EPSRC-SFI: Table Top Manufacturing of Tailored Silica for Personalised Medicine [SiPM]

近年来，个性化医疗（PM）受到了极大的关注，因为它有可能改变地球仪的医疗保健，摆脱“一刀切”的模式，利用个人情况，病史和需求提供个性化的治疗。目前的批量制造技术无法满足这些需求中的大多数，因为它们对变化的反应缓慢，资本密集，使用不可持续的方法，并且不能灵活地满足PM需求。EPSRC资助的医疗保健再分配制造最近的一份白色论文指出，迫切需要小规模、本地化、高速和自动化的制造平台来实现PM。他们确定这种“盒子里的工厂”应该：-能够按需制造，-灵活地提供具有所需特性的多种产品，-可持续（节能和使用温和的条件），以及-能够使用数据科学工具集成各种单元操作。 考虑到未来对PM的需求，最近的研究努力已经指向使用先进的方法如微流体、热熔挤出或3D打印将活性药物成分（API）及其制剂重新定义为例如用于口服剂量的片剂的制造。然而，由于药物是API与非活性组分（如赋形剂或药物递送系统（DDS））的精心设计的制剂，因此制造PM的非活性组分的挑战也同样重要，但尚未得到解决。非活性成分改善了原料药的理化性质和生物利用度。在其许多形式中，二氧化硅是许多当前和未来的API配方中最常用的组分之一，但它们的制造不存在以满足PM要求。具体而言，尽管在药物制剂中使用二氧化硅方面取得了巨大进展，但目前，它们的按需、自动化和灵活的制造以生产具有PM所需性质的二氧化硅是不存在的。一个关键原因是，绝大多数有前途的二氧化硅需要的合成条件是禁止任何有意义的规模扩大和实施在一个“工厂在一个盒子”的平台。因此，尽管最近在API和制剂的生产方面取得了进展，但这一缺失部分仍然是实现PM的重大障碍。我们已经展示了生物启发二氧化硅（BIS）作为替代药物输送系统的潜力，该系统具有可扩展性、经济性和可持续性，是按需和灵活制造的理想候选产品。这项研究将依赖于计算建模和实验合成之间的密切协同作用。申请人的绿色合成工艺和对强化反应器的研究将被用作起点。一系列强化反应器和基于高斯过程的建模将用于实现颗粒制造过程的过程强化。综合模型将用于创建数字孪生的流体设备和配方的绿色合成二氧化硅颗粒使用这些设备。将开发基于反应器模拟结果的机器学习方法，以将生产的二氧化硅的质量属性与关键工艺和操作参数相关联。将对器械几何结构和过程参数进行操作，以实现预期的关键质量属性（CQA）。这项工作将有助于革新PM，并帮助医院和药店提供桌面制药设备。最终，其影响将包括治疗和生活质量的显著改善，以及组建新公司来建造这些单位。

Personalised medicine (PM) is gaining significant attention in recent years as it has the potential to transform healthcare across the globe by moving away from the "one-size-fits-all" model to utilise personal circumstances, medical history and needs to deliver individually suitable treatment. Current bulk manufacturing technologies are unable to meet most of these demands as they are slow in responding to changes, capital intensive, use unsustainable methods and are not flexible to meet PM needs. A recent white paper from the EPSRC funded Redistributed Manufacturing in Healthcare has identified that small-scale, localised, high-speed and automated manufacturing platforms are urgently needed to realise PM. They identified that such "factory-in-a-box" should be: - able to manufacture on-demand,- flexible to deliver multiple products with desired properties, - sustainable (energy efficient and using mild conditions) and - able to integrate various unit operations using data science tools. Given the future needs for PM, recent research efforts have been directed towards redefining the manufacturing of active pharmaceutical ingredient (API) and their formulations into e.g. tablets for oral dosages using advanced methods such as microfluidics, Hot Melt Extrusion or 3D printing. However, as a medicine is a carefully designed formulation of an API with non-active components such as excipients or drug delivery systems (DDS), challenges in manufacturing of the non-active components for PM are also equally important, but have not been addressed. The non-active components improve physicochemical properties and bioavailability of APIs. In its many forms silica is one of the most commonly used component of many current and future API formulations, yet their manufacturing to meet the PM requirements do not exist. Specifically, despite tremendous progress made on the use of silica in pharmaceutical formulations, currently, their on-demand, automated and flexible manufacture to produce silica of desired properties for PM is non-existent. A key reason for this is that the vast majority of promising silicas require synthesis conditions that are prohibitive for any meaningful scale-up and for implementation in a 'factory in a box' platform. Hence, this missing piece, despite the recent developments in manufacturing of API and formulations, creates a significant barrier to making PM a reality. We have shown the potential of bioinspired silica (BIS) as an alternate drug delivery system, which is scalable, economical and sustainable - an ideal candidate for on-demand and flexible manufacturing. This research will rely on a close synergy between computational modelling and experimental synthesis. Green synthesis processes and research on intensified reactors by the applicants will be used as a starting point. A range of intensified reactors and Gaussian Process-based modelling will be used to achieve process intensification of particulate manufacturing processes. Comprehensive models will be used to create digital twins of fluidic devices and recipes of green synthesis of silica particles using those devices. Machine learning approaches based on results of simulations of reactors will be developed to relate quality attributes of silica produced with key process and operating parameters. Device geometry and process parameters will be manipulated to achieve the desired Critical Quality Attributes (CQAs). The work will contribute to revolutionising PM and help deliver table top pharmaceutical manufacturing equipment in hospitals and pharmacies. Ultimately, the impact will include significant improvements in treatments and quality of life as well as the formation of new companies to build such units.