Addressing the yearly $53B problem for pharma companies using groundbreaking AI and ‘Patient- on-a-chip’ technology
Isaac Bentwich, CEO, Quris
Quris pioneers a groundbreaking Bio-AI platform, merging advanced AI with innovative biology to revolutionise drug discovery. Partnering with major pharma companies like Merck, its patient-on-a-chip technology and real-time nano-sensing precisely predict drug safety, aligning with modern legislative changes and setting new standards in pharmaceutical innovation and clinical safety prediction.
Drug development stands at a pivotal juncture, struggling with the challenges of slow progress and soaring costs. With an average investment exceeding US$2.6 billion and a timeline stretching over 12-15 years per drug, bringing new medicine to market is fraught with many financial and timely hurdles. However, at the heart of this challenge lies a fundamental dilemma:the inability to predict which drug candidates will prove safe and effective for the human body.
The gap between preclinical testing and human trials remains a major hurdle in drug development. Notably, the high failure rate in clinical trials (around 92 per cent) suggests limitations in current methods, particularly considering the effectiveness of animal testing. This inefficiency can be costly, with estimates suggesting upwards of US$60 billion lost annually in failed oncology trials alone.
For decades, the pharmaceutical industry has heavily relied on animal testing as a means of predicting clinical outcomes. Yet, despite its widespread use, animal testing has proven to be an unreliable indicator of human response. Studies have revealed an alarming fact - nearly 89 per cent of drug candidates that successfully navigate animal testing fail to deliver the expected results in human clinical trials. This unfortunate reality emphasises the urgent need for innovative approaches to drug development.
Fortunately, recent regulatory changes have set the stage for a paradigm shift in drug development practices. The FDA Modernization Act 2.0, enacted in 2022, represent a landmark departure from the reliance on antiquated animal studies. Instead, the focus has shifted towards embracing modern, more effective methodologies that hold the promise of revolutionising drug development for the better good.
The Rise of Disruptive Technologies
At the forefront of this revolution are several disruptive technologies that offer new approaches to tackling the clinical prediction challenge. Chief among them is machine learning (ML), a field that has witnessed remarkable advancements in recent years. ML algorithms have demonstrated remarkable capabilities in various facets of drug discovery and development, from identifying novel drug targets to predicting the safety profiles of candidate compounds.
One of the strengths of ML lies in its ability to analyse vast amounts of data and recognise complex patterns that may escape human perception. Supervised, unsupervised, and reinforcement learning algorithms are deployed to sift through massive datasets, extracting valuable insights that inform decision-making processes at every stage of drug development.
Organoids and organ-on-chip
Traditionally, new drugs were first tested in animals. However, this doesn't always translate perfectly to how a drug will work in humans. Organoids and organ-on-chip technology are essentially miniaturised, simplified versions of human organs grown in a lab. These tiny organs offer a more realistic testing ground for potential drugs, allowing scientists to see how they might be absorbed, processed, and affect the body in a way that's closer to what happens in humans. This can help to identify potential problems earlier and avoid wasting time and resources on drugs that might not be effective or safe.
Stem cell biology
Stem cells are the body's master builders, with the potential to develop into many different cell types. This exciting field of science is leading to the development of patient-specific drug testing models. By using stem cells from individual patients, scientists can create miniaturised versions of their organs. This allows for a more personalised approach to drug development, where researchers can see how a drug might affect a specific patient's unique biology. This holds promise for the future of personalised medicine, where treatments can be tailored to each individual's needs.
Real-time nano-sensing
Imagine being able to watch a drug interact with cells in real-time. That's the potential of nano-sensing technologies. These tiny sensors can be used to monitor how cells respond to drug treatment, giving scientists a much clearer picture of how a drug is working and its potential effects. This real-time data can help researchers make better decisions about which drugs to pursue and how to optimise their effectiveness.
The Bio-AI Safety Prediction Approach
At the nexus of these disruptive technologies lies the Bio-AI safety prediction approach, a novel framework that integrates the power of machine learning with high-throughput experimentation on patient-on-chip platforms. This approach represents a paradigm shift in drug development, moving away from traditional animal testing models towards more accurate and personalised methodologies.
Central to the Bio-AI approach is the concept of high-throughput experimentation, enabled by patient-on-chip platforms that replicate the complexity of human physiology in vitro. These platforms allow researchers to conduct thousands of experiments simultaneously, generating vast amounts of data on drug responses across diverse patient populations.
ML algorithms are then deployed to analyse this data, identifying patterns and correlations that can predict drug safety and efficacy with unprecedented accuracy. By leveraging the power of ML, researchers can uncover subtle tradirelationships between drug properties and biological responses, leading to more informed decision-making in drug development.
Real-time sensing technologies further enhance the predictive capabilities of the Bio-AI approach by providing continuous monitoring of cellular responses to drug treatments. By capturing dynamic changes in metabolite levels and biomarker expression, these technologies offer valuable insights into drug mechanisms of action and potential adverse effects.
Stem cell genomic diversity plays a crucial role in the Bio-AI approach, allowing researchers to test drug responses across genetically diverse patient populations. By generating patient-specific stem cell models, researchers can simulate the variability observed in human populations, enabling more accurate drug safety and efficacy predictions.
Implications for Drug Development
The Bio-AI approach holds immense promise for revolutionising drug development, offering a more efficient, costeffective, and personalised alternative to traditional methodologies. By harnessing the power of machine learning, highthroughput experimentation, and realtime sensing, researchers can overcome the limitations of animal testing and accelerate the pace of drug discovery.
Moreover, the Bio-AI approach could potentially transform clinical trial design by enabling more precise patient stratification and selection. By leveraging patient-specific models, researchers can identify subpopulations that are most likely to benefit from a given treatment, thereby improving clinical trial outcomes and reducing the likelihood of trial failures.
The implications of the technology are substantial – it doesn't just potentially save drug developers considerable resources by mitigating the risk of a 90 per cent clinical trial failure rate, it also expands the realm of possible treatments, benefiting patients with all sorts of conditions, including those suffering from rare diseases, which are often marginalised due to financial and trial limitations. Advanced Bio-AI platforms have proven the ability to simulate clinical trials and a real human body’s reaction to drugs to effectively predict drug toxicity by leveraging a patented organ-on-chip system through stemcell-derived tissue and AI. Bio-AI can better predict which drug candidates will safely work in humans—avoiding the tremendous risks and costs of failed clinical trials, ending the reliance on ineffective animal testing, and accelerating speed to market.
Challenges and Opportunities
Despite its transformative potential, the widespread adoption of the Bio-AI approach faces several challenges. Foremost is the need for robust validation and standardisation of patient-onchip platforms and machine-learning algorithms. Ensuring the reliability and reproducibility of experimental results will be essential for building trust in the predictive capabilities of this approach.
Furthermore, integrating these technologies poses logistical and technical challenges that must be addressed. From optimising experimental protocols to developing interoperable data analysis pipelines, researchers must overcome numerous hurdles to realise the full potential of the Bio-AI approach.
However, with these challenges come unprecedented opportunities for innovation and collaboration. By bringing together experts from diverse fields, including biology, engineering, and computer science, researchers can leverage their collective expertise to overcome barriers and drive progress in drug development.
Summary
The Bio-AI approach represents a paradigm shift in drug development, offering a more efficient, predictive, and personalised alternative to traditional methodologies. By integrating disruptive technologies such as machine learning, high-throughput experimentation, and real-time sensing, researchers can overcome the limitations of animal testing and accelerate the pace of drug discovery.
Moreover, the Bio-AI approach holds promise for transforming clinical trial design, advancing our understanding of disease mechanisms, and improving patient outcomes. While challenges remain, the potential benefits of this approach are too significant to ignore, making it a compelling avenue for future research and innovation in drug development.
For more information about the Bio-AI technology, visit www.quris.ai
Author Bio
ISAAC, CEO of Quris-AI, is a physician and entrepreneur who has previously founded three transformative bio-AI companies in medicine and genomics. These ventures laid the foundation for the breakthroughs and innovations in drug development at Quris-AI.
