Mapping the Existing Body of Knowledge on New and Repurposed TB Vaccine Implementation: A Scoping Review

Study design overview

We employed a narrative synthesis approach¹⁰, sourcing literature from PubMed, Medrxiv, PLOS journals, and expert suggestions. Articles were selected using a systematic method, and data extraction and analysis were carried out iteratively using the Roadmap⁸ and ECVP⁹.

Search Methods

The narrative synthesis approach involved three systematic search components: (1) database search, (2) search on internet sites, and (3) identification of relevant articles by the team. The search strings for the database queries combined the following concepts: TB, TB vaccination, and vaccine preparedness. A combination of Medical Subject Headings (MeSH) terms and separate keywords relating to the research questions were included to create a broad and inclusive search strategy. All concepts included variations of the keywords. Filtering removed non-human, infant/children, clinical, and immunological studies, limiting to English articles published between 2013-2023. Using the search strings, article retrieval was carried out on 30^th April 2023. The detailed search strings are provided in Annex A.

1. Databases: Searches were conducted using PubMed and Medrxiv.

2. Internet sites: PLOS journals were identified as having relevant literature related to this topic, however not all PLOS journals were indexed in PubMed or Medrxiv at the time of conducting the search. Therefore, a quick literature search was conducted within the PLOS journal.

3. Identification of additional literature: to accommodate the identification of articles currently not covered within the above search components, additional literature identified by the internal team of experts on TB vaccine readiness were also included.

Inclusion and exclusion criteria

Scientific literature (including preprints) related to the implementation of a new and repurposed TB vaccines for adults and adolescent was included (Table 1). We focused on studies conducted in or concentrating on LMICs and we focused on articles published between 2013 and 2023. Expanding the timeframe was not anticipated to yield additional relevant articles, given that the development of TB vaccine candidates has been a more recent occurrence. Studies focusing on vaccine candidates moving to/are in phase 3 trials were included as were generic vaccines that align with the WHO Preferred Product Characteristics (PPC)⁸. Literature covering other types of vaccines, (clinical/non-human) trials, and immunological studies were excluded.

Studies on BCG vaccines (other than explicitly related to a revaccination strategy) were also excluded as were studies focused on the efficacy/effectiveness, and safety of new and repurposed TB vaccines. The primary focus of our review was on collecting primary data regarding the implementation of new and repurposed TB vaccines. Consequently, we chose not to include a grey literature search, as our expectation was that this type of literature might not predominantly contain primary data.

Selection of articles

The literature assessment occurred in a sequential process. Initially, articles from PubMed and Medrxiv were imported into Zotero, where their abstracts and titles were evaluated for relevance. Simultaneously, PLOS journal articles were examined directly on its website, and those meeting the relevancy criteria were individually added to Zotero, unless already detected through the other searches. Subsequently, a comprehensive full-text screening was conducted to in-or exclude articles, while removing duplicates. In instances where articles reported secondary results, the initial paper was added. JSB undertook the initial selection, screening, and assessment. PTP checked a random sample of articles to ensure consistency in the inclusion process. After title and abstract screening, JSB checked articles for their relevance through a full-text assessment. JSB consulted PTP to discuss articles for potential inclusion in cases of uncertainty. After full-text inclusion, the data were subtracted and collected in an extraction sheet using Excel 2023. A thorough assessment (snowballing) of the references was performed on both the articles cited within the identified reviews and the final selected articles to ascertain any potentially relevant literature that was not captured within the initial search strategies. Similarly, any articles identified through the review of references and reviews followed the same structured inclusion process, as did any additional articles identified by the team.

Data extraction and analysis

The AIGHD/EDCTP Roadmap and WHO’s Evidence Considerations for Vaccine Policy^8,9 provided key implementation factors for ensuring public health impact in shaping our analytical framework. Four pivotal implementation research components were evaluated: (1) Epidemiological impact (e.g. cases and deaths averted by vaccine), (2) Economic impact – on health systems, societies, macroeconomic-, value for money (e.g. cost per death or DALY averted, budget impact and impact on equity and social protection, including costing and cost-effectiveness, (3) Acceptability by potential target population (i.e. the likely acceptability, including barriers and enablers, of the vaccine in the target populations and other key stakeholder groups), and (4) Implementation feasibility by potential target population (i.e. practicality of vaccine implementation, considering the logistics, delivery, and program related considerations). We adapted the conceptual framework from Atun ¹¹ based on these key components (Figure 1).

• Download figure
• Open in new tab

Figure 1. Conceptual framework of factors influencing implementation of TB vaccination among adults and adolescents^9,11

We listed the number of studies conducted within each component in an Excel sheet by country and evaluated each study based on the four components. Results were reported for countries separately, outlining the vaccine characteristics, anticipated vaccination coverage, and time spans of the projected TB vaccine roll out. Implementation feasibility was further assessed through the lens of WHO’s health systems building blocks¹² : health service delivery, health workforce, health information systems, accessibility, health financing, and leadership and governance.

Articles output

We identified 1,807 articles of which 682 were from PubMed and 1,125 from Medrxiv (Figure 2). Following title and abstract screening of these initial articles and after removal of duplicates, 81 articles were included for full-text screening. The search strategy for PLOS journals resulted in the identification of two additional articles, which were included for full-text screening. Of the 83 articles for full-text inclusion that were reviewed, 66 articles did not meet the inclusion criteria of which seven articles referenced other articles with primary data. Five articles were identified through snowballing and one additional article was added by the team, resulting in 23 total articles for inclusion. Of the 23 articles, 18 were peer-reviewed, and 5 were scientific preprints.

• Download figure
• Open in new tab

Figure 2. Flow diagram of article selection.

Study characteristics

A total of 23 articles were included in the review. Table 2 provides an overview of the number of included studies per country, vaccine readiness focus, vaccine candidate or profile, and vaccine endpoints. Most of the research was conducted in/ or focused on India (9 studies), followed by high burden LMICs in general (7 studies), and articles specific by country: South Africa (7 studies), China (6 studies), Indonesia (1 study), and Cambodia (1 study). Among the key research topics on TB vaccine implementation, epidemiological impact was the most frequently assessed topic (16 studies), followed by costing, cost-effectiveness, and/or economic impact (11 studies). One article addressed acceptability within the context of implementation feasibility, with a specific focus on exploring implementation strategies and health system readiness while five articles assessed implementation strategies. Various vaccine candidates and profiles were covered in the literature, including M72 /AS01e (8 studies), BCG revaccination (5 studies), and generic vaccines (15 studies). Endpoints for the generic vaccines, most frequently encompassed prevention of disease (PoD) (14 studies), while prevention of infection (PoI) (6 studies) was less common; no study modelled a prevention of recurrence (PoR) endpoint (see table 3).

Epidemiological and Costing, Cost-effectiveness, and/or Economic Modelling Studies

A total of 21 studies estimated the epidemiological and/or costing, cost-effectiveness, and/or economic benefits of new and repurposed TB vaccines as shown in table 4 & 5. The studies were conducted across multiple LMICs, with India (8 studies) leading, followed by South Africa, China, Indonesia, and Cambodia. Epidemiological impact projection took precedence, as 16 studies estimated the effects of vaccines on health outcomes. Additionally, 11 studies explored the costing, cost-effectiveness, and/or economic implications of vaccine implementation, emphasizing the comprehensive evaluation of benefits. The studies examined various vaccine candidates and profiles, with generic vaccines being the most common (14 studies), followed by M72/AS01E (PoD) (7 studies) and BCG revaccination (PoI) (3 studies). The primary endpoints of the generic vaccine profiles modelled with were PoD (13 studies) and PoI (5 studies) (see table 6 & 7).

Epidemiological impact

Epidemiological impact studies were mostly conducted in India (6) or in LMICs/various high burden countries in general (4), China (4), and South Africa (4). Vaccine candidates and profiles modelled with were generic vaccines (11), M72/AS01E (5), and BCG revaccination (3). Of the studies modelling with a generic vaccine, 9 studies used a PoD endpoint, 5 used PoI, and none used PoR. Most studies (11) focused on the general adult/adolescent population, 1 on adolescents only, 1 on the general population, 1 on miners, 1 on people with diabetes mellitus (DM), and one within a geographic hotspot with a high TB incidence (see table 6).

Overall, the epidemiological impact of a new and repurposed PoD TB vaccine was found to be substantial compared to a no TB vaccine scenario (beyond BCG vaccination in neonates), with the highest health gains projected within the groups/countries with the highest TB burden (see Table)^13–17. Based on studies focusing on a selection of LMICs, the delivery strategy with the highest projected epidemiological impact consisted of routine vaccination with an instantly rolled out mass vaccination campaign, and concurrent with other public health interventions such as improved diagnostics, treatment, and reduced delay in health-seeking^13–17. A new or repurposed PoD TB vaccine with an efficacy of 60% was found to help reach the End TB targets if a minimal coverage of 72% could be achieved in South Asia, Central Asia Republics and Europe (CAR/EU) and 40% in Sub-Saharan countries. This difference for Sub-Saharan countries is due to the projected impact of TPT uptake in PLHIV¹³.

Country-specific studies revealed differences in impact as vaccine characteristics, target population and implementation strategy, timing of roll out, TB burden, and vaccination coverage rates vary. In China, a PoD vaccine targeting older people, protective in both mycobacterium tuberculosis (Mtb.) infected and recovered populations, could deliver the highest impact¹⁸. In South Africa, Jayawardena et al.¹⁹ projected the largest epidemiological impact with a strategy implementing a sequence of two mass vaccination campaigns and a second South African study found that a PoD vaccine could be most impactful if effective in both uninfected and infected people, and safe for use among PLHIV²⁰. Moreover, Shrestha et al.²¹ found a greater impact of a miner-targeted vaccine strategy compared to a community-targeted strategy in associated labour sending communities in South Africa. This was due to the higher concentration of adult men in the miner target population. In India, Silva et al. found that a vaccination program’s effectiveness targeting high incidence geographic areas depends on interaction of populations between geographic areas. Vaccination of people in a well-connected area with high TB incidence could reduce incidence 1.6-fold or more²². Another study showed in Indonesia and India that vaccinating either all ages or 18-49 year olds could bring substantial health and economic gains, and both health and economic gains would be more substantial with vaccinating all age groups²³.

Costing, Cost-effectiveness, and/or Economic impact

Eleven studies investigated costing, cost-effectiveness, and economic impacts (see table 5 & 7). The distribution of studies included general LMICs (4) and studies in India (5), South Africa (3), China (2), and Indonesia (1). Across the studied countries, the M72/AS01E vaccine candidate was most frequently studied in India (4 studies), whereas generic PoD vaccines were investigated across India (5), South Africa (3), China (2), and Indonesia (1). Costing, cost-effectiveness, and economic measurement outcomes differed across studies, with cost-effectiveness estimated for all except one country (Indonesia). Overall, adults and adolescents from the general population were the most frequently modelled target populations.

In studies focused on LMICs, adult/adolescent TB vaccine strategies could be economically beneficial compared to situations without new and repurposed TB vaccines. However, this is dependent on the TB burden, prevalence of HIV co-infection, vaccine program approaches, and ultimately the efficacy and duration of protection of the vaccine^16,17,30,31. An adult/adolescent TB vaccine was found to be cost-effective in most countries and cost-saving in 58 LMICs³⁰, with higher TB incidence countries gaining the most benefit. Furthermore, an adult/adolescent vaccine program could reduce catastrophic costs by 40%, with the highest gains in the lowest income quintiles¹⁶.

Country-specific estimates also illustrate that costing, cost-effectiveness, and economic impact is dependent on vaccine characteristics, TB burden, country-specific context and thresholds for cost-effectiveness, coverage, and timing of a program. In India, M72/AS01E is 7-times more cost-effective than BCG revaccination despite cost-effectiveness being evident across all scenarios²⁸. Additionally, Harris et al. found that only a vaccine which includes efficacy in uninfected people was cost-effective. This was due to the epidemic in India and their specific thresholds for cost-effectiveness³². In South Africa, a PoD vaccine could be highly cost-effective and even the least effective PoD vaccine scenario (80% coverage of 10 year olds, post-infection protection) remained below the cost-effectiveness threshold³². The inclusion of HIV-related costs was a relevant contextual factor in South Africa as mortality caused by TB among PLHIV will be prevented by the vaccine. In China, Harris et al. found that a PoD vaccine effective in both TB infected and uninfected people could be most cost-effective based on RR-TB reductions²⁴. Another study by Harris et al. found that in India and South Africa, the costs of vaccinating 18 year olds with 50% coverage was equal to the costs of vaccinating 15 year olds with 80% coverage³². Lastly, Silva et al. demonstrated that for India and Indonesia, a delay in implementation from 2020 to 2025 and the subsequent loss in economic impact could be recovered by vaccinating all ages instead of only a specific age group (18-49 year olds)²³.

Acceptability

One study assessed the acceptability of TB vaccine implementation in India, China, and South Africa³⁴. This article evaluated acceptability among high level vaccine policy and TB experts (Table 8. Factors influencing acceptability and demand of a new and repurposed TB vaccine). In all three countries, barriers included TB-related stigma affecting coverage during a mass vaccination campaign as well as affordability. Barriers to uptake specific to China included vaccine hesitancy and perceived level of protection. In India, experts perceived negative publicity of side effects as barrier to acceptability by the public and were hesitant about BCG revaccination resulting from previous trials showing no effects among Indian children. South African specific barriers included vaccine hesitancy and the need for a second dose of a vaccine candidate. The facilitators identified across countries included low price per dose, and public awareness of TB disease and the new and repurposed TB vaccine. Moreover, routine vaccination strategies, engaging communities/civil societies, and familiarity with the vaccine (in the case of BCG revaccination) were considered facilitators in the Chinese and South African contexts.

Implementation feasibility

Five articles contained information on key considerations for implementation feasibility, including implementation strategies of an adult/adolescent TB vaccine in LMIC, and health system readiness ^{19,21,22,34,35}. shows the number of implementation feasibility studies by country. Three countries were investigated, with South Africa being the most represented.

Implementation strategies by target population

Five articles contained information related to country-specific implementation strategies, in the context of China, India, and South Africa (Table 9)^{19,21,22,34,35}. Various strategies for TB vaccine implementation were identified, which are summarized in table 10. Target populations that were similar across the countries were adolescents, adults, socially vulnerable groups, PLHIV and other people with immunocompromised systems, and healthcare workers^34,35. The elderly and high risk contacts were target populations evaluated in India and China³⁴ and in South Africa the mining-community was evaluated as a target population²¹. In the Indian context, a focus on high-risk populations by targeting high incidence geographic ‘hotspots’ was also assessed²².

Health system readiness

Two articles discussed health system readiness for new and repurposed TB vaccine implementation in India, China, and South Africa (Table 11. Overview of studies assessing health system readiness)^34,35. For all three countries, there was a described need for an extensive monitoring and surveillance system to track potential adverse events. In India, it was mentioned that a mass vaccination campaign could be hindered by new and repurposed cold chain requirements, while routine vaccination could be more easily integrated within the existing public health system. Additionally, testing for Mtb. infection was considered too expensive and unfeasible in India. In South Africa, testing for Mtb. infection and the high prevalence of HIV infection-which may limit which vaccines could be safely used-were also potential challenges. In China, storage capacity was an additional factor that could delay effective implementation, yet the current procurement and supply chain could manage the integration of a new and repurposed vaccine program. In South Africa and China, insufficient staff capacity could hinder any future vaccination campaign.