The pandemic set high expectations for how quickly new vaccines and therapeutics can make it onto pharmacy shelves. Even if clinical trials and the approval process can be accelerated, bottlenecks in manufacturing capacity and capabilities and the use of legacy technologies can slow the overall time to market.

More innovative ways of producing vaccines, therapeutics, and advanced modalities are needed to deliver the gains in efficiency that the industry must achieve to keep pace with new opportunities. These innovations, categorized as being essential components of advanced manufacturing, include closed and continuous processing, automation and process analytics, digitalization, and predictive modeling.

In this article, we explore the key trends driving the biopharmaceutical industry towards advanced manufacturing and what is essential to create highly agile and flexible facilities of the future. We also discuss the evolution towards smart manufacturing which takes advanced manufacturing a step further to connect disparate data sets across manufacturing processes and sites, creating a fully interconnected, data-driven, and highly efficient production and supply chain ecosystem. This connectivity will allow both contract development and manufacturing organizations (CDMOs) and biopharmaceutical companies to capture real-time data from their production chain and use it to optimize and streamline operations, make better decisions on how to allocate resources, and predict possible deviations before they occur.

Begin with the business opportunity or challenge

While advanced technologies are a hallmark of the facility of the future, implementing technology for technology’s sake is not the guiding principle. The roadmap to the CDMO facility of the future must start with an understanding of the dynamics and trends influencing today’s biopharmaceutical industry and the resulting business opportunities and challenges. The CDMOs best positioned to help clients take full advantage of this environment and overcome associated obstacles will thrive among their peers. 

In its updated technology roadmap published in 2021, the BioPhorum organization identified several drivers shaping the future of the biopharmaceutical industry (Figure 1).¹ These drivers include the need for speed, quality, value, agility, and sustainability and continue to push the industry towards next generation bioprocessing and influence significant shifts in manufacturing technologies. 


From the CDMO perspective, these drivers may manifest as initiatives to increase productivity from an existing facility and create greater flexibility to better serve clients. Gains in efficiency, quality, flexibility, and lower costs will require integration of new, enabling technologies into the production workflow and CDMOs may be best suited to bring about this vision for the industry given their broad and deep experience.

The impact of enabling technologies

Enabling technologies that will support advanced manufacturing include those for process integration and intensification, closed processing, process analytics, control and automation, and the use of data to support more rapid decision making. These technologies will be essential to not only create more efficient and cost-effective processes for novel modalities such as cell and gene therapies, but also for mature classes of therapeutics such as monoclonal antibodies (mAbs). While templates for mAb production are, to a large extent, optimized to the level they can be using available processes and technologies, a step change is needed to move beyond conventional approaches to deliver against these critical business drivers.

Process closure, integration, and intensification

Process integration and intensification can be used to address many business challenges including the need to accelerate timelines from process development to the clinic, maximize capacity within existing facilities, and/or ensure new facilities have the most streamlined, economically efficient processes. Inherent to this approach is that once unit operations are intensified and connected, robust process control is essential.

While automation has been in place for many years, related standards are evolving and becoming fit-for-purpose in this new connected and convergent environment. Our organization collaborates with institutions such as the National Institute for Innovation in Manufacturing Biopharmaceuticals (NIIMBL) and the User Association of Automation Technology in Process Industries (NAMUR) to participate in the evolution of bioprocessing automation standards.

We are also collaborating with many of our customers to implement technologies for process integration and intensification to increase efficiency and reduce costs.

One example of process intensification is a flow-through polishing step we developed in partnership with China-based Transcenta, a clinical stage biopharmaceutical company developing antibody-based therapeutics.

In addition to developing effective and differentiated biologics for patients, Transcenta sought to ensure their biomanufacturing platform was able to address challenges associated with significant drug pricing pressure and demand uncertainty, while at the same time, minimizing the upfront costs of building a manufacturing facility.

Company leadership determined that the best approach to decrease footprint and facility costs while maximizing plant output was to intensify the production process via continuous perfusion and integration with hybrid continuous single-use downstream processing. The continuous perfusion platform has been shown to increase productivity by more than 10-times when compared to a conventional fed-batch platform. This strategy also enables smaller and highly flexible facilities to match the output of much larger facilities, thereby significantly lowering the cost of goods, increasing speed to the clinic and market, and improve process and product control.
 
In collaboration with Transcenta, we co-developed an industry-first fully automated, single-use flow-through polishing system and associated flexware that integrates four downstream unit operations and can support both perfusion and fed-batch workflows. With this novel technology, Transcenta will be able to significantly decrease cycle time and increase downstream purification process throughput in support of their own and their CDMO clients’ drug development.

Our organization was also part of a multi-national collaboration funded by the European Union’s Horizon 2020 program to optimize and implement an integrated manufacturing platform for monoclonal antibodies.² The platform includes continuous antibody capture using multi-column chromatography, continuous viral inactivation systems, single-use technologies, and advanced analytical tools for in-line monitoring of quality attributes. A comparison of traditional downstream processing and the intensified, continuous process is shown in Figure 2.


In the optimized process, mAbs were produced in a 1000L bioreactor, operated in fed-batch mode. Cell culture harvest was pre-treated with a flocculant and clarified using depth filtration followed by a sterilizing-grade filter. This clarified pool was continuously loaded onto Protein A resin utilizing a multi-column capture system and elution peaks were continuously sent to viral inactivation. The viral inactivation skid included 3L single-use tanks with the capacity to continuously decrease the pH, hold the solution for the desired time, and re-adjust the pH prior to polishing. The process feedstream was then polished in flow-through mode and sent to subsequent operations.

Continuous operations were performed using interconnected systems to trigger process actions and send feedback alarms in case deviation occurred (Figure 3). Post-column UV sensors triggered the switch between loading columns based on a breakthrough detection and the switch between waste and fraction during elution peak detection. A valve was installed on the inlet of the virus inactivation skid to divert fractions to one tank or another based on tank levels. Feedback alarms such as tank overloading could be sent back to interrupt the multi-column capture system. The level of the last tank of the inactivation skid controlled the start of the polishing chromatography system.


This 1000L, fully connected, single-use downstream platform was used to produce more than 3.3 kg of mAb in 2.5 days, with similar product quality and purity to that obtained in a traditional batch process.

Use of continuous multi-column capture chromatography considerably reduced the volume of protein A resin required, and thus, the overall process cost. The platform footprint was less than 30 m², offering a significant improvement of productivity and a reduced environmental impact. This proof of concept is one of many that are paving the way for the manufacturing facility of the future.

Process analytics

Process analytical technology (PAT) is another foundational element of advanced manufacturing and the facility of the future. The goal of PAT is to build quality into biopharmaceutical processes through real-time monitoring and control. Through the identification and control of critical process parameters (CPP) and critical quality attributes (CQA) within a specified design space, PAT enables quality to be built into the bioprocess, rather than being tested into the products. Process analytics can be used to drive better process understanding, make real-time decisions, and enable real-time release. 

In the biopharmaceutical setting, PAT relies on chromatographic, spectroscopic and/or mass spectrometric sensors that are integrated into unit operations to enable real-time monitoring and control of the process. One example of a PAT tool is in-line Raman spectroscopy which is an analytical technique that recognizes the chemical fingerprint of a sample by molecule differentiation.³ When used in upstream processes, Raman spectroscopy can provide in-line and real-time measurement of a variety of CPPs including glucose, lactate, and ammonium, as well as performance indicators such as total and viable cell densities.

As a PAT tool, Raman spectroscopy provides the real-time analytical data necessary to better understand and control processes which can translate into increased reproducibility, yield, and quality.

Facility of the future: digital & smart

The BioPhorum group recently published a guide to their Digital Plant Maturity Model (DPMM) that describes the stages of digital maturity from simple paper-based facilities to the fully automated and integrated adaptive, “lights-out” smart manufacturing plant of the future (Figure 4).⁴ The model emphasizes that the maturity process is a stepwise evolution and necessitates use of new digital tools to support advanced manufacturing technologies, data-driven decision-making, and highly efficient production, with full end-to-end supply chain integration. 


This evolution in digitalization offers several important advantages including:

• Improved product quality and greater consistency;
• More cost-effective manufacturing;
• Elimination of manual interventions that could increase the risk of operator errors and contamination;
• Greater insight into process parameters and anticipation of deviations;
• Reduced risk of non-compliance and batch failures; and
• Increased flexibility to reconfigure processes and respond to new modalities.

From our perspective, this digital maturity model is best viewed alongside an organization’s process intensification maturity level, as digital and process evolution are inextricably linked and dependent upon each other. With increasingly connected and continuous processing, each production batch has the potential to generate in excess of one billion data points. Given the amount of data that must be managed, interpreted, and leveraged, paper-based systems, data silos, incompatible technologies and data types present major obstacles to establishing connectivity and a digital backbone across biomanufacturing. 

Fortunately, when it comes to data management and use in a regulated environment, digital technologies are relatively agnostic in terms of the therapeutic modality being manufactured. The same fundamental approaches can be used whether a mAb, vaccine, or novel modality is being produced. In addition to advanced software, automation and analytics, technologies such as the Internet of Things (IoT), machine learning (ML), virtual or augmented reality (VR/AR), digital twins and artificial intelligence (AI) are now making inroads into biopharmaceutical manufacturing.

It is important to note, however, that while maturity models of digitalization and process intensification provide invaluable guidance and milestones, facilities of the future will be individualized—the concept of “one size fits all” does not apply. While the same basic data will be collected, and analytics will be applied, the actual design of the plant may be vastly different, with each facility aligned with corporate goals and objectives, geographies, therapeutic modalities, pipelines, and patient population.

Ensuring a successful journey

Remarkable advances in therapeutic modalities, manufacturing technologies, and digitalization are ushering in a new era of biopharmaceuticals. Market trends and business drivers create a wealth of opportunities and choices for drug companies and CDMOs alike, and once corporate and process objectives have been defined, the pathway to the smart manufacturing facility of the future begins to crystallize.

This journey will be facilitated by keeping a few critical success factors in mind:

• Seek experienced collaborators who understand process dynamics to help define options for problem solving;
• View the manufacturing holistically versus optimizing each unit operation in a silo. Be mindful that intensification of one process may create or exacerbate a bottleneck in another; and
• Work closely with regulatory authorities to demonstrate the quality and reliability of new processes and process changes.

While the pathways taken to achieve a new manufacturing vision will vary among biopharmaceutical companies and CDMOs, the ultimate goal is the same—a future in which patients are able to get faster access to the high-quality medicines they need. 

References
1. https://www.biophorum.com/phorum/technology-strategy/biophorum-technology-roadmap-vision-2-0/
2. Kervennic D,  Delagrée C, Hribar G. Multi-Column Capture Prototype Tests for Monoclonal Antibody Capture During Project nextBioPharmDSP (Horizon 2020). Application Note
3. In-Line Real-Time Monitoring of CHO Cell Culture Process Parameters using Raman Spectroscopy. Application Note
4. A Best Practice Guide for Using the Biophorum Digital Plant Maturity Model and Assessment Tool. BPOG publication. https://www.biophorum.com/wp-content/uploads/bp_downloads/BPOG-DPMM-Best-Practice-for-Plant-Assessments-May-2018.pdf

---

Merrilee Whitney is vice president of digital experience at MilliporeSigma, the U.S. and Canada Life Science business of Merck KGaA, Darmstadt, Germany.

Willem Kools is head of the technology pioneering group at MilliporeSigma, the U.S. and Canada Life Science business of Merck KGaA, Darmstadt, Germany.