Kristin Brooks, Managing Editor, Contract Pharma04.10.24
Many new drug candidates in development today are highly complex and creating robust and efficient manufacturing processes is a challenge. Efforts to accelerate promising therapies through discovery and development run the risk of being hindered by clinical and commercial manufacturing hurdles.
With the rise in manufacturing costs, especially as additional antibody drug conjugates and cell and gene therapies are brought to market, the pharma industry is increasingly leveraging digital tools in manufacturing to help keep pace with progress in drug discovery and clinical development. AI and ML show the most promise in accelerating R&D and reducing manufacturing costs by speeding up target identification, improving processes, and simplifying the supply chain.
Barry Heavey, M.D. at Accenture, and global lead for Life Sciences Manufacturing & Quality Clinical and Commercial Supply Chain, provides insight on how the industry can leverage digital tools to achieve some of these benefits and reduce manufacturing costs.
Contract Pharma: How can the pharma industry leverage digital tools in manufacturing to improve drug discovery and clinical development?
Barry Heavey: The pharmaceutical industry can leverage digital tools in manufacturing to improve drug discovery and clinical development by helping to accelerate our understanding of the complexities of biology. Drugs bind to proteins in the body to elicit their effects and artificial intelligence is providing faster insights into what proteins must be targeted to treat various disease. To find drugs that will bind these proteins it is helpful to know the three-dimensional structure of these target proteins and AI systems have been developed that can predicts a protein's 3D structure from its amino acid sequence, saving scientists time and experimental effort.
Machine learning (ML) and artificial intelligence (AI) can accelerate research and development by helping to identify patients who are likely to respond favorably to a drug, or not have serious side effects, by helping to predict the biology of drug metabolism, drug response and potential off target binding of the drug to other proteins that could trigger side effects in patients with different genetic heritage. The net result is that target identification, drug discovery and clinical development can be executed more quickly and efficiently using AI.
In addition, tools like digital twins enable in-silico experimentation on the complex manufacturing, purification and formulation processes used to supply new drugs at large scale. By simulating scientific experiments, these tools can help improve manufacturing processes, increase yield, process robustness /consistency and simplify the supply chain. This can result in significant long-term cost savings and reduce environmental impact.
Furthermore, digital tools can enhance productivity and improve the overall experience of scientists and engineers involved in developing manufacturing processes. By automating specific tasks and providing real-time data and insights, digital tools can streamline workflows and enable more efficient decision-making.
Contract Pharma: Which digital tools show the most promise in reducing manufacturing costs, including AI and ML?
Barry Heavey: AI and ML are among the digital tools showing promise in reducing manufacturing costs in the pharma industry. More and more detailed data can be collected from manufacturing process, either using “in-line” sensors embedded in reactor vessels (such as RAMAN spectroscopy) or “off-line” testing of samples from the manufacturing process using high tech laboratory systems like whole genome sequencing of producer cells or mass spectroscopy analysis of complex mixtures. However, with advanced data analytics tools, these new sensors can leave scientists “data rich but insight poor”. AI and ML can help analyze large amounts of data and identify patterns and insights that humans may not be able to detect. By leveraging AI and ML, companies can improve manufacturing processes, increase efficiency, and reduce costs.
For example, AI can be used to analyze data from previous manufacturing processes and identify areas for improvement. It can also predict and prevent potential quality issues on “in-flight” batches of drugs, reducing the risk of costly delays, rejected batches or recalls. Using AI to reduce delays on batches due to quality issues is becoming increasingly important for hyper-personalized medicines like CAR-T cells, where the cancer patient has provided their blood for modification and urgently need their cells to be modified in a manufacturing facility and shipped back to the hospital for infusion. For very sick cancer patients, every minute can count and so minimizing delays and risks in manufacturing is incredibly impactful.
Because they can simulate specific steps and end-to-end manufacturing processes virtually, digital twins can identify bottlenecks, improve resource allocation, and enhance efficiency, saving money and increasing productivity. In short, digital twins have the potential to revolutionize the manufacturing industry by providing a safe, cost-effective, and efficient way to test and optimize manufacturing processes.
Contract Pharma: What's driving the rise in manufacturing costs?
Barry Heavey: The rise in manufacturing costs in the pharmaceutical industry can be attributed to the much greater mix of new modalities emerging in recent years and the fact that most companies have not yet achieved economies of scale with these products. In contrast, solid oral dose (tablet manufacturing) has been established for over a century and companies have invested heavily in facilities and technology during that time to achieve the scale to drive down cost-per-tablet. More recently monoclonal antibodies had extremely high cost per dose in the 1990’s but this has come down in the last three decades due to investment in additional manufacturing capacity and new technology to improve yield from the biological production process. New modalities like AAV, oligonucleotides, CAR-T, antibody drug conjugates are more complex, less stable or more variable than older modalities and their manufacturing processes are less mature and hence less productive/higher cost.
It can be expected that costs for these new modalities will decrease over time through investment in technology, but greater use of AI has the potential to move these modalities down the cost curve faster than previously seen with older modalities. Using advanced analytics to improve manufacturing productivity quickly will be important because there is less potential to rely on simple economies of scale (large production vessels and heavy automation). With all these new modalities and the fact that new drugs are often targeted against smaller patient populations, factories are moving from producing a high-volume of a small number of drugs to producing a high mix of drugs all at low volumes. Production vessels will be smaller, automation more challenging because of the ever-changing mix of products and workers will need to be highly skilled to execute a wide variety of processes.
The various production processes for various drugs require expensive equipment (increasingly single use equipment), patient protected and therefore expensive raw materials. Additionally, regulatory requirements are often more stringent and variable from market to market for new and complex modalities where regulators are less familiar with the products and processes and hence risk averse – this too adds to the cost of maintaining manufacturing that is compliance in all markets. It is estimated that, process development and manufacturing can account for 20–25% of the total cost of bringing a new drug to market and this is likely to rise for new modalities – however it is a vital investment to ensure that manufacturing complexity doesn’t become a bottleneck to bring exciting new medicines to all patients in a sustainable manner.
Barry Heavey is the managing director and the Accenture Life Sciences Industry X and Supply Chain global lead. He has over 25 years of experience in the pharmaceutical industry, with his early career focusing on the development of Biologics and stem cell therapies. Barry has a Ph.D. in genetics from the University of Vienna and an MBA from Edinburgh University.
With the rise in manufacturing costs, especially as additional antibody drug conjugates and cell and gene therapies are brought to market, the pharma industry is increasingly leveraging digital tools in manufacturing to help keep pace with progress in drug discovery and clinical development. AI and ML show the most promise in accelerating R&D and reducing manufacturing costs by speeding up target identification, improving processes, and simplifying the supply chain.
Barry Heavey, M.D. at Accenture, and global lead for Life Sciences Manufacturing & Quality Clinical and Commercial Supply Chain, provides insight on how the industry can leverage digital tools to achieve some of these benefits and reduce manufacturing costs.
Contract Pharma: How can the pharma industry leverage digital tools in manufacturing to improve drug discovery and clinical development?
Barry Heavey: The pharmaceutical industry can leverage digital tools in manufacturing to improve drug discovery and clinical development by helping to accelerate our understanding of the complexities of biology. Drugs bind to proteins in the body to elicit their effects and artificial intelligence is providing faster insights into what proteins must be targeted to treat various disease. To find drugs that will bind these proteins it is helpful to know the three-dimensional structure of these target proteins and AI systems have been developed that can predicts a protein's 3D structure from its amino acid sequence, saving scientists time and experimental effort.
Machine learning (ML) and artificial intelligence (AI) can accelerate research and development by helping to identify patients who are likely to respond favorably to a drug, or not have serious side effects, by helping to predict the biology of drug metabolism, drug response and potential off target binding of the drug to other proteins that could trigger side effects in patients with different genetic heritage. The net result is that target identification, drug discovery and clinical development can be executed more quickly and efficiently using AI.
In addition, tools like digital twins enable in-silico experimentation on the complex manufacturing, purification and formulation processes used to supply new drugs at large scale. By simulating scientific experiments, these tools can help improve manufacturing processes, increase yield, process robustness /consistency and simplify the supply chain. This can result in significant long-term cost savings and reduce environmental impact.
Furthermore, digital tools can enhance productivity and improve the overall experience of scientists and engineers involved in developing manufacturing processes. By automating specific tasks and providing real-time data and insights, digital tools can streamline workflows and enable more efficient decision-making.
Contract Pharma: Which digital tools show the most promise in reducing manufacturing costs, including AI and ML?
Barry Heavey: AI and ML are among the digital tools showing promise in reducing manufacturing costs in the pharma industry. More and more detailed data can be collected from manufacturing process, either using “in-line” sensors embedded in reactor vessels (such as RAMAN spectroscopy) or “off-line” testing of samples from the manufacturing process using high tech laboratory systems like whole genome sequencing of producer cells or mass spectroscopy analysis of complex mixtures. However, with advanced data analytics tools, these new sensors can leave scientists “data rich but insight poor”. AI and ML can help analyze large amounts of data and identify patterns and insights that humans may not be able to detect. By leveraging AI and ML, companies can improve manufacturing processes, increase efficiency, and reduce costs.
For example, AI can be used to analyze data from previous manufacturing processes and identify areas for improvement. It can also predict and prevent potential quality issues on “in-flight” batches of drugs, reducing the risk of costly delays, rejected batches or recalls. Using AI to reduce delays on batches due to quality issues is becoming increasingly important for hyper-personalized medicines like CAR-T cells, where the cancer patient has provided their blood for modification and urgently need their cells to be modified in a manufacturing facility and shipped back to the hospital for infusion. For very sick cancer patients, every minute can count and so minimizing delays and risks in manufacturing is incredibly impactful.
Because they can simulate specific steps and end-to-end manufacturing processes virtually, digital twins can identify bottlenecks, improve resource allocation, and enhance efficiency, saving money and increasing productivity. In short, digital twins have the potential to revolutionize the manufacturing industry by providing a safe, cost-effective, and efficient way to test and optimize manufacturing processes.
Contract Pharma: What's driving the rise in manufacturing costs?
Barry Heavey: The rise in manufacturing costs in the pharmaceutical industry can be attributed to the much greater mix of new modalities emerging in recent years and the fact that most companies have not yet achieved economies of scale with these products. In contrast, solid oral dose (tablet manufacturing) has been established for over a century and companies have invested heavily in facilities and technology during that time to achieve the scale to drive down cost-per-tablet. More recently monoclonal antibodies had extremely high cost per dose in the 1990’s but this has come down in the last three decades due to investment in additional manufacturing capacity and new technology to improve yield from the biological production process. New modalities like AAV, oligonucleotides, CAR-T, antibody drug conjugates are more complex, less stable or more variable than older modalities and their manufacturing processes are less mature and hence less productive/higher cost.
It can be expected that costs for these new modalities will decrease over time through investment in technology, but greater use of AI has the potential to move these modalities down the cost curve faster than previously seen with older modalities. Using advanced analytics to improve manufacturing productivity quickly will be important because there is less potential to rely on simple economies of scale (large production vessels and heavy automation). With all these new modalities and the fact that new drugs are often targeted against smaller patient populations, factories are moving from producing a high-volume of a small number of drugs to producing a high mix of drugs all at low volumes. Production vessels will be smaller, automation more challenging because of the ever-changing mix of products and workers will need to be highly skilled to execute a wide variety of processes.
The various production processes for various drugs require expensive equipment (increasingly single use equipment), patient protected and therefore expensive raw materials. Additionally, regulatory requirements are often more stringent and variable from market to market for new and complex modalities where regulators are less familiar with the products and processes and hence risk averse – this too adds to the cost of maintaining manufacturing that is compliance in all markets. It is estimated that, process development and manufacturing can account for 20–25% of the total cost of bringing a new drug to market and this is likely to rise for new modalities – however it is a vital investment to ensure that manufacturing complexity doesn’t become a bottleneck to bring exciting new medicines to all patients in a sustainable manner.
