November 1, 2018 (Vol. 38, No. 19)

The biologic drug market is projected to reach $390 billion within the next two years,1 and the need for efficiency and speed may be outpacing the market. The advent of personalized medicine requires smaller batch sizes, while new drug modalities and new production technologies emphasize the need for adaptable manufacturing environments.

Biopharmaceutical manufacturers are seeking dynamic processing economics that address the paradigm shift from blockbuster drugs with established, large-batch production schemes to small-scale manufacturing and innovative technologies.

From Innovation, to Acceptance, to Industry Standard

The promise of flexibility and speed that single-use technology brings to a marketplace in continuous change has been realized, and the wide acceptance of single-use systems is evidenced by the growth in this technology. Changeover time between batches with single-use bioprocessing are up to five times faster than the CIP/SIP process required in a stainless-steel environment, enabling production of as many as seven more batches per year.2 Ballroom or dance-floor implementations with closed-process single-use bioprocessing have further streamlined the logistics of single-use processing by reducing cleanroom requirements for most of the process train.

Following early implementations of single-use bioprocessing systems, initial regulatory hurdles were cleared, and guidance for single-use systems has been developed. Regulatory acceptance and familiarity, in conjunction with the success of early adopters, have driven steady growth in the adoption of single-use systems. The ability to rapidly create and expand capacity by deploying modular single-use production facilities has proved attractive to emerging markets,3 as well as to start-ups. There is little doubt that single-use production will be part of every biomanufacturer’s strategy moving forward.

The challenge now is to gain greater efficiency with smaller scale, flexible production schemes. Driving for process intensification has become the new operating standard. Given that a 1% improvement in overall process yield can lead to as much as $2.8 million in additional revenue per batch (for a bioreactor titer of 5 g/L with a dosage size of 50 mg), investing in processes and technologies that will increase efficiency is imperative.

What’s Next? BioPharm 4.0

Efficiency gains with intelligent biomanufacturing have tremendous potential to further advance the benefits of single-use manufacturing. Bioprocessing with stainless-steel equipment and fixed piping has been automated, to varying degrees, for decades. The growth in single-use bioprocessing presents new challenges in how to leverage flexibility, optimize changeovers, and incorporate automation and its associated data to gain efficiency.

When a transition from a stainless-steel environment occurs, the simplicity of a single-use process can be stunning. Gone are the hundreds of feet of stainless piping, transfer panel jumpers, and numerous automated ported isolation valves. Connecting one bioreactor to another is a single piece of disposable tubing. However, unlike stainless-steel–based bioproduction, single-use bioprocessing requires operators to route and secure tubing, make proper connections (such as sterile connector assemblies and tube welds), remove tubing clamps, and complete a host of other actions.

Automation in single-use becomes a combination of automated process control augmented by guiding the operator with their manual operations. Automated platforms simplify interactions between process equipment and reduce the possibility that human error will influence product quality. Equipment behaves predictably, reliably, and consistently. Automated operator reminders and confirmations can improve the consistency of the required manual operations. The real-time control of equipment through automation allows the system to dynamically adjust for detected variability, assuring that the process performs optimally and that the end product is consistent. These improvements lead to better yields, less waste, and higher quality in the final product.

Bioprocessing Joins the World of Big Data

The benefits of automated control are well understood, but we are just scratching the surface of utilizing the data that automated systems provide. big data analytics is a transformative, enabling technology that is penetrating every aspect of modern life. The science that drives precision medicine would not be possible without advanced data analytics. The application of big data analytics to biomanufacturing is still in the developmental stages.

Beyond real-time control and reporting, automated solutions enhance the ability to collect and analyze historical product, process, operational, and performance data. Analysis of this data can result in several useful outcomes, such as predicting equipment failures before they happen to allow scheduled maintenance, as well as controlling process values to optimize product yields.

The Big Leap: Machine Learning

The big leap forward will come with the utilization of big data to develop and validate analytic models that can predict and optimize process parameters and sequencing, resulting in better product yields. GE Healthcare has developed scale-agnostic machine learning (ML) infrastructure to address yield-efficiencies in bioprocess production facilities. With select biopharma clients, the GE Healthcare data science team has used advanced ML algorithms to produce process-parameter values designed to predictively maximize yield of each production run, developing sophisticated ML/AI algorithms to drive efficiency and improve product yields in biologics manufacturing.

For this purpose, GE Healthcare data science teams, in collaboration with on-site manufacturing teams, collect all relevant bioprocess data: these datasets include online bioprocess (per minute) data, manufacturing execution systems (MES) data, laboratory information management systems (LIMS) data, raw material manufacturing and QC data, manual entries, and information about corrective and preventive actions (CAPA) for all bioproduction scales used to manufacture biologic therapeutics. A proprietary data-ingestion pipeline then converts, digitizes, and standardizes the datasets. ML models are then trained on all noncorrelated input features with labeled output parameters such as batch yields and/or viable-cell-density values at each step of the biomanufacturing process.

Directed by the ML models, feature-set correlation analytics typically results in the reduction of the input parameter sets from several hundred to a few dozen, specifically down to the few dialable parameters. This produces actionable input values that the operator can dial into the process to achieve high-yield runs on a regular basis. Even with the expected production variance of raw materials, GE Healthcare’s ML models typically demonstrate accuracy metrics of greater than 90%. This end-to-end rigorous approach produces ML models that can reliably and accurately predict batch-specific yield efficiencies.

Data analytics in bioprocessing will continue to grow in importance. Much of the innovation associated with automation technologies relates to the capture, communication, analysis, and utilization of data. Bioprocessing will incorporate technologies drug makers are confident will improve their business objectives, including those with either no regulatory impact or those that provide improved regulatory compliance. Stable, robust technologies that reduce or eliminate risk without introducing risk of their own will be critical. Continuous improvement driven by Big Data will drive greater efficiency and higher yields. The future of process optimization and cost reduction involves new players—your automation team, your data science team, and your existing process historian database.

Michelle Stafford is global marketing leader, Pete Genest is director of automation strategy, and Randal Goomer, Ph.D., is staff data scientist at GE Healthcare Life Sciences.

Previous articleEarly Liver Cancer Detection through Glycolytic Genes
Next articleNip, Tuck, CRISPR: Gene Editing Could Give Plastic Surgery a Lift