Background

While regulatory endpoints for pneumococcal conjugate vaccines (PCVs) routinely include radiologically- or vaccine serotype (VST) confirmed disease, public health decision-making benefits from consideration of all disease prevented regardless of diagnostic or etiological confirmation.

Methods

We followed PRISMA guidelines to perform a systematic literature review for Phase III/IV efficacy trials of PCVs from 1997 through 2019. We estimated study-specific vaccine-preventable disease incidence (VPDI) (control group minus intervention group incidences) for all-cause disease versus radiologically- and/or etiologically-confirmed outcomes. VPDI ratios between the broadest clinical and most narrowly defined outcomes were calculated [PROSPERO registration, CRD42019145268].

Results

Ten studies met the criteria. In children < 5 years, VPDI ratios ranged from 0.6 to 3.7 for otitis media (clinical versus VST etiologically-confirmed); from 1.3 to 1.8 for pneumonia (clinical versus radiologically-confirmed); 3.1 and 19.0 for pneumonia (clinical versus bacterial or VST); and 3.8 in one study of invasive pneumococcal disease (non-laboratory to laboratory-confirmed). In adults, VPDI ratios ranged from 2.2 to 2.9 for pneumonia (clinical to pneumococcal or VST).

Conclusions

Relying on only radiologically-confirmed or etiologically-confirmed outcomes substantially underestimated PCVs’ public health benefits. Broader clinical outcomes should be considered by vaccine technical committees when making decisions for pediatric and adult PCV use.

Background

In the Netherlands randomized controlled trial (RCT) of 13-valent pneumococcal conjugate vaccine (PCV13) among persons age 65+ years, the primary outcome was chest x-ray (CXR) confirmed, vaccine-type (VT) primarily non-bacteremic Community Acquired Pneumonia (CAP), the latter determined almost entirely by serotype-specific urinary antigen detection (SSUAD). The sensitivity and specificity of SSUAD and CXR for identifying primarily non-bacteremic pneumonia preventable by PCV13 is unknown.

Methods

In the Netherlands RCT, we compared vaccine-preventable-disease-incidence (VPDI) for all episodes of clinically defined CAP to that for CXR-confirmed VT-CAP. VPDI was calculated as the incidence rate difference between controls and vaccinated persons.

Results

VPDI per 100,000 person-years of follow-up for all episodes of clinical CAP vs. VT-CAP were: total population, 72.2/25.1 (2.9-fold); at-risk condition for pneumococcus other than immunosuppression, 121.4/34.3 (3.5-fold); and not at-risk, 35.5/17.6 (2.0-fold) (Table).

Conclusions

PCV13 prevented 2.0 to 3.5 more episodes of clinical CAP than CXR-confirmed VT-CAP, reflecting that SSUAD was designed for the regulatory outcome of vaccine efficacy, which relies on test specificity, and that CXR may have imperfect sensitivity for CAP in elderly persons. Assessing PCV13’s true public health value will require use of rate reductions or adjustment factors to account for clinically defined CAP.

Background

Quantification of Streptococcus pneumoniae (Sp) carriage density and identification of serotypes in contacts contribute to understanding Sp transmission and thus evaluating pneumococcal conjugate vaccine impact.

Methods

266 lytA-positive nasopharyngeal samples (NPS) taken during season 1 (of 2, October-December 2017) of the Transmission of Pneumococcus (TOP) study (see abstract 715) were analysed by microarray. NPS were collected from 120 families, 2-weekly over 2 months.

Results

83/120 index children (2-year-olds) were lytA positive at baseline (visit 1): the three most common serotypes were 11A, 15B and 35F (n=83) (Graph 1).

In 17/28 families (28 index children, 45 contacts: 26 <16-year-olds and 19), index children shared ≥1 strain/serotype with ≥1 household contact.

40% (107/266) of lytA-positive NPS analysed had multiple serotypes/strains.

Sp was not detected in some lytA-positive samples but rather related streptococcal species, particularly from participants aged ≥5 years (Table 1).

Conclusions

NP carriage of the same Sp serotypes by multiple household members suggests that household Sp transmission occurs.

Multiple serotype carriage was observed in a significant proportion of participants.

Among young children, the age group most likely to transmit Sp, lytA was more predictive of true Sp infection than it was among older persons.

Background

Pneumococcal disease is a major public health concern globally and particularly in Burkina Faso where PCV13 was introduced nationwide into the routine immunization schedule in 2013. We sought to evaluate vaccine impact on all-cause pneumonia hospitalizations among children < 5 years.

Methods

We retrospectively collected hospitalization data over 10 years (January 1^st 2009-December 31^st 2018) in 4 rural district hospitals, using medical records to extract data on relevant variables. We used interrupted time series and segmented regression to estimate the effect and impact of PCV13 on the rates of pneumonia hospitalizations. Severe acute malnutrition and unintentional injury were used as control conditions.

Results

We found a vaccine effectiveness of 34% (95% CI : 16-49, p=0·001), 24% (95% CI : 2-41, p=0·032) and 50% (95% CI : 30-64, p<0·001) against all-cause pneumonia hospitalizations among children < 5 years, < 2 years, and 2-4 years, respectively. By October 2018, PCV13 had led to an absolute reduction in the pneumonia hospitalization rate of 348 cases per 100,000 among children < 5 years. No decline was observed for control conditions.

Conclusions

Our estimates point to a substantial public health impact of PCV13 against pneumonia hospitalizations among children < 5 years in Burkina Faso.

Background

While regulatory endpoints for pneumococcal conjugate vaccines (PCVs) routinely include radiologically- or vaccine serotype (VST) confirmed disease, public health decision-making benefits from consideration of all disease prevented regardless of diagnostic or etiological confirmation.

Methods

We followed PRISMA guidelines to perform a systematic literature review for Phase III/IV efficacy trials of PCVs from 1997 through 2019. We estimated study-specific vaccine-preventable disease incidence (VPDI) (control group minus intervention group incidences) for all-cause disease versus radiologically- and/or etiologically-confirmed outcomes. VPDI ratios between the broadest clinical and most narrowly defined outcomes were calculated [PROSPERO registration, CRD42019145268].

Results

Ten studies met the criteria. In children < 5 years, VPDI ratios ranged from 0.6 to 3.7 for otitis media (clinical versus VST etiologically-confirmed); from 1.3 to 1.8 for pneumonia (clinical versus radiologically-confirmed); 3.1 and 19.0 for pneumonia (clinical versus bacterial or VST); and 3.8 in one study of invasive pneumococcal disease (non-laboratory to laboratory-confirmed). In adults, VPDI ratios ranged from 2.2 to 2.9 for pneumonia (clinical to pneumococcal or VST).

Conclusions

Relying on only radiologically-confirmed or etiologically-confirmed outcomes substantially underestimated PCVs’ public health benefits. Broader clinical outcomes should be considered by vaccine technical committees when making decisions for pediatric and adult PCV use.

Background

Despite the indirect effects of routine pediatric pneumococcal conjugate vaccine (PCV) use, invasive pneumococcal disease (IPD) persists in older adults. This analysis describes IPD coverage of current PCVs (PCV10/PCV13) and next generation, PCV15 and PCV20.

Methods

Serotype counts for adults ≥60/≥65 years were obtained from national or regionally-representative IPD surveillance systems in high-income countries. Up to the three most recent years of data were included. Excluded countries reported <20 cases overall or only reported data prior to pediatric PCV introduction into the National Immunization Program (NIP). Coverage was calculated for PCV10 (1,4,5,6B,7F,9V,14,18C,19F,23F), PCV13 (PCV10-types,3,6A,19A, plus 6C, which is highly-related to 6A), PCV15 (PCV13-types, 22F,33F), and PCV20 (PCV13-types, 8,10A,11A,12F,15B, plus 15C, which is highly-related to 15B) as the number of cases in PCVs divided by total number of cases (excluding missing and non-typeable) reported for each country.

Results

Twenty-eight of 80 (35%) high-income countries met inclusion criteria, reporting >25,000 cases that were included in the analysis. Coverage by current and next generation PCVs varied by country and World Health Organization (WHO) region (Table).

Conclusions

PCV20 offers substantial coverage above current PCVs and next generation PCV15 and could address the considerable remaining IPD burden among older adults in high-income countries globally.

Background

In Canadian adults aged ≥65, routine pneumococcal-polysaccharide-vaccine (PPV23) use has been recommended for several decades, and 13-valent pneumococcal conjugate vaccine (PCV13) on an individual basis - since 2016. When PCV13 is administered, sequential PPV23 is recommended to broaden serotype coverage.

We aimed to assess the proportion of PCV13-, PPV23/non13-, PCV15-, and PCV20-type IPD among Canadian older adults.

Methods

Proportions of IPD due to PCV13, PPV23/non13, PCV15 and PCV20 serotypes were calculated from case counts reported, by serotype and age group, from the National Microbiology Laboratory between 2010 and 2017.

Results

Among all IPD, the contribution of PCV13-type declined from 50% (487/967) to 23% (287/1,238). PPV23/non13-type IPD increased from 25% (240/967) to 39% (487/1,238). In 2017, IPD proportions due to PCV15- and PCV20-types were 36% (447/1,238) and 52% (646/1,238) (Figure 1).

Conclusions

Despite routine PPV23 program, increasing trend in cases and proportions of PPV23/non13-type IPD was noted during the observation period. Initial decline in PCV13-type IPD, likely an indirect effect of pediatric program, leveled off in recent years supporting the need for broad PCV13 use in older adults. PCV20, which shares 19 serotypes with PPV23, could address limitations of the PPV23 program in the ability to control vaccine-type IPD.

Background

In the Netherlands randomized controlled trial (RCT) of 13-valent pneumococcal conjugate vaccine (PCV13) among persons age 65+ years, the primary outcome was chest x-ray (CXR) confirmed, vaccine-type (VT) primarily non-bacteremic Community Acquired Pneumonia (CAP), the latter determined almost entirely by serotype-specific urinary antigen detection (SSUAD). The sensitivity and specificity of SSUAD and CXR for identifying primarily non-bacteremic pneumonia preventable by PCV13 is unknown.

Methods

In the Netherlands RCT, we compared vaccine-preventable-disease-incidence (VPDI) for all episodes of clinically defined CAP to that for CXR-confirmed VT-CAP. VPDI was calculated as the incidence rate difference between controls and vaccinated persons.

Results

VPDI per 100,000 person-years of follow-up for all episodes of clinical CAP vs. VT-CAP were: total population, 72.2/25.1 (2.9-fold); at-risk condition for pneumococcus other than immunosuppression, 121.4/34.3 (3.5-fold); and not at-risk, 35.5/17.6 (2.0-fold) (Table).

Conclusions

PCV13 prevented 2.0 to 3.5 more episodes of clinical CAP than CXR-confirmed VT-CAP, reflecting that SSUAD was designed for the regulatory outcome of vaccine efficacy, which relies on test specificity, and that CXR may have imperfect sensitivity for CAP in elderly persons. Assessing PCV13’s true public health value will require use of rate reductions or adjustment factors to account for clinically defined CAP.

Background

In Sweden, pneumococcal serotype distribution in adults with community-acquired pneumonia (CAP) and potential coverage of licensed and developmental pneumococcal conjugate vaccines (PCVs) are unknown.

Methods

2016-2018, consecutive patients aged ≥18 years hospitalized with chest x-ray positive (CXR+) CAP were enrolled at Skåne University Hospital. Streptococcus pneumoniae (Spn) blood culture isolates were serotyped by multiprime PCR and Quellung reaction. Urine was tested by the pan-pneumococcal urinary antigen test (BinaxNOW®) and Pfizer’s proprietary serotype-specific urine antigen detection assays (UAD1/UAD2). UAD1 detects serotypes in PCV13, UAD2 detects additional serotypes in PCV15 and PCV20 plus serotypes 2,9N,17F and 20.

Results

Of 567 enrollees, 518 had CXR+CAP and urine sample collected and were included in analysis. Spn serotypes were identified by UAD or blood culture isolates.

Table CXR+CAP by age group and vaccine serotype categories.

*PCV13:1,3,4,5,6A/6C,6B,7F,9V,14,18C,19A,19F,23F

PCV15:PCV13+22F,33F

PCV20:PCV15+8,10A,11A,12F,15B/C

Conclusions

In the context of robust pediatric PCV immunization, PCV13 serotypes were relatively common in adult CXR+CAP, emphasizing the limits of relying on indirect protection. PCV20 will further increase the ability of direct vaccination to reduce adult pneumonia morbidity.

Background

Interruption of vaccine serotype transmission underlies effectiveness of pneumococcal conjugate vaccine (PCV) programmes at population level. Streptococcus pneumoniae (Sp) nasal carriage density is affected by upper respiratory tract viral infections and PCV. The impact of carriage density on transmission is unknown.

In a multi-centre prospective randomised stepped-wedge trial, we have used the Live Attenuated Influenza Vaccine (LAIV) to modulate Sp carriage density in 2-year-olds and assess impact on household transmission.

Methods

410 families were recruited across 10 UK sites. Families were randomised 1:1 to early or late (4 weeks later) index-child LAIV; saliva and nasopharyngeal samples (NPS) were collected 2-weekly over 2 months. Samples are being analysed for Sp by real-time quantitative PCR (lytA), culture amplification and microarray.

Results

SP nasal carriage rate, densities and density range are highest in our index children (Table 1 for baseline (visit 1) and Figure 1 for visits 1-5).

Conclusions

The range of Sp carriage density is very wide with rates and densities in keeping with the literature. The impact of density on Sp transmission will be analysed when full results are available, June 2020.

Background

The serotype distribution of invasive pneumococcal disease (IPD) informs the coverage of next generation pneumococcal conjugate vaccines (PCVs). This analysis describes IPD coverage by the current PCVs (PCV10/PCV13) and next generation, PCV15 and PCV20.

Methods

Serotype counts for children <5 years were obtained from national or regionally-representative IPD surveillance systems from high-income countries. Up to the three most recent years of data were included. Excluded countries reported <20 cases overall or only reported data prior to pediatric PCV introduction into the National Immunization Program (NIP). Coverage was calculated for PCV10 (1,4,5,6B,7F,9V,14,18C,19F,23F), PCV13 (PCV10-types, 3,6A,19A, plus 6C, which is highly-related to 6A), PCV15 (PCV13-types, 22F,33F), and PCV20 (PCV15-types, 8,10A,11A,12F,15B, plus 15C, which is highly-related to 15B) as the number of cases in PCVs divided by the total number of cases (excluding missing and non-typeable cases) reported for each country.

Results

Twenty-seven of 80 (34%) high-income countries met inclusion criteria, reporting >5,000 cases that were included in the analysis. Coverage by current and next generation PCVs varied by country and World Health Organization (WHO) region (Table).

Conclusions

PCV20 offers substantial coverage above current PCVs and next generation PCV15 and could address the considerable remaining IPD burden among children in high-income countries globally.