The painstaking journey to bring one new drug to market almost spans a decade and costs an average of $2.6 billion. The AI as well as computing revolution, however, happens to be dynamically shifting the pharma industry, getting into a new era of drug development that is all set to be better, faster, as well as affordable.
As one could see at the J.P. Morgan Healthcare Conference in January 2024, which is a major event in the healthcare industry, one of the most important topics of discussion was the usage of AI to speed up health innovation. Deep Pharma Intelligence reports a 27-fold rise in the amount of capital that has been funded in AI-driven pharma companies since 2015, with the leading 800 such companies getting $59.3 billion in terms of investment as of December 2022.
It is well to be noted that innovative AI approaches are not just happening within the biotech sector. Traditional big pharma companies happen to also be actively getting involved in AI drug discovery, either by way of partnering with small AI biotech companies to speed-up the exploration of new therapies and/or creating in-house AI drug discovery units. For instance, Pfizer partnered with IBM’s Watson so as to speed up drug discovery within immuno-oncology, and Sanofi got engaged with Exscientia so as to make use of AI to identify metabolic-disease drug targets.
AI companies have gone on to use existing data from prior experiments, robotics, as well as images rather than going ahead with traditional lab-based scientific research approaches in order to discover new drug targets at speeds that are unprecedented. These findings have gone on to bring energy to the sector and have offered very initial evidence of the usage of AI when it comes to biopharmaceutical innovation.
Although the way drugs have been discovered has gone on to evolve with tech as well as computing, the way success gets measured for any drug happens to remain the same by way of successful clinical trials that lead to FDA nods.
Separating AI Reality from the Hype
It is worth noting that although the sector is indeed clamoring for tangible success metrics so as to prove the utility of AI, the rhetoric when it came to the J.P. Morgan Healthcare Conference still leaned towards the promise of what more is yet to come. The past 18 months have been a cautionary tale when it comes to the limits of AI capacities within drug discovery, with some expected AI-designed drug candidates being failed in clinical trials.
Industry leaders happen to be now realizing the fact that AI hype has led to some unrealistic anticipations for what it can go ahead and achieve. AI-based drug discovery sans the patient biology, especially the samples as well as wet lab experiments, happens to be very risky and also unlikely to be successful on its own.
2024 goes on to represent an opportunity for firms that have quietly gone on to prioritize biology-first approaches towards AI so as to progress in clinical trials with drug candidates that have been developed with the help of proprietary AI platforms while at the same time leveraging AI algorithms so as to define target populations. One believes the success of companies in making the utmost use of AI in drug discovery will transition the narrative from tech hype to real value when it comes to patients.
Apparently, the winners in AI-powered drug development must go on to recognize the following truths about opportunity when it comes to AI so as to improve drug discovery as well as development.
AI happens to be a tool when it comes to enhancing drug discovery and not a replacement for it.
One has finally gone on to reach the slope of enlightenment, where organizations that are doing AI well are starting to see the success of their work within the clinic, whereas those who are not doing it well are witnessing clinical failures. It is worth noting that savvy biopharma companies happen to be using AI as a guide through the very complex wealth of patient data that is developed by way of using real biological samples and not just public databases.
When making potential use of real biology, AI platforms can indeed be incredibly effective when it comes to identifying promising drug targets, methodologies used for targeting complex biological pathways, as well as ways to make full use of later-stage clinical trials that are based on the characteristics of responders in previous stage trials. This synergistic relationship of AI and biology goes on to offer the potential to reshape drug discovery spectrum, thereby converging data analysis by way of biological insights.
It is not just about drug targets that are validated.
It is well to be noted that AI-driven drug discovery starts with taking in to account and also validating drug targets, which are similar to uncovering treasures that are hidden in a complex biological landscape, however, it does not end there. Precision when it comes to syncing biological profiles by way of clinically relevant patient data is mandatory, much like decoding distinct genetic patterns that go on to shape an individual’s health journey, and AI can go ahead and serve as a very lucrative tool when it comes to refining this understanding.
But the foundation for effective drug discovery goes on to remain rooted in robust biological understanding. Integration of AI’s computational power along with biology goes on to act as a catalyst, thereby enabling the identification as well as validation of potential drug targets that are critical when it comes to groundbreaking therapeutic advancements. It can also be made use of to better understand the traits of the people who will go on to respond to the therapies that are being tested. Making the utmost use of AI models in combination with actual biological observations is indeed the next frontier when it comes to applying AI to the gamut of drug development.
Not all AI happen to be equal.
Beginning with a biology-first approach, the AI spectrum in drug discovery goes on to unleash a diverse array of available types, like machine learning, neural networks, as well as Bayesian AI, among others.
Among them, initiating Bayesian AI goes on to offer hypothesis-free discovery and, at the same time, holds the potential to redefine conceptualization, discovery, as well as the development of drugs. Neural AI steps in after that to decode the complex relationships between genetic factors as well as common diseases, thereby aiding critical decision-making within the drug development pathways. Varied AI modules should be made use of in varied aspects of discovery, like health and clinical analytics, as there is indeed no effective one-size-fits-all kind of approach.
The Path forward is being paved
The fact is that AI revolutionizes drug development by way of boosting efficiency, elevating data analysis, and also reshaping trial structures, that address rising costs as well as high failure rates within drug development. Apparently, a biology-first AI approach can go ahead and elevate patient specificity, thereby enabling faster identification of viable candidates in terms of clinical trials and fostering rapid success.
After years of successes as well as setbacks, all this marks the start of a phase that is focused on tangible outcomes by way of AI-developed drugs as well as diagnostics that are validated through emerging clinical data, thereby moving past the era of just hype.
