Trends, challenges and priorities

A team of 25 organizations has scoped a proof of concept for a fully automated, perfusion, single-use bioreactor demonstrating automated control of selected attributes, automation of seed train inoculation and harvest, and MVDA (multi-variant data analysis) orthogonal verification in combination with multiple integrated PAT tools for real-time process monitoring/control and automation of control actions.

A host site for the proof of concept has been selected and equipment components are being confirmed to ensure state of the art.

Single-use bioreactors and advanced perfusion systems are available and already well developed as off-the-shelf solutions in the market. However, the use of integrated process analytical technology (PAT) tools for dynamic and autonomous control is still lagging, especially in terms of widespread implementation. Industry guidance on how to implement and validate these tools will be a key deliverable for this work which when implemented will result in the realization of cost savings through increased automation and reduced batch deviations and lead times, improved consistency of product quality and shortening of technology transfer timelines.

Initially established as a Robotics Community of Practice, the Automated Materials Transfer team is proactively investigating opportunities to automate / remove human intervention areas for transfer of all materials (single use, systems, PPE, raw materials etc) from warehousing across Clean Non Classified (CNC) to Grade D areas.

Comprising biomanufacturers, software suppliers and robotic manufacturers, the team has concluded its discovery phase of as-is process mapping and will produce a User Requirement Specification outlining the visionary to-be process.

The intention is that this will act as an enabler to reaching full lights-out automation within facilities.

The biopharmaceutical industry is risk averse when considering the adoption of new technologies onto production lines. A cross-industry team of end-users and suppliers has started writing a set of best practice resources for new technology adoption. The resources include the
following key elements:

• Defined key strategic drivers for senior leaders to adopt
• A process for technology adoption
• Elements to support best practice adoption, such as case studies, templates and guidelines
• Benchmarking via a technology adoption maturity assessment is being considered.

The data required to improve process performance and make better manufacturing operational decisions exists today but is scattered across multiple systems, in a wide variety of formats and is difficult to assemble into context for evaluation, reporting and decision support. Production / operations / quality management personnel spend most of their time in gathering and contextualizing data rather utilizing consolidated data in context for operational improvements.

The Big Data to Smart Data team are building on the opportunities first highlighted in the Automated Facilities Roadmap by developing an industry standard taxonomy / indexing approach for data sources and proposing an architectural solution to add context to data when first created. This delivers direct benefits by reducing investigation and analysis timelines.

A Proof of Concept was run at BTEC, North Carolina to show that contextualizing data and using a simple ontology gives much more value to the manufacturing data and associated decisions.

The proof of concept is being written up, with the paper to be published by end Q1 2023.

The workstream is now considering what future topics are of value for collaboration. Once opportunities are identified an updated charter will be presented to the TR Steering Committee for approval

This collaboration achieved the design, build and test of a full-scale cGMP buffer stock blending (BSB) system in collaboration with the National Institute for Innovation in Manufacturing Biopharmaceuticals (NIIMBL), as well as the open-source release of the full design documentation and performance data under a freedom to operate agreement for any company to review, build and implement their own BSB system.

Testing has demonstrated the skid performs well against the expectations laid out in the initial white paper (November 2019), with the BSB system’s speed described as a ‘gamechanger’ in allowing the development of an entire buffer stock in 1 hour (60 litres produced per minute). This will enable more effective resource utilisation and use of facility space, reduction in cost of buffer per litre (from $78 to $30) and support on-demand buffer delivery to a downstream process.

2023 will see the launch of Phase 2 for the team, exploring challenges surrounding raw material transfer in to the buffer preparation process.

The objective of this workstream is to produce a digital technology roadmap for the biomanufacturing industry, looking to 2030 and beyond. It is aimed, primarily, at strategic decision makers and practitioners in the field to better understand the technologies and capabilities needed and the value to be realised at industry and member company level.

Part one of the digital technology roadmap (problem statement) has been published and describes the current state and commonly seen industry barriers. Part two (making it happen) has been planned as a series of mixed media content, that seeks to improve engagement through sharing best practice and case studies to support the development of digital strategies with convincing value propositions.

Examples of future part two content include:

• Best practice paper on use of biomanufacturing digitalization readiness model
• Podcast with QA and Regulatory SMEs on getting them involved earlier on in digitalization initiatives
• Positioning paper on data governance and the use of FAIR principles
• Critical persona template with ten worked examples e.g., Automation Manager, Process Engineer, Pharmaceutical Development Scientist, Service Technician, QA Specialist, Procurement Leader, Validation Scientist, Facility Manager, CMC Lead, and MSAT Lead
• Webinar on the topic of data governance

The objective of this workstream is to show how the plug and play concept can be adopted by biopharma manufacturing. Driving increased equipment and automation standardization will help improve efficiency and reduce costs as biopharma uses increasingly modular manufacturing approaches.

To date, the workstream has published documents about validation strategy, audit trail strategy, stirred tank unit specification.

Documents are being written for filtration equipment specification, an overview document and a “business case” for building a test-bed facility to support the adoption of plug and play capability in biopharma.

In 2023 we will hold a face-to-face event to finalize guidance on building and operating the test-bed facility. Hopefully, this output will be brought to the TR Steering Committee to seek support to set this up. Alongside this event, we plan a showcase to help drive adoption of the plug and play principles by both biopharma and suppliers.

A BioPhorum survey completed by biopharmaceutical end-users revealed that the biggest hurdles to adopting new and rapid alternative adventitious agent detection methods are:

• The regulatory acceptance of these alternative methods
• Filing a regulatory change control
• The time taken to validate a non-compendial method

Further, there was consensus that detection technologies were continuing to evolve but the provision of a guidance document that enabled sharing of best validation practices within the industry (including multi-site and multi-company) and including regulatory involvement would have the greatest impact on widespread industry adoption of new methods.

To meet this need, the Rapid Methods for Adventitious Agents team has developed technology-agnostic user requirement specifications (URS) for rapid mycoplasma and viral tests, which have been rubber stamped by related Phorum workstreams. A best practice approach for adventitious agent ARMM evaluation, validation and implementation based on an existing Fill Finish document has been published . The team’s best practice approach for Mycoplasma NAT-based detection methods is close to publication and will incorporate member company case studies. A third paper dealing on virus detection is in the planning stage.

These deliverables will help speed up product release from 35 to less than two days for mycoplasma and virus testing. They will enable the faster adoption of new rapid methods suitable for lot-release testing and provide regulatory agencies with the view of a single, consolidated validation package.

Concluding in 2022, this workstream explored the approach of modular and mobile engineering and construction approaches for facility design, and applying this approach to a multi-product facility to enable reduced front-end engineering design time in construction.

This culminated in the worked example of a late stage clinical / early commercial viral vector facility, demonstrating how to implement modular capacity concepts to allow phasing of CAPEX investment, and reducing both financial and business risk.

2023 will see the kick-off of the Ultimate Facility Design workstream, which will explore how to apply these concepts to a multi-modality design.