As the number of biologic drugs has grown in the pharmaceutical pipeline and market, aseptic manufacturing capabilities have steadily grown in importance. Assuring sterility across many different modalities and product formats/sizes/volumes (e.g., vials, cassettes, prefilled syringes/pens, microdosing systems, intranasal delivery systems, on-body devices, bags, etc.), many of which are increasingly potent, has also become more challenging. Innovation in processing technologies has been essential to address the evolving needs of manufacturers. Key areas of development have included digitalization and automation solutions.

Repeatability and speed with improved quality and yield

Digitalization and automation provide numerous benefits to aseptic processing operations. Two of the most important, according to Alex Strauss, senior principal process equipment engineer with Curia, are greater speed combined with improved repeatability. “The key to automation is having a recipe-driven approach that is qualified. When the operator selects the recipe, the process is then repeatable. In addition, processes such as visual inspection can be completed much more rapidly than is possible by human operators and with incredible accuracy,” he explains.

New filling systems, adds Tom Clemens, senior director of engineering and facilities at Lifecore Biomedical, perform non-destructive weight checks of individual products immediately after filling, improving quality and increasing batch yields. Many also include automated camera verification systems to check vial crimping and stopper placement. “When an issue is detected, the system can automatically stop the line so the problem can be investigated immediately, avoiding costly material losses and potentially much longer production delays,” he says.

Many newer biologic products, including antibody-drug conjugates, radio-immunoconjugates, and cell and gene therapies, require high containment solutions to ensure operator safety. The diversity of products with high activity produced in multiproduct facilities has also created the need for effective solutions that reduce cross-contamination risks.

Regulations such as Annex 1 requirements in the European Union (1) have also been an important driver toward greater adoption of isolator technology to increase product safety by minimizing potential human impacts on sterile drug products, according to Clemens. Many manufacturers have therefore been shifting to the use of isolators for aseptic processing.

“Due to the nature of these closed, isolator-based filling systems, automation and robotics have been incorporated to perform necessary processing steps within the sterile environment,” Clemens observes. For instance, he notes that while older systems may have required an operator to remove a lid from a ready-to-use (RTU) tub of vials or syringes, many new isolator fillers use automated systems that heat the lid adhesive and then remove the lid from the tub with a variety of rollers. “This approach,” Clemens states, “reduces human contact and potential particle generation.”

Visual inspection operations are another area that has benefited significantly from advances in automation and digitalization technologies. These automated visual inspection systems use high-speed cameras and advanced software to automatically check for particulates and defects and can operate at fairly high speeds, according to Strauss. The Syntegon system deployed at Curia’s Albuquerque, NM facility, for instance, can inspect 400 vials per minute, matching the filling system’s filling speed.

As importantly, newer automated visual inspection systems are now available that can reliably evaluate highly viscous formulations (in excess of 100,000 centipoise) in vials and syringes, which they were previously not able to do. “Traditional automated visual inspection systems spin a vial/syringe and rely upon a camera to enable evaluation of the internal movement of the product to identify contaminants, which is not a suitable approach for viscous formulations because they do not flow in the same manner as lower viscosity solutions,” Clemens explains. Newer systems, he comments, incorporate additional cameras and advanced virtual image processing to identify particulates and defects and can even improve performance over time by incorporating machine learning algorithms.

Robotic systems do not just support the incorporation of aseptic filling processes within isolators and visual inspection activities. They have been widely implemented in sterile drug manufacturing processes to automate many different repetitive tasks, including vial filling, sealing, and labeling. By automating these activities, human interaction with sterile drug products is minimized, reducing contamination risks. Furthermore, robotic systems minimize product losses and enable the accurate filling of products that require only very small volumes.

Robotic systems can also eliminate some of the steps required when human operators are more involved in the aseptic filling process. For instance, the VarioSys Flex Line system at its Albuquerque, NM facility uses robotics to remove vials from a pre-sterilized tub, eliminating the need for a vial washer and sterilizing tunnel, according to Eric Schneider, director of global equipment engineering at Curia.

“In manually operated systems, it is not possible to aseptically remove primary containers without obstructing the first air (uninterrupted unidirectional filtered airflow used to provide a sterile environment) and/or subjecting the vials to scratches, etc., thus creating the need for additional steps,” Schneider says. With the VarioSys Flex Line, operator interventions are minimized, simplifying the process while also affording improved quality and safety outcomes.

Schneider does caution, however, that even the best robotics and automation systems still rely on human operators. “Training programs for operators are essential,” he emphasizes. To address this need, Curia operates Sterile University, a program designed to provide both training for new employees and continuing education for all manufacturing and quality staff through a simulated manufacturing environment that replicates the facilities, equipment, operations, and protocols found at the company.

Improvements in automated and digitized analytics

Automation and digitalization of analytics that support aseptic processing has several benefits as well. One of them, according to Jon Kallay, senior scientific portfolio specialist for microbial solutions with Charles River Laboratories, has been the facilitation of the adoption of rapid testing techniques that enable faster product release while still ensuring product quality.

One challenge with complex biologics highlighted by Kallay is the tendency for these products to appear turbid during routine sterility testing due to their physicochemical properties. Identification of such turbidity during manual visual inspection operations typically prompts a lengthy evaluation process to confirm the turbidity is not due to contamination.

Rapid test methods that evaluate multiple attributes, such as detection of adenosine triphosphate, respiration, and other microbial indicators, when combined with automated analysis of test results, provide accurate indicators and eliminate the subjective aspect of test results, Kallay contends. “Analysts no longer need to shake a bottle and determine if it fits their understanding of ‘turbid.’ Results are automatically analyzed and determined to be positive or negative based on the standards established during the validation of the technology,” he says. “Objective results eliminate unnecessary procedures while allowing quick release of product to patients,” Kallay concludes. In addition, more rapid access to test results allows users to respond quickly to any contamination events that could impact the facility and other product lots.

Digitizing the test process also improves data integrity. Documentation errors are not uncommon when test results are managed manually, according to Kallay. With digital test technologies, however, test materials and results are immediately documented and automatically transcribed to the laboratory information management system (LIMS) data storage system, avoiding opportunities for operator error.

The ability to collect real-time manufacturing data during aseptic processing has been, according to Clemens, a real game-changer, as it has made it possible to monitor quality in a proactive rather than a reactive manner. Lifecore Biomedical, for instance, uses a dashboard approach displayed on monitors throughout the facility to keep a watchful eye on manufacturing operations real-time and quickly identify and resolve line stoppages. “Trending and identification of variances, such as between shifts, also support identification of specific groups of operators that could benefit from additional training or opportunities for improving the mix of operator experience levels within shifts,” Clemens notes.

A specific example of an important development in real-time monitoring, according to Laura Choteau, quality control manager at Catalent Biologics, has been the introduction of real-time monitoring for visible particles. “Today, automation allows us to read culture plates from the start of incubation through to the end, with full traceability of the plate’s location and condition throughout. As a consequence, we can detect contamination risks much earlier and respond immediately. That reduces the likelihood of recurring contamination and helps us get much closer to identifying the root cause,” she comments.

Other notable examples of process analytical technology tools improving aseptic processing operations, according to Jean-François Boé, senior director of drug product development for Catalent Biologics, include onboard mass spectrometry and Raman spectroscopy systems embedded into compounding or fill/finish units. “The data generated from these instruments give us real-time visibility into critical process parameters. In the long term, this level of control could reduce the need for traditional final release testing, shifting more quality assurance upstream and closer to the point of manufacture,” he says.

Like Schneider, Boé emphasizes the role still played by human operators in many monitoring activities. For instance, air monitoring systems in Grade A areas still often require operators to insert culture plates, typically using glove ports. This need does carry a small but real risk of contamination, as well as the potential for plate insertion to be accidentally skipped, which would create a gap in monitoring and raises questions during batch review. These issues could be avoided, Boé says, with the use of robotics to install the plates, which would “add a layer of reliability to aseptic operations and increases confidence in the process.”

New data management software equally impactful

Automation in analytics has been around for some time and continues to improve with advances in technologies. What has driven a real shift, according to Choteau, has been how the generated data are managed and used. “Data management plays a central role in helping us act faster and make smarter decisions in aseptic environments. Modern LIMS platforms now allow us to automate calculations and perform trend analysis continuously, which provides more frequent and reliable insights, helps ensure accuracy, and allows quality teams to concentrate on investigations when abnormal trends appear,” she explains.

Off-the-shelf, user-friendly visualization and reporting software, meanwhile, has enabled engineering teams to use data in a way that would have required significant involvement of information technology personnel in the past, Clemens notes. For example, Lifecore Biomedical tracks all line interventions digitally. “Having operators identify reasons for interventions in a consistent manner from shift to shift enables our engineering teams to identify trends and take corrective actions before real issues arise,” he says.

Furthermore, both process efficiency and control are also increased. “Automation brings greater control and traceability. It minimizes sampling errors, confirms that all required samples have been taken, and provides a clear record for each action. This helps maintain compliance and strengthens confidence in our processes,” Choteau states.

Staff training, like with other aspects of digitalization and automation of aseptic processing, remains an essential component, however. Operator readiness to respond in the event of system failures must be maintained, and staff must be able to manage operations manually when needed, according to Choteau. Data security, she adds, is equally important. “Protecting data from cyber threats and ensuring its integrity remains a top priority as more systems become connected and automated,” Choteau observes.

Digital twins support effective process simulation

Modeling is an important aspect of digitalization that is benefiting many process development and manufacturing activities, including sterile production operations. Digital twin technology, for instance, is becoming increasingly valuable in drug substance manufacturing and beginning to show potential in aseptic fill/finish operations, according to Boé.

Digital twins in the pharmaceutical manufacturing context are virtual models of processes that are continuously updated with real-time data, allowing assessment of performance. “By creating a digital replica of equipment and processes, digital twins help us simulate scenarios, predict outcomes, and optimize parameters before moving to physical implementation,” he says. Boé finds this approach especially useful during process transfer, because it allows teams to anticipate risks, reduce variability, and avoid costly delays.

Artificial intelligence has growing value

Digital twins are not the only example of how automation and digitalization are being combined with artificial intelligence (AI), machine learning (ML), and other advanced computing capabilities to further improve aseptic processing. When leveraged together, these myriad technologies are accelerating timelines and enhancing process control, according to Boé.

As an example, Boé points to the benefits AI-driven models can provide for freeze-drying (lyophilization) processes. In particular, he notes they can be used to optimize cycle development, significantly reducing the time required to reach a robust process and positively impacting project timeliness and resource efficiency.

In fact, Boé expects AI and ML to play an even larger role and provide growing value as complexity—product, process, and equipment-related—continues to increase. “AI and ML help with not only data analysis and trending, failure prediction, and workflow optimization; they also support more informed decision making. All of these benefits ultimately lead to greater efficiency and reliability of aseptic manufacturing operations,” he concludes.

1. EC. Annex 1. Manufacture of Sterile Products. 

EudraLex–Volume 4–Good Manufacturing Practice (GMP) Guidelines


Vol. 49, No. 8
October 2025
Pages: 18–20

When referring to this article, please cite Challener, C.A. Advances in Digitalization and Automation of Aseptic Processing. 

Articles

The increasing prevalence of biologic drugs has heightened the importance of aseptic manufacturing, necessitating innovations in processing technologies. Digitalization and automation have enhanced aseptic processing by improving speed, repeatability, and quality. The adoption of isolator technology and robotics has minimized human contact, reducing contamination risks. Automated visual inspection systems and real-time monitoring have further improved quality control. Data management advancements have facilitated faster decision-making and process optimization. Digital twins and AI are increasingly valuable for simulating processes and enhancing control, ultimately improving efficiency and reliability in aseptic manufacturing.

Manufacturing

The increasing diversity and complexity of injectable drug products is driving innovation.

Advances in Digitalization and Automation of Aseptic Processing

Advanced drug discovery tools are leading the identification of increasingly complex small molecules with physicochemical properties that lend promising biologic and pharmacologic activity while simultaneously creating tremendous challenges in developing formulations capable of delivering the drug to its intended site of action. Many different molecular properties acting alone and interdependently with each other and different formulation excipients and buffers influence the pharmacokinetics and pharmacodynamics of APIs. Using traditional empirical methods for investigating and identifying potential formulations for these complex molecules requires excessive time and resources at a time when reducing cost and accelerating development are essential to success. Formulation scientists are turning to predictive modeling approaches and platform formulation processes that leverage artificial intelligence (AI) and machine learning (ML) to more rapidly and comprehensively evaluate complicated formulation spaces including complex molecules and advanced drug delivery technologies (1–4).

Moving from academia to industrial application

The use of AI/ML for industrial pharmaceutical formulation development is at the nascent stage. “Much of the progress so far has come from academic research, which often seeds future industry innovation, as well as from emerging startups and the steady improvement of open-source models,” says Pauric Bannigan, chief science officer and cofounder of Intrepid Labs.

“Interest is growing rapidly, fueled by increasing recognition of AI/ML’s potential to navigate high-dimensional formulation design spaces, optimize excipient selection, and predict stability or release performance before extensive lab work,” Bannigan observes. Given these developments and the rapid advances in lab automation currently being made, he believes it is only a matter time before AI and ML become widely applied to formulation development.

How are growing formulation challenges overcome?

Small-molecule drug developers face several challenges, and many of them are well-suited to being addressed using AI/ML approaches, according to Bannigan.

Top of the list, says Daniel Joseph Price, head of the Excipients Business and working within Process Solutions in the Life Science business of Merck KGaA, Darmstadt, Germany, is the poor solubility of drug substances, with approximately 70–90% of drugs currently in the pipeline suffering from this problem (5). “This challenge is pronounced for newer drug candidates that often surpass Lipinski’s Rule of Five (6), which sets foundational criteria for oral bioavailability. In this context, overcoming solubility issues to ensure effective oral drug delivery and hence enhancing patient acceptance remains paramount,” Price emphasizes.

Other small-molecule formulation challenges highlighted by Bannigan include achieving precise release kinetics for controlled- and sustained-release systems, identifying robustness early for scale-up and manufacturability, and minimizing API waste during screening. “Each of these challenges is a multidimensional problem where dozens of formulation components and process parameters interact in nonlinear ways,” he says.

Formulations that address these key issues often involve multiple excipients. Different combinations of excipients must thus be investigated along with different formulation concentrations and processing parameters (e.g., mixing speed and time, temperature control, compression force, drying method and rate, etc.) to identify the optimum formulation design. Even with tight constraints, the exploration space can be enormous, according to Bannigan.

Pharmaceutical formulation, Bannigan adds, is a high-dimensional, sparse-data challenge where dozens of excipients and process parameters interact in nonlinear ways. As an example, he notes that with three excipients, five possible concentrations, and five process parameters at three settings each, the total number of possible unique formulations would be 3,645,000. And that would be a conservative case. “In reality, formulations often use more excipients in varying ratios with additional process parameters, pushing the possibilities into the tens or hundreds of millions,” he says. Exploring them all is simply not possible using conventional approaches, including predictive modeling systems leveraging conventional statistical methods.

Without AI/ML, in fact, Bannigan observes that countless breakthrough formulations, and more importantly the medicines they could enable, will remain buried in the noise, invisible to traditional trial-and-error approaches. “AI/ML can cut through this complexity by modeling these interactions, learning from each experimental result, and steering exploration toward the most informative and promising regions of the design space,” he contends.

Specifically, Bannigan notes, AI/ML can map complex, multi-objective relationships between components and performance; identify promising solutions that human intuition might overlook; and adapt in real time as new experimental data arrives through active learning. “By efficiently navigating vast combinatorial spaces, AI/ML not only accelerates the discovery of robust, scalable formulations, it reduces API waste, increases the probability of hitting target product profiles, and brings hidden, high-potential medicines into view,” he observes.

The application of AI/ML in formulation development offers a transformative approach to current formulation challenges, agrees Price. “By using AI/ML, scientists can streamline the formulation process, effectively narrowing down the design space. This allows them to concentrate on formulations and excipients that are likely to have a significant impact, rather than getting lost in exhaustive trial-and-error testing.”

“Furthermore, AI/ML enables researchers to explore a wider array of options, facilitating innovative solutions without the extensive time investment typically required for traditional formulation trials,” Price states. For instance, he anticipates companies harnessing AI/ML technologies to investigate broader ranges of chemicals and solutions by trying out things that go beyond the obvious, fostering greater innovation in drug formulation. “This strategic shift could ultimately lead to more effective therapies and improved patient outcomes,” he says.

Improving predictive formulation modeling

Formulation development has traditionally relied on mechanistic or simplified models to guide decision-making and optimize outcomes, according to Price. Examples include the use of the Arrhenius equation to predict the stability of drug substances under accelerated conditions, or Lipinski’s Rule of Five, which provides heuristic understanding for assessing the drug-likeness and oral bioavailability of small molecules. These models, which are rooted in physical chemistry and empirical rules, have delivered significant value—often requiring limited data to produce accurate and interpretable results for well-understood systems, Price notes.

The greater complexity of current drug candidates combined with the increased volume and variety of available formulation data is revealing the limitations of mechanistic models to solve some formulation challenges, Price says. Many new drug molecules fall outside the “rule-of-five” chemical space and exhibit behaviors that are not easily captured by conventional equations.

“In this evolving landscape, AI/ML models offer distinct advantages. ML algorithms can process and learn from large, multidimensional datasets, capturing subtle, nonlinear relationships between formulation variables and outcomes such as solubility, stability, and bioavailability,” Price comments. Unlike mechanistic models, he adds, ML approaches are not constrained by predefined assumptions and can uncover patterns that may be invisible to human experts, enabling formulation scientists to explore a far broader design space, accelerate optimization cycles, and address formulation challenges that would otherwise be intractable using traditional methods. Thus, while mechanistic models remain foundational tools in formulation science, Price believes AI/ML models provide powerful complementary capabilities, especially for novel drug development scenarios.

It is also advantageous, according to Bannigan, that AI/ML algorithms can be slotted directly into existing predictive modeling approaches by turning what is usually a static experimental plan into a living, adaptive process. “Instead of running a fixed set of tests designed at the start using design-of-experiment approaches, AI/ML begins with a small number of experiments, learns from the results, and instantly updates its recommendations for the next round. This cycle repeats, zeroing in on the most promising parts of the formulation space while skipping unproductive paths,” Bannigan explains. The advantages over traditional approaches include speed, efficiency, and reach. “Predictive models leveraging AI/ML enable identification of high-potential formulations faster using less API, and often find solutions that conventional methods never get close to,” he concludes.

Enhancing platform formulation strategies

AI/ML can also be used to improve platform processes for formulation development. One example, says Bannigan, is the use of these technologies as decision-making engines within self-driving labs and automated formulation platforms. “In this case, the AI/ML system analyzes incoming experimental data in real time, recommends the next most informative experiments, and identifies multiple viable formulation paths rather than prematurely converging on a single lead,” he explains. The system also standardizes workflows so that every project benefits from cumulative platform learning.

When integrated into automated laboratory platforms, AI/ML technologies can, adds Price, not only help reduce the manual effort required and variability associated with human operation, but enable the generation of large volumes of high-quality, consistent experimental data that can be leveraged to train and refine machine learning models, enabling the discovery of complex patterns and relationships that might otherwise go unnoticed.

Price also notes that the integration of ‘active learning’ strategies such as Bayesian Optimization, which works in tandem with ML models to iteratively select the most informative experiments that will most efficiently reduce model uncertainty and accelerate the search for optimal formulations, enhances this process.

“This synergy between automation, AI/ML, and active learning allows researchers to maximize the impact of each experiment, accelerate formulation optimization, and tackle challenging formulation development problems,” Price says. He adds that “these advances enable drug developers to innovate faster, reduce costs, and make more confident, evidence-based decisions throughout the formulation lifecycle.”

“The tightly integrated, data-rich, closed-loop cycle created by combining automation, AI/ML, and active learning not only accelerates the design–build–learn process and boosts experimental efficiency, it also reduces API consumption and increases the likelihood of finding scalable, high-performance formulations,” Bannigan contends. This intelligent, iterative process, he believes, will help companies identify novel formulation approaches and, ultimately, new life-altering medicines that would otherwise go undiscovered.

Increasing the likelihood of candidate success

Accelerated formulation development and identifying new formulation strategies are not the only advantages of leveraging AI/ML technologies when exploring formulation spaces. By rapidly focusing development on formulations with the highest predicted likelihood of meeting target product profiles, AI/ML also cuts down both early attrition and late-stage failures from chemistry, manufacturing, and controls-related issues, according to Bannigan.

 correlation from the start, ensures scalable formulations are identified sooner, and conserves API so teams can explore backup candidates in parallel. The result is a higher probability that promising small-molecule drugs will make it through development and into the hands of patients, essentially turning potential therapies into real, market-ready medicines faster and with less risk,” Bannigan contends.

One industrially relevant example is a predictive model developed by Merck KGaA, Darmstadt, Germany for accurately and reliably predicting cocrystal coformers for a range of APIs without the need for any prior experiments. Optimum cocrystals, according to Price, often exhibit significantly higher solubilities than the original API, but identifying them can be challenging. The AI-based model was trained using a carefully curated, large dataset of test drugs and coformers covering a large chemical space and supplemented with targeted experimental data and has been shown to dramatically improve the likelihood of finding the optimum coformer (by a factor of three compared to the trial-and-error approach) (2).

While hindrances to adoption of AI/ML in formulation development do exist, such as fragmented, inconsistent, and proprietary historical data; cultural and regulatory caution that demands interpretability; and the lack of integrated automation infrastructure to operationalize models effectively, Bannigan anticipates AI/ML becoming a standard part of formulation workflows within the next one to three years. He sees such systems becoming integrated with high-throughput platforms, driving faster design–build–learn cycles, and using active learning to adapt in real time with model design and experiment selection guided by large language models (LLMs) capable of rapidly synthesizing literature, historical data, and domain expertise.

In the longer term (3–10 years), Bannigan predicts foundation models, federated learning, and AI-driven excipient design will transform formulation development from trial-and-error to precision engineering, with LLMs evolving into specialized scientific copilots. “In this future, more therapies will advance successfully from development to the patients who need them,” he concludes.

Huanbutta, K.; et al. Artificial Intelligence-Driven Pharmaceutical Industry: A Paradigm Shift in Drug Discovery, Formulation Development, Manufacturing, Quality Control, and Post-Market Surveillance. European Journal of Pharmaceutical Sciences, 2024 203, 106938. DOI: 10.1016/j.ejps.2024.106938

Merck KgaA, Harnessing AI To Speed Up Drug Formulation.EMD Group Science Space Blog, emdgroup.com.

Bannigan, P.; et al. Machine Learning Directed Drug Formulation Development. Advanced Drug Delivery Reviews, 2021 175, 113806. DOI: 10.1016/j.addr.2021.05.016

Konagurthu, S. Revolutionizing Drug Development: AI-driven Solutions for Poor solubility and Bioavailability. Patheon Blog, Patheon.com, April 15, 2024.

Ting, J.M.; et al., Advances in Polymer Design for Enhancing Oral Drug Solubility and Delivery. Bioconjugate Chemistry, 2018 29 (4), 939-952. DOI: 10.1021/acs.bioconjchem.7b00646

Halford, B. Wrestling with Lipinski’s Rule of 5. C & E News, 2023 101 (8). https://cen.acs.org/pharmaceuticals/drug-discovery/Wrestling-Lipinski-rule-5/101/i8

Cynthia A. Challener, PhD, is a contributing editor to 


Vol. 49, No. 8
October 2025
Pages: 14-17

When referring to this article, Challener, C. A. Predictive Modeling for Small-Molecule Formulation Development Using Advanced Algorithms. 

Advanced drug discovery tools are identifying complex small molecules with promising biologic and pharmacologic activity, posing challenges in formulation development. Traditional empirical methods are time-consuming and resource-intensive, prompting a shift towards AI/ML-driven predictive modeling and platform formulation processes. These technologies optimize excipient selection, predict stability, and enhance formulation efficiency. AI/ML can navigate complex formulation spaces, reduce API waste, and improve the likelihood of successful drug development. Despite challenges, AI/ML is expected to become integral to formulation workflows, transforming drug development from trial-and-error to precision engineering.

Development

Artificial intelligence and machine learning are anticipated to boost success rates.

Predictive Modeling for Small-Molecule Formulation Development Using Advanced Algorithms 

Bioanalysis is used in drug development to determine pharmacology, bioavailability, and bioequivalence and includes pharmacokinetic, toxicokinetic, or biomarker concentration evaluation (1). Regulators require bioanalytical methods to be validated to ensure their effectiveness (1). Advanced tools, such as artificial intelligence (AI) and machine learning (ML), are helping drug developers evaluate the complex data involved in these types of studies, especially for novel therapeutics, and generate new insights from these data.

Evolution in the field of bioanalysis techniques is often driven by the need to address analyte modalities, specifically biologic drugs, metabolites, and biomarkers, according to Steve Lowes, senior director of Scientific Affairs at IQVIA Laboratories. “Sensitivity is the typical challenge, heightened by the interest to achieve needed quantification from a smaller sample as possible,” he explains. “We, thus, have three compounding factors. One is more sensitivity. Second is a smaller sample we’re expected to work from, and third is an increasing complexity of the analyte modalities themselves.”

How are new technologies making bioanalysis more accurate, more efficient, and quicker? And what is driving, or challenging, their implementation?

While early use of AI/ML was in reports, quality control of assay validation, and study sample analysis, these technologies are being adopted more widely in bioanalysis for validated applications in development, according to Mark Arnold, owner and principal, Bioanalytical Solution Integration. “As the capabilities of the AIs have advanced, so have the applications in bioanalysis,” he says. “New applications are looking at taking GLP [good laboratory practice] or GCP [good clinical practice] protocols and bioanalytical contracts to write bioanalytical plans, perform quality control checks that integrate ELN [electronic lab notebook], instrument, and LIMS [laboratory information management system] data on a daily basis to catch errors early and identify trends that lead to failure before they happen.”

AI/ML help automate data processing and integrate complex datasets from analytical platforms, enhancing assay precision and decision-making, according to Stuart McDougall, principal research fellow, at Arcinova. “AI-driven predictive models support real-time quality control, while machine learning algorithms identify trends and anomalies to ensure regulatory compliance,” he notes. “Together, they reduce development timelines, improve operational efficiency, and support scalable, data-driven innovation across the drug development lifecycle. When integrated into systems like LIMS, these insights are directly tied to scientific context, not just numbers on a screen.”

Yasmin Emamgholi, PhD, product manager, LabVantage Solutions, adds, “Instead of spending hours combing through assay data or chromatograms, these solutions can pick up subtle trends and patterns in real time. In development, models can suggest how a compound might behave, flag outliers, or even predict interactions before they show up in the lab. On the manufacturing side, AI can spot early signals in quality control, so issues do not turn into costly failures.”

Emamgholi also points to agentic AI, which can analyze data, as well as perform tasks across the bioanalysis workflow. “For example, agentic AI solutions can help in assay development and optimization with autonomous experiment design and closed-loop optimization,” Emamgholi explains. “Such solutions can also help in manufacturing and quality control by autonomously analyzing in-process bioassay data, tracking anomalies, and helping with predictive troubleshooting, including detecting early assay or equipment issues and triggering preventative maintenance before failures occur.”

These technologies can also support predictive modeling, according to Theo de Boer, PhD, principal scientist–scientific director LCMS, at Ardena, enabling quick detection of complex biomarker patterns and enhanced pharmacokinetic and pharmacodynamic modeling.

Long Yuan, director, Department of Drug Metabolism and Pharmacokinetics, Biogen, also sees AI/ML as beneficial for data interpretation. “For example, AI can help to do the automated peak detection integration, calculate and generate the concentration data, and even integrate with the downstream report generation process,” he states, adding that AI/ML can generate study protocols and validation reports.

“Ultimately, biotech and pharma companies will be using AI to write the bioanalytical sections of filings,” Arnold assures. “Health authorities are also looking at AI as a tool to analyze data in submissions and during on-site or remote inspections of bioanalytical labs. The resulting analyses will highlight both compliance with regulations and detect problems and errors that need further review.”

Arnold cautions that controls must be put in place for AI use in regulated bioanalysis to prevent “hallucinations,” or the creation of data that have not been provided. “Not knowing if the report content is correct is not something bioanalytical labs can risk and would result in utilizing the same or greater QC [quality control] and QA [quality assurance] resources to make sure of the report’s accuracy,” he says. “Which is contrary to what AI is supposed to do: reduce human involvement and improve accuracy.” Audit trails that demonstrate how/if the data reached the correct location can help prevent this problem, Arnold suggests.

Bioanalytical workflows can be enhanced, and regulatory compliance can be supported, by incorporating large language models (LLMs) in bioanalysis, says McDougall. LLMs can automate documentation and interpret complex datasets.

“They streamline reporting by generating method summaries, validation protocols, and audit-ready records from structured and unstructured data,” McDougall says. “LLMs also assist in protocol optimization by analyzing historical assay performance and suggesting refinements. Integrated with lab systems, they enable natural language querying of experimental data, accelerating decision making. Additionally, LLMs support training and knowledge transfer by summarizing scientific literature and [standard operating procedures (SOPs)], improving consistency and reducing onboarding time. Their ability to contextualize and communicate scientific insights makes them valuable tools in modern bioanalysis.”

Yuan believes that LLMs can improve regulatory compliance for QC and QA practices by helping review ELNs and audit trails and create SOPs. “[LLMs] can be more interactive with humans. It’s more like a real human. So, [a person] can… get guidance and assistance on troubleshooting and regular bioanalytical workflows,” Yuan suggests.

“LLMs can help create a conversational digital twin by providing a natural-language interface to bio-analytical workflows, enabling scientists to converse with their assay datasets,” says Panchali Roychoudhury, senior director, TCG Digital (a part of LabVantage’s parent company, The Chatterjee Group). LabVantage analytics uses a hybrid approach to integrate semantic knowledge graphs with retrieval-augmented generation (RAG)-based GenAI models to create “a more accurate, comprehensive, contextually aware, and scalable system, with substantial reduction in hallucinations,” according to Roychoudhury. “Responses are also traceable and linked to related experiments or compounds. This combination reduces noise, expedites decision making, and strengthens confidence in the bioanalytical workflows.”

De Boer notes that “LLMs could play a role in automating data interpretation, supporting audit readiness, and improving knowledge management across projects. While their adoption in bioanalysis remains limited today, ongoing advancements suggest LLMs will increasingly help accelerate research, reduce administrative burdens, and enhance the integration of complex scientific data.”

Key to developing safe and effective drug products is the ability to perform precise quantification and structural characterization of complex molecules, according to McDougall, which can be achieved with high-resolution mass spectrometry (HRMS). He also points to the use of automated liquid handling systems and microfluidics to enhance sample processing. “These instruments are unique for their integration with digital systems, compatibility with AI-driven analytics, and ability to handle biologics and multiplexed assays—making them essential for accelerating development timelines and ensuring robust, regulatory-compliant bioanalytical data,” McDougall says.

According to Yuan, HRMS provides exceptional specificity. “For example, for oligonucleotides, sometimes the commonly used quadrupole mass spectrometer may not be able to differentiate the parent and metabolites. HRMS had a unique advantage to provide additional specificity for these types of applications,” Yuan says. He adds that HRMS can also obtain quantitative and qualitative information in the same run, allowing for data mining at a later date that can avoid sample reassessment.

Liquid chromatography (LC)–mass spectrometry (MS), automated enzyme-linked immunosorbent assay) (ELISA) (ELLA), Meso Scale Discovery (MSD), ddPCR, and flow cytometry may be used for pharmacokinetics and/or biomarker analysis, according to de Boer. Robust, reproducible results for complex modalities may be achieved by combining hybrid immunoaffinity LC–MS techniques with automated sample preparation, de Boer explains.

Although LC–MS can be more expensive and not as sensitive, Lowes sees it as a alternative to ligand binding assays (LBAs). “The benefits of mitigating the challenges and the reliance on critical reagents associated with immunoassays leads to some interesting cost and efficiency assessments of LC–MS versus LBA,” he explains. “With increasing frequency, we’re seeing LC–MS to be the best tool for the job, including in biologic bioanalysis for protein-based biologic drugs and ligand oligonucleotide therapeutics. Integral to both are also the conjugated drug structures in many oncology applications for which an antibody is directed to a tumor-associated antigen to deliver a cytotoxic payload. These are inherently complex molecules, and subsequently, so are the assays that we develop to address them. That is, we often need a free payload drug assay, a conjugated payload, and then the total antibody measurement as well.”

According to Arnold, “the improved flow cytometry instruments, with some instruments having 30 and 40 color detection ability, are being used for cellular characterization, and the improved sensitivities allow them to be applied to circulating exosome biomarkers that can detect and characterize many diseases. Mass cytometry is being used in research, and once suitable applications are identified, likely in measuring biomarkers, it will move into regulated bioanalysis.”

How do changing regulations impact bioanalysis advancements?

Development and implementation of new technologies is directly impacted by bio/pharmaceutical development and manufacturing regulations. Good manufacturing practices are often necessary to follow, and each country and/or region has its own rules. The International Council for Harmonisation (ICH) works to harmonize these regulations to help both manufacturers and regulators navigate a complex system. Bioanalysis, in particular, says McDougall, is guided by ICH M10 (2).

“Having a standard—[M10]—accepted in many major markets eliminated the need for bioanalysts to keep up with changes in each of the individual countries and adjust their practices, especially when there were conflicting expectations,” Arnold explains.

De Boer adds that “in 2025, FDA released updated guidance on bioanalytical method validation for biomarkers, emphasizing a fit-for-purpose approach that ensures sensitivity, specificity, and reproducibility (3). These evolving standards are accelerating adoption of standardized SOPs, automation, advanced assay platforms, and comprehensive documentation, enabling compliant, reproducible, and high-quality bioanalysis that supports precision medicine and faster regulatory approvals.”

FDA’s initiation of new approach methodologies to reduce animal testing in pharmaceuticals has also impacted bioanalysis (4). “The FDA announcement has certainly catalyzed ongoing efforts and support for in silico, in vitro, and ex vitro methods,” Lowes says. “The evolving developments around the organoids and organ-on-a-chip platform technologies I find particularly exciting. How bioanalysis fits into this space is still to be determined, but I suspect it will have a lasting influence on how we conduct pre-clinical stages of drug development.”

“Regulatory updates from agencies like FDA and [European Medicines Agency] demand greater precision, sensitivity, and documentation, driving innovation in instrumentation and data management,” McDougall says. “Compliance with good laboratory and clinical practices ensures data integrity and ethical standards. As expectations evolve, bioanalytical teams must adapt workflows and technologies to meet stricter requirements, ultimately improving the reliability and regulatory acceptance of drug development data.”

McDougall highlights a paper authored by members of the European Bioanalysis Forum (EBF) AI team, published in July 2025, that questions the adequacy of traditional validation frameworks for AI in bioanalysis (5). The paper “argues that static, checklist-driven validation is ill-suited for adaptive, learning systems,”
McDougall summarizes. “Instead, they propose ‘adaptive qualification,’ a dynamic, context-sensitive approach grounded in scientific oversight and continuous engagement. AI is reframed not as a tool but as a trainee, requiring lifecycle monitoring rather than one-time validation. EBF emphasizes the need for scientific stewardship, collaborative ownership across disciplines, trust built on transparency and relevance, and a shift from rigid compliance to thoughtful, fit-for-purpose oversight.”

“To be successful in delivering data accepted globally, bioanalysts must be aware of a variety of regulations and how they apply to the purpose of the analysis, type of samples they are analyzing, and the technologies,” Arnold warns. “When no regulations exist, scientists must continue to collaborate on science-based best practices.”

Bioanalytical Method Validation, Guidance for Industry

M10 Bioanalytical Method Validation and Study Sample Analysis

Bioanalytical Method Validation for Biomarkers, Guidance for Industry

FDA. FDA Announces Plan to Phase Out Animal Testing Requirement for Monoclonal Antibodies and Other Drugs. Press Release. April 10, 2025.

Timmerman, P.; et al. Why Traditional Validation May Fall Short for Artificial Intelligence in Bioanalysis: A Perspective from the European Bioanalysis Forum. 

 17(13), 835–837. DOI: 10.1080/17576180.2025.2535219

®
Vol. 49, No. 8
October 2025
Pages: 10–13

When referring to this article, please cite it as Haigney, S. Enhancing Bioanalysis and Drug Development with Advanced Tools and Technologies. 

Advancements in bioanalysis are crucial for drug development, focusing on pharmacology, bioavailability, and bioequivalence. AI and ML are increasingly used to handle complex data, improve accuracy, and streamline processes. These technologies enhance assay precision, decision-making, and regulatory compliance. LLMs further optimize workflows by automating documentation and data interpretation. New instrumentation like HRMS and LC-MS are pivotal for precise quantification and structural characterization. Regulatory changes, such as ICH M10 and FDA guidelines, drive innovation and necessitate adaptation in bioanalytical practices to ensure compliance and data integrity.

Cover Story

Artificial intelligence, machine learning, and novel instruments are helping drug developers evaluate complex data from bioanalytical studies.

Enhancing Bioanalysis and Drug Development with Advanced Tools and Technologies 

Q. My company just launched a novel, small-molecule drug into the market. For how long do we need to gather real-world evidence to monitor its safety and efficacy?

A. Real-world evidence is part of pharmacovigilance activities, and these are obviously mandatory for all marketed medicines. Real-world evidence collation wasn’t really required in the past but it is now, and is likely to stay given the benefits that real-world evidence delivers.

Real-world evidence for the product must be gathered for as long as you keep the product on the market. The difficulty with real-world evidence is that it may be coming from patients, who are lay persons, who interpret the effect that the product has on them. Someone may find that an effect is horrible for them; however, a different person may interpret it as slight annoyance. This is difficult for the pharmaceutical expert to really understand if the side effect is really that bad, or was it not? There is a lot of room for interpretation of what you get.

Q. My company is investigating options for a novel cell therapy for the oncology market. How can we leverage real-world evidence from our own research and from post-market drugs already on the market to choose which direction to go in?

A. If a drug developer wants to find out what is the best indication for a new or existing drug product that could be applied to adults only or to adults and adult adolescents, and so on, I think there are two aspects where real-world evidence can be beneficial.

One is to gain regulatory approval to conduct clinical trials, and hopefully at a later stage, to also get marketing approval. There is also a need to have this evidence when it comes to reimbursement at a later stage, because in the end, you don’t just want to have an approved product, you also want to be able to sell it at an appropriate price. Health technology assessments will have to be performed, and hopefully you will get a positive outcome.

Cell therapy products, by their very nature, are typically quite difficult and complex to produce and manufacture, and therefore they are often also very expensive. To get the approval to even run the trial, you have to have good reasons and good evidence. And then, of course, you may have a limited patient group market. In the end, you may have a very expensive product, and the more evidence you can provide to support your claims and the very need for the product, the better it is.

Siegfried Schmitt, PhD, is vice-president, Technical at Parexel.

When referring to this article, please cite it as Schmitt, S. Real-World Evidence as Part of Pharmacovigilance. 

Real-world evidence (RWE) is essential for pharmacovigilance and must be collected for the entire market life of a drug. It provides insights into safety and efficacy from patient experiences, though interpretation can be challenging. For novel cell therapies, RWE aids in regulatory approval, clinical trial design, and reimbursement strategies. It supports health technology assessments and justifies the high costs associated with complex cell therapy products. Comprehensive RWE strengthens claims and demonstrates the necessity of the product, facilitating market access and pricing strategies.

Ask the Expert

Siegfried Schmitt, PhD, vice-president, Technical at Parexel, answers questions on the use and benefits of real-world evidence for small-molecule and large-molecule drug development.


Real-World Evidence as Part of Pharmacovigilance

®’s “Ask the Expert” column explored the implications of FDA’s January 2025 draft guidance 

Considerations for Complying With 21 CFR 211.110

 and how it impacts the approach that has been taken, up to now, in complying with 21 

“Allowing flexibility to use new technological advances to determine drug product integrity and batch uniformity during manufacturing could reduce product reject rates because in-process results could be reported sooner (perhaps in real time) and manufacturing corrections could be made to ensure the batch integrity,” column authors Susan J. Schniepp and Rona LeBlanc-Rivera, PhD, wrote at the time (1). “Drug products manufactured using advanced in-process techniques may have non-traditional quality considerations to determine the suitability of the product, and this guidance allows for flexibility and latitude on how to implement these controls effectively.”

Between the issuing of the draft guidance and the closing of the comment period on April 7, 2025, which came after our “Ask the Expert” feature was published, numerous industry stakeholders weighed in on their perceived effectiveness of the guidance in its final version, suggesting certain changes to verbiage or focus. Here, we present a summation of those comments, as an update on the progress of this document.

Two comments were directed to FDA by experts on their own behalf. Rodolfo Romanach, professor of Chemistry at the University of Puerto Rico at Mayagüez, said he was happy with FDA’s efforts to define the term “advanced manufacturing,” but he was concerned with the proposed definition as worded in this and other documents (4,5).

“I believe that the current definition is confusing pharmaceutical manufacturing research with advanced manufacturing,” Romanach wrote, contending that the definition restricts advanced manufacturing only to that which “has the potential to improve the reliability and robustness of the manufacturing process and supply chain and increase timely access to quality medicines for the American public” (4).

He added that “many companies in the [United States] export products (4). Medicines could contribute to good relations between countries. The timely access should be to all patients, not just the American public.”

Jitendra Pund, Quality Management System head at Shilpa Medicare, suggested alterations to the section “General Considerations for In-Process Sampling and Testing,” including ensuring that in-process parameters are selected based on drug product quality attributes and that critical process parameters be controlled and monitored in process validation studies (6,7).

“When continuous process verification is used, the continuous manufacturing system performance and output material quality [should be] continuously monitored, such that the real-time data collected demonstrate the maintenance of a state of control and production of output material with the desired quality for the runtime duration,” Pund wrote in the comment to FDA (6).

While Pund submitted an individual comment, Bristol Myers Squibb, as a whole, submitted a response to FDA that sought to address the risk of using in-process control strategies that rely solely on process models (8).

“We think that there could be a challenge in demonstrating continued model validity via process monitoring or testing via alternate tool (PAT) during routine manufacture,” reads the Bristol Myers Squibb comment (8). “It will be beneficial to acknowledge in the guidance that this can be alternatively achieved through a model maintenance [or] model assumption verification plan [or] protocol.”

To achieve this, Bristol Myers Squibb suggested the use of a tablet PAT method to support model maintenance and the continued validity of assumptions (8).

The asset optimization software company Aspen Technology also submitted a comment touching upon several topics in the draft guidance, including statistical validation in multivariate models, mitigating unplanned disturbances, and the role of engineering principles (9).

“When used effectively, statistical metrics help ensure that the process model’s underlying assumptions remain valid during routine manufacturing,” Nithiya Parameswaran, Aspen Technology vice-president of Product Management, wrote in the company’s statement (9). “They also provide a means for detecting when an assumption is no longer valid, triggering the need for physical retesting as confirmation. Simulated data derived from rigorous models... ensures that multivariate process models maintain control over manufacturing operations using the aforementioned statistical methods.”

Professional organizations share their views

The International Society for Pharmaceutical Engineering (ISPE) raised concerns about a 17-line passage within the document which, as they wrote, “categorically excludes the use of process models solely for compliance with the requirements of 21 CFR 211.110” (10).

“It is curious why this interpretation is issued at this point [versus] continuing to rely on the scientific and risk-based approach outlined, for example, in lines 86, 121, and especially line 130 about the spirit of flexibility that regulations provide,” Carol Winfield, senior director, Regulatory Operations, wrote on behalf of ISPE (10). “The somewhat categorical exclusion of the use of process models as the main or only means of control could have a significantly detrimental effect on the development of advanced manufacturing control strategies for process models in general.”

While the Parenteral Drug Association (PDA) provided proposed changes to the background information provided in the draft guidance, as well as additional suggested references, like Pund, the group devoted much of its attention to the “General Considerations for In-Process Sampling and Testing” section (11). Similarly to Romanach, PDA raised concerns related to the term “advanced manufacturing,” saying that a sentence “highlighting ‘innovative methods’ is not consistent with the recently issued guidance on advanced manufacturing technologies ... [and] moves away from the idea that innovative methods are superior” (11).

The Biotechnology Innovation Organization (BIO) registered some of the most pointed concerns about considerations for batch uniformity (12).

“Overall, it is not more difficult to perform in-process evaluation in a CM [continuous manufacturing] process; it just requires different approaches,” E’Lissa Flores, director, Science & Regulatory Affairs, commented on behalf of BIO (12). “The draft guidance as written implies that you have more opportunity to evaluate the quality of a batch product.”

Comments submitted by the Pharmaceutical Research and Manufacturers of America (PhRMA) drilled down on manufacturing controls and the use of process models but also provided recommendations on the scope of the guidance in general (13).

“PhRMA would appreciate more discussion of the balancing necessary to increase the use [of] innovative approaches and the need to meet regulatory expectations,” the group wrote (13). “Specifically, we would appreciate additional guidance on how a sponsor can ensure a suitable redundancy in testing without an excessive burden on the control strategy.”

In a best-case scenario, as Schniepp and LeBlanc-Rivera wrote in April, expansion of the guidance that clarifies the flexibility that sponsors may be afforded will be a welcome step in the finalized document. “This new guidance offers flexibility to manufacturers, including contract manufacturing organizations, to introduce new, more efficient methodologies to measure batch uniformity and product integrity required to determine product suitability to ensure continued access to necessary pharmaceutical products,” they wrote (1). “It is up to [the] pharmaceutical industry to work with FDA and determine the best implementation strategy for realizing the benefits of modernizing manufacturing processes.”

Draft Guidance for Industry, Considerations for Complying With 21 CFR 211.110

 Title 21, 211.110 (Government Printing Office, Washington, DC) 169–170.
3. Schniepp, S. J. and LeBlanc-Rivera, R. Drug Shortages and Complying with FDA’s 21 CFR 211.110 Guidance. 

, 49 (3) 34.
4. FDA. Comment from Rodolfo Romanach. 

, Jan. 22, 2025 (accessed Sept. 23, 2025).
5. LinkedIn. Rodolfo Romanach. 

, accessed Sept. 23, 2025.
6. FDA. Comment from Jitendra Pund. 

, Apr. 8, 2025 (accessed Sept. 23, 2025).
7. LinkedIn. Jitendra Pund. 

, accessed Sept. 23, 2025.
8. FDA. Comment from Bristol Myers Squibb. 

, March 18, 2025 (accessed Sept. 23, 2025).
9. FDA. Comment from Aspen Technology. 

, April 7, 2025 (accessed Sept. 23, 2025).
10. FDA. Comment from International Society for Pharmaceutical Engineering (ISPE). 

, March 24, 2025 (accessed Sept. 23, 2025).
11. FDA. Comment from Parenteral Drug Association (PDA). 

, April 2, 2025 (accessed Sept. 23, 2025).
12. FDA. Comment from Biotechnology Innovation Organization (BIO). 

, April 4, 2025 (accessed Sept. 23, 2025).
13. FDA. Comment from Pharmaceutical Research and Manufacturers of America (PhRMA). 

, April 4, 2025 (accessed Sept. 23, 2025).

The FDA's January 2025 draft guidance on 21 CFR 211.110 has sparked diverse feedback from industry stakeholders, focusing on advanced manufacturing and in-process controls. Concerns include the definition of advanced manufacturing, the exclusion of process models for compliance, and the need for flexibility in implementing new technologies. Stakeholders like Bristol Myers Squibb and Aspen Technology emphasize model validation and statistical metrics. Professional organizations, including ISPE and PDA, highlight the importance of innovative methods and batch uniformity. The guidance aims to modernize manufacturing processes while ensuring product integrity and quality.

Quality/Regulations

The FDA draft guidance "Considerations for Complying with 21 CFR 211.110" raises points to consider regarding drug products made using advanced manufacturing, batch uniformity, drug product integrity, and how manufacturers can incorporate process models into control strategies.

Stakeholders React to FDA's Push for Quality, Integrity, and Uniformity

Digital technologies are evolving rapidly and offer benefits for bio/pharmaceutical manufacturers and developers who seek digital transformation. New artificial intelligence (AI) and machine learning (ML) tools help manage, transform, and analyze data—including data coming from process analytical technology (PAT) tools—so it can be better used to develop process insights or for process control. Early adopters of these technologies have been process development laboratories, but implementation in commercial manufacturing is emerging.

“Historically, PAT has been used to define processes and build control strategies, but its integration into manufacturing has often been limited,” says Michalle Adkins, director of life sciences strategy at Emerson. She points to the emergence of inline spectral PAT, which makes it easier to embed spectral waveform data directly into control systems, as a significant advance.

“We’re also starting to see OEMs [original equipment manufacturers] including PAT in their offerings with their bioreactors, which signals a broader readiness to integrate advanced analytics not only in process development but also in commercial manufacturing,” Adkins says. “These initial steps are building the foundations to lead to these technologies rolling out to bigger scale solutions in the future.”

Currently, AI-based tools are often used to create process insights that people on operations teams can use to improve processes, but eventually AI tools will allow more automated control, suggests Ryan Thompson, senior specialist, Industry 4.0 at CRB.

“Huge strides are being made in bioreactor productivity (e.g., higher titers) and chromatography efficiency using these tools,” Thompson says. “The next big step will be [to take] the person out of the middle and have truly AI-based closed-loop control.”

The convergence of AI and ML with advanced process control is helping to address some particularly difficult measurement and control areas, adds Adkins.

“Additionally, regulatory bodies are increasingly open to and encouraging the use of these more automated approaches, particularly as companies are demonstrating success in both process development and targeted projects,” Adkins says. “With early successes demonstrated in bulk drug substance processes such as fermentation and purification, along with pilot-scale manufacturing, broader adoption is expected across the manufacturing lifecycle.”

Challenges and solutions for commercial manufacturing

Moving from process development to commercial manufacturing requires technology to be robust and designed for use by operators on the factory floor.

“If new technologies add uncertainty or additional steps, they’re unlikely to be widely adopted,” notes Adkins. “Historically, that has been a challenge, but we’re starting to get over those complications as the technologies themselves mature. Today’s solutions are far more robust and designed for high-performance environments.”

Technology transfer is another concern. “Teams need to be sure that when they change the scale, it won’t change the outcomes,” Adkins says. “They need collaborative software solutions designed specifically to help navigate scale-up and scale-out.”

Data handling is also a crucial issue. “Process data on the factory floor is generally unstructured and lacks context,” says Thompson. “It takes a lot of time and effort to search, organize, and clean data before [they are] useful for AI/ML algorithms.”

Large quantities of data are needed, however, because AI/ML accuracy and insights improve with the quantity of data.

“Data modeling is a necessary first step to meaningful AI/ML, where process data and modern architectures, such as a unified namespace, should be considered,” Thompson suggests.

“One of the most critical—and often underestimated—elements of effective data analytics in pharma manufacturing is context,” agrees Adkins. “Teams must be empowered to generate and preserve context as data flows across the enterprise. Whenever context needs to be reconstructed, the value of that data diminishes. It becomes harder to use, more expensive to manage, and slower for driving meaningful action. That’s not just a technical challenge—it’s a strategic one.”

Emerson’s new digital platform will enable integration of disparate industrial automation technologies, according to Adkins. A centralized data model will allow teams to collect full context and share the master data across solutions without needing redundant definitions in separate systems (e.g., the distributed control system, the manufacturing execution system, and others), Adkins says.

“The enterprise operations platform will eliminate that fragmentation by creating a common data platform that distributes consistent equipment information across all solutions via a shared data fabric,” Adkins says. “This not only simplifies data management but also strengthens the foundation for more autonomous and scalable operations.”

Using the approach of managing data as a product, the platform is designed to make data findable, accessible, interoperable, and reusable (i.e., the FAIR principles).

“This approach enables scalable, AI-ready data environments that support both operational efficiency and innovation,” says Adkins.

Digital technologies, particularly AI and ML, are transforming bio/pharmaceutical manufacturing by enhancing data management and process control. Early adoption has been seen in process development, with commercial manufacturing beginning to integrate these technologies. AI tools are improving bioreactor productivity and chromatography efficiency, with potential for fully automated control. Challenges include technology robustness, data handling, and technology transfer. Digital twins and PAT integration offer benefits in process control and efficiency. The pharmaceutical industry is gradually embracing AI, with significant potential for innovation and efficiency gains.

Advances in digital technologies offer effective data handling for bio/pharma manufacturing.

Pharmaceutical Technology October 2025

Advances in Digitalization and Automation of Aseptic Processing

Predictive Modeling for Small-Molecule Formulation Development Using Advanced Algorithms

Enhancing Bioanalysis and Drug Development with Advanced Tools and Technologies

Real-World Evidence as Part of Pharmacovigilance

Stakeholders React to FDA's Push for Quality, Integrity, and Uniformity

Optimizing Data Use for Digital Transformation