Introduction
Quality-by-design (QbD) is drastically leading to improvements in drug development and manufacturing. The enhancements in technology, global health demands, and strict regulatory frameworks are directing the industry towards fast innovation (Zhu, 2025). Its biggest impact is on quality throughout a drug’s lifecycle, with a focus on quality assurance from earlier on in development. Pharmaceutical manufacturers can stay more consistent with product performance, reduce variability, and enhance overall quality assurance practices. (Romero-Obon et al., 2025). Prior to the implementation of QbD, quality control was heavily limited by the end-product testing and empirical development approaches (Yang et al., 2025). The traditional quality-by-design framework comprises the Critical Quality Attributes (CQAs), Critical Process Parameters (CPPs), and Critical Material Attributes (CMAs). Components include risk assessment, experimental design, design space definition, process optimization, and monitoring and continuous improvement (Romero-Obon et al., 2025). The QbD technique allows for improvement to both analytical techniques and manufacturing processes. In order to successfully use the concept of quality-by-design in pharmaceutical research and development, an in-depth understanding of formulation and manufacturing process variables is needed (Kovacs et al., 2021).
Key Aspects of QbD
QbD emphasizes that planning, risk mitigation, and predictive modeling lead to improved quality. One element of quality-by-design is Target Product Quality Profile, or TPQP, which combines patient-reported outcomes, real-world evidence, and regulatory frameworks to stay in the flow of current clinical focuses. TPQP is important because of its attention to clinical efficacy, safety, and focus on the patients. Critical Quality Attributes include the characteristics that directly affect product safety, efficacy, and patient acceptability. Lastly, QbD also involves looking at Critical Process Parameters and Material Attributes (Yang et al., 2025).
Methods of QbD
Pi-Buckingham Theorem
An apparent challenge in QbD integration for pharmacy is the need for an efficient experimental design and risk assessment. The Pi-Buckingham theorem allows for decreased variables in the path to a solution, such as composition of the input materials, process parameters, environmental conditions, and human manual factors. With its aid in comprehending product dynamics and process dynamics, it plays a vital role in efficiently applying QbD (Romero-Obon et al., 2025).
AI-Integration
Artificial Intelligence (AI) drives quality-by-design towards personalized medicine, particularly in formulation design. This permits QbD to be more tailored to individual patient needs, as opposed to being trial-and-error for a general population (Yang et al., 2025). AI was introduced to the QbD method for a variety of reasons. The first reason was the requirement to improve the efficiency and reliability of drug development and manufacturing. The addition of AI to QbD organizations can focus on product quality and regulation. Lastly, the reliability of the end product can be continuously monitored. With all these factors, a more dynamic model can be observed for manufacturing and the end product. The most common artificial intelligence techniques utilized are machine learning and deep learning models (Zhu, 2025).
Tools for Quality-by-Design
Risk assessment is a good tool to anticipate any issues in product quality. Throughout time, various methods have been researched to help evaluate, identify, and control potential hazards to product quality. The ICH Q9 guidelines highlight five important steps: risk initiation, risk assessment, risk control, risk communication, and risk review (Kovacs et al., 2021). Failure Mode and Effects Analysis (FMEA) plays a more proactive role by investigating any possible flaws in products, the cause of the problem, and the final impact. Fishbone Diagrams allow for additional visual tools for root cause analyses. With current methods of analyzing data, such as machine learning, optimization of the process is possible (Yang et al., 2025). Process analytical technology (PAT) and continuous manufacturing (CM) are also tools in quality by design.
Barriers to QbD in Pharmacy
Applying QbD in pharmacy faces technical and regulatory barriers. Data integrity and validating multivariate processes face challenges due to the complexity of the current systems. Agreement and harmonization are difficult to achieve among various regulatory bodies (Yang et al., 2025). When it comes to AI integration, there is potential algorithmic bias, cybersecurity risks in the PAT systems, increased computational dependency and initial implementation costs (Zhu, 2025).
Discussion
The pharmaceutical industry is moving from the traditional methods of drug development and manufacturing to a more unique quality-by-design method. This method allows for testing throughout the product’s process, rather than performing “trial and error” at the end. QbD allows for an increased focus on the individual patient. An addition of AI into QbD will further enhance the ability of innovation.
References
Kovacs, B., Peterfi, O., Kovacs-Deak, B., Szekely-Szentmiklosi, I., Fulop, I., Baba, L. I., & Boda, F. (2021). Quality-by-design in pharmaceutical development: From current perspectives to practical applications. Acta Pharm, 71(4), 497-526. https://doi.org/10.2478/acph-2021-0039
Romero-Obon, M., Sancho-Ochoa, V., Rouaz-El-Hajoui, K., Perez-Lozano, P., Sune-Pou, M., Sune-Negre, J. M., & Garcia-Montoya, E. (2025). Scale-Agnostic Models Based on Dimensionless Quality by Design as Pharmaceutical Development Accelerator. Pharmaceuticals (Basel), 18(7). https://doi.org/10.3390/ph18071033
Yang, S., Hu, X., Zhu, J., Zheng, B., Bi, W., Wang, X., Wu, J., Mi, Z., & Wu, Y. (2025). Aspects and Implementation of Pharmaceutical Quality by Design from Conceptual Frameworks to Industrial Applications. Pharmaceutics, 17(5). https://doi.org/10.3390/pharmaceutics17050623
Zhu, Z. (2025). Intelligent information management enables quality-by-design in pharmaceutical production. Sci Rep, 15(1), 44201. https://doi.org/10.1038/s41598-025-27879-w