A Familiar Debate
Every pharmaceutical company eventually has this discussion.
The process development laboratory is expanding. A new building is being designed. Equipment is being purchased. Someone asks whether the laboratory should operate under GMP.
Within minutes, the room divides.
Quality argues that instruments should be qualified, procedures controlled, and changes documented. Scientists counter that process development exists to explore, learn, and iterate—not to manufacture commercial product. Engineering simply wants to know which standards to design around.
Interestingly, everyone is trying to solve the same problem, yet they are answering different questions.
After nearly two decades working across process development, manufacturing, manufacturing science and technology, and commercial operations, the author has come to believe that the industry is debating the wrong issue.
The question is not whether process development should be GMP. The real question is much simpler.
When does science become GMP?
Current regulations define expectations for commercial manufacturing remarkably well. Activities that directly support commercial manufacturing or product release—and have the potential to affect product quality and patient safety—are clearly governed by GMP. At the same time, ICH Q8(R2) recognizes pharmaceutical development as the stage where process and product understanding are established and enhanced. Process characterization, scale-down model qualification, process optimization, and analytical method development are all integral to this phase of the product lifecycle.
What remains less clear is the quality framework that should govern the laboratories performing these activities. Process development laboratories generate the scientific knowledge that underpins technology transfer, control strategy development, process validation, and ultimately commercial manufacturing. Yet, unlike commercial manufacturing operations, their primary purpose is to generate scientific understanding rather than manufacture or release product intended for patients.
This distinction has resulted in remarkably different practices across the industry. Some organizations apply commercial GMP-like controls throughout process development, including equipment qualification, formal change control, and quality oversight. Others rely on robust scientific quality systems built around calibration, preventive maintenance, documented procedures, trained personnel, and data integrity. Both approaches share the same objective—producing reliable, scientifically defensible data—but reflect different interpretations of the level of oversight appropriate for development activities.
Consider a simple pH meter. During an exploratory chromatography study, few would expect every calibration excursion to trigger a formal GMP deviation. Use the same instrument the following day to adjust the pH of a commercial manufacturing batch, however, and full GMP expectations rightly apply. The instrument has not changed; only the intended use of the data has.
From GMP vs. Non-GMP to Fit-for-Purpose Quality
Rather than asking whether a process development laboratory is GMP or non-GMP, we should ask whether its quality system is appropriate for the intended use of the data it generates. Data intended to build process understanding should not necessarily be governed in the same way as data used to support commercial manufacturing, batch release, or decisions that directly affect product quality and patient safety.
This is not an argument for reducing quality. On the contrary, process development laboratories require rigorous scientific quality systems to ensure reliable, reproducible, and traceable data. The question is not whether quality is required, but whether the level of quality oversight should be proportionate to the purpose and regulatory reliance of the work being performed.
Looking Ahead
The pharmaceutical industry has continually evolved its approach to quality as science and technology have advanced. As process development assumes an increasingly important role in process characterization, technology transfer, and lifecycle management, it may be time to reconsider how quality systems are applied. Rather than relying on a binary GMP or non-GMP distinction, a fit-for-purpose, risk-based framework that aligns quality expectations with the intended use of the data could provide a more consistent and scientifically appropriate approach for the next generation of pharmaceutical development.
