Integrated BioAI Platforms Address Drug Development Bottlenecks

From artificial intelligence to patients-on-chip, new technologies decrease reliance on animal testing

Photo portrait of ISAAC BENTWICH, MD
Isaac Bentwich, MD
Photo portrait of ISAAC BENTWICH, MD

Isaac is the founder and CEO of Quris, where he and his team are using a bioAI approach to disrupt the drug development process. Prior to Quris, Isaac founded and led three bioAI technology companies, each of which led revolutions in medicine, genomics, agriculture, and conservation. He is a physician and entrepreneur with a passion for leading interdisciplinary teams of scientists and technologists to tackle impactful challenges in the intersection between machine learning and life sciences, and to leverage and commercialize the resulting solutions.  

ViewFull Profile
Learn about ourEditorial Policies.
Published:Dec 06, 2022
|3 min read
Register for free to listen to this article
Listen with Speechify
0:00
3:00
Photo portrait of ISAAC BENTWICH, MD
Isaac Bentwich, MD, is the founder and CEO of Quris, where he and his team are using a bioAI approach to disrupt the drug development process. Prior to Quris, Isaac founded and led three bioAI technology companies, each of which led revolutions in medicine, genomics, agriculture, and conservation.

A major impediment for drug discovery is our limited ability to successfully predict the clinical safety and efficacy of drugs from animal testing before they are taken to clinical trials with human participants. Almost 90 percent of drug candidates that successfully pass animal testing fail in clinical trials, and these failed trials translate to a loss of billions of dollars. With increased efforts to minimize animal testing, including through the FDA Modernization Act, which was unanimously approved by the Senate, and increased regulations by the European Union, how will these changes impact drug development? 

Moving beyond animal models of disease

Moving beyond animal models and the data tied to those models will be a major undertaking for the pharmaceutical industry. Artificial intelligence (AI), including machine learning (ML), has become a leading frontier for pharma innovation and has shown significant success in various stages of drug discovery and development. 

However, despite the adoption of new technologies like AI and organ-on-a-chip devices, there is at present no platform that can accurately predict the clinical safety of drug candidates. Why? AI pharma companies rely primarily on two-dimensional biology, which leads to poor predictions of drug safety, while current organ-on-a-chip devices overlook the value of AI/ML.

Traditional data, based on two-dimensional (2D) biology (combinations of publicly available data, data collected by other pharma and health organizations, and proprietary data they generate) that AI pharma companies use to train their models cannot predict clinical safety for humans, as the data itself is only based on animal studies. Traditional animal testing is so inaccurate at predicting what will actually work in the human body that 90 percent of drugs that enter clinical trials fail. AI/ML platforms are only as strong as the data they are trained on. 

Patient-on-a-chip technology requires AI/ML integration

The transition to non-animal data relies on new, more accurate testing methods. Early 3D organoid technologies proved more accurate than conventional 2D biology in testing potential drug toxicity, and the jump to organ-on-a-chip models allowed drug developers to even better mimic specific organs like the liver. 

While sophisticated, 3D miniaturized organ technologies have matured dramatically in the last few years, their lack of scalability is a major barrier to meeting the demand for data. In addition, these technologies often still rely on manual analysis of “end-point” data collected at the end of a single experiment rather than collected in real-time during the experiment. 

The next innovation, patient-on-a-chip (i.e., multiple, interconnected organs-on-a-chip) technology, adds even more complexity and expense—severely limiting any move toward implementing this technology on industrial scales without AI/ML as part of the process. Unlocking the full potential of patient-on-a-chip technology requires AI/ML.

The benefits of integrated, advanced technology platforms

BioAI platforms, which fully integrate critical new technologies including AI/ML, organ- or patient-on-a-chip platforms, and real-time nano-sensing, while applying stem cell-derived genomic diversity, offer a clear path forward for advancing drug discovery. 

With ML at the center, integrated BioAI prediction platforms open the door for performing thousands of patient-on-a-chip experiments in parallel, generating massive amounts of predictive, real-time, nano-sensing data that can be used to train ML algorithms to accurately predict drug safety. 

For example, stem cell experiments performed on thousands of different patient-on-a-chip platforms can be used to capture a population’s genomic diversity and personalize drug selection. The future of clinical prediction lies in the ability to reliably and cost-effectively run thousands, even millions, of such clinical prediction experiments.

Enhancing drug development with integrate, advanced technology platforms

Well-timed with current regulatory pressures, the creation of integrated, advanced technology platforms will create a surge in drug innovation, while ensuring improved drug safety, better patient outcomes, and more efficient drug development. These benefits have the potential to span all disease areas, accelerating a drug candidate’s path to market, enabling rare disease drug discovery, and repurposing existing drugs, thus enhancing drug development and accessibility for the general population.


Isaac Bentwich, MD
Isaac Bentwich, MD

Isaac is the founder and CEO of Quris, where he and his team are using a bioAI approach to disrupt the drug development process. Prior to Quris, Isaac founded and led three bioAI technology companies, each of which led revolutions in medicine, genomics, agriculture, and conservation. He is a physician and entrepreneur with a passion for leading interdisciplinary teams of scientists and technologists to tackle impactful challenges in the intersection between machine learning and life sciences, and to leverage and commercialize the resulting solutions.  


Tags:

Personalized MedicinePrecision MedicineInformaticsArtificial IntelligenceMachine Learningmodels and simulationsorganoidsorgan-on-a-chip
Top Image:
Moving beyond animal models and the data tied to those models will be a major undertaking for the pharmaceutical industry.
iStock, shuoshu