Other age groups than children need to be considered as carriers of Streptococcal pneumoniae serotypes

ABSTRACT

We need to raise the issue that focus on children as the only carriage group for pneumococci is not optimal; we need to consider that other age groups might also be carriers of pneumococcal serotypes causing invasive pneumococcal diseases (IPD) in unvaccinated age groups. The pneumococcal conjugate vaccines (PCV) have successfully removed IPD from vaccinated children. Studies have shown an effect of PCV reducing the pneumococcal carriage of PCV serotypes in children. The status for several countries having used PCV for many years is that they do not see PCV serotypes neither carried nor as a cause of IPD in children.

PCV vaccination of children has shown a herd protection effect in unvaccinated groups as a reduction in IPD cases caused by PCV serotypes. However, not all PCV serotypes have disappeared as the cause of IPD in the unvaccinated age groups. The author therefore believes that if we are to see PCV serotypes disappear as a cause of IPD in unvaccinated age groups, we need to perform further carriage studies to examine carriage in other age groups. Alternatively, all age groups should be vaccinated against pneumococci to eliminate IPD caused by PCV serotypes from possible hidden carriers.

Introduction

For many years, Streptococcus pneumoniae (pneumococci) has been a worldwide cause of morbidity and mortality among children and the elderly.¹ Invasive pneumococcal disease (IPD) is one of the most frequent types of bacteraemia and meningitis in infants globally.² However, since the introduction of pneumococcal conjugate vaccine (PCV) PCV-7 in 2000,³ the epidemiology of pneumococcus has changed particularly in children. Until recently, PCV vaccination has mainly been focused on children, partly because they are considered to be the main carriers of pneumococci.^4-6

The author's purpose with this commentary is to raise the issue that focusing on children as the only carriage group for pneumococcus is not optimal. It is necessary to consider more widely that other age groups might also be carriers of the pneumococcal serotypes causing IPD in other age groups than children.

Pneumococcal conjugate vaccines (PCV)

The PCV was introduced to children, and effective protection has been provided in children against IPD caused by the serotypes included in the vaccines.⁷ The first pneumococcal conjugate vaccine was Prevenar 7 (PCV-7) (Pfizer Vaccines) covering 7 different serotypes and it was licensed in the USA in 2000.³ Since then, the 10-valent pneumococcal Haemophilus influenzae protein D conjugate vaccine (PCV-10) (Synflorix, GlaxoSmithKline Biologicals) and Prevenar 13 (PCV-13) (Pfizer Vaccines) have been introduced, which expanded the vaccine coverage to 10 and 13 serotypes, respectively.^7,8 Further vaccines are underway expanding the coverage further with the PCV-15 as the most recent example.⁹ The 15-valent PCV vaccine, containing the PCV-13 serotypes plus serotypes 22F and 33F,⁹ is being tested in clinical trials in the USA.

In general, the focus for these vaccines has been children less than 5 y of age; not until recent years, the focus has changed to also include the rest of the population and in particular the elderly. A major example is the Community-Acquired Pneumonia immunisation Trial in Adults (CAPiTA) performed in the Netherlands, where the study focuses on IPD in adults and their PCV vaccination status.¹⁰

In many countries, PCV vaccination of children is a part of the national vaccination programmes resulting in high vaccine coverage. Pneumococcal vaccination of adults is generally recommended; however, there is no systematic vaccination of the elderly, and data on their vaccination status is limited.^11,12 However, with the expanded licensing of the PCV-13 vaccine for all age groups,¹³ vaccination strategies including all age groups have been suggested,^13,14

PCV vs. IPD

In many countries, vaccination with PCV (PCV-7, PCV-10 and PCV-13) of children has been followed by a successful reduction in IPD cases caused by the vaccine serotypes in children.^7,8,15

It has furthermore been shown that the protection of the vaccinated children is accompanied by a herd protection effect for the unvaccinated age groups, particularly for the age group 65+ years.^4,16 Thus, a reduction in IPD due to serotypes within the PCV's has been shown in age groups who have not received the PCV. Particularly, the introduction of the PCV-7 showed a pronounced reduction of IPD caused by PCV-7 serotypes in unvaccinated groups.⁷ Also PCV-10 and PCV-13 have shown this effect although the effect has not been as pronounced as with the PCV-7.^7,17

PCV vs. carriage

Several studies have shown that the PCV has an effect on the carriage of PCV serotypes in children.^18-23 It has been shown that children act as reservoirs and carriers of different pneumococcal serotypes,^4,5,6 and it is generally believed that children are the main carriers and thus the most important source of transmission of IPD serotypes to the elderly.^4,18 In a review study on the effect of PCV vaccination schedules on the carriage in children, it was shown that PCV-7 serotype carriage was reduced when using these vaccination schedules: 2+0, 2+1, 3+0 and 3+1.²⁴ So although there is an ongoing discussion of whether the PCV reduces the carriage in children, data have been presented showing a positive effect of PCV on the carriage of vaccine serotypes.

PCV vaccine failure for the protection of IPD

PCV vaccine failures in children are reported around the world. However, they are considered to be seldom and often in combination with suspected immunosuppression.^25,26 A Danish study found vaccine failure in 3 cases with PCV-7 over a 31-month surveillance period.²⁷ A general vaccine failure in children for some of the serotypes does not seem to be the explanation for the continued appearance of IPD due to PCV serotypes in other age groups.

What about PPV-23 on the IPD and carriage?

The pneumococcal polysaccharide vaccine (PPV-23) includes 23 serotypes (1, 2, 3, 4, 5, 6B, 7F, 8, 9N, 9V, 10A, 11A, 12F, 14, 15B, 17F, 18C, 19F, 19A, 20, 22F, 23F and 33F) and is solely based on a capsular polysaccharide combination²⁸ in contrast to the PCV vaccines. The effect of PPV-23 on IPD has been discussed for many years and is still an issue,^28,14 although it is generally accepted that PPV-23 does reduce IPD in immunocompetent adults.²⁹ It has been recommended that the 65+ years age group should be vaccinated with the PPV-23; however, systematic vaccination data on elderly patients are limited, and therefore the coverage rate of pneumococcal vaccination of the elderly is difficult to assess. The general problem with PPV-23 is that the vaccine coverage in adults is low.^11,12,30

With regard to carriage in the vaccination group, it has been shown in several studies that PPV-23 does not seem to reduce carriage.^24,28 In a review study by Fleming-Dutra et al.,²⁴ they found that the use of PPV-23 as a vaccination booster in children did not have any effect on carriage compared to using no booster.

Why have PCV serotypes not disappeared? (Table 1)

Table 1.

This table shows carriage and IPD data of the serotypes included in the PCV-7, PCV-10 and PCV-13 for a selected number of countries. The presented data are partial data from the mentioned references and do not represent the number of cases, but only the appearance of serotypes.

Study		PCV-7 (4,6B,9V,14,18C,19F,23F)	PCV-10 (1,5,7F)	PCV-13 (3,6A,19A)	Vaccine started	Samples collected
Esposito et al. 2016	Carriage study (adult >64) in Italy	4,9V,19F	1	3,6A,19A	Not mentioned	2015
Azzari et al. 2016	Carriage study (children (0–60 months) in Italy	4,14,19F,23F	1,5,7F	3,19A	PCV-7 since 2006 and PCV-13 since July 2010	2009–2010
Azzari et al. 2016	IPD study (adults >18 year) in Italy	4,14,19F,23F	1,5,7F	3,19A	PCV-7 since 2006 and PCV-13 since July 2010	2007–2014
Steens et al. 2015	Carriage study (< 77 months) Norway	18C,19F	1,7F	3,6A,19A	PCV-7 July 2006 and PCV-13 April 2011	2013
Steens et al. 2013	IPD study (< 5 years) (Norway)	9V	1	None	PCV-7 July 2006 and PCV-13 April 2011	2012
Steens et al. 2013	IPD study (≥ 5 years) (Norway)	4,6B,9V,14,18C,19F,23F	1,7F	3,6A,19A	PCV-7 July 2006 and PCV-13 April 2011	2012
Gounder et al. 2014	Carriage (< 5 years) Alaska, USA (n=938)	19F	7F	3,6A,19A	PCV-7 2001 and PCV-13 April 2010	2011–2012
Gounder et al. 2014	Carriage (≥ 18 years) Alaska, USA (N=521)	19F	7F	3,6A,19A	PCV-7 2001 and PCV-13 April 2010	2011–2012
Bruce et al. 2015	IPD (< 5 years) Alaska, USA	19F	7F	3,19A	PCV-7 2001 and PCV-13 April 2010	2010–2013
Bruce et al. 2015	IPD (all ages) Alaska, USA	4,14,19F	1,7F	3,6A,19A	PCV-7 2001 and PCV-13 April 2010	2010–2013
van Hoek et al. 2014	Carriage (< 5 years) England	19F	None	3	PCV-7 September 2006 and PCV-13 April 2010	2012/2013
van Hoek et al. 2014	Carriage (≥ 5 years) England	19F	7F	19A	PCV-7 September 2006 and PCV-13 April 2010	2012/2013
Waight et al. 2015	IPD study England and Wales (< 5 years)	Yes (not specified)	1,7F	3,19A	PCV-7 September 2006 and PCV-13 April 2010	2013/2014
Waight et al. 2015	IPD study England and Wales (≥ 5 years)	Yes (not specified)	1,7F	3,6A,19A	PCV-7 September 2006 and PCV-13 April 2010	2013/2014
Slotved et al. 2016	IPD study (< 5 years) (Denmark)	none	1,7F	none	PCV-7 October 2007 and PCV-13 2010	2014
Slotved et al. 2016	IPD study (≥ 5 years) (Denmark)	4,6B,9V,18C,19F,23F	1,7F	3,6A,19A	PCV-7 October 2007 and PCV-13 2010	2014
Bosch et al. 2016	Carriage (≤ 24 months) Netherlands	6B,19F	1,7F	3,6A,19A	PCV-7 June 2006 and PCV-10 March 2011	2012/2013
Bosch et al. 2016	Carriage (parents) Netherlands	none	7F	3,19A	PCV-7 June 2006 and PCV-10 March 2011	2012/2013
Knol et al. 2015	IPD (≤ 4 years) Netherlands	Yes (not specified)	Yes (not specified)	No information	PCV-7 June 2006 and PCV-10 May 2011	2013–2014
Knol et al. 2015	IPD (> 4 years) Netherlands	Yes (not specified)	Yes (not specified)	No information	PCV-7 June 2006 and PCV-10 May 2011	2013–2014
van der Linden et al. 2015	IPD (≤ 4 years) Germany	6,14,19F	1,7F	3,19A	PCV-7 licensed 2001/intro July 2006, PCV-10 April 2009 and PCV-13 December 2009	2013–2014
van der Linden et al. 2015	IPD (> 4 years) Germany	4,6B,9V,14,18C,19F,23F	1,7F	3,6A,19A	PCV-7 licensed 2001/intro July 2006, PCV-10 April 2009 and PCV-13 December 2009	2013–2014

Table 1 presents extracted data from both IPD and carriage studies from a selected number of countries, which have performed PCV vaccination of children for many years. The presented studies have collected IPD/carriage samples within recent years; for details see Table 1.

Studies show that IPD due to PCV serotypes are disappearing from the vaccinated children. However, this is not the case for some of the PCV serotypes when looking at the elderly in particular.^12,17 Studies have shown that there does not seem to be a hidden circulation of PCV serotypes carried in vaccinated infants, while PCV serotypes still contribute to IPD in unvaccinated individuals.³¹ It has also been shown that some of the additional PCV-13 serotypes are still a major cause of IPD in the group of elderly persons, and that vaccination of children with PCV-13 has not shown a reduction in IPD with these serotypes in this age group.^12,17 Serotypes 3, 7F, and 19A are examples of PCV serotypes that do not seem to be reduced in the elderly group, even though children have been vaccinated with PCV-13 for several years.^12,17 All these studies support the possible presence of a S. pneumoniae reservoir outside vaccinated children, an observation also made by other authors.^12,17,31

Non-PCV serotypes

In recent years, the major focus has been on predicting which non-vaccine IPD serotypes will emerge, and several studies have shown data on the most common non-PCV serotypes in specific regions.^4,12,15 Following the introduction of the PCV vaccination program for children, the majority of PCV serotypes became rare in both children and adults due to a combination of immunity and herd protection.^4,7,15,18 In a study by Slotved et al.,¹² it was shown that the incidence rate for 10 non-PCV serotypes showed an age-dependent difference. In general, the IPD incidence of these serotypes was low in the age groups 0–4 y and 5–64 years, while it was relatively high in the 65+ years age group. This was also observed in other studies,^4,5 where serotypes not included in the vaccines were more often the cause of IPD in older children and adults than in young children.

A carriage study in children (10 to 77 months of age) in Norway did not find serotypes 8, 12F and 20, which are often found in IPD among patients over 65 y of age in Denmark¹⁸ and in patients over 5 y of age in Norway.¹⁵ Thus, the carriage study from Norway did not provide information on whether children are the main reservoir of these non-PCV serotypes (8, 12F, 20), or whether other age groups are more important.

Conclusion

It is the author's belief that although children have been found to be main carriers of the pneumococcal serotypes included in the PCV-13 vaccines, this is probably not the case for many of the non-vaccine serotypes. It is also the author's belief that several of the PCV-13 serotypes may also be carried in other age groups than children. This may explain why the level of IPD caused by pneumococcal serotypes such as serotypes 3, 7F, and 19A has not been reduced by the herd protection effect, but continues to be a problem in the elderly. It is the author's opinion that if we are to obtain maximum benefit from pneumococcal vaccination, carriage studies should be performed for all age groups showing where the serotypes are carried; based on these data, nationwide pneumococcal vaccination strategies ought to be planned. Alternatively, the strategy of only vaccinating children should be changed to include the entire population of a country. Selection of a specific pneumococcal vaccine should be based on carriage and IPD rate data on the different age groups, and vaccine recommendations for a country may be a combination of several different pneumococcal vaccines for different age groups.

Abbreviations

IPD

Invasive Pneumococcal Diseases

PCV

Pneumococcal Conjugate Vaccines

PPV-23

Pneumococcal Polysaccharide Vaccine includes 23 serotypes

Disclosure of potential of conflicts of interest

H-C Slotved has participated in a project supported by Pfizer and is employed at Statens Serum Institut, which until recently included a subdivision producing antisera for the typing of pneumococci.

Acknowledgments

Marian Jørgensen, Karen A. Krogfelt, and Helle B. Konradsen are thanked for their helpful comments to this manuscript.

Author contributions

Hans-Christian Slotved has drafted the manuscript and approved the final manuscript.