Challenge
Meet the demands of an increasingly competitive market, even as global pharmaceutical regulatory bodies deepen quality and clinical documentation requirements.
Solution
Celltrion has made a commitment to extensive Quality by Design (QbD) applications in product development and process control. JMP® statistical software is an indispensable tool.
Results
A QbD culture facilitates the consistent production of the high-quality, safe and affordable products upon which Celltrion has built its reputation for industry leadership.
Competition in today’s global pharmaceutical industry is fierce, and regulatory demands are increasingly exacting. Rising to the challenge, market leaders have turned toward robust Quality by Design (QbD) practices as a highly systematic means of guaranteeing that safety standards are consistently met – all without incurring additional costs that could raise prices for consumers.
At the forefront of this industry initiative is Celltrion, Korea’s preeminent biopharmaceutical company. Celltrion specializes in the research, development and manufacture of biosimilars and other innovative drugs, including in the breakthrough antibody biosimilar category. Headquartered in Incheon, this pharmaceutical innovator ensures the quality, safety and affordability of its products by manufacturing them in state-of-the-art mammalian cell-culture facilities designed to meet and surpass global regulatory standards.
Celltrion works closely with the Korean Ministry of Food and Drug Safety to advance systematic product development and is likewise meeting the demands of regulatory agencies around the globe. Remsima was the world’s first biosimilar monoclonal antibody to be approved by the EMA (2013) and the FDA (2016), which indicated for the treatment of eight autoimmune disease including rheumatoid arthritis and inflammatory bowel disease. In 2018, Herzuma (biosimilar of trastuzumab) and Truxima (biosimilar of rituximab) also received the FDA’s approval for HER2-overexpressing breast cancer and three NHL indications, respectively. Celltrion has now been developing several biosimilar candidates at various stages of development, and plans to market at least one new biosimilar product a year.
In order to bring these and other cutting-edge therapies to market in a timely fashion, Celltrion has embraced statistical methods – not least of all QbD – through a commitment that innovation cycles be driven by data.
Meeting global regulatory demands
“Celltrion uses QbD extensively for product development,” says Choo MinJoo, head of Celltrion’s Chemistry, Manufacturing and Control (CMC) Statistics team. Choo and her team address a wide range of statistical issues across all projects in Celltrion’s diverse product portfolio. They’re known within the company as problem solvers – a go-to resource for all kinds of data challenges. And in the R&D space, that means the team submits all drugs in development to the rigors of QbD.
“QbD is a method of establishing a drug's quality goals – of developing products systematically based on risk assessment through a clear understanding of all input factors,” Choo explains. “It’s about fully leveraging statistics in product development. This is how we approach our work.” For Choo and her team, QbD begins with predefined objectives, emphasizes a full understanding of all inputs and establishes process control based on sound science and informed risk management. Quality is assured proactively by researching and controlling all variables.
Choo emphasizes that QbD plays an equally important role at Celltrion in product management: “To ensure product quality, you must have systems and procedures to measure critical quality attributes and performance in a timely manner, and to design, analyze and control manufacturing processes.”
This analytics-driven philosophy was not always standard practice at Celltrion, Choo says, as like all its competitors, the company previously adhered to a one-factor-at-a-time methodology. Since that time, however, Choo attests that efficiency has undergone dramatic improvement with the introduction of quality engineering practices. “Our experiments can be expensive,” she says. “So we must apply the appropriate statistical method for the data characteristics.”
QbD emphasizes decision making that’s founded in a depth of process knowledge; designed experiments are used to comprehensively investigate processes and product factors. When conducted properly, there’s less reliance on inspection of the final product. QbD, Choo explains, is a more quality-effective and cost-effective approach.
And that is exactly where JMP® comes in. “The tool we rely on most is JMP,” Choo explains. “We use it to optimize method development and processes. JMP is easy to use and essential for our data analysis.”
Design of experiments: A practical, efficient approach
In Pharmaceutical Quality by Design Using JMP, Rob Lievense – formerly a research fellow of global statistics at American pharmaceutical company Perrigo – writes that the data visualization capabilities in JMP, coupled with its built-for-purpose quality engineering applications, help to create a product that robustly meets requirements for the entire life cycle. Analysis conducted in JMP, Lievense argues, provides an excellent foundation for regulatory submissions when it comes to both products and processes. “Submissions supported with robust statistics tend to have fewer deficiencies,” he writes. “Regulatory deficiencies that occur can be better answered with data visualizations and statistics, which tend to also increase the speed of product approvals.”
Design of experiments (DOE) tools in JMP are critical to this success. DOE is a practical approach to exploring multifactor opportunity spaces, and JMP offers considerable design and analysis capabilities that are easily accessible. Products developed using these tools may reach the market as much as two times faster than those developed with principal science and subject matter expertise alone. Models are used to find the optimum input settings, thereby reducing the chance of costly mistakes.
Choo agrees that these tools are critical to her team’s ability to meet its goals. Using DOE capabilities in JMP, she says “we can get a lot of information from the experimental data, such as the interaction of each factor or curvature. DOE is systematic. We can conduct a small number of experiments to determine what factors affect quality characteristics, build a mathematical model for prediction and then find optimal conditions based on that model.”
Top-quality data visualization is essential for well-designed experiments, she continues, “and JMP offers a variety of excellent plots.” She cites the Prediction Profiler, 3D scatterplot and correlation as being particularly illuminating.
“I'm very attracted to the fact that I can edit the results of statistical analysis using JSL scripts,” Choo says. JMP Scripting Language is an interpreted language for recreating analytic results. It includes a set of functions that make building interactive dialogs simple. “I’m very happy that I can easily change or add a result output using JSL.”
‘Numbers don’t lie’
Choo says that when she joined Celltrion eight years ago, there were no CMC statistics teams in the pharmaceutical field. Today, “most global companies have created teams that specialize in CMC statistics in R&D and manufacturing.” But she’s confident that over the next decade, increasingly more emphasis will be placed on applied statistics, adding: “Numbers are objective indicators. Numbers don't lie."
At Celltrion, support for these quality initiatives comes from the top. CEO Kee WooSung, Choo notes, consistently emphasizes the importance of applying statistics in every aspect of operations. The CMC statistics team creates and teaches curricula for employees to improve their understanding of statistical methods and applications. And the CMC team is itself ever eager to advance its knowledge. “Everyone on our team loves to study,” Choo says. “It’s always very exciting to find new ways to apply new statistical methods when doing new work.”
And JMP affords those opportunities. “JMP is just so intuitive, so easy to use,” Choo says. “It’s a great asset for our business.”
The results illustrated in this article are specific to the particular situations, business models, data input and computing environments described herein. Each JMP customer’s experience is unique, based on business and technical variables, and all statements must be considered nontypical. Actual savings, results and performance characteristics will vary depending on individual customer configurations and conditions. JMP does not guarantee or represent that every customer will achieve similar results. The only warranties for JMP products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with JMP as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of JMP software.