Automation in drug product development - efficiency and quality

anglais. This work evaluates the efficiency and quality of drug substance and product automated sample preparation for liquid chromatography analysis in comparison to the manual preparation technique. Despite the potential advantages of process automation in improving efficiency, accuracy, and precision while reducing time and resources, sample preparation in the early phases of drug development in the pharmaceutical industry is predominantly performed manually. Moreover, recent literature comparing manual and automated methods in this context is limited. Additionally, assessing the uncertainty inherent in analytical measurements is often overlooked, yet crucial for ensuring data quality and reliability. This work aims to evaluate the equivalence of manual and automated sample preparation methods and to assess the associated measurement uncertainties, with the goal to improve efficiency and quality in analytical workflows by implementing automated systems. An automated workflow for sample preparation of drug substance and product was established using the Tecan Fluent liquid handling system. A comprehensive method validation, including specificity, accuracy, precision, linearity, limit of quantification, and stability in solution, was conducted for both manual and automated sample preparation of drug substance and drug product. Furthermore, a thorough uncertainty analysis was performed to judge reliability of the manual and automated method. The manual and automated sample preparation and analysis were both successfully validated, yielding equivalent results, although the manual procedure exhibited higher accuracy. Implementation of the automated workflow resulted in a 72% reduction in hands-on analyst time, and a 69% reduction in required solvent volume. The uncertainty analysis revealed a higher uncertainty for the automated approach compared to the manual sample preparation. While the manual method delivers results with better quality, the automated procedure demonstrated superior efficiency and additional optimization strategies could further enhance the quality and efficiency of the automated workflow.

anglais. Discovering and developing new medicines is crucial for public health, and verifying their quality at every step of this process is equally important. This involves analysing drugs in detail, often using advanced techniques that require meticulous preparation of the drug samples, ensuring that, for example, a tablet preparation can be effectively analysed using specific instruments. While automated systems have improved this process, the pharmaceutical industry has been slow to fully adopt them, leading to the continued use of manual sample preparation methods, which can be time consuming and resource intensive. Several studies suggest that automated sample preparation could offer equivalent results to manual methods while saving time, costs, and resources. However, understanding the precision (how close measurement results in a series are to each other) and accuracy (how close measurement results are to the true value) of automated approaches compared to manual ones remains an area of research. This work aims to address this gap by establishing an automated sample preparation workflow and comparing it with the manual procedure. This includes an evaluation of both methods for their suitability in drug analysis, investigating various parameters, including accuracy and precision, according to regulatory validation guidelines. Additionally, the measurement uncertainty, describing the level of doubt associated with the exact value of a result due to factors like equipment limitations or variations in conditions, for both the manual and automated sample preparation are assessed to judge the quality and reliability of the measurements. The study has shown that both methods generally yielded similar results, with the automated workflow proving suitable for drug sample preparation and analysis. However, the manual method demonstrated slightly higher accuracy and a lower uncertainty, indicating higher quality. Automation, on the other hand, reduced the hands-on analyst time and solvent usage, leading to significant gains in efficiency, environmental friendliness, and safety of the work environment. While moving forward, the choice between manual and automated sample preparation should be made on a case-by-case basis, considering the advantages and drawbacks of both approaches, the work also highlights the potential for additional improvement in the automated sample preparation workflow to enhance both its efficiency and quality further.