Claudia Zylberberg, Chair of the Board and Co-Founder, Kosten Digital
The pharmaceutical industry is increasingly leveraging IT innovations from sectors like artificial intelligence (AI), cloud computing, and automation to optimize drug development and production. AI and machine learning models are accelerating the discovery phase by predicting molecule interactions and outcomes in silico, significantly reducing time and cost. Cloud-based platforms improve collaboration across global teams and streamline data management, while automation in manufacturing enhances precision and scalability. Digital clinical site management creates opportunities for close-look and on-time monitoring. Our company supports life sciences by implementing custom-built digital solutions that integrate these innovations, facilitating smoother processes from clinical trials to production while ensuring compliance and quality control.
Søren Flygenring Basset, Executive Director, R&D, Lonza
The pharmaceutical industry is witnessing a transformative shift driven by the surge of innovations in new technologies, including Ai and machine learning (ML).
AI/ML has made a tremendous impact in revolutionizing drug development by accelerating target identification and the creation of novel biomolecules. These advancements enable researchers to efficiently analyze vast data sets through sophisticated algorithms to pinpoint biological targets and binding partners. These new technologies are also helping the pharmaceutical industry develop new-to-nature (novel) biomolecules to bind to such targets.
Building upon the advancements in target identification and biomolecule creation, AI/ML is also revolutionizing the later stages of drug development. In biologics process development, AI/ML is used to build computational models predicting the best cell line DNA to express a given peptide/protein and the optimal process conditions. At Lonza, process analytical technologies (PAT), including spectroscopic methods like Raman, are used in combination with ML technology to monitor and control critical process parameters, such as nutrient levels during fermentation, without taking manual samples, providing valuable process knowledge and stringent process control.
As digital transformation reshapes and accelerates innovation in the pharmaceutical industry, we can also expect enterprise-level AI implementations and digital co-pilots to enhance research and operational efficiencies in drug development and manufacture.
Stella Vnook, Chief Executive Officer, Likarda
The pharmaceutical industry is benefiting immensely from cross-sector innovations, which are leading to more efficient drug development pipelines, personalized treatments, and sustainable production.
One prominent area of integration is the use of AI and ML from the tech sector. These technologies are now being applied to drug discovery, allowing for faster identification of viable drug candidates. AI algorithms can rapidly sift through vast data sets to identify patterns and predict which compounds may be most effective, reducing the time and cost associated with traditional methods.
Another is materials science, particularly in the development of new drug delivery systems. Nanotechnology, for instance, has enabled the creation of hydrogel-based delivery mechanisms that can target specific tissues or cells more precisely, improving therapeutic outcomes and minimizing side effects. This cross-industry collaboration is revolutionizing treatments for diseases like cancer and autoimmune disorders, where targeted therapies can drastically enhance patient outcomes.
Sustainable manufacturing practices from industries focused on green technologies are helping pharmaceutical companies reduce their environmental footprint. Biomanufacturing processes that use renewable resources and generate less waste are being adopted to create more sustainable production lines, which aligns with the growing demand for eco-friendly solutions across industries.
Forrest Brown, President, GermFree
Within the pharmaceutical industry, there is a huge push for Pharma 4.0, which is more or less an extension of the Industry 4.0 operating model that accounts for the increased regulatory scrutiny we now face. Pharma 4.0 centers around digitization of processes and connecting systems to create a "smart" plant that allows heightened control over process and quality. This will ultimately allow lifesaving products to be produced at a lower cost with higher levels of product quality and patient safety. There are undoubtedly challenges associated with implementing this plan, especially in a regulated environment. I am confident that we will see the fruits of Pharma 4.0 first in the drug development process and later in clinical and commercial production as we progress through the change curve.
One of the latest global trends is AI, and I think everyone is starting to realize the power that it holds. As we push toward Pharma 4.0, the amount of data gathered will increase drastically, and all these data will hold valuable information that can be unlocked through AI. I'm excited to see it help the industry develop and manufacture better drugs at a lower cost to improve the human condition worldwide.
Josh Ludwig, Global Director Commercial Operations, ScaleReady
For too long, the cell and gene therapy space has been stymied by an inefficient manufacturing paradigm inherited from other pharma modalities, such as antibody production. Processes sourced from academia are developed for quick clinical assessment, a primary academic objective. This results in little to no planning for manufacturing at commercial scale. This often results in therapies with potential curative effects that very few patients can access or afford.
Recent efforts focused on automation have thus far resulted in little improvement to scalability. Other industries have recognized that investing in complex machinery without the right methodology can leave drug developers with a “monument” — a piece of technology that becomes obsolete without ever proving useful, ultimately just taking up expensive space.
As a field, we’re starting to learn from industries like fast food and automobile production, which look less like the batch manufacturing used for scaling antibodies and more like CAR-T cell therapies. Lean manufacturing — a methodology famously developed by Toyota — focuses first on simplifying processes to maximize efficiency. Automation becomes most impactful only after a process is optimized and greatly simplified, allowing for lean manufacturing techniques like continuous improvement and assembly line production — and avoiding the collection of monuments.
Nathan Givoni, Chief Executive Officer and Co-Founder, Gelteq
Innovation in the pharmaceutical sector leverages clinician-led fields but also shares formulation strategies across fields as diverse as food, nutrition, and materials science. Oral therapies will continue to overlap with nutrition and the health of the microbiome with therapeutics that both manipulate the gut and also use dietary additives to change bioavailability.
Manufacturing technologies are looking for advancements just as much as for the environmental footprint, while in the intellectual property space, smarter chemistry will likely be seen in unique delivery systems, prodrugs, and semisynthetic substances that give targeted activity with less invasive administration. We can see food production in recent years also contributing to drug development.
Looking further ahead, we expect a greater emphasis will be placed on the impacts across the whole body and the environment for active substances, where systems built using natural ingredients will have a demonstrable advantage.
Life sciences in general is leveraging emerging technology and we believe that the pharmaceutical industry will also begin to more actively rely on areas such as AI, machine learning, and robotics to advance drug development and production more efficiently.
Timmy Besel, Senior Manufacturing Engineer, Lifecore Biomedical
With deep experience in highly viscous parenteral formulations in excess of 100,000 centipoise, one of the tools that we see helping address an industry challenge relates to new technology for automated visual inspection (AVI). For less viscous products, the typical approach is for a system to spin each individual product and then use a camera system and software to monitor the flow and identify contaminants. However, this approach does not work with viscous formulations because they do not flow in the same manner. Now, like tools in other industries, some AVI systems incorporate ML and sophisticated virtual image processing to identify particulates and defects that would normally have to be identified by a human visual inspector. Also, over time, some systems can utilize a database of classified rejected products to train ML algorithms to improve performance and reduce false reject rates. As a CDMO that manufactures a variety of viscous formulations, we’re pleased to see advances in AVI that target this specific issue.
Svea Cheeseman, Ph.D., Director, Product Marketing Management, Refeyn
Personalized therapies have the potential to revolutionize healthcare by tailoring treatments to an individual's unique genetic and molecular profile. The success of these therapies requires scientific advances but also a completely different system of production, as each treatment is unique and often only a small batch is needed. It calls for localized biomanufacturing near the patient — as opposed to the large-scale, centralized production used for traditional medicines. The analytics needs for this type of medicine are also radically different. Personalized medicines, such as cell or gene therapies and cancer vaccines, require complex viral vectors that deliver tailored genetic material directly into patients' cells. Consequently, there is a need for innovative analytical methods that are suited to the vectors’ complexity and can handle tiny sample volumes.
One such technology is mass photometry, rapidly adopted by the AAV field due to its fast and accurate virus characterization on very little sample. These emerging analytics must ensure the purity and potency of these treatments — accurately quantifying viral particles and identifying impurities. They must also assess the integrity of the genomic cargo, determine the stability of the treatments over time, and monitor consistency and reproducibility across batches. Failing to adequately address these challenges will significantly impact the cost and accessibility of personalized medicines. In summary, we envision several promising innovations in analytics and data processing to emerge over the next 10 years that address the challenges above and their adoption will be a crucial step towards realizing the promise of widely accessible and affordable personalized treatments.
Jordi Robinson, Chief Commercial Officer, Navin Molecular
The pharmaceutical industry has been traditionally slow to adopt new technologies; however, it has historically looked to other manufacturing industry sectors for innovations to improve its processes where cost and production efficiencies can be maximized.
Continuous processing is an obvious area where pharmaceuticals has lagged behind other industries, whereas it is commonly employed in automotive, processed food, and electronics manufacturing. This is beginning to change with the growth of newer modalities such as oligonucleotides, peptides, and biologics, where processing lends itself to automation, but in traditional, small molecule drug manufacturing, the vast majority of processes are carried out in batches, where the chemistry is fully understood and robust, and the use of automation is limited. Automation has increased as technologies allow, and provides the ability to capture more data within manufacturing campaigns, but is a long way from being classed as seamless and fully continuous.
Another area of increased interest for pharmaceuticals is in the leverage of sustainable manufacturing techniques. The last decade has seen a marked increase of the use of enzymes in place of traditional metal catalysts, which offer potentially cleaner and higher yielding reactions, but involve greater amount of process development work to optimize their efficiency.
Heejeong Kim, Vice President and Head of Drug Substance Manufacturing, Samsung Biologics
The regulation-laden pharmaceutical industry is reasonably conservative when adopting innovative technologies from adjacent industries. A slight divergence from the Current Good Manufacturing Practices (CGMP) rules that apply to the entire drug development/manufacturing cycle, ranging from developing cell lines to packaging filled/finished vials, could cost a patient’s life. Any new technologies must be seamlessly woven into the CGMP standards, followed by a standard operating procedure (SOP). On the flip side, the success of drug development/manufacturing projects, among other equivalences, hinges on speed. Arguably a water-and-oil relationship, the industry’s want for business sustainability often comes at odds with the industry’s commitment to patients.
To overcome the odds and bring life-essential drugs to patients in time without compromising quality, the pharma industry wants to catch up with other industries, such as auto and energy, regarding technology innovation. The underlying challenge in catching up is that pharma companies with rich scientific expertise often lack the resources, capacity, and/or operational know-how to bring new technologies into their labs and facilities. And this is where resource-rich contract development and manufacturing organizations (CDMOs) can step in and help.
An ultimate goal of integrating innovations is process optimization, which when done right leads to operational excellence. Pharma 4.0 and even now 5.0 technologies have been talked about among the industry players in recent years. At the crest of Pharma 4- or 5.0 is digital twin (DT) technology empowered by AI and ML tools.
Samsung Biologics, a global CDMO, defines the output of DT integration enables CDMOs to offer agile solutions, as well as flexibility to pharma companies. At large, bioprocessing built on DT is broken down into three critical elements:
Real-time monitoring and multivariate analysis of bioreactor
Recent advances in PAT for real-time monitoring, such as Raman spectroscopy, enable CDMOs to measure multiple process attributes and parameters in the bioreactor without sampling, leading to a comprehensive collection of cell culture data and multivariate monitoring.
Product and process data management
The historical and real-time data collected in production are processed and stored in a process data management system (PDMS). The PDMS system connects the physical and virtual parts of the bioprocess DT, enables data sharing with clients in real time, and thus facilitates transparency between CDMOs and their clients.
Mechanistic/statistical models for bioprocesses
First principle–based approaches with cell growth kinetics and fluid dynamics theoretically solve the cellular growth rate and hydrodynamic quantities at specific time points. In more detail, data-driven approaches, such as ML, help scientists identify hidden parameters, which could be vital assets for comprehensive and precise product analysis and ultimately operational excellence. The hybrid model combines first principle–based and data-driven approaches and is a powerful tool. It replicates the physical processes of cells and bioreactors, allowing for the mechanical simulation of their operational conditions and dynamic behaviors at reactor scales.
At the heart of DT integration is, again, process optimization that leads to consistent product quality and also timely regulatory approval. We at Samsung Biologics have proactively improved and leveraged DT in our operations. As a tangible output of our innovative efforts, we have so far earned more than 335 regulatory product approvals, and we believe the approval number will continue to increase as our proactive efforts to adopt newer technologies continue.
Aki Ko, Chief Executive Officer, Co-Founder, and Board Chairman, Elixirgen Therapeutics
Currently, autologous cell therapy manufacturing involves a complex, multi-step process often confined to specific, centralized manufacturing facilities, which can be far from the patients these therapies are meant to treat. This geographic separation leads to lengthy wait times and limiting access for patients in urgent need.
The technology industry has faced similar challenges with data storage and providing digital services and has continued to address them through decentralization.
A decentralized approach in cell therapy manufacturing can alleviate key bottlenecks in the supply chain, offering several significant advantages. This model can reduce dependency on centralized supply chains, potentially lowering costs for these therapies as well. By investing in technologies that enable more localized manufacturing, we can significantly enhance the efficacy of cell therapies and improve the overall patient experience.
Simon Brunner, Ph.D., Head of Platform, Quotient Therapeutics
Innovations from genome sequencing, big data, and artificial AI industries are revolutionizing how we discover novel drug targets.
Drug development is firmly in its genetics era: drugs supported by genetics are 2.5 more likely to succeed in the clinic, with further innovations anticipated to accelerate this trend. Recent breakthrough research revealed that each of our cells is genetically distinct due to somatic mutations acquired throughout life. This collective somatic genome can be studied using cutting-edge technologies with the sensitivity and accuracy to reveal cell-to-cell genetic variation. Understanding and identifying beneficial and disease-causing mutations can uncover a rich new set of druggable targets for the development of medicines.
Determining the functional impact of somatic mutations within human disease tissue de-risks target insights at discovery. By leveraging AI tools to link genotype to histology data and cellular function, we can identify within-tissue variations that aren’t visible to the human eye. Resulting data sets are large; however, their analysis is significantly accelerated by using scalable cloud infrastructure, capable of storing and processing large amounts of data. The combination of these technologies enables the mapping of every possible mutation in the human genome at a pace unthinkable just a few years ago.
Ludovic Brellier, President Hardware Solutions, Cytiva
Car manufacturers have successfully used in silico design to refine their blueprints — positioning themselves ahead of potential risks even before creating a prototype. The early introduction of simulation saves considerable time and costs by eliminating guesswork. This in turn paves the way for speedier innovation and rise of options because workflows are accelerated, and we can home in on products for swifter development.
Biopharma has similarly embraced in silico modeling — from early development right through the product life cycle — to simplify drug development and mitigate risks associated with trial-and-error methods. In silico modeling drastically cuts experimentation lead times, fueling speedier innovation; it propels workflows to supply therapies faster and fine-tunes products for the highest yield and quality.
To fully benefit, biopharma companies need to: invest in technology by allocating resources to build robust computational infrastructure and acquire top-notch software tools; upskill scientists and researchers in computational modeling techniques to maximize their expertise; integrate approaches by combining in silico strategies with traditional experimental methods to create a seamless workflow; and lastly, perform regular updates to refine models using experimental data to ensure accuracy and precision.
By leveraging in silico design, biopharma can streamline drug development, reduce costs, and, most importantly, deliver high-quality therapies to patients more efficiently.
Sabrina Yang, Ph.D., Cofounder and Chief Innovation Officer, Empress Therapeutics
AI is one of the most profound recent innovations and refers to many specific technologies and applications. One domain that has been transformed is the interpretation and manipulation of data, evidenced by the generative AI explosion with ChatGPT and other large language models.
The biopharmaceutical industry relies heavily on the ability to interpret and manipulate a different type of language (genetic code) to discover new targets and medicines. The industry has a number of examples where companies are leveraging AI to generate better therapeutics.
At Empress Therapeutics, we are applying AI in a unique way, transforming genetic code into therapeutic compounds, novel targets, and disease association. Our ChemilogicsTM platform leverages the central dogma that DNA encodes for RNA that encodes proteins but extends this further to identify starting points for novel small molecule drugs from the chemistry that some proteins (enzymes) catalyze. This new capability allows, for the first time, the ability to find chemical compounds inside the human body that have a genetic basis and can be causally linked to health or disease.
Empress has already discovered 15 drug leads across multiple diseases and target classes in 24 months, highlighting a new use of AI in drug development.
Anil Kane, Ph.D., Executive Director, Global Head of Technical & Scientific Affairs, Thermo Fisher Scientific
Next-generation technologies, such as AI, augmented reality (AR) and virtual reality (VR), are transforming business strategies and operational efficiency across industries, allowing companies to tackle complex challenges with greater speed and precision. In the pharmaceutical industry, AI is being leveraged to overcome longstanding challenges of data silos and supply chain optimization. Through predictive analytics and access to real-time data, AI is enhancing demand forecasting and inventory management, thus improving decision-making, reducing delays, and creating overall efficiencies.
Another critical aspect of the drug development and production process that is leveraging cutting-edge technological solutions is the acquisition and retention of talent.
Thermo Fisher Scientific’s Training Center and Aseptic Academy in Monza, Italy, uses a tech-forward approach to employee training by implementing AR/VR training programs for both hands-on and simulated GMP environments. The outcome is significantly decreased training time and time to competency — up to a 50% reduction — while enhancing knowledge retention through structured learning environments, expert guidance, and practical experience.
These next-generation technologies create greater agility and responsiveness in tackling evolving challenges throughout the pharmaceutical supply chain, and these examples illustrate how the industry is adopting innovations from other sectors to streamline drug development and production processes to get therapies to patients swiftly.
Alan Marcus, Chief Growth Officer, LabVantage Solutions
AI is a prime example. AI has modernized sectors like finance, manufacturing, and healthcare by automating processes, improving data analysis, and enabling predictive modeling. Similarly, in the pharma industry, AI can now analyze vast data sets of molecular and biological information to expedite drug discovery, identifying potential drug candidates more efficiently than traditional methods.
In addition to drug discovery, AI is revolutionizing drug development. AI-powered tools can optimize clinical trial design, predict patient outcomes, and identify adverse events early on. This not only improves the efficiency of clinical trials but also enhances the safety and efficacy of new drugs.
AI-enabled laboratory information management systems (LIMS) and analytics are also transforming laboratory workflow and dataflow management in drug development. Automated data capture and processing reduce manual errors and accelerate research timelines. ML models help search and interpret complex biological data, aiding in decision-making and optimizing experimental designs. This streamlines the laboratory workflow, enhances data integrity, and speeds up the overall development process. In manufacturing, AI-driven systems enhance production processes, predict equipment maintenance needs, and ensure quality control, leading to cost savings and higher product quality.
Perhaps most importantly, AI is driving the development of personalized medicine. By analyzing patient genetic data and other factors, AI can identify the most effective treatments for individual patients.
Maksymilian Karczewski, Ph.D., Project Manager, SciY
The pharmaceutical industry is increasingly investing in technologies driven by AI to accelerate the drug discovery process and reduce the environmental impact of producing APIs.
While computational methods are known to support candidate selection in the drug discovery process, the design of synthetic pathways has traditionally relied on chemists' expertise. Shifting this task to computers represents a major paradigm shift. Despite investments in AI tools for synthesis planning, purely data-driven models still face critical limitations. They rely on biased data sets dominated by positive outcomes and lack negative examples, which hampers their ability to identify incompatible functional groups or improbable structural motifs. This often results in synthetic plans that propose unfeasible reactions or intermediates, limiting their practical use.
The future lies in AI-hybrid systems, which combine AI with expert knowledge. These systems can overcome the limitations of data-driven models by integrating human expertise, allowing them to avoid synthetic pitfalls. Additionally, thanks to the expert insights, AI-hybrid systems can adapt synthetic routes based on the scale of production, designing pathways for both small-scale laboratory work and larger-scale industrial applications.
Beyond retrosynthesis, AI-hybrid systems can generate libraries of synthesizable analogs when provided with a lead compound scaffold, streamlining early-stage drug discovery. These libraries can be refined using neural networks to predict binding affinities and drug-likeness, enhancing the precision of candidate selection and bringing the "Lab of the Future" closer to reality.
Deepak Bahl Ph.D., Global Head of Applied Sciences – Pharma, Roquette
The pharmaceutical industry is one of the most regulated in the world — but it doesn’t exist in a vacuum. Our market draws influence from several related sectors, most notably food production, industrial manufacturing, and — increasingly — digital technologies.
Beginning with drug product manufacturing, many of the processes employed by the pharmaceutical industry were first pioneered in other industries. Film coating, spray-drying, hot-melt extrusion, and continuous manufacturing are all examples of processes adopted from the food industry. Even more surprising, roller compaction was embraced from the world of coal mining!
Innovations in the field of robotics have also helped to facilitate drug discovery and product manufacturing by accelerating hit and lead generation, as well as reducing execution times for excipient compatibility studies. Additionally, for formulation development, functionality excipients adapted from the polymer and food industries support manufacturers to deliver targeted product profiles incorporating quality-by-design (QbD) principles. This ensures the drug product formulations are so robust that even slight variations will not impact a drug’s clinical performance.
Finally, we cannot deny the seismic impact of digitization, including the use of AI and other smart machine learning technologies on virtually every industry in recent years. From trend monitoring and supply forecasting to accelerated drug discovery, testing, regulatory filings, and approval —` digitalization and AI may once have belonged to the realm of science fiction, but it and advancements like it are now firmly part of today’s pharma reality.
Andrew Anderson, Vice President, Innovation and Informatics Strategy, ACD/Labs
My experience has been that when non-pharma folks join the pharmaceutical industry, there is an effort to lift and adapt technology and know-how from other industries. They are first met with enthusiasm, which is represented in budget and resources to assess why/when/how to implement such technologies, and pharma innovation teams will establish a formal procedure for their implementation and review.
We've certainly seen some incredible successes in technology adoption in pharma — of commercial supply chain technologies from the commodity chemicals and materials industry, in product lifecycle management (PLM) tech from the automotive and aerospace industries — and in electronic health records (EHR) for patients in healthcare practitioner engagement.
In R&D, cross-fertilization projects have been more challenging. Pharma generates some of the most complex and prolific varied data which requires the juxtaposition of materials and life sciences data. Thought leaders seeking to cross fertilize technologies should consider "integratability" as a technology readiness factor. Scientific data is nuanced and rich, and no single informatics systems can handle it all. Pharma must develop an environment with well-integrated software tools that handle the variety of specialized data well.
Anna Codina, Ph.D., Senior Director Biopharma & Strategic Market Development, Bruker BioSpin
The pharmaceutical industry continues to invest time and money in optimizing the process of discovering new medicines and bringing them to market. We have seen spectacular results during COVID, resulting from a concerted effort, but perhaps unsustainable in the long term. Today, we see the start of a new revolution, with the first results revealing how the adoption of AI is changing drug discovery, development, and advanced manufacturing.
We are embarking on a new era where drugs are discovered in silico, in ‘dry labs’ via the utilization of intelligent engines trained with experimental data and digital twins. The caveat is that the models used to predict are only as good as the data used to train them. While data and metadata are now being collected in vast amounts, it is imperative that the data are in the correct format. Companies adopting scientific data platforms able to standardize the harvested data and prepare it for AI are overcoming one of the major bottlenecks that has been hindering the digital transformation. They are paving the way for the new medicine-making paradigm that is revolutionizing drug discovery and development for faster delivery of more personalized medicine to patients.
