Top Use Cases for Artificial Intelligence, Machine Learning, and Blockchain within Drug Development
Frank Leu, Founder and Managing Member, BioPharMatrix LLC
When artificial intelligence (AI), machine learning (ML), and blockchain are used cohesively and correctly, they could ameliorate the existing complexity, inefficiency, lengthy, and costly drug discovery to market. Here we will address a few executive-level questions within the context of these rapidly growing disruptive techs in drug development.
The drug development process, from discovery to market, is an overly complex, inefficient, lengthy, and costly journey that involves multiple phases of research, development, and regulatory reviews. This process consists of four major steps, with each step having significant internal and external inefficiencies that could potentially be best addressed using disruptive techs. The costliest, while not revenue generating, and most effort-intensive in validating a drug’s usefulness are the first two: discovery to preclinical testing (1-6 years) and clinical trials (6 to 7 years). Discovery involves first the identification of potential targets and compounds. High-throughput screening, artificial intelligence (AI), machine learning (ML), and blockchain, collectively referred to as the “disruptive techs” and methods can accelerate this phase and move to the preclinical examination of the candidate via tests on cell cultures and animals to assess the safety and biological activity. This phase is critical as it determines whether a compound is safe for human testing while demonstrating efficacy in at least two animal models.
The next step would be clinical trials, commonly conducted in three major phases:
Phase I: Test a drug on a small group of people (20-100) to evaluate its safety, determine a safe dosage range, and identify side effects.
Phase II: The drug is given to a larger group of people (100-300) to see if it is effective and to evaluate its safety further.
Phase III: Conducted on large groups of people (1,000-3,000) to confirm its effectiveness, monitor side effects, compare it to commonly used treatments, and collect information that will allow the drug to be used safely.
Once the candidate is deemed to be safe enough and effective by the developer for its intended disease indication, it moves on to the third step, which is to seek regulatory review by submitting a New Drug Application (NDA) or Biologics License Application (BLA) with the intent to seek approval to market by the Food and Drug Administration (FDA) and its similar agencies around the world. The final and seemingly perpetual step is post-market surveillance, as the drug of interest continues for ongoing monitoring for any adverse effects that may not have been detected during the clinical trials.
What is the cost estimated for a drug from discovery to market without full-scale implementation of disruptive technologies, such as AI, ML, and blockchain?
The cost of developing a new prescription medicine that gains marketing approval is estimated to be over US$2.5 billion, according to a study by the Tufts Center for the Study of Drug Development. This figure includes the cost of most drug candidates that fail during development (the cost of failure is factored into the overall cost of successful drug development). It's important to note, however, that this is an estimated average and can vary widely depending on the drug, the disease it targets, the complexity of clinical trials, and the regulatory path it follows. The estimated cost of US$2.5 billion encompasses discovery, preclinical development, clinical trials, and the regulatory submission process. It also includes opportunity costs (the cost of capital) and the costs associated with research and development activities that do not lead to a successful new medicine. As one can imagine, the rampant inefficiencies in the current drug developmental process can be minimised or eliminated by using disruptive techs.
In general, how could the implementation of disruptive technologies such as AI, ML, and blockchain enhance the process of drug development?
Many major factors often affect the timeline and cost during a legacy drug development process. Adopting AI, ML, and blockchain could potentially reduce significant amounts of time and cost by facilitating the identification of promising drug candidates more efficiently and enhancing the design of clinical trials. These disruptive techs can be used to directly tackle novel drug mechanisms and to properly reduce the complexity of the process. Next, the same techs could also be used to comply with an ever-changing regulatory environment, which would then reduce the duration and cost of the clinical trials. Ultimately, they can improve clinical trial outcomes and reduce the need for any unnecessary additional clinical trials.
How can disruptive tech accelerate the drug discovery process?
Blockchain ensures the integrity and immutability of research data, clinical trial results, and manufacturing records, preventing data tampering, and enhancing trust; accomplished through the improvement of data integrity and immutability. Through the application of AI and ML drug companies could quickly and intelligently analyse vast datasets to identify potential drug candidates much faster than traditional methods, significantly reducing the time and cost of the discovery phase. With good blockchain-encapsulated data blocks, this high-confidence information can be used by AI and ML to predict how drugs will interact with specific targets in the body, helping scientists understand the potential efficacy and safety of a drug before it goes to clinical trials, which would be a much-improved understanding of drug-target interaction. Additionally, by implementing the fitting AI algorithms in their drug design, pharma companies can derive novel drug molecules with desired properties and predict their synthesis paths, streamlining the drug design process by leaning toward in silico. Furthermore, ML models can be deployed here and used to analyse historical data to predict the outcomes of clinical trials, identify the most promising drug candidates, and optimise trial designs to improve drug candidates’ success rates.
How could disruptive tech support collaborative drug research?
Disruptive techs enable secure and transparent sharing of research data and intellectual property among multiple relevant stakeholders, fostering real-time collaboration while protecting proprietary information on a need-to-know basis. Can blockchain improve the management of intellectual property in drug development? Yes, it provides a secure and transparent platform to register and manage intellectual property rights, facilitating licensing and collaboration. Furthermore, algorithmic approaches such as natural language processing (NLP) can be used by AI and ML to extract relevant data and information from collaborators’ databases and documentation, public scientific literature, and pre-clinical and clinical reports, automating and expediting the research process and identifying new drug candidate opportunities.
How could disruptive techs enhance human clinical trials?
Blockchain can first facilitate patient consent and data sharing by securely managing and simultaneously verifying consent, this allows stakeholders, such as sponsors, hospitals, CROs, and patients to control their data during clinical trials. This is accomplished by providing stakeholders with complete control down to the individual level (i.e. patient), with enhanced transparency, trust, and engagement during clinical trials. Subsequently, the Al algorithms and ML models can be used to analyse historical and real-time collected clinical data, identifying and designing the most promising drug candidates and optimising the best trial design fitting to determine candidate drugs’ efficacy accurately.
How do disruptive techs enable personalised medicine during clinical trials?
Blockchain could be used to capture and facilitate patient consent and personal/medical data during a clinical trial, hence allowing the managing and sharing of data in a secure and verifiable manner. This allows patients to have control over their data and provide direct consent to participate in any clinical trial. AI and ML could then be applied to analyse patients’ genetic data and other biomarkers to develop personalised drug regimens with precision, ensuring higher efficacy with fewer side effects, possibly to achieve the best outcome in a personalised manner.
How would disruptive techs transform the regulatory aspect of drug development?
Blockchain can ensure that clinical trial and medical data are captured and stored according to ethical guidelines, with patient consent recorded in real-time in a trustless (meaning with the highest trust) manner with immutability, thus forging best ethical clinical testing practices. This enables downstream processing of the clinical data to be organised for regulatory compliance. Essentially, blockchain can securely store and manage regulatory submissions and communications with high fidelity, ensuring traceability and thus compliance with the best regulatory standards. Equipped with these highfidelity data compiled using blockchain, AI and ML then can be deployed to best analyse personalised clinical data with precision to identify disease markers, track disease progression, and evaluate potential drug efficacy. Simultaneously, adverse drug reactions and interactions are also closely monitored, ultimately enhancing drug safety for patients.
How could disruptive techs help to maintain the integrity of the management of the supply chain and optimise the drug manufacturing process?
Blockchain's traceability features can verify the authenticity of drugs at every step of the supply chain, which can significantly reduce the prevalence of counterfeit drugs. This is accomplished by providing a transparent and secure way to track the production, distribution, and delivery of pharmaceuticals. Subsequently, AI and ML could then be used to optimise drug manufacturing processes by predicting and controlling the parameters required for best drug manufacturing processes to ensure quality, reduce waste, and lower costs via implementing on-demand production.
Despite the apparent usefulness of disruptive techs, the challenge that remains is the proper and consensus integration of blockchain techs throughout drug development. Technical complexity, scalability, regulatory acceptance, and the need for industry-wide standardisation are significant challenges that must be addressed before these techs can become useful. Human resistance, the fear of being replaced and the cost of replacing legacy systems also create a chasm in their adoption. However, the cost of not immediately devising a plan to adopt will be devastating for late adopters. For example, once the blockchain, AI, and ML plans are readying to integrate, not only will they cut tremendous costs to facilitate drug development, but also add tremendous value to reliably analyse market data, patent information, and competitive landscape to predict the commercial viability of drug candidates.
In conclusion, a major bottleneck for adopting disruptive techs in an organisation is the actual commitment to bringing drug development into the future. The decision to adopt is not “if”, rather it is only a matter of “when”. When Steve Jobs saw the graphic interface at Xerox, he knew that technology would best serve the future of personal computing. But his executives told him, “We can’t do that. It would blow up the company.” Jobs replied, “Better we should blow it up than someone else.” When your entire business model is challenged and under great duress in the face of AI, ML, and blockchain revolution, would you be willing to blow up your legacy system to survive and out compete your competitors? If you have any doubts, the history of technological disruption has provided valuable lessons.
Author Bio
Frank Leu is the Founder and managing member of BioPharMatrix LLC, providing intelligence in the areas of innovative processes and platforms for biotech, pharmaceutical, and life sciences. Frank has combined around a hundred speaking presentations and publications on disruptive technologies and processes in drug development.
