Summerize the article It takes two to tango: Bridging the gap between country need and vaccine product innovation

are thoughtful, and analyze the content or question asked make connections to other content and real-life situations extend discussions already taking place, or pose new possibilities or opinions not previously voiced

Background Despite a growing global commitment to universal health coverage, considerable vaccine coverage and uptake gaps persist in resource-constrained settings. One way of addressing the gaps is by ensuring product innovation is relevant and responsive to the needs of these contexts. Total Systems Effectiveness (TSE) framework has been developed to characterize preferred vaccine attributes from the perspective of country decision-makers to inform research and development (R&D) of products. A proof of concept pilot study took place in Thailand in 2018 to examine the feasibility and usefulness of the TSE approach using a rotavirus hypothetical test-case. Methods The excel-based model used multiple-criteria decision analysis (MCDA) to compare and evaluate five hypothetical rotavirus vaccine products. The model was populated with local data and products were ranked against decision criteria identified by Thai stakeholders. A one-way sensitivity analysis was performed to identify criteria that influenced vaccine ranking. Self-assessment forms were distributed to R&D stakeholders on the usability of the approach and were subsequently analysed. Results The model identified significant parameters that impacted on MCDA rankings. Self-assessment forms revealed that TSE was perceived as being able to encourage closer collaboration between country decision makers and vaccine developers.

Introduction All United Nations Member States have agreed to accelerate progress towards universal health coverage (UHC) by 2030 [1]. Today, at least half of the worlds population do not have full access to essential health services. Each year, 20 million infants do not receive their full course of recommended vaccines, resulting in approximately 1.5 million deaths from vaccine preventable diseases. Over 50% of these infants live in six countries with the weakest health systems infrastructure [2]. In terms of inequitable health coverage, it is estimated that 100 million people live in extreme poverty due to out of pocket expenditure on health [3]. Effective global immunisation is essential to attaining the UHC goal, however to achieve this, vaccines must be designed suitable for use in resource constrained settings. To have the greatest public health impact, the product attributes of vaccines that are intended for deployment in low-and middle- income countries (LMICs) must be informed by the needs of LMIC vaccine delivery systems. LMICs are confronted with unique challenges when implementing successful immunisation programmes [46]. Persisting difficulties include but are not limited to: logistical complexity, barriers and gaps in delivery systems, constraints in human resource and challenges in prioritisation of vaccines over other competing interventions. Issues of ensuring timely delivery and appropriate storage conditions to preserve vaccine potency and safety prevail. In addition, the emergence and introduction of new vaccines, improved vaccine products and novel delivery technologies requires that policy and decision makers have the information and tools to assess and select the products that will be of greater impact for their programmes, after product licensure as well as during product development, to ensure that product characteristics are aligned with country need. The above challenges faced by each country may vary significantly based on their geographical, healthcare, and political constraints [7]. This impacts the priorities and goals for vaccine programs of every country and differentiated vaccine product characteristics may be desirable to overcome these constraints [8]. Despite wide acknowledgment of these practicalities, there is no clear methodological approach of identifying the vaccine product attributes that would address these context specific barriers, and communicating these country priorities to inform vaccine R&D by manufacturers. R&D describes the discovery and technical advancement of a new or existing technology [9]. The development of novel vaccine products and innovative delivery mechanisms is vital to address to the current programmatic challenges and combat the threats of emerging and existing infectious diseases. Vaccine R&D is not an easy endeavor; vaccines are complex and costly biological preparations. The development and introduction of one new vaccine product requires immense amounts of money, time, and resources and there is no guarantee of a return on this investment [10, 11]. Aligning the vaccine design and development initiatives with the government vaccine product preferences will likely help to incentivise and accelerate the development of vaccines that are optimal for local contexts, but these vaccines may also have broader potential for uptake and impact at the regional or global level. Designing new vaccine products that better anticipate market demand and challenges and are more responsive to government needs, will decrease the risk in investment, and increase the potential for return. Early stage Health Technology Assessment (HTA) is the evaluation of new technologies in development to determine their potential value and provides an opportunity to facilitate dialogue between government and market actors. Performing such an evaluation with inputs from key stakeholders including decision makers, immunisation programme managers, clinical experts, and manufacturers can help to identify preferred vaccine attributes, thereby informing R&D about the key needs of innovation and accelerating product uptake [12]. The World Health Organization (WHO) and partners are developing the concept of Total Systems Effectiveness (TSE), which aims to strengthen structured and transparent vaccine product selection processes at the country level and to build a platform for stakeholders in LMIC immunisation programmes to shape vaccine supply and R&D. TSE has since been renamed Capacity-led Assessment for Prioritisation on Immunisation (CAPACITI) [13]. The aim of the TSE approach is to foster a stronger collaboration between industry and country decision makers to develop products that are more closely aligned with the needs and context of LMICs.

Methods Three workshops were held during the pilot to better understand criteria of importance in national decision-making and the perspective of different stakeholders towards TSE. An initial workshop in May 2018 was convened to obtain preliminary feedback from stakeholders involved in vaccine policy decision making and research in Thailand. In addition to WHO, HITAP and NUS, 15 participants were in attendance and included representatives from: Subcommittee of the National List of Essential Medicines, The National Immunization Technical

Advisory Group (NITAG), academia, pharmaceutical industry and other ministerial departments. During the workshop, stakeholders were asked to complete an open-ended survey questionnaire to nominate top criteria important for vaccine product selection. The survey template was used previously by the WHO in a stakeholder meeting convened in Indonesia. The survey instrument was subsequently fine-tuned and modified for this study context. The survey instrument is available in S1 File. Responses were collected from 15 respondents and the top five criteria were selected as the decision-making criteria for the pilot study in Thailand. For the purpose of the study, an excel-based model for rotavirus product selection, based on the TSE approach, was modified for the Thai context. The TSE rotavirus product selection model included a generic set of decision criteria, with defined indicators, and enabled comparison between five hypothetical rotavirus vaccine products (RVV1 to RRV5, product characteristics detailed in S1 Table). The final list of decision criteria for the Thai model was identified by stakeholders in Thailand. The final five criteria and associated outcomes for the Thai model are as follows: 1) health outcomes (i.e. cases averted, hospitalisation cases averted, and deaths averted due to the vaccination); 2) cost estimates (total programme costs, healthcare costs, and a five-year budget impact); 3) safety data (intussusception cases); 4) budget impact; and 5) cost effectiveness. Since budget impact and cost-effectiveness had not been included in the generic TSE rotavirus model, they were added for the purpose of the exercise. The excel-based model utilized methodology that has been used in established models such as UNIVAC and the Vaccine-Technology Impact Assessment (V-TIA) tool to calculate the outcomes defined for each criterion, using country and product specific data inputs [17]. Locally relevant parameter inputs were generated from government reports, published literature and expert opinions. They also included local inputs on socio-economic status, coverage of vaccine programs, vaccine efficacy, and schedule, costs for storage, training and administration, and other epidemiological data on birth cohort, disease burden and resource use (S2 Table). Following MCDA methodology, the outcomes for each criterion were transformed into a score based on a common absolute scale ranging from 0100, to calculate an aggregate score for each option, resulting in a final ranking of each options according to performance across all criteria [18]. A vaccine with a higher total score was considered to be the better choice. Equal weights were used for all criteria, as the research team were unable to collect information on stakeholder preferences for weighting. A full description of the model and how the model outcomes were generated is available elsewhere [17]. A one-way sensitivity analysis was conducted on the vaccine characteristics, to identify the thresholds of each characteristic that could modify the overall ranking of products. It is hypothesised that this could form the basis for identifying minimum and aspirational bounds for target product profiles that meet country needs. We chose the vaccine product which is currently being ranked the second in the MCDA analysis, and via the sensitivity analysis, investigated if by varying the vaccine characteristics, it could become the top ranked vaccine. The sensitivity analysis was restricted to base and best case values for vaccine characteristic for identifying the variable changes which could make the second best product become the best ranked product. The base case was determined by the input values that reflected the most likely scenario and the input values for the best case represented the most optimistic scenario. The details on the inputs for the vaccine related variables for the base case and the best case can be found in the S3 Table. To identify the usefulness of such an approach from the R&D perspective, a separate workshop was held with stakeholders working in R&D in LMICs from both public and private sector organisations in Indonesia, Thailand, and Vietnam.The results from the rotavirus product MCDA evaluation were disseminated to illustrate the principle thatwith information on country priorities for vaccine productsmanufacturers could model a pipeline product against those of competitors to identify important attributes and thresholds given uncertainty. Key challenges and priorities in the countries were presented, and discussions were encouraged to deliberate the relevance and usability of TSE. Qualitative data was also collected from 17 selfassessment forms on the applicability and usefulness of this approach to inform R&D. The self-assessment form included a mixture of closed and open-ended questions to gather participant opinions of TSE and is available in S2 File. The form was developed for the purpose of this meeting and participation was completely voluntary. Participants were made aware that the forms were anonymized and that responses were to be kept strictly confidential and reflect personal opinions rather than the positions of their employers. Thematic analysis was deployed to analyse main patterns in the responses within the questionnaires. The forms were read repeatedly to enhance the overall understanding, then coded and classified. Then codes were grouped into emerging themes. A third workshop was convened to present the pilot study results back to the stakeholders who attended the initial meeting.