Using Pneumococcal Vaccines in Combination for Maximum Protection From Ear and Lung Infections in First 3 Years of Life

BACKGROUND:

PCV13 is now the recommended PCV in Australia, as 3+0 for non-Indigenous children and as 3+1 for Indigenous children in Western Australia, Queensland and Northern Territory. The decision to recommend PCV13 over PHiD-CV10 has been primarily driven by the increase in cases of invasive pneumococcal disease due to serotype 19A(1). However for the NT, in the last 3 years there has been one case of serotype 3, 6A or 19A invasive pneumococcal disease (IPD) per year in Indigenous children less than 2 years of age. By contrast, almost 20% of young Indigenous children have perforated tympanic membranes of which 60% are associated with NTHi, compared to 3% having serotypes 3, 6A or 19A.

The PREV-IX_COMBO trial is a randomised controlled trial (RCT) of these two pneumococcal conjugate vaccines currently licensed in Australia. Participating Indigenous infants living in remote communities in 4 regions of the NT are randomised between 28 and 38 days of age and allocated to one of three groups: PHiD-CV10 at 2, 4, 6 months or PCV13 at 2,4,6 months, or an investigational group of a 4-dose combination schedule of PHiD-CV10 at 1,2, and 4 months plus PCV13 at 6 months.

PREV-IX_BOOST builds on PREV-IX_COMBO to address questions not answerable by a booster trial in the general population:

The 7 month visit is the final assessment included in the PREV-IX_COMBO trial. In October 2011 the standard PCV schedule for all Indigenous babies in the NT switched from PHiD-CV10 to PCV13 at 2, 4, 6 and 18 months of age. Thus, all babies in PREV-IX_COMBO will go on to receive PCV13 at 18 months of age as a part of standard care. An alternative and innovative option is to take this opportunity to answer these questions: should PCV13 or PHiD-CV10 vaccine be the booster vaccine for infants in the three PREV-IX_COMBO groups, and at what age should the booster be given? Secondary questions: By randomising the three different primary course groups to booster dose of either PHiD-CV10 or PCV13, a number of secondary questions can be proposed. Outputs include the comparative benefit (immune and carriage correlates) of schedules such as 3+1 PCV13, 3+1 PHiD-CV10, and mixed vaccine schedules such as single dose PCV13 or single dose PHiD-CV10 boosters following alternate vaccines in the primary course, and booster options following the early and mixed 4-dose primary course. Furthermore the design will allow for an evaluation at 18 months of age, of a 12 month boost versus no boost of specific vaccine components unique to each of these vaccines. For example, a comparison of the booster effect of HiD (in PHiD-CV10) with no HiD boost (PCV13) on antibody concentration and nasopharyngeal carriage at 18 months. By combining some of these groups it will be possible to compare more subtle differences such as differences in immunogenicity of serotypes shared by the two vaccines (1,4,5,6B,7F,9V,14,18C,19F,23F) but with different carrier proteins and thus anticipated differences in immunogenicity.

PCVs in the NT: PCV7, PHiD-CV10 (Synflorix) and PCV13 (Prevenar13):

Prevention of invasive pneumococcal disease (IPD) is the primary goal of pneumococcal conjugate vaccination in infants. Effectiveness against overall IPD has generally remained high although IPD caused by non-PCV7 types has increased in some high risk groups and regions, particularly associated with serotypes 19A, 6A, and 33F. These increases have been described as a 'replacement phenomenon'. In addition there are ongoing outbreaks of IPD due to non-PCV7 types; in the NT and WA serotypes 1 and 12F account for resurgence in IPD cases in otherwise low risk groups. Effectiveness of PCVs for all-cause pneumonia and otitis media has been greatly compromised by replacement serotypes and the broad diversity of non-pneumococcal respiratory bacterial pathogens which continue to cause high rates of disease, such as non-typeable H. influenzae (NTHi). Replacement serotypes, particularly 19A(2,3) and NTHi, are the targets of new generation vaccines. In Australia two new PCVs are licensed: 10-valent pneumococcal-Haemophilus influenzae protein-D conjugate vaccine, PHiD-CV10 (Synflorix, GSK) and 13-valent PCV, PCV13 (Prevenar13, Pfizer). The two vaccines have 10 pneumococcal serotypes in common (1,4,5,6B,7F,9V,14,18C,19F,23F), however these are conjugated to different carrier proteins and immune correlates indicate potential vaccine differences in efficacy for individual serotypes. PHiD-CV10 has protein D of NTHi as the carrier for some serotypes and PCV13 is non-toxic diphtheria protein (CRM197)-conjugated, as was its precursor PCV7, and has three additional serotypes (3, 6A and 19A) but not HiD. There have been no efficacy trials that directly compare these vaccines; licensure has been based on immune correlates of protection in comparison with PCV7. Evidence of HiD efficacy has come from a single RCT in which otitis media was a primary outcome. In that trial of the original formulation (PHiD-CV11) there was ~55% efficacy for vaccine type AOM, and 35% efficacy for NTHi AOM following a 3+1 schedule.(3) This trial also demonstrated a 43% (95% Confidence Interval (CI): -17 to 72, ns) reduction in the carriage of vaccine serotypes and a 43% (95% CI: 1-67) reduction in the carriage of H. influenzae. This finding has not been replicated in later trials, partially due to very low baseline rates of NTHi carriage (<10%) and thus low power to achieve statistical significance. The PREV-IX_COMBO trial will therefore be very important in providing data on immunological correlates and carriage for different PHiD-CV10 schedules. However, an extended follow-up and comparison of booster dose (this PREV-IX_BOOST proposal) will provide further evidence for or against inclusion of PHiD-CV10 as a booster (compared to the NT standard PCV13 at 18 months).

In the NT, booster pneumococcal vaccines are recommended at 18 months of age. This is largely due to the previous schedule which included 23-valent pneumococcal polysaccharide vaccine (23PPV) as a booster dose at 18 months. A 3+1 PHiD-CV10 schedule replaced 3+1 PCV7-23PPV in 2009, and the booster age remained at 18 months. In Queensland (QLD) and in Western Australia (WA) the vaccination schedules allow for PCV13 boost at 12 months of age for high risk and Indigenous children.

At present there are 4 intra-muscular immunizations (IMIs) at 12 months of age for Indigenous infants in the Northern Territory childhood vaccination schedule. A combined booster dose vaccine (e.g. Menitorix) became available in Australia prior to commencement of first enrolment into the BOOST study. This reduces the number of IMIs at 12 months of age and provides a 'space' for PCV at 12 months. Administration of PCV booster, originally scheduled for 13 months of age, will be at 12 months of age for the trial.

Surveillance of otitis media in remote communities shows high rates of disease from weeks after birth throughout early childhood, particularly tympanic membrane perforations in the second year of life, and thus supports the need for a booster dose. Between 2008 and 2010, in PCV7-vaccinated children, mean age of tympanic membrane perforation was 18 months; for children with AOM with perforation (AOMwiP), the mean age was 13 months and for chronic suppurative OM (CSOM) 21 months. IPD cases occurring in Indigenous and non-Indigenous children less than 2 years of age for all serotypes; in 2009 there were 7 cases; in 2011 8 cases. The majority of IPD cases in children under 2 years occurred between 12 and 18 months.

In summary, the precedent of a 12 month PCV booster for Indigenous children in WA and QLD, and the high incidence of IPD and OM in the first half of the second year of life support the decision to choose a 12 month booster for this study. The mixed schedules will still allow us to compare boost versus no boost for HiD, 3,6A and 19A.

Antibody decay post-primary PCV7 to pre-booster:

Furthermore, published data(4) from the PCV7 era which show post-primary antibody levels at mean age 15 months (range 0.32 to 0.83 microg/mL for all but serotype 14) have declined almost to those detected in cord blood (range 0.3647 to 1.1797 microg/mL). The clinical implications of this decay in terms of memory response and duration of protection from mucosal infections, pneumonia and otitis media, are not known, however vaccine efficacy is lower for mucosal compared to invasive infection(5). Which vaccine for the booster dose?

Mucosal disease (OM and respiratory illness):

Culture of ear discharge from these perforations shows 60% of specimens from AOMwiP cultured NTHi and 26% cultured pneumococcus. For CSOM, these figures were 47% and 23% respectively. Serotyping identified 19A or 6A in 20% of pneumococcal isolates, which translates to 5% AOMwip and 4% CSOM specimens. NTHi is also associated with exacerbations of chronic suppurative lung disease (CSLD), or bronchiectasis, in young Indigenous children.(6) Percentage of ear discharge specimens with OM pathogens pneumococcus (Spn), 19A, NTHi, M. catarrhalis or neither Spn nor NTHi.

Evidence for efficacy of single dose PHiD-CV10 boost \& inclusion in Australian PCV13 3+0 schedule:

A single dose of PHiD-CV10 might be valuable for many Australian children now receiving the 3+0 PCV13 schedule, to provide some immune protection from NTHi. Group 2 of the PREV-IX_COMBO trial receive either PCV13 at 2,4,6 months and will be randomised in this study to PCV13 (4th dose) or PHiD-CV10 (1st dose) at 12 months of age. Vesikari et al.(7) evaluated a single dose of PHiD-CV as catch-up (not booster) in 2 to 5 year old children. Following 1 dose in the 2 to 5 years group ELISA pneumococcal responses were lower than the reference group (reference group: 3 primary doses with booster at 12 to 15 months) for several serotypes. Responses following 1 dose in children 2 to 5 years of age suggested that 2 doses may be preferable. Unfortunately, HiD Geometric Mean Concentrations (GMCs) were not reported. A trial underway in Queensland will evaluate a double dose of PHiD-CV10 in preventing exacerbations of CSLD in older children. The BOOST trial in non-CSLD children will complement those data.

Invasive disease due to PCV13 serotypes 3, 6A and 19A:

In 2009, there were 89 cases of IPD in the NT population, 59 in 2010 and 114 in 2011; 25 (10%) of these 262 cases were the unique PCV13 serotypes (3, 6A and 19A) and 17(7%) of these were Indigenous, four (1.5%) were Aboriginal children < 2 years of age - about one per year (unpublished, Centre for Disease Control, NT Health).

Episodes of AOM and respiratory illness to 3 years of age:

A very high rate of clinic presentations occurs for babies and young children in remote communities. Our antibiotic trial for prevention of OM randomised infants at mean age 5 months(9,10) and recorded episodes of illness in placebo and active groups during approximately 6 months (unpublished). Around 9 episodes of primary respiratory illness or AOM (8.2 + 0.8) were recorded in notes per child year in the second 6 months of life. There are no published data for illness in the second and third years of life for this population. Whilst the BOOST study is unlikely to be sufficiently powered to detect group differences in rates of illness, these will be measured and reported.