Issue link: https://www.e-digitaleditions.com/i/1543258
In this opinion article, the authors discuss potential benefits of utilizing both empirical methods and in silico models for inhaled drug development. Traditional inhaled formulation development can be a costly, slow and iterative process, with no assurance of clinical or commercial success. On average, the development of a successful new drug formulation takes 14+ years and costs $1.5 billion. 10 February 2026 Inhalation Point of View: The role of in silico models in accelerating and de-risking inhaled drug development In this opinion article, the authors discuss potential benefits of utilizing both empirical methods and in silico models for inhaled drug development Victoria Legh-Land, PhD a ; Dan Hardy, PhD a ; Katja Petkau-Milroy, PhD a and Mark Sanders, MBA a, b a Microsol Ltd b University of Hertfordshire Introduction Traditional inhaled formulation development can be a costly, slow and iterative process, with no assurance of clinical or commercial success. On average, the development of a successful new drug formulation takes 14+ years and costs $1.5 billion [1] . However, as few as 3% of developed inhaled drug formulations eventually reach the market, with the majority fail- ing during clinical trials due to efficacy or safety issues [1, 2]. Furthermore, even for formulations that reach patients, drug deposition and uptake are often subop- timal, with the efficiency of inhaled drug delivery typ- ically remaining below 50% for most devices [3-5]. e development process itself is inherently risky. Even minor changes in manufacturing or formula- tion composition can have unforeseen effects on drug deposition and therapeutic outcomes. Furthermore, external market pressures, such as environmental considerations around propellant use [6], can sig- nificantly impact drug formulation development. Reducing these risks and enhancing the prospect of clinical and commercial success within shorter timelines is a critical objective for many inhaled for- mulation development teams. While experienced formulators bring invaluable expertise that can expe- dite the process and lead to successful solutions, using traditional approaches might overlook promis- ing novel technologies or development routes. Teams might be unfamiliar with such alternatives or may previously have had limited success with them, even if those approaches could offer faster, better or lower- cost outcomes. Consequently, integrating predictive in silico methods and artificial intelligence (AI) to guide drug development may offer a more efficient and informed path forward. From a regulatory perspective, in silico modeling is gaining traction as an important tool within the drug development pathway, and regulatory agencies are increasingly recognizing its potential as a com- plementary approach to experimental methods. e United States Food and Drug Administration (FDA), in particular, has been a strong proponent of computational modeling as part of its commitment to modernize drug development under the 21st Cen- tury Cures Act and the FDA Modernization Act 2.0 [7]. ese initiatives actively encourage the adop- tion of model-informed drug development (MIDD) approaches, which utilize in silico models to support formulation optimization, bioequivalence assess- ments and risk mitigation strategies [8]. According to the FDA's 2018 guidance on MIDD, computa- tional models can provide critical insights that may reduce the need for extensive clinical studies, thereby accelerating regulatory review and approval processes [8]. In silico tools can provide supplementary data and may reduce the need for extensive clinical test- ing, particularly in demonstrating bioequivalence or supporting formulation changes [9]. e European Medicines Agency and other global regulatory bodies have also acknowledged the poten- tial of in silico methods [10]. Given this clear regu- latory push towards embracing in silico methods, it is timely to explore how these tools can assist in the development of inhaled drug formulations.