Artificial intelligence (AI) is revolutionising the pharma and biotech sectors. By accelerating drug discovery, optimising clinical trials, and facilitating personalised medicine based on genetic data, AI is a key player enabling the development of more precise and tailored treatments. These advancements are significant for biotech companies such as phage specialist Nexabiome, whose work in phage applications for human, animal, and harvest health can be greatly enhanced by AI. However, as the industry embraces AI, it also grapples with ethical considerations. In this article, we will delve into the advantages and challenges of AI in pharma and biotech, and how it could support Nexabiome’s to improve human, animal, and harvest health with safe and effective bacteriophage-based products.
What is AI?
AI refers to the simulation of human intelligence in machines. That is the ability of computers or machines to perform tasks associated with cognitive function, such as understanding natural language, recognizing patterns, solving problems, and making decisions. Over the past two decades, there have been huge advancements in the power of AI, attributable to increases in computational power, vast amounts of data, and refined algorithms. Advancements in AI have integrated into the pharmaceuticals and biotechnology industries, driving innovation and efficiency. Some of the most transformative advancements encompassing AI include accelerating the drug discovery process, streamlining clinical trials, and improving personalised medicine.
AI in Drug Discovery
Drug discovery is renowned for being an expensive, inefficient, and uncertain process riddled with challenges and an extraordinarily high failure rate. However, recent technological advancements involving AI are improving the drug discovery pipeline and increasing the success rate.
Understanding protein structure allows scientists to design drugs that specifically bind to the target and disrupt its function. Previously, this involved a tedious, costly experimental workflow that often failed to yield high-resolution structures. However, AI-based protein structure prediction, utilising machine learning (ML) algorithms such as AlphaFold and RoseTTAFold have improved the accuracy, speed, and success of this process.
In drug design, AI algorithms can be used for molecular modelling, which enables accurate prediction of how molecules will behave and provides insights into how drugs can be modified to improve efficacy. AI can rapidly screen thousands of compounds to predict their likelihood of being effective as drugs. By utilising big data resources, ML and AI approaches enable systematic identification of drug repurposing leads, reducing the risk involved in drug development and significantly reducing the time and cost involved compared to developing a new drug from scratch.
Similarly, AI can be leveraged in the development and optimisation of phage therapies. AI models can detail phage-bacteria interactions, predicting therapy efficacy against specific bacterial strains. By analysing big datasets, AI can pinpoint the most promising phage candidates to combat specific bacterial infections.
AI in Clinical Trials
Clinical trials are a crucial step of the drug development process and are also incredibly costly in terms of both money and time. Despite this, clinical trial failure is a devastatingly common occurrence; only 1 in 10 compounds that enter trials make it to the market. Clinical trial failure is a multifactorial problem, involving challenges with patient cohort selection, recruitment, and retainment, patient monitoring, and safety issues.
Recently, AI has been employed in the process of clinical trial design, and innovative AI platforms have proven useful, safe, and effective tools for optimising clinical trials and lessening some of the challenges that ultimately lead to failure. AI-based tools can optimise and streamline patient recruitment, monitor trials effectively, and predict outcomes, thus expediting the trial process and increasing its efficiency. After the trial’s completion, AI can be used to efficiently process and analyse trial data, enabling quicker insights into the efficacy and safety of therapeutics.
Within bacteriophage therapy, the lack of robust clinical trial data is a huge issue contributing to the shortfall in regulatory approval of bacteriophage-based therapies, stemming from the inherent complexities of designing and running phage therapy trials. Thus, integrating AI-based tools and platforms into designing optimal clinical trials could be crucial for proving the safety and efficacy of phage therapies and accelerating their approval.
AI in Personalised Medicine
Personalised medicine, also known as precision medicine, refers to the tailoring of medical decisions on prevention and intervention strategies to the individual, based on the patient’s individual genetic profile as well as environmental and lifestyle factors. It is an exciting realm of medicine that provides opportunities for more safe, effective treatments, rather than the traditional “one size fits all” approach.
AI is facilitating personalised medicine by enabling quick analysis of vast genomic datasets to characterise a patient’s genomic profile, as well as mining health records to extract insights into patient history, treatment responses, and risk factors. Together, this data enables safe, effective treatments to be selected. Moreover, AI can perform predictive analyses to predict disease risk or progression and response to treatment and identify novel biomarkers which enable better prognosis prediction and guide treatment decisions.
These capabilities are advantageous for phage therapies since they enable the identification of the specific bacterial strains causing infections, ensuring the selection of the most apt phage treatment. AI’s rapid genomic data analysis tracks bacterial evolution, ensuring phage therapies evolve in tandem. The technology can guide the development of potent phage cocktails tailored to individual infections, and obtain insights from patients’ historical data to fine-tune treatments.
Despite its numerous advantages, there are certain ethical considerations surrounding AI as a tool in pharma and biotech. AI’s ability to mine large datasets is invaluable, however, with it comes concerns surrounding data privacy. That is, these datasets will usually include patient information and data, and ensuring the privacy and security of this data is paramount. Without proper safeguards, there’s a risk of data breaches or misuse, potentially leading to personal information being exposed or exploited. In addition, informed consent must be obtained relating to exactly how patient data is likely to be used. Moreover, there is a need for clear regulatory guidelines surrounding the use of AI in healthcare settings, and it needs to be ensured that AI recommendations can be easily and accurately interpreted by human users for safety reasons.
Conclusion and Future Perspectives
In summary, developing AI tools and technologies are proving hugely beneficial to the pharma and biotech industries. AI is accelerating drug discovery, optimising clinical trials, and facilitating more personalised treatment strategies. In terms of phage therapies, AI is likely to prove invaluable, especially considering the existing issues surrounding the lack of robust clinical trial data and the individualised nature of bacteriophage treatments. However, data privacy concerns remain a hurdle to overcome to allow AI to be employed to its full potential within patient-centric industries.