DATA STORIES | AI & MACHINE LEARNING | DEEP DIVES
AI algorithms to advance drug efficacy for patients
AI algorithms can facilitate the scalability of production within reduced time and manufacturing cost, thereby decreasing the cost of healthcare
What is an API?
In the world of pharmaceuticals, the key player is the active pharmaceutical ingredient (API) — the powerhouse behind a drug’s therapeutic effects. APIs work their magic by interacting with the body in targeted ways to combat diseases and conditions.
The efficacy of a drug hinges heavily on the quality and potency of its API. Take, for instance, a painkiller: its ability to alleviate discomfort rests on the potency of its active ingredient. This vital component, known as the API, is the linchpin in the effectiveness of countless medications.
According to the Pharmaceutical definitions, active pharmaceutical ingredients (APIs) come in two main types: synthetic chemicals and natural chemicals. Crystallization plays a crucial role in defining the physicochemical properties of APIs, making it a vital process in the pharmaceutical industry. Pharmaceutical formulation involves multiple steps, where the active drug is combined with other components, considering factors such as particle size, polymorphism, pH, and solubility. This comprehensive process results in the creation of the final medicinal product, ready to provide its intended benefits to patients.
The success of pharmaceutical formulation hinges on four fundamental components: the benefits and limitations of active pharmaceutical ingredients (APIs), the value of excipients, the interactions between them, and the manufacturing procedure.
Issues of Concern
Pharmaceutical formulation entails the development of a pharmaceutical product, encompassing its chemical properties, formulation, and details of the treatment protocol for clinical application. Despite substantial annual investments, the list of medications remains stagnant due to minimal growth in the number of new molecular entities (NMEs) approved by the FDA each year. This stagnation arises from uncertainties inherent in drug design, wherein the target molecule’s shape and charge must align with those of the drug. Drug design relies on sophisticated computer modeling techniques and bioinformatics approaches to identify suitable interactions between molecules.
Computational methods frequently serve to improve the affinity, selectivity, stability, and efficacy/potency of the drug with its target molecule.
Where can AI help in drug manufacturing?
Indeed, there are areas of scarcity where robust AI-based computational methods for analyzing crystalline structures can play a pivotal role in ensuring drug safety and efficacy during the formulation phase. Let’s delve into an innovative and sophisticated technique driven by machine learning and computer vision algorithms. This method facilitates the rapid processing, classification, and visualization of crystalline structures, thereby safeguarding the safety and efficacy of medications for patients globally.
According to a report by Statista on “Artificial intelligence (AI) technologies market share globally in 2022”, AI adoption in the pharmaceutical and healthcare industry stands at a mere 15%. However, a study published by the same source highlights significant cost-saving opportunities, with potential savings of up to 30% through AI utilization in drug discovery worldwide as of 2023. Additionally, experts surveyed, combined with an analysis of scientific publications, suggest that AI-enabled workflows could reduce the time required to bring a new molecule for a difficult or poorly understood target to the preclinical candidate (PCC) stage by up to 40%. These findings underscore the transformative potential of AI in revolutionizing drug discovery processes and enhancing efficiency in the pharmaceutical industry.
Stages of Drug Development
Note: A drug goes through 5 critical stages before it is available to a patient for effective treatment, and each of the stages contributes to drug effectiveness for a disease and the safety of the drug for human consumption. In this article, we will focus on the first step where AI is significantly helping in accelerating the process.
Discovery and development
At this juncture, scientists embark on the critical task of identifying a specific molecule — be it a DNA sequence, RNA molecule, protein, or metabolite — that assumes a pivotal role in a disease state and holds the potential to be targeted by a drug, yielding therapeutic benefits. With a plethora of compounds under consideration, researchers delve into a series of experiments to assess their efficacy and suitability as the ultimate drug substance. Factors such as absorption, administration, side effects, and potential interactions are meticulously evaluated in this process. Following the completion of these experiments and the isolation of the most promising compounds, the preclinical research phase is poised to commence, marking another significant stride in the quest for novel therapeutics.
Can we trust AI algorithms to advance drug development?
During the phase of drug discovery, the analysis and comprehension of crystalline structures in solid dispersion formulations of active pharmaceutical ingredients emerge as an essential endeavor. Presently, scientists tackle this task manually, grappling with its time-intensive nature and susceptibility to manual errors. Consequently, the process spans years before the identification of compounds deemed effective for drug development reaches fruition.
Note: Pharmaceutical formulation development currently relies on trial-and-error methods, which aid in predicting optimal formulations. However, this approach is marred by its time-consuming and expensive nature, requiring a substantial amount of labor for development and monitoring.
Following successful experiments on compound selection, the API process development entails optimizing the manufacturing process for the drug’s active ingredient. This optimization aims to ensure that the API can be produced at scale while meeting all required quality standards.
Alternatively, AI can aid in the development of technologically advanced tools by employing image processing techniques on 3D CT scan data of drug molecules’ microscopic structures, converting them into high-resolution images and video data. These data can then be utilized in automated deep learning-based analysis algorithms to explore structural changes, shape rotations, tolerance to enlargement, and stability. In addition to optimizing various formulation parameters, deep learning AI algorithms can facilitate the scalability of production within a shorter timeframe and at a reduced manufacturing cost, thereby contributing to the decreased cost of healthcare.
Highlight: Pharmaceutical giants such as AbbVie are championing AI’s role in drug development and manufacturing, evident in their substantial investment of $7.8 billion into Research and Development in 2023, as reported by Abbvie.
Note: According to recent public announcements, the company is harnessing technology to expedite scientific advancements by constructing cutting-edge technologies and integrating them throughout the entirety of Research and Development. This underscores the gradual ascent of AI in earning the trust of industry leaders like AbbVie in the pharmaceutical realm, facilitating rapid, secure, and cost-effective drug development processes.
Conclusion
As discussed in the AI market share analysis across industries, many sectors have already experienced significant transformation due to AI advancements, while others await a similar level of transformation. Among these, the pharmaceutical industry appears to be relatively advanced in embracing this transformative process. The realization of the full potential of AI will come to fruition as more industry leaders place their trust in AI solutions. A recent article in the Future Journal of Pharmaceutical Sciences by Springer titled “Influence of artificial intelligence in modern pharmaceutical formulation and drug development” sheds light on the immense potential of AI in streamlining the drug formulation and development pipeline.
The future is here: “AI is a potent pharmaceutical formulation and development tool, allowing researchers to analyse vast amounts of data, optimize drug formulations, and streamline clinical trials. As technology develops, experts anticipate that AI will increasingly show a crucial part in drug development, enabling faster, more efficient, and more effective treatments for various diseases.”
Resources
Cite: Advanced Technologies by Abbvie
Cite: News Center by Abbvie
Cite: Influence of artificial intelligence in modern pharmaceutical formulation and drug development by Springer
Cite: Time and cost savings through AI usage in drug discovery worldwide by scenario 2023 by Statista
Cite: Market share of AI technologies worldwide 2022, by industry by Statista
