Decreasing the cost and time of drug development while advancing differentiated therapies with improved efficacy and safety is a priority for drug makers. Automation of activities across all aspects of drug development and manufacturing can help achieve these goals. Indeed, “automation can help the API industry with improved safety, productivity, product quality, and optimized costs, making it an essential tool for the industry to remain competitive and meet regulatory requirements,” states Raghavendar Rao Morthala, chief scientific officer, Navin Molecular. These benefits can be realized at all stages of development as well as across development and manufacturing.
Many applications for automation in API development and manufacturing
Automation in many forms has, in fact, been used for API development and manufacturing for many decades, according to Morthala. He points to examples such as electronic lab notebooks and electronic batch manufacturing records, statistical software for design-of-experiment studies employed in process optimization efforts, control systems for API manufacture in both batch and continuous mode, and inventory management and material movement solutions.
“Automation is used extensively in API synthesis,” emphasizes David Ford, senior director of chemistry at Snapdragon Chemistry, a Cambrex company. “While batch manufacturing processes are primarily executed under direction of a human following a batch record, automation is used for temperature and flow control (reagent dosing), as well as for safety interlocks. Continuous flow manufacturing processes use automation more extensively, as automation can be used for reliable start-up and shut-down of the reactor, clean-in-place operations to recover from a clog, and to divert reactor effluent to waste if the quality is expected to be poor,” he explains.
Automation in the development lab
The above forms of automation, Ford says, make work more convenient, but an operator is still required to be present to initiate operations and complete the batch record. “In contrast, automation in the R&D lab can be used to run processes completely without operator intervention,” he observes. In this context, Ford explains, whether batch or flow processes, all steps including reagent charges, sampling, on-line or off-line liquid chromatography, and storage of product for further analysis can be automated.
Such automation tools can also be coupled with statistically powered experimental design or auto-optimization algorithms. “This type of automation can accelerate tasks that can be labor-intensive, such as process optimization and process characterization. Automation can, in addition, free a process development chemist or engineer to research more broadly in the literature or execute more high-level analysis of the data,” Ford comments.
In addition to accelerating physical experiments, literature search engines are available that continuously update comprehensive databases of identified and potential chemical substances, chemical transformations, and newer methodologies. “Using these automated systems, research scientists can quickly identify desired organic transformations and optimum reaction conditions, as well as gain insights into reactant and product properties and raw material sourcing options,” says Morthala.
Phase-appropriate techniques important
It is important, cautions Jens Schmidt, process expert and high-throughput experimentation (HTE) manager at Lonza, to apply phase-appropriate automation techniques to ensure highly effective use of the technology across the entire lifecycle of a project. In the initial phase of a project, according to Schmidt, automation is particularly potent and extensively utilized in process development, commonly in the form of HTE.
“This technique facilitates the miniaturization and parallelization of chemical reactions, allowing for the rapid testing of a broad spectrum of reagents, solvents, catalysts, and conditions. It enables the assessment of various routes, thereby increasing the likelihood of identifying optimal conditions early in the project’s timeline,” he explains.
For example, Ford highlights the use of automation for screening of reagents, catalysts, or solvents to find beneficial combinations and automated workflows for measuring solubility and partitioning in liquid-liquid extraction to investigate isolation strategies for intermediates and APIs. “Such a screening approach can enable identification of useful combinations that would escape attention if experiments are performed manually because a much larger number of combinations can be screened in an automated context,” he notes.
As a project matures, HTE continues to play a crucial role, albeit in a slightly different capacity. The main goal is to aid in process optimization and characterization, thereby enhancing process robustness, notes Schmidt. He adds that automation during this phase also encompasses the standardization and automated execution of individual experiments, with the objective of delivering reliable and high-quality data.
“Upon reaching the commercial stage, automated experiment execution or HTE supports a diverse range of activities, including the optimization of established processes and the enhancement of process understanding,” Schmidt comments. Automation also contributes to the attainment of sustainability goals, the securing of supply chains, and the resolution of unforeseen issues. Of course, automation of large-scale manufacturing is a significant aspect that warrants consideration due to the numerous advantages it affords.
“Automation is transforming the landscape of API development and manufacturing. By automating tasks like route exploration, reaction optimization, and data analysis, researchers can significantly accelerate the identification of optimal synthetic pathways and fine-tune processes for maximum efficiency. These benefits translate to faster development times, reduced errors, and more consistent API production. For drug developers, that means reduced risk and uncertainty in API development, enabling a greater focus on clinical trials and regulatory approvals,” Schmidt concludes.
Using automation in synthetic route development
The concept of automated multicriteria evaluation of new process routes, combined with route development, is based on data mining, according to Morthala. “There is an increasing prevalence of chemical processes being developed based on automated experiments that make use of data science, which can provide multiple numeric criteria for scientists to reach a balanced decision on the suitability of a novel synthetic route,” he says.
Information generated via data mining can be used for chemical route development and evaluation, including planning of hypothetical synthetic routes, Morthala contends. Such efforts can take into consideration not only process conditions, but environmental indicators, such as energy consumption, E-factor values, solvent scores, reaction reliabilities, and route efficiencies allowing for multi-criteria environmental sustainability evaluations, he observes. “This type of automated workflow is based on deep data mining and algorithmic decision-support tools to enable multi-criteria decision making,” he says.
Artificial intelligence in synthetic route development
Small-molecule APIs are becoming more and more complex, and the resulting longer synthetic pathways pose challenges for process chemists to develop efficient API manufacturing processes. Overcoming these challenges is crucial to avoid delays in investigational new drug applications and expenses in transitioning new drug candidates from early development to commercial production, according to Simon Wagschal, associate director of advanced chemistry technologies, Lonza Small Molecules. “As the industry faces new drug candidates requiring over 20 synthetic steps on average, AI [artificial intelligence]-enabled predictive route design technologies have become essential for creating efficient retrosynthetic routes,” he says.
Indeed, AI-enabled API development tools are already being used by several R&D companies for the identification of optimized synthetic routes, according to Raghavendar Rao Morthala, chief scientific officer, Navin Molecular. “An AI platform has the potential to design and synthesize APIs in a matter of weeks or months, compared to the years it can take using conventional methods,” he says.
Currently, however, the application of AI to API synthetic route design is in its early stages, contends David Ford, senior director of chemistry at Snapdragon Chemistry, a Cambrex company. Synthetic route design by AI is used primarily at the brainstorming stage of route development, he says, with identified options narrowed down by humans. “Given the importance of the choice of a route or routes and the quantity of resources used to pursue them, the role of the synthetic chemist remains central in route design,” he states.
Wagschal adds that AI tools have traditionally been used for early-phase discovery, often generating routes not suitable for commercial-scale production.
Other exciting applications for AI in process development, Ford says, lie in its ability to manage knowledge. “Recent advances in large language models have made it possible to index and explore diverse sets of documents like weekly updates, PowerPoint presentations, technical reports, batch records, and campaign summaries,” he explains. These AI models allow for the content to be searched semantically, based on the meaning of the search, instead of by keywords.
“Given the volume of information trapped in these documents, it is very likely that work is forgotten, and hard lessons then need to be re-learned,” Ford observes. “We see great opportunity in helping researchers make connections between their work and the work that has been done in the past,” he notes. Generative AI models may also prove useful for preparing first drafts of templated documents like standard operating procedures, qualification protocols, and batch records, according to Ford.
Lonza, meanwhile, is one company not deterred by the limitations of early AI programs developed for scouting of synthetic routes for small-molecule APIs. “The introduction of computer-aided synthesis planning (CASP) technologies with route prediction capabilities has significantly influenced the route selection process,” says Wagschal.
Nearly all CASP tools employ a form of machine learning called Monte Carlo Tree Search to rapidly prioritize computationally predicted synthetic routes, Wagschal explains. This heuristic search method evaluates the likelihood of reaction success at each step of the proposed pathway, from starting materials through intermediates to the final API. The prioritized routes are then reviewed and refined by process chemists. “This approach, guided by our experts, has increased the number of potential strategies and led to more efficient API synthesis routes,” he states.
To overcome the limitations of early AI-based models for route development with respect to commercial practicality, Lonza has combined leading AI-powered predictive route design models with proprietary supply-chain data, including raw material costs, availability, and detailed supply chain intelligence, according to Wagschal. “This integrated technology allows process chemists to quickly identify not just the shortest pathways, but also the most commercially viable ones by considering real-world supply chain constraints. These advanced AI solutions provide a holistic view that streamlines route scouting and accelerates the transition from research and development to robust manufacturing processes, resulting in significant time and cost savings in developing complex APIs,” Wagschal concludes.
Access to data essential
There are challenges to using automation tools for API development, though. Automation systems require substantial investment in hardware, software, and infrastructure, and thus can be expensive to commission, Morthala observes.
Access to relevant, high-quality data is also important, Morthala adds. For automated route development in particular, wide implementation of automated development and evaluation methodologies requires recognition of the value of reaction data currently being held in the cheminformatics community by the wider chemistry and chemical engineering community and publication of that data in a manner that ensures it is accurate and available electronically. “If this state can be achieved, it will be possible to leverage automation for process development on a wider scale,” contends Morthala.
One of the biggest challenges to benefitting from automation in API development in Ford’s opinion is that any automation tools need to at least perform as well as the highly trained and experienced process development scientists and engineers currently executing much of this work manually. “These experts have developed very efficient ways of doing their work and can easily adjust to accommodate a very wide range of process conditions, something that is difficult to achieve in an automated platform. Given that challenge, the key to succeeding in using automation is to identify areas where automation will have a fighting chance of adding value,” he comments.
A holistic approach is best
Beyond having access to the right data and tailoring automation strategies to the specific stage of the project lifecycle, best practice for using automation in API development activities is to take a holistic approach. “First and foremost,” underscores Schmidt, “is to define clear objectives that allow for prioritization of tasks and processes that will most benefit from automation.”
It is also important, according to Schmidt, to ensure that automated workflows are reliable and thoroughly understood. Applying appropriate analytical techniques and where possible integrating advanced analytical tools into automation solutions will enable monitoring and optimization of processes. “Finally,” Schmidt observes, “promoting interdisciplinary collaboration to leverage diverse expertise and ensure cohesive integration of automation solutions, including those for data handling and treatment, across different phases of API synthesis will allow for full realization of the benefits of automation.”
About the author
Cynthia A. Challener, PhD, is a contributing editor to Pharmaceutical Technology®.
Article details
Pharmaceutical Technology®
Vol. 48, No. 8
August 2024
Pages: 14–17
Citation
When referring to this article, please cite it as Challener, C.A. Automating Development of Small-Molecule APIs. Pharmaceutical Technology 2024 48 (8).
