Where has the organ-on-a-chip of the target replacement animal gone?

Source: Investment Community

The last time organ-on-chips "advanced by leaps and bounds" was about two years ago. In 2022, the FDA took frequent actions: data including organ-on-a-chip was used for the first time in the FDA's new drug application; Promoted by the FDA, pharmaceutical companies and biotech companies, from the Senate to the House of Representatives, the United States Congress finally wrote non-mandatory animal testing into the bill. With several big moves by the FDA, market sentiment is rising step by step. Around 2022, it is also the time when domestic organ-chips begin to emerge, and several organ-on-chip companies have received financing.

In the past two years, with the resurgence of the AI wave, the development of organ-on-a-chip has undergone tremendous changes, and one of the changes* is the introduction of AI and machine learning. AI is being used to analyze the massive amounts of data generated by these systems, optimize chip designs, predict cell behavior, and simulate complex biological interactions. The pharmaceutical industry has increased trust in organ-on-a-chip models during drug development, expanding the scope of collaborative experiments.

On a global scale, organ-on-a-chip is becoming more and more like "TechBio" rather than "biotechnology" in the pure sense.

01 AI is a must

The introduction of AI technology has greatly improved the accuracy and efficiency of organ-on-a-chip technology. AI is able to process large amounts of biological data obtained from organ-on-a-chip to identify complex biological response patterns.

The data generated by organ-on-a-chip is complex and varied, including cell activity, metabolite levels, and electrophysiological signals. Artificial intelligence, especially deep learning algorithms, is able to process these multi-dimensional data efficiently. AI can identify subtle patterns that are difficult for humans to detect, such as subtle changes in cell behavior. It also enables the integration of different types of data, such as gene expression, protein levels, and metabolite concentrations, to draw comprehensive conclusions.

In addition, AI excels at anomaly detection, being able to identify data points that deviate from the normal state, which may indicate early signs of disease or side effects of medications. Through this comprehensive and in-depth data analysis, AI provides researchers with unprecedented insights that enable them to better understand complex biological systems.

Using machine learning algorithms, researchers can build complex predictive models. These models are able to predict the likely response of new drugs on organ-on-a-chip based on historical data, greatly accelerating the drug development process. At the same time, these models can also simulate the development of diseases in organ-on-a-chip and predict biomarker changes at different stages. What's more, AI models are able to predict the long-term effects of drugs or environmental factors, overcoming the time constraints of organ-on-a-chip experiments.

The ability of the AI system to monitor the organ-on-a-chip 24 hours a day greatly improves the accuracy and reliability of the experiment. It can monitor multiple physiological indicators at the same time, such as pH, oxygen concentration, cell morphology, etc. When any anomalies are detected, the system sends out an immediate alert, allowing researchers to intervene in a timely manner. More advanced AI systems can even automatically adjust experimental conditions based on monitoring data to maintain a state, making long-term, complex experiments more feasible and reliable. According to the needs of the cells, the AI can dynamically adjust the media composition and supply rate to ensure that the cells grow in the * state.

In addition, AI can also simulate various in vivo stimuli, such as mechanical forces and electrical signals, which can be used to study organ responses to these stimuli. This highly automated and intelligent control not only improves the accuracy and reproducibility of experiments, but also makes long-term, complex experiments more feasible.

For example, Yaosu Technology's AI algorithm can automatically analyze image data and accurately predict the effect of drugs on organs, greatly shortening the drug development cycle and improving the reliability of experimental results.

Through continuous innovation based on cell morphology AI technology, Yaosu Technology has proposed a series of application results in the field of computer vision and image analysis, and has published relevant research results at conferences such as CVPR (Computer Vision and Pattern Recognition Conference), ECCV (European Conference on Computer Vision) and SBI2 (International Society for Bioimaging and Optics). These studies demonstrate how AI technology can be used for high-throughput drug screening and organ function simulation, making it possible to perform large-scale automated analysis on a chip.

Then, organ-on-a-chip technology generates a wide variety of complex data types, including time-series data, spatial data, and multimodal data. The heterogeneity and complexity of this data pose a huge challenge to AI algorithms. When processing this data, AI systems need to perform complex pre-processing and analysis to ensure that the biological information contained in the data can be accurately captured and interpreted. For example, time-series data may reflect the dynamics of cells over time, while spatial data may reveal cell-to-cell interactions. AI algorithms must be able to integrate these different types of data to extract meaningful patterns and insights from them.

In addition, AI models, especially deep learning models, are often regarded as "black boxes", and their decision-making processes are difficult to understand and explain. This is particularly problematic in the biomedical field, where researchers need to fully understand the basis for the model's decision-making to ensure that its results are biologically meaningful and interpretable. As a result, developing explainable AI models has become a key challenge. Researchers need to design AI systems that provide a clear decision-making path that enables scientists to understand how models arrive at specific conclusions, thereby increasing trust in AI-assisted research results.

Effectively integrating AI technology with organ-on-a-chip systems requires close collaboration between biomedical engineers, computer scientists, and clinicians. This interdisciplinary collaboration requires not only breakthroughs at the technical level, but also consensus-building on methodologies and professional language. For example, computer scientists need to have a deep understanding of biological processes, while biologists need to grasp the fundamentals of AI. This cross-domain knowledge exchange and integration is a continuous process that requires the establishment of effective communication mechanisms and collaboration platforms. In addition, there is a need to develop specialized tools and frameworks to facilitate effective collaboration between experts from different disciplinary backgrounds to ensure that AI technologies can truly serve the needs of biomedical research.

02 The gap between China and the rest of the world may have widened

The United States Organ-on-a-Chip Research Phase I can be traced back to 2012, developing the most basic microarray and organoids and cells; The Phase II study began in 2015 by combining chips with cells and initiating drug testing collaborations with 40 pharmaceutical companies, including GlaxoSmithKline; Since 2017, the Phase III study has constructed a variety of disease models and screened them with drugs, and the Phase III study is nearing completion in 2022.

Although the domestic organ-on-chip industry started late, it has experienced a period of rapid development. Especially in the second half of 2022, the biotechnology track has survived the winter as a whole, but many organ-on-a-chip companies have received financing one after another, and even domestic companies have begun to take the lead in formulating various standards for organoid models and organ-on-chips.

However, from 2023 to 2024, the development of the organ-on-chip industry on an international scale has entered a new stage. On the one hand, organ-on-chip companies in Europe and the United States have seen a surge in funding, such as CN Bio, a veteran organoid and organ-on-a-chip company, which has received $21 million in financing, and organ-on-a-chip newcomer Quris, which has developed an AI-driven "patient-on-a-chip" platform that combines artificial intelligence and organ-on-a-chip technology to simulate individual patients' responses to drugs on chips.

Another emerging organ-on-a-chip company, Vivodyne, raised $38 million in a seed round. Vivodyne's technology platform enables high-throughput drug testing on thousands of functional human tissues and leverages AI for data analysis. Vivodyne can test a wide range of therapies from small molecules to advanced biologics, including cutting-edge therapies such as mRNA nanoparticles and cell therapies.

On the other hand, the cooperation between MNC and organ-on-chip companies in Europe and the United States has also increased.

Johnson & Johnson has deepened its partnership with CN Bio Innovations. The collaboration focuses on liver-on-a-chip models to improve the predictive power of drug-induced liver injury. The two companies have been working to reduce preclinical failures and enhance drug metabolism research with human-relevant organ-on-a-chip models.

AstraZeneca has expanded its partnership with Emulate. The collaboration is focused on drug safety testing using Emulate's liver and lung microarray models. This collaboration aims to address drug toxicity issues at an early stage.

Merck has partnered with Hesperos to use a multi-organ platform for cancer drug testing. These models provide a more accurate simulated microenvironment, allowing Merck to test the efficacy and toxicity of novel cancer therapies. It is worth mentioning that one of the three key technologies for future new drug research and development that Merck has invested heavily in is organoids and organ-on-chips, and the other two are AI and laboratory automation.

As one of the first MNCs to try organoids and organ-on-a-chip, Roche has continued to make efforts in the field of organ-on-a-chip technology in recent years, mainly focusing on cardiac and oncology applications. For example, Roche has partnered with TARA Biosystems to assess the cardiotoxicity of drugs using its "heart-on-a-chip" model. Roche's organoid research institute has also expanded from 50 two years ago to nearly 200 R&D personnel.

In addition, Sanofi, GSK, Pfizer, Novo Nordisk and other MNCs all use organoids and organ-on-a-chip technology in drug development. In addition, the FDA has established the OASIS Consortium, which includes 17 multinational pharmaceutical companies and 8 biotechnology companies, including Yaosu Technology, to jointly develop AI-based, multi-omics, and multimodal evaluation tools and standards for the hepatotoxicity of next-generation drugs, including organ-on-a-chip.

In contrast, the scale of organ-on-chip financing in China in the past two years has been relatively small, making the investment in the development of this original technology relatively scarce.

"Adequate capital support is a necessary condition for startups to carry out long-term technology research and development, so as to avoid being forced into the dilemma of pursuing short-term profits. In the past two years, the focus of many domestic organoid and organ-on-chip companies has shifted to areas that are relatively easy to obtain revenue, such as sales of reagents. To some extent, this shift reflects the reality of current financing difficulties. Some practitioners in related industries told Arterial Network.

In addition, it can be seen from a number of cooperation with MNC that organ-on-a-chip technology in the European and American markets is mainly used in the field of new drug research and development, which reflects its high technical level and the specific needs of pharmaceutical companies. In contrast, the domestic market is more likely to use organoid technology in relatively mature business areas such as drug susceptibility testing.

However, there is no shortage of pioneers in China, and several new drugs using organ-on-a-chip data have been approved for clinical trials.

For example, Hengrui Pharmaceutical entrusted Suzhou Medical Device Research Institute of Southeast University and Jiangsu Aiweide Biotechnology Co., Ltd. to carry out in vitro screening, and successfully screened hundreds of compounds in more than 8 months using human heart organ-on-a-chip, and the screened HRS-1893 was approved by the State Food and Drug Administration to enter clinical research.

and the QLF3108 for injection of bispecific antibody class I new drugs for tumors developed by Qilu Pharmaceutical has obtained clinical approval, and the target indication is advanced solid tumors. In the process of research and development, Beijing Big Oak Technology Co., Ltd. uses the self-developed IBAC O2 chip to build a tumor organoid immune co-culture model, which highly restores the human tumor immune microenvironment, and evaluates the efficacy of the QLF3108, providing a more effective method for the development and evaluation of drugs.

Overall, it is expected that more IND and practical application cases based on organ-on-a-chip technology will emerge in the next 3 to 5 years, especially in the field of drug safety evaluation. As the FDA's collaborative project nears completion, industry standards and regulatory access processes will become more standardized and unified to provide clear guidance for technology adoption. In terms of technology application, drug repositioning studies based on high-throughput organ-on-a-chip screening may make breakthroughs.

Therefore, from a rational point of view, non-animal technology is still in its infancy, and it is still necessary to rely on traditional means to promote the development of non-clinical research. In addition, everyone is actively embracing innovation, hoping to better verify the reliability of non-animal technologies through clinical data and animal data in the future. In the past two years, the combination of organ-on-a-chip technology with cell therapy and tumor immunology will accelerate the marketization process of related therapies and provide strong support for the development of more innovative treatment options.

[This article is authorized to be published by Arterial Network, a partner in the investment community, and this platform only provides information storage services. If you have any questions, please contact ([email protected]) the investment community.