Abstract

BACKGROUND: Widespread use of pneumococcal conjugate vaccines (PCVs) has reduced vaccine-type invasive pneumococcal disease (IPD). We describe the serotype distribution of IPD after extensive use of ten-valent PCV (PCV10; Synflorix, GSK) and 13-valent PCV (PCV13; Prevenar 13, Pfizer) globally. METHODS: IPD data were obtained from surveillance sites participating in the WHO-commissioned Pneumococcal Serotype Replacement and Distribution Estimation (PSERENADE) project that exclusively used PCV10 or PCV13 (hereafter PCV10 and PCV13 sites, respectively) in their national immunisation programmes and had primary series uptake of at least 70%. Serotype distribution was estimated for IPD cases occurring 5 years or more after PCV10 or PCV13 introduction (ie, the mature period when the serotype distribution had stabilised) using multinomial Dirichlet regression, stratified by PCV product and age group (<5 years, 5-17 years, 18-49 years, and ≥50 years). FINDINGS: The analysis included cases occurring primarily between 2015 and 2018 from 42 PCV13 sites (63 362 cases) and 12 PCV10 sites (6806 cases) in 41 countries. Sites were mostly high income (36 [67%] of 54) and used three-dose or four-dose booster schedules (44 [81%]). At PCV10 sites, PCV10 serotypes caused 10·0% (95% CI 6·3-12·9) of IPD cases in children younger than 5 years and 15·5% (13·4-19·3) of cases in adults aged 50 years or older, while PCV13 serotypes caused 52·1% (49·2-65·4) and 45·6% (40·0-50·0), respectively. At PCV13 sites, PCV13 serotypes caused 26·4% (21·3-30·0) of IPD cases in children younger than 5 years and 29·5% (27·5-33·0) of cases in adults aged 50 years or older. The leading serotype at PCV10 sites was 19A in children younger than 5 years (30·6% [95% CI 18·2-43·1]) and adults aged 50 years or older (14·8% [11·9-17·8]). Serotype 3 was a top-ranked serotype, causing about 9% of cases in children younger than 5 years and 14% in adults aged 50 years or older at both PCV10 and PCV13 sites. Across all age and PCV10 or PCV13 strata, the proportion of IPD targeted by higher-valency PCVs beyond PCV13 was 4·1-9·7% for PCV15, 13·5-36·0% for PCV20, 29·9-53·8% for PCV21, 15·6-42·0% for PCV24, and 31·5-50·1% for PCV25. All top-ten ranked non-PCV13 serotypes are included in at least one higher-valency PCV. INTERPRETATION: The proportion of IPD due to serotypes included in PCVs in use was low in mature PCV10 and PCV13 settings. Serotype distribution differed between PCV10 and PCV13 sites and age groups. Higher-valency PCVs target most remaining IPD and are expected to extend impact. FUNDING: Bill & Melinda Gates Foundation as part of the WHO Pneumococcal Vaccines Technical Coordination Project.