%), CC19 (10%), CC12 (11%), and CC1 (7%) isolates.
Among the Alps, Rib was most prevalent (59%), followed by Alp1 (20%). Among the capsular CPS serotypes, serotype III (57%) was most prevalent, followed by serotype Ia (20%; Supplementary Table 2). The hvgA gene was detected in 47% of the iGBS isolates, all of which belonged to CC17.
Some intraclade variation was observed in determinants of the Alps and serotypes. However, overall homology within clades was high, particularly among CC17 and CC23 isolates (Fig. 1).
A detailed description of the genomic characteristics of the 529 GBS isolates, including CC, Alp, and serotype distributions, as well as homology within CCs, is provided in the Supplementary Information (Supplementary Tables 2–3 and Supplementary Sections 2–4).
GBS clones
GBS clones were defined as phylogenetically clustered GBS isolates (Fig. 1) belonging to the same CC and sharing identical Alp and CPS serotype profiles (CC/Alp/serotype).
Among the 529 characterized GBS isolates, the majority belonged to the CC17/Rib/III clone (48%), all of which carried the hvgA gene (Fig. 1). The second most common clone was CC23/Alp1/Ia, accounting for 16% of isolates.
GBS isolates belonging to CC12 and encoding the AlphaC comprised 10% of all the isolates and were associated with serotypes Ib (62%) and II (38%). These isolates were grouped as a single clone, CC12/AlphaC/Ib–II.
Less prevalent GBS clones included CC19/Rib/III (8%) and CC1/Alp2/3 (6%). The remaining GBS isolates (13%) consisted of specific CC/Alp/serotype combinations, each observed in ten or fewer isolates (n ≤ 10) and were grouped into “other GBS clones”. Individual clone combinations and sequence types for all included GBS isolates are listed in Supplementary Table 3.
GBS clone distributions by gestational age
The distribution of GBS clones differed substantially by GA as presented in Table 2. Among term infants, the predominant clone was CC17/Rib/III (50%), followed by CC23/Alp1/Ia (16%).
In moderate preterm infants, CC17/Rib/III was equally common (50%), whereas CC19/Rib/III occurred more frequently (17%) than among term infants (6%).
Among very preterm infants, clone distribution was even more diverse than among term and moderate preterm infants. Although CC17/Rib/III remained the most common clone (28%), its relative prevalence was substantially lower, whereas CC23/Alp1/Ia (26%) and CC12/AlphaC/Ib-II (15%) accounted for considerably larger proportions of isolates.
GBS clone distributions by gestational age in infants with iGBS disease onset during day 1-3 of life
Among infants with iGBS disease onset during day 1-3 of life, the distribution of GBS clones did not differ substantially between term and moderate preterm infants (Fig. 2a). In term infants, the most prevalent clone was CC17/Rib/III (38%), followed by CC23/Alp1/Ia (22%). Among moderate preterm infants, CC17/Rib/III was also most common (31%), followed by CC12/AlphaC/Ib-II (16%).
Fig. 2. Distribution of GBS clones by gestational age among infants with invasive GBS disease according to postnatal age at disease onset.
a Distribution among infants with disease onset at days 1–3 of life (n = 253; 10 infants were excluded because of missing gestational age). b Distribution among infants with disease onset at days 4–89 of life (n = 231; eight infants were excluded because of missing gestational age). “GBS clones not reportable” indicates GBS clones (specific CC/Alp/serotype combinations) with fewer than three isolates (n < 3) within a gestational age category that could not be reported in accordance with Danish data protection regulations. GBS: Group B Streptococcus.
Only a small number of very preterm infants (n = 8) presented with iGBS disease during day 1–3 of life, and detailed clone distribution by GA was not reported in accordance with Danish data protection practices.
GBS clone distributions by gestational age in infants with iGBS disease onset day 4-89 of life
In contrast to iGBS disease onset during day 1-3 of life, the clone distribution varied markedly by GA among infants with disease onset during day 4-89 of life (Fig. 2b). Among term and moderate preterm infants, CC17/Rib/III accounted for 71% of the iGBS isolates. However, a higher prevalence of CC19/Rib/III was observed among moderate preterm infants (19%) than term infants (7%).
Among very preterm infants, a markedly more diverse GBS clone distribution was observed. Although CC17/Rib/III remained the most frequent clone, it accounted for only 31% of the isolates compared with 71% among term and moderate preterm infants, whereas CC23/Alp1/Ia (27%) and CC12/AlphaC/Ib-II (19%) represented a substantially larger proportion of disease in the very preterm infants compared with term and moderate preterm infants (Fig. 2b).
Given the small number of infants with disease onset between 90 and 365 days of life (n = 27), GBS clone distributions stratified by GA were not reported for this group in accordance with Danish data protection practices.
GBS clone distribution by postnatal age at iGBS disease onset within gestational age categories
Overall, GBS clone distributions differed significantly between infants with disease onset on days 1–3 and on days 4–89 of life, even after stratification by gestational age.
Specifically, among term infants, the proportion of CC23/Alp1/Ia was significantly more prevalent in GBS isolates from onset at age 1-3 days compared with onset at age 4-89 days (22% vs. 6%, p < 0.001).
In contrast, this pattern was not observed among moderate preterm infants, in whom the distribution of CC23/Alp1/Ia was similar in the two onset categories (9% vs. 10%, p = 0.967).
Sensitivity analyses
Exclusion of GBS isolates from infants with postnatal age at disease onset of six months or older (n = 6) did not affect the overall distribution of the Alps and serotypes.
In addition, alternative categories of postnatal age at iGBS disease onset (ages 1–6 and 7–89 days) showed distributions of CCs, Alps, and serotypes similar to those of the primary categories (1–3 days vs. 1–6 days and 4–89 days vs. 7–89 days).
Discussion
Main findings
This nationwide study provides new insights into infant iGBS disease by demonstrating that GA at birth modifies the association between postnatal age at disease onset and the genomic characteristics of the disease-causing GBS isolates, within the context of a universal, risk-based IAP strategy.
Findings in relation to previous studies
Our findings reveal distinct patterns in the genomic characteristics of GBS isolates in relation to both postnatal age at disease onset and GA at birth. Whereas previous studies have primarily described associations between GBS genomic characteristics and postnatal age at onset, without stratification by GA, our results demonstrate that these associations differ substantially between term and preterm infants.
Term and moderate preterm infants presented with iGBS disease at similar frequencies during days 1–3 and days 4–89 of life. In contrast, very preterm infants rarely developed disease in the first days of life and predominantly presented with iGBS disease after day 3 of life. This pattern is likely influenced by routine administration of IAP before 35 weeks’ gestation, which reduces iGBS disease in the first days of life but does not provide sustained protection later in infancy24.
Previous studies have reported that CC23/Alp1/Ia was more frequently associated with iGBS disease presenting during the first days of life, whereas the hypervirulent CC17/Rib/III clone predominates among infants with iGBS later in infancy21,23. By stratifying analyses by GA, we show that this pattern is primarily driven by term infants. Among term infants, CC23/Alp1/Ia was more prevalent in disease presenting during day 1–3 of life than at later onset, whereas this association was not observed among moderate preterm infants and could not be assessed in the very preterm infants because iGBS disease during the first days of life was rare in this group.
Among infants presenting with disease after day 3 of life, CC17/Rib/III predominated, consistent with previous reports21–23.
However, our findings extend current knowledge by demonstrating that the genomic epidemiology of iGBS among very preterm infants differs from that of term and moderate preterm infants in two important ways. First, very preterm infants exhibit substantially greater clonal diversity. Second, this diversity is accompanied by a marked redistribution of dominant clones, with a considerably lower relative prevalence of the hypervirulent CC17/Rib/III clone and increased representation of CC23/Alp1/Ia and CC12/AlphaC/Ib-II. Notably, the GBS clone distribution among very preterm infants with disease onset at days 4-89 of life resembled that observed among infants with disease onset during days 1–3, suggesting a GA-related shift in postnatal age at disease onset that may be influenced by IAP.
Together, our findings indicate that previously reported associations between GBS clones and postnatal age at disease onset may largely apply to term infants but not to preterm infants. The broader distribution of GBS clones observed among very preterm infants may reflect host susceptibility due to limited transplacental transfer of maternal IgG and immaturity of the neonatal immune system12. In this context, susceptibility to iGBS disease in very preterm infants appeared less dependent on a specific hypervirulent GBS clone and more related to host vulnerability to a wider range of GBS clones.
Potential vaccine coverage
The overall distribution of Alps and serotypes among infant iGBS isolates in this study is consistent with reports from the USA, South Africa, and Korea21–23,25. Studies from, e.g, the USA and South Africa stratified the GBS isolates by patient age21–23, while other studies did not25, when reporting distributions. This may limit comparability because of known epidemiological differences between neonatal and adult invasive GBS infections. These differences should be considered when interpreting cross-study comparisons.
Based on the distributions observed in our study, the AlpN vaccine (targeting Rib, AlphaC, Alp1, Alp2-3) would theoretically have covered 100% of infant GBS isolates, while the hexavalent GBS6 conjugate vaccine (targeting CPS serotypes Ia–V) would have covered 98% of the GBS isolates in the Danish setting and in settings with similar approaches to GBS prevention.
By describing Alp and serotype distributions across GA and postnatal age at disease onset, our findings suggest that the theoretical antigen coverage of these vaccines would remain high among moderate and very preterm infants, also. However, antigen coverage alone does not necessarily translate into clinical effectiveness, particularly in preterm populations. Because transplacental IgG transfer is reduced early in gestation, very preterm infants have substantially lower IgG concentrations at birth, and it remains uncertain whether maternal vaccination early enough in pregnancy can generate protective antibody concentrations in this group12. Nevertheless, partial transplacental transfer of vaccine-induced IgG earlier in pregnancy, together with potential reduction in maternal colonization, may still contribute to protection against iGBS disease later in infancy, particularly infections acquired through horizontal transmission12 Evaluation of immunogenicity, optimal timing, and clinical effectiveness of maternal GBS vaccination therefore warrants further investigation.
Strength and limitations
A major strength of this study is the linkage of genomic characteristics of GBS isolates with GA at birth, using high-quality data from the Danish Medical Birth Registry, enabling assessment of associations between GA and GBS genomic characteristics. Another strength is the nationwide, population-based design, which includes GBS isolates cultured from the blood or cerebrospinal fluid of Danish infants over a 24-year period. A high proportion of isolates (80%) were available for WGS, with many originating from recent study years (2018–2023). Consequently, the genomic characterization presented is likely to reflect contemporary GBS isolates.
In addition, infants with and without available GBS isolates for WGS were similar with respect to postnatal age at onset, specimen type, and sex distribution, supporting the external validity of the findings to comparable settings.
Importantly, sensitivity analyses using conventional epidemiological categorizations of postnatal age at iGBS disease onset (days 1–6 and 7–89) yielded consistent findings, supporting the robustness of the observed associations and comparability with previous studies using these categorizations.
A limitation of this study is that individual-level information on IAP administration, timing, and dosage was not available. Consequently, we could not determine whether infants with iGBS disease had been exposed to IAP, whether IAP was complete or incomplete, or whether the disease occurred despite adequate IAP. Assessment of IAP effectiveness and clinical risk factors for iGBS disease was beyond the scope of this genomic epidemiological study.
Nevertheless, iGBS disease, despite current prevention strategies, may occur for several reasons, including non-indicated IAP, insufficient time between antibiotic administration and delivery, rapid decline in neonatal antibiotic concentrations after birth, or postnatal acquisition through horizontal transmission. These mechanisms could not be assessed in the present study and warrant further investigation in studies with detailed clinical data on IAP exposure and transmission pathways.
The generalizability of our findings is primarily limited to high-income settings with established GBS prevention strategies. In these settings, prevention of GBS disease during the first days of life typically relies on IAP guided by either universal microbiological screening or risk-based approaches26.
In contrast, many low- and middle-income countries lack national GBS screening policies and have limited implementation of IAP. Globally, 91% of high-income countries report having an IAP policy, compared with only 13% of low-income and 25% of lower-middle-income countries. In addition, maternal GBS screening coverage remains low in several low- and middle- income countries, with reported rates below 15%27. Differences in prevention strategies, healthcare infrastructure, and disease burden should therefore be considered when extrapolating our findings across different settings.
Interpretation of clone distribution among infants presenting with iGBS disease in the first days of life is also influenced by IAP exposure. Consequently, clone distributions for iGBS disease onset at days 1-3 of life may not reflect underlying distributions in settings without widespread IAP. Furthermore, GBS clone distribution among very preterm infants could not be robustly assessed because of the small number of infants presenting with iGBS disease at age 1-3 days of life. Investigations in settings with limited or inconsistent IAP implementation are therefore warranted.
Finally, infants without available GBS for WGS had a slightly higher proportion of missing information on GA. Although these infants could theoretically differ in clone distribution, the number affected and the magnitude of the difference were small, and GA was considered missing at random. It is therefore unlikely that this limitation influenced our overall findings.
In conclusion, this nationwide genomic analysis of infant iGBS demonstrates that GA substantially modifies the associations between the genomic characteristics of disease-causing GBS isolates and postnatal age at disease onset.
Notably, the GBS clone distribution among very preterm infants, who predominantly presented with iGBS disease during day 4–89 of life, resembled that observed among term and moderate-preterm infants with disease onset during day 1–3. This suggests that very preterm infants may experience a delay in the onset of iGBS disease, potentially influenced by IAP.
These findings highlight that established associations between GBS clones and postnatal age at disease onset do not apply uniformly across GA. GA-stratified genomic surveillance should therefore play a central role in future investigations of GBS pathogenesis and disease and in the evaluation of preventive strategies, including maternal GBS vaccination strategies, particularly for very preterm infants.
Methods
Ethics and approvals
According to Danish legislation, this registry-based study using pseudonymized microbiological and health registry data did not require approval from a research ethics committee or written informed consent. The requirement for informed consent was therefore waived under Danish regulations because no direct contact with participants or intervention was involved. Approval for the processing and linkage of registry data was obtained from the Region of Southern Denmark (No. 21/69301) and Statistics Denmark (No. 708661). The processing of personal data was notified to, approved by, and recorded in the internal register of the Region of Southern Denmark (No. 22/9764). Sequenced GBS isolates were linked to individual-level data from Danish health registries using pseudonymized identifiers, and analyses were conducted on de-identified data to minimize the risk of re-identification. This study is reported in accordance with the Strengthening the Reporting of Molecular Epidemiology for Infectious Disease (STROME-ID) guideline.39 The completed checklist provided in Supplementary Information (p. 5).
Study design and participants
Infants with iGBS disease and collection of GBS isolates
Infants younger than 1 year of age with invasive Group B Streptococcus (GBS) disease, defined by a positive blood or cerebrospinal fluid (CSF) culture, identified in Denmark between Jan 1, 2000, and Dec 1, 2023, were eligible for inclusion. By contacting all the Danish departments of clinical microbiology and the State Serum Institut, we identified all infants with positive GBS cultures from blood and cerebrospinal fluid (CSF) in Denmark during the study period. GBS isolates were retrieved from the relevant departments of clinical microbiology or the State Serum Institute, provided that a viable isolate was available. Postnatal age at disease onset was defined by the sample date of the first positive GBS culture. GBS cultured from both blood and CSF were categorized by CSF sample. Postnatal age at iGBS disease onset was categorized as day 1-3, day 4-89, or ≥90 days of life.
All infants with iGBS disease and collection of GBS isolates were identified using the same population-based collection method, minimizing the risk of differential ascertainment.
Infant gestational age and sex
GA was estimated during pregnancy by routine early ultrasound and obtained from the Danish Medical Birth Registry. Using the unique personal identification number assigned at birth, information on GA and sex assigned at birth was linked to the included infants. GA was categorized as term ( ≥ 37 gestational weeks), moderate preterm (32–36 gestational weeks), and very preterm ( < 32 gestational weeks). Sex was included as a descriptive characteristic of the study population and is reported in Table 1. Sex-stratified analyses were not performed because there is no established evidence that the genomic characteristics of iGBS disease differ by sex, and investigation of sex differences was not an objective of this study.
Genomic analyses
WGS was performed on all available viable GBS isolates. Genomic DNA was extracted, libraries were prepared using the Illumina Nextera XT DNA Library Preparation Kit, and paired-end sequencing was performed on Illumina MiSeq or NextSeq 550 platforms. Sequence reads were quality controlled, assembled, and analyzed to determine multilocus sequence types (MLSTs), clonal complexes (CCs), alpha-like surface protein (Alp) genes, capsular polysaccharide (CPS) serotypes, and the hvgA gene. Detailed descriptions of the sequencing workflow, quality control, genome assembly, and bioinformatic analyses are provided in the Supplementary Information (Supplementary Section 1)28. A threshold of ≥ 95% sequence coverage relative to the reference sequence was required to report presence of a gene. Descriptions of the MLSTs, CCs, Alps, CPS serotypes, and hvgA, and their role in GBS virulence are provided in Supplementary Information (Supplementary Section 2).
Based on single-nucleotide polymorphism analysis of the core genome of sequenced isolates, a phylogenetic tree was constructed29. Maximum-likelihood phylogenetic inference was performed using IQ-TREE (http://www.cibiv.at/software/iqtree; version 2.3.4) with model selection by ModelFinder and branch support assessed using 100 bootstrap replicates30. The tree was visualized using the Interactive Tree of Life online tool (https://itol.embl.de/; iTOL version 6.9.1)31.
Definition of GBS clones
A GBS clone was defined as iGBS isolates belonging to the same CC and sharing identical Alp and CPS serotype profiles (CC/Alp/serotype). Phylogenetic clustering based on maximum-likelihood analysis (Fig. 1) supported the genetic relatedness of isolates within these groups. Combinations of CC/Alp/serotypes observed in fewer than ten iGBS isolates were grouped as “other GBS clones”. The list of individual clone combinations and sequence types for all included GBS isolates are provided in Supplementary Table 4.
Statistical analysis
Observations are presented as absolute numbers (n) and as percentages (%), with percentages calculated using the total number of available (non-missing) observations as the denominator. For variables with missing data, the number of missing observations (n) is reported. Results are presented stratified by categories of GA and postnatal age at iGBS disease onset.
Comparisons of categorical variables, including GBS clone distributions across gestational age groups and postnatal age categories, were performed using the χ² test with two-sided p values. A p value < 0.05 was considered statistically significant. Each participant contributed a single observation; therefore, no adjustment for non-independence was required. In accordance with Danish data protection practices, cell counts of fewer than three observations (n < 3) could not be reported. Statistical analyses were performed using Stata version 18 (StataCorp, College Station, TX, USA).
Missing data
Missing GA data were handled using multiple imputations with chained equations32. The imputation model is described in Supplementary Information (Supplementary Section 3).
A small number of infants with iGBS disease were assigned a replacement personal identifier at the time of sampling, e.g., in case of foreign maternal citizenship or electronic hospital system failure. Consequently, linkage to health registries was not possible and data on GA nor imputation was available.
All sequenced GBS isolates (n = 529), including those from infants with missing GA (n = 19) were included in overall descriptive analyses of genomic characteristics and in relation to postnatal age at disease onset. Isolates from infants with missing GA (n = 19) were excluded from analyses stratified by GA.
Sensitivity analyses
To assess the robustness of the findings with respect to potential maternal GBS vaccine–relevant disease windows, sensitivity analyses were conducted by restricting the analyses to the Alps and CPS serotypes.
In addition, alternative categorizations of postnatal age at iGBS disease onset (days 1–6 and days 7–89 of life) were applied to align with definitions used in previous studies and to facilitate comparison of results.
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Supplementary information
Acknowledgements
We thank the Danish departments of clinical microbiology and Statens Serum Institut for their essential contributions to national microbiological surveillance and for collecting and providing invasive Group B Streptococcus isolates for this nationwide study. During the preparation of this manuscript, the authors used ChatGPT (OpenAI, San Francisco, CA, USA) for language and grammar editing. The authors subsequently reviewed and edited the manuscript and take full responsibility for its content.
Author contributions
M.M., T.B.H., J.K.M., H.C.S. and S.Y.N. conceived and designed the study. M.M. and H.C.S. acquired the GBS isolates and genomic data. M.M., T.B.H., J.K.M., H.C.S. and S.Y.N. contributed to data analysis and interpretation. MM drafted the manuscript. T.B.H., M.A., K.P., J.H.E., S.E.E., H.L.N., F.B.H., N.N.L., M.K., J.K.M., H.C.S. and S.Y.N. critically revised the manuscript for important intellectual content, approved the final version, and agreed to be accountable for all aspects of the work. All authors had full access to the study data and had final responsibility for the decision to submit the manuscript for publication.
Peer review
Peer review information
Nature Communications thanks Laura Bonofiglio and Juan Lifor their contribution to the peer review of this work. A peer review file is available.
Funding
M.M., H.C.S., and S.Y.N. disclose support for the research of this work from MinervaX. T.B.H., M.A., K.P., S.E.E., H.L.N., F.B.H., N.N.L., M.K., and J.K.M. declare no relevant funding. The funder had no role in the study design, data collection, data analysis, data interpretation, manuscript preparation, or the decision to submit the manuscript for publication.
Data availability
The whole-genome sequencing data (FASTQ files) generated in this study have been deposited in the European Nucleotide Archive (ENA) under accession code PRJEB80515. The reference genome used for SNP analysis is available in the European Nucleotide Archive under accession code PRJEB43628 (https://www.ebi.ac.uk/ena/browser/view/PRJEB43628). Individual-level clinical and registry data used in this study are not publicly available because of Danish data protection legislation and restrictions imposed by Statistics Denmark and the Danish health data authorities. Access to aggregated data may be granted upon request to the corresponding author.
Competing interests
The authors declare the following competing interests: H.C.S. is a consultant for MinervaX. M.M. and S.Y.N. are employed at the Department of Clinical Microbiology, Lillebaelt Hospital, which has received research funding from MinervaX, including funding supporting the present study and other research projects. These relationships did not influence the design, conduct, analysis, interpretation, or reporting of the present study. MinervaX had no role in the study design, data collection, data analysis, data interpretation, manuscript preparation, or the decision to submit the manuscript for publication. The remaining authors declare no competing interests.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
These authors contributed equally: Hans Christian Slotved, Stine Yde Nielsen.
Supplementary information
The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-76999-y.
References
- 1.Madrid, L. et al. Infant Group B Streptococcal Disease Incidence and Serotypes Worldwide: Systematic Review and Meta-analyses. Clin. Infect. Dis.65, S160–S172 10.1093/cid/cix656 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Seale, A. C. et al. Estimates of the Burden of Group B streptococcal disease worldwide for pregnant women, stillbirths, and children. Clin. Infect. Dis.65, S200–S219 10.1093/cid/cix664 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Gonçalves, B. P. et al. Group B streptococcus infection during pregnancy and infancy: estimates of regional and global burden. Lancet Glob. Heal.10, e807–e819 10.1016/S2214-109X(22)00093-6 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.World Health Organization. Group B Streptococcus Vaccine Full Value of Vaccine Assessment. (2021).
- 5.Horváth-Puhó, E. et al. Mortality, neurodevelopmental impairments, and economic outcomes after invasive group B streptococcal disease in early infancy in Denmark and the Netherlands: a national matched cohort study. Lancet Child Adolesc. Heal.5, 398–407 10.1016/S2352-4642(21)00022-5 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Andersen, M., Matthiesen, N. B., Murra, M., Nielsen, S. Y. & Henriksen, T. B. Early-onset neonatal infection and epilepsy in children. JAMA Netw. Open8, e2519090 10.1001/jamanetworkopen.2025.19090 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Andersen, M., Matthiesen, N. B., Murra, M., Nielsen, S. Y. & Henriksen, T. B. Early-onset neonatal infection and attention deficit hyperactivity and autism spectrum disorder: a nationwide cohort study. Paediatr. Perinat. Epidemiol.39, 588–597 10.1111/ppe.70036 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Manuel, G. et al. Group B streptococcal infections in pregnancy and early life. Clin. Microbiol. Rev. 38, e0015422 10.1128/cmr.00154-22 (2025). [DOI] [PMC free article] [PubMed]
- 9.Puopolo, K. M. et al. Group B Streptococcus Infection in Extremely Preterm Neonates and Neurodevelopmental Outcomes at 2 Years. Clin. Infect. Dis. An Off. Publ. Infect. Dis. Soc. Am.75, 1405 10.1093/cid/ciac222 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Nanduri, S. A. et al. Epidemiology of Invasive Early-Onset and Late- Onset Group B Streptococcal Disease in the United States, 2006 to 2015: Multistate Laboratory and Population-Based Surveillance. JAMA Pediatrics173, 224–233 10.1001/jamapediatrics.2018.4826 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Karampatsas, K. et al. Clinical risk factors associated with late-onset invasive group b streptococcal disease: systematic review and meta-analyses. Clin. Infect. Dis.75, 1255–1264 10.1093/cid/ciac206 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Palmeira, P., Quinello, C., Ucia Silveira-Lessa, A. L., Zago, C. A. & Carneiro-Sampaio, M. IgG placental transfer in healthy and pathological pregnancies. Clin. Dev. Immunol.2012, 13 10.1155/2012/985646 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Baker, C. J. & Kasper, D. L. Correlation of maternal antibody deficiency with susceptibility to neonatal group b streptococcal infection. N. Engl. J. Med.294, 753–756 10.1056/NEJM197604012941404 (1976). [DOI] [PubMed] [Google Scholar]
- 14.Panneflek, T. J. R. et al. Intrapartum antibiotic prophylaxis to prevent Group B streptococcal infections in newborn infants: a systematic review and meta-analysis comparing various strategies. eClinicalMedicine74, 102748 10.1016/j.eclinm.2024.102748 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kyohere, M. et al. Seroepidemiology of maternally-derived antibody against Group B Streptococcus (GBS) in Mulago/Kawempe Hospitals Uganda - PROGRESS GBS. Gates Open Res.4, 155 10.12688/gatesopenres.13183.2 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Le Doare, K. et al. Association between functional antibody against Group B Streptococcus and maternal and infant colonization in a Gambian cohort. Vaccine35, 2970–2978 10.1016/j.vaccine.2017.04.013 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Pawlowski, A. et al. A group B Streptococcus alpha-like protein subunit vaccine induces functionally active antibodies in humans targeting homotypic and heterotypic strains. Cell Reports Med.3, 100511 10.1016/j.xcrm.2022.100511 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gonzalez-Miro, M. et al. Safety and immunogenicity of the group B streptococcus vaccine AlpN in a placebo-controlled double-blind phase 1 trial. iScience26, 106261 10.1016/j.isci.2023.106261 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Saukkoriipi, A. et al. Association between anti-capsular IgG levels at birth and risk of invasive group B streptococcus disease in Finnish newborns: a retrospective case–control study. The Lancet Microbe5, 689–696 10.1016/S2666-5247(24)00038-7 (2024). [DOI] [PubMed] [Google Scholar]
- 20.Perme, T. et al. Genomic and phenotypic characterisation of invasive neonatal and colonising group B Streptococcus isolates from Slovenia, 2001–2018. BMC Infect. Dis.20, 958 10.1186/s12879-020-05599-y (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.McGee, L. et al. Multistate, population-based distributions of candidate vaccine targets, clonal complexes, and resistance features of invasive group b streptococci within the United States, 2015-2017. Clin. Infect. Dis.72, 1004–1013 10.1093/cid/ciaa151 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Gent, V. et al. Surface protein distribution in Group B Streptococcus isolates from South Africa and identifying vaccine targets through in silico analysis. Sci Rep.14, 22665 10.1038/s41598-024-73175-4 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ntozini, B. et al. Molecular Epidemiology of Invasive Group B Streptococcus in South Africa, 2019–2020. J. Infect. Dis.231, e697–e707 10.1093/infdis/jiae633 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Nielsen, S. Y. et al. Timing and dosage of intrapartum prophylactic penicillin for preventing early-onset group B streptococcal disease: assessing maternal and umbilical cord blood concentration. Arch. Dis. Child. Fetal Neonatal Ed.110, F128–F132 10.1136/archdischild-2024-326986 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Lee J. H., et al. Genotypic Distribution of Alpha-Like Proteins in Group B Streptococcus Strains Isolated in Korea: Implications for Vaccine Coverage 10.3947/ic.2024.0127 (2025). [DOI] [PMC free article] [PubMed]
- 26.Hasperhoven, G. F. et al. Universal screening versus risk-based protocols for antibiotic prophylaxis during childbirth to prevent early-onset group B streptococcal disease: a systematic review and meta-analysis. BJOG127, 680–691 10.1111/1471-0528.16085 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lawn, J. E. et al. Every country, every family: time to act for group B streptococcal disease worldwide. Clin. Infect. Dis.74, S1–S4 10.1093/cid/ciab859 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lambertsen, L., Ekelund, K., Skovsted, I. C., Liboriussen, A. & Slotved, H. C. Characterisation of invasive group B streptococci from adults in Denmark 1999 to 2004. Eur. J. Clin. Microbiol. Infect. Dis.29, 1071–1077 10.1007/s10096-010-0941-z (2010). [DOI] [PubMed] [Google Scholar]
- 29.Slotved, H. C., Johannesen, T. B., Stegger, M. & Fuursted, K. Global pneumococcal sequence cluster lineage for invasive pneumococcal isolates in Denmark from summer 2019 to 2023. Sci. Rep.15, 24566 10.1038/s41598-025-10149-0 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Nguyen, L.-T., Schmidt, H. A., von Haeseler, A. & Minh, B. Q. IQTREE: a fast and effective stochastic algorithm for estimating maximum-likelihood phylogenies. Mol. Biol. Evol.32, 268–274 10.1093/molbev/msu300 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Letunic, I. & Bork, P. Interactive Tree Of Life (iTOL) v5: an online tool for phylogenetic tree display and annotation. Nucleic Acids Res.49, W293–W296 10.1093/nar/gkab301 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.von Hippel, P. T. How many imputations do you need? a two-stage calculation using a quadratic rule. Sociol. Methods Res.49, 699–718 10.1177/0049124117747303 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The whole-genome sequencing data (FASTQ files) generated in this study have been deposited in the European Nucleotide Archive (ENA) under accession code PRJEB80515. The reference genome used for SNP analysis is available in the European Nucleotide Archive under accession code PRJEB43628 (https://www.ebi.ac.uk/ena/browser/view/PRJEB43628). Individual-level clinical and registry data used in this study are not publicly available because of Danish data protection legislation and restrictions imposed by Statistics Denmark and the Danish health data authorities. Access to aggregated data may be granted upon request to the corresponding author.