Be part of launching a new scientific field - immunostimulation/ immunodynamic (IS/ID) mathematical modelling to accelerate vaccine development

疫苗一旦开发出来，就是拯救人类生命的最有效方式之一。但是，开发一种新的疫苗可能需要数十年的时间，成本约为10亿美元。目前用于确定合适疫苗剂量的科学方法是经验性的和过时的。相比之下，基于数学模型的药物开发(PK/PD)经常被用来预测使用的最佳药物剂量大小和时间表。药物动力学模型用于模拟药物在体内的浓度随时间的变化，一般考虑药物的吸收、分布、代谢和排泄4个阶段。药效学模型侧重于药物在吸收和排泄之间对人体的影响。这些工具可用于计算剂量，以将物质维持在有效浓度而不达到有害水平。此外，可能存在曲线下面积很重要的情况，这意味着这些身体不仅受当前浓度的影响，还受整个给药历史的影响。根据不同的因素，不同的模型和不同的患者效果可能与药物模型相关，疫苗剂量模型可能需要考虑类似的问题。腺病毒(Ad)载体疫苗是一种新型的疫苗抗原传递途径。作为载体的Ad的有效性主要来自于产生T细胞和抗体反应，这种疫苗的有效性已经在临床上进行了各种疾病的试验。腺病毒载体和改良的安卡拉痘苗病毒载体经常被用来诱导T细胞反应。然而，在了解联合使用疫苗时的最佳剂量方面开展的工作很少，而且对疫苗接种的时机以及是否应该给予一个或多个初级疫苗或加强疫苗的理解很差。主要的重点应该放在从药物开发到腺病毒疫苗开发的建模技术上。如果成功，这可能会开启一个新的科学领域--免疫刺激/免疫动力学(IS/ID)建模。将创建新的统计和机制模型，使用来自Vaccitech小鼠疫苗实验和文献的经验数据进行参数化，以帮助设计在Vaccitech进行的新的临床前和临床研究。这个项目有几个研究目标，这些目标将与Vaccitech植入的集合相一致。目标1：首先，我需要继续将我的数学知识应用于药代动力学和药效学模型，并对以前的PK/PD工作和免疫反应建立更好的理解。这将主要通过围绕该主题的文献来实现。目标2：第二，我们将使用现有的实验数据来创建和参数化新的统计/机制模型，以预测小鼠的疫苗剂量/免疫反应关系。目的2：第三，我们将通过预测新的经验性小鼠实验的最佳设计来改进统计/机制模型的剂量-反应预测，以最大化获得的剂量-反应信息并最小化使用的小鼠数量。目标3：第四，使用异速生长比例假设和统计/机制模型，我们将基于小鼠的数据，对人类最具免疫原性的剂量-反应关系做出初步预测。这些预测将在Vaccitech进行的新的经验性临床实验中进行评估。目的4：在本博士论文期间，我们将创建的模拟证据将用于改进Vaccitech腺病毒构建物和改良的痘苗病毒安卡拉病毒载体的人体试验设计。除了创造新的知识外，该项目还有一条影响疫苗发展政策的途径，以及更广泛的经济和社会影响。如果这一领域被证明是有前途的，它可以极大地减少开发疫苗所涉及的财政和时间负担。

Vaccines, once developed, are one of the most effective ways of saving human lives. But, developing a new vaccine can take decades, and cost around a billion dollars. The current scientific methods used to identify the right amount of vaccine dose are empirical and antiquated. In contrast, mathematical model-based drug development (PK/PD) is regularly used to predict the best size and schedule of drug dose to use. Pharmacokinetic models are used to model the concentration of drug in the body over time and generally consider the 4 stages of absorption, distribution, metabolism and excretion of the drug. Pharmacodynamic models focus on the effects that a drug can have on the body between absorption and excretion. These tools can be used to calculate dosage to maintain the substance at an effective concentration without reaching harmful levels. In addition, there may be cases where the area under curve is important, meaning that these body is not only affected by this present concentration but is influenced by the entire history of dosing. Dependant on various factors a different model and varying patient effects may be relevant in drug models, and vaccine dosage models would likely need to consider similar issues. Adenovirus(Ad) vector vaccines are a novel approach to vaccine antigen delivery. The effectiveness of Ad as a vector comes primarily due to the production of T-cell and antibody responses, and the efficacy of such vaccines have been clinically trialed for various diseases. Adenoviral constructs alongside modified vaccinia Ankara, two viral vectors, are often used in combination to induce T cells responses. However, minimal work has been carried out to understand the best doses of vaccines when used in combination, and there is a poor understanding of the timing of vaccination, and if one or more primes or boosts should be given. The main focus should be on modelling techniques from drug development to adenovirus vaccine development. If successful this could launch a new scientific field - immunostimulation/ immunodynamic (IS/ID) modelling. Novel statistical and mechanistic models will be created, parameterised using empirical data from Vaccitech mouse vaccine experiments and the literature, to help design new pre-clinical and clinical studies performed at Vaccitech. This project has several research aims that will coincide with a collection of Vaccitech placements. Aim 1: First, I will need to continue to adapt my mathematical knowledge towards pharmacokinetic and pharmacodynamic models and establish a greater understanding of previous PK/PD work and of the immune response. This will be primarily achieved through literature surrounding the topic.Aim 2: Second, we will use existing experimental data to create and parameterise novel statistical/mechanistic models to make predictions for the vaccine dose/immunological response relationship in mice. Aim 2: Third, we will improve the statistical/mechanistic model dose-response predictions, by making predictions for the best design of new empirical mouse experiments, to maximise the dose-response information gained and minimise the number of mice used. Aim 3: Fourth, using an allometric scaling assumption and statistical/mechanistic modelling, we will make initial predictions for the most immunogenic dose-response relationship in humans, based on the mouse data. These predictions will be evaluated in new empirical clinical experiments carried out at Vaccitech. Aim 4: Within the period of this PhD, the modelling evidence we will create will be used to improve the design of a human trial of the Vaccitech adenoviral constructs and Modified vaccinia Ankara viral vectors. In addition to the creation of new knowledge, the project has a route to impact on vaccine development policy alongside wider economic and societal impact. If this field proves promising, it could greatly reduce the financial and time burdens involved in developing vaccines.