Optimal age targeting for pneumococcal vaccination in older adults; a modelling study

Abstract BackgroundInvasive pneumococcal disease (IPD) risk increases with age for older adults whereas the population size benefiting from pneumococcal vaccines and the robustness of the immunogenic response to vaccination decline. This poses a conundrum for identifying the optimal age for vaccination. We estimate how demographics, vaccine efficacy/effectiveness (VE), and waning VE impact on optimal age for single-dose pneumococcal vaccination. MethodsAge- and vaccine-serotype-stratified IPD incidence from routine surveillance of adults ≥55 years old (y) in Brazil, England, Blantyre (Malawi), and South Africa, ≥4-years after infant-pneumococcal vaccine introduction and before 2020, was used to parameterise exponential growth models of increasing IPD risk with age. A piecewise-constant model estimated VE and waning VE from prior studies. All estimates were then combined in a cohort model to assess the vaccine preventable IPD burden of delivering 13-, 15-, 20-valent conjugate or 23-valent polysaccharide vaccines at various ages. FindingsIn Brazil, Malawi, South Africa and England 51%, 51%, 54% and 39% of adults older than 55y were younger than 65y old. A smaller share of IPD was reported among adults <65y old in England (4,657 annual cases; 20% <65y) compared to Brazil (186; 45%), Malawi (4; 63%), or South Africa (134, 48%). PCV13- and PPV23-serotypes were estimated to cause 39% and 85% of IPD cases on average across settings. Vaccination at 55y in Brazil, Malawi, and South Africa, and at 70y in England had the greatest potential for IPD prevention with 38%, 31%, 27%, 8% of IPD preventable if using PCV13, 44%, 32%, 30%, 13% with PCV15, 65%, 69%, 53%, 37% with PCV20, and 30%, 29%, 20%, 14% with PPV23, respectively. Vaccination efficiency or cost effective use was optimal in 60-65y in Brazil, Malawi and South Africa, and in 80-85y in England. InterpretationIn low/middle-income countries, pneumococcal vaccines may prevent a substantial proportion of the residual IPD burden if administered earlier in adulthood than is typical in high-income countries. FundingUK National Institute for Health and Care Research. Research in contextEvidence before this studyA recent review conducted for the World Health Organisation (WHO) Scientific Advisory Group of Experts (SAGE) working group on pneumococcal vaccines reviewed the evidence on the efficacy, effectiveness and potential impact of pneumococcal vaccines in older adults. At the time, little evidence was identified to support recommendations for the age of pneumococcal vaccination in older adults, particularly in LMICs. We did an updated search of PubMed, Medline, Embase and Web of Science for studies on age targeting for pneumococcal vaccination in older adults published from Jan 1, 2003 to June 30, 2022, using the search terms “(optimal age targeting OR age targeting OR timeliness) AND (pneumococcal vaccination OR pneumococcal vaccines) AND (adults OR older adults OR elderly)”. We identified one study on the role of timeliness in the cost-effectiveness of older adult vaccination in Australia. This study explored the impact of the timeliness of pneumococcal conjugate vaccination (PCV) in older adults and found that more hospitalisations and deaths can be prevented if PCV13 was given to 70y old than as recommended at 65y. On the other hand, most low-income countries (LICs) and middle-income countries (MICs) do not have routine older adult pneumococcal vaccination programmes, despite often less indirect protection from childhood PCV programmes and high burden of vaccine preventable disease. The implications of differences in population demographics and risk for IPD compared to high income countries on the optimal age for pneumococcal vaccination in LICs/MICs are unknown.Added value of this studyThis multicountry study uses data from long-standing older adult pneumococcal surveillance programmes at the national/sub-national level, to explore the optimal age-targeting for a single-dose pneumococcal vaccination against vaccine-serotype-IPD in older adults living without human immunodeficiency virus in LIC (Malawi), MICs (Brazil, South Africa), and HIC (England). We show that despite mature pneumococcal infant vaccination programmes there is a substantial burden of vaccine preventable pneumococcal disease remaining in the four countries considered. Vaccinating at 55y in the LICs/MICs was optimal whereas vaccinating at 70y prevented most IPD cases in England than vaccinating at other ages. While the number of older adults needed to vaccinate to prevent an IPD case may be higher for programmes vaccinating older individuals, vaccination efficiency or vaccine cost effective use was optimal in 60-65y in LICs/MICs, and in 80-85y in England.Implications of all available evidenceLICs/MICs with a mature infant immunisation programme for pneumococci and a remaining high burden of vaccine preventable pneumococcal disease in older adults may consider implementation of a vaccination programme for older adults, albeit at a younger age than typical in HICs. Affordability and cost-effectiveness of an adult vaccination programme as well as the indirect effects following potential changes to higher valency infant PCV vaccination programmes will also need to be considered.