Fermion AI Biotech Secures Two BIC Pipeline IND Approvals in Early 2024: An Interview with Dr. Deng Daiguo

In 2023, the world's top ten MNCs all emphasized the important role of AI in innovative drug development in their annual reports. They continued to invest strategically in AI-driven drug discovery and constantly explored commercial validation of AI in pharmaceuticals. Meanwhile, multiple global AI drug pipelines have gradually entered clinical trials, and the AI pharmaceutical industry is transitioning from the clinical validation phase to product commercialization. Breakthrough advancements in large models and generative AI have brought new opportunities for the development of the AI drug discovery sector.

Guangzhou Fermion Technology Co., Ltd. ("Fermion"), which began operations in 2019, is a leader in China's AI pharmaceutical sector. As an innovative drug development company driven by AI and data, Fermion leverages its self-developed Drug Studio AI drug discovery platform to achieve breakthroughs in ultra-high target selectivity and tissue targeting. By reducing drug off-target effects and enhancing precise distribution in target tissues, it improves the safety of clinical medications. The company has realized industrial application and built an integrated AI-driven drug R&D solution. Since its establishment, Fermion has developed 13 differentiated BIC/FIC small molecule drug pipelines focusing on early innovative targets in the central nervous system (CNS) and autoimmune fields.

In October last year, Fermion signed an agreement with Joincare Pharmaceutical Group Industry Co., Ltd. ("Joincare"), granting the listed company Joincare exclusive rights to FZ008-145, a novel analgesic independently developed by Fermion, in Greater China, achieving the commercial closed-loop of the AI Biotech model.

Since the beginning of this year, Fermion's non-addictive analgesic pipeline Nav1.8 inhibitor FZ008-145 and autoimmune pipeline TYK2 JH2 inhibitor FZ007-119 have successively received IND approval. Thus far, since its establishment, Fermion has obtained three IND approvals, demonstrating extremely high efficiency in new drug development. Among them, the fastest progressing non-addictive chronic pain pipeline SSTR4 inhibitor FZ002-037 has already completed Phase I clinical trials by the end of 2023 and is about to enter Phase II clinical POC (proof of concept) studies this year.

Looking back at the past few years, AI has demonstrated its role in empowering drug research and development across various scenarios.，The penetration of AI technology is gradually broadening the boundaries of application, but controversies persist in the AI pharmaceuticals field. How can the business model of AI pharmaceutical companies succeed? What are the choices for disease areas in AI pharmaceutical enterprises? How to rapidly advance commercial implementation? What are the innovation trends in the AI pharmaceuticals sector?...with curiosity about the "cultivation method" of Fermion, an AI Biotech that has recently received consecutive IND approvals,VCBeat had the honor of conducting an exclusive interview with Dr. Deng Daiguo, founder of Fermion.

Dr. Deng Daiguo, Founder of Fermion

(Photo generated by Fermion AI technology)

VCBeat: How do you view the role of AI in new drug development? What kind of AI technology is needed to support the innovative drug field?

Dr. Deng Daiguo: Looking globally, the exploration of AI in new drug development scenarios has involved multiple aspects, such as molecular generation and optimization screening, dosage form design, clinical trial design optimization, clinical outcome prediction, virtual clinical trials, and real-world research. However, in terms of application results, to be honest, the current maturity of AI pharmaceuticals still has a noticeable gap compared to expectations.

Life systems are very complex, and the research and development of innovative drugs is also a systematic project, involving numerous disciplines and a large number of basic scientific research problems that need to be explored and solved. Regardless of what new technology it is, the essence of drug development remains unchanged: to provide patients with more effective and safer treatment options. The development of technology is always incremental. At this stage, the most effective way for AI in innovative drug R&D isIntegrate into the biopharmaceutical ecosystem, integrating multi-disciplines such as artificial intelligence, biology, medicinal chemistry, and medicine, making AI a powerful tool to empower innovative drug research and development. The biggest challenge for AI pharmaceutical companies is not about which technology is used, but how to create high-value products, solve practical problems with AI, and ultimately develop commercially competitive products.

Fermion chose small molecule innovative drug development because the theory of small molecule drugs is relatively mature, with large-scale and easily standardized data available for AI to learn. In the drug discovery phase, AI technology is combined with traditional drug research and development through a combination of dry and wet experiments. AI optimizes molecular selectivity and tissue targeting, solving drug safety issues by reducing off-target effects and improving precise distribution in target tissues.

VCBeat: As a serial entrepreneur in the application of AI technology, what aspects do you think AI pharmaceutical companies need to focus on during their development?

Dr. Deng Daiguo: As a team of serial entrepreneurs, we have our own business judgment on AI + new drug development: the value of drug pipelines is relatively low in the early stages of new drug research and development, and the core commercial value is often obtained in the later stages. In 2019, when Fermion began formal operations, it explored and iterated through three business models—AI SaaS, AI CRO, and AI Biotech (selling software, services, and drug development). By 2020, it decided to position itself as an AI Biotech company. The characteristics of this industry dictate that maximum value can only be captured by engaging in the backend processes. AI Biotech spans the preclinical and clinical trial stages of drug development. Therefore, after analyzing from multiple perspectives, such as creating commercial value and user value, AI Biotech is the best choice.

At the same time, any emerging technology, including the application of AI, has its boundaries. The system for innovative drug research and development is complex. For AI-driven drug discovery, it is necessary to identify the application scenarios of AI technology based on an understanding of unmet clinical needs, while considering the competitive landscape and potential transactional collaborations within specific disease areas. The key to the commercial implementation of Fermion's AI technology lies in a small entry point and rapid iteration—focusing on mining knowledge and data within specific fields.To make AI technology and business closely integrated, more vertical to form a stronger core competitivenessThe advantages of AI technology can be utilized to select potential indications and targets for drug development that address unmet clinical needs, which will be more beneficial for the commercial development of AI.

VCBeat: In your opinion, what is the core reason that Fermion has been able to achieve IND in a short period of time?

Dr. Deng Daiguo: Since its operation in 2019, Fermion has established a model that combines AI technology with traditional drug development. Our self-developed Drug Studio AI drug discovery platform has achieved industrial application, supported by a complete theoretical and technical foundation. Specifically, after identifying a target, we use our proprietary molecular OCR tool, Img2Chem, to automatically extract real-world molecules from patents and journals related to the target, constructing a structured database for that target. Focusing on target selectivity and tissue targeting, we then generate molecules, perform patent conflict detection, virtual screening, and property prediction to obtain the top 10 molecules with optimal comprehensive parameters. We emphasize talent density, maintaining a team size of around 20 people. Core resources are allocated to computation and pipeline design, while experimental work is conducted in collaboration with well-known domestic and international CROs, enabling rapid validation and iteration of dry and wet experiments, efficiently screening multiple PCCs (preclinical candidate compounds) that meet the set technical criteria.

Currently, more than ten pipelines of small molecule drugs focusing on ultra-high selectivity and tissue targeting have been developed. The R&D process to deliver PCC takes about 10 months, with costs around 4.5 million yuan. Compared with the traditional R&D model, the cycle is shortened by 50%, costs are reduced by 70%, and preclinical development efficiency is significantly improved. In the IND-enabling stage and later clinical development, we also closely collaborate with high-quality external resources to fully advance project development. Our TYK2 JH2 inhibitor FZ007-119, which was approved in early February, received IND approval in just over two years from its initiation in May 2021, showcasing Fermion's extremely fast new drug R&D speed.

VCBeat: Why does Fermion focus on CNS and autoimmune fields? What pain points in current disease treatment are the existing R&D products expected to solve?

Dr. Deng Daiguo：The development cycle of innovative drugs is long and the investment is high. For innovative pharmaceutical companies, selecting a disease area is crucial. Fermion has chosen to focus on the CNS field for three main reasons: First, compared to other disease areas, there are limited therapies in the CNS field. The existing drugs are mostly older ones that have been on the market for a long time, and their efficacy and safety urgently need improvement. There is a huge unmet clinical demand. Additionally, influenced by factors such as aging and living environments, the number of CNS disease patients worldwide is increasing annually, offering significant growth potential in both domestic and international markets. Second, our AI platform's underlying algorithms have undergone extensive development and accumulation specifically targeting the safety of CNS products. We possess unique algorithmic advantages, especially in constructing target knowledge graphs, optimizing candidate compounds, and evaluating safety screenings. We have developed molecular heterogeneous graph algorithms with higher prediction accuracy for ultra-high selectivity and substructure heterogeneous graph algorithms based on structural information and physicochemical properties for tissue targeting, which can efficiently screen out candidate molecules with prominent target selectivity and tissue targeting, thereby enhancing the safety of clinical medications. Third, most of our team members have a background in CNS drug research and development, possessing certain "know-how" and experience, with over 20 project development experiences in this field, equipped with comprehensive preclinical and clinical development capabilities.

Our fastest progressing non-addictive chronic pain pipeline, FZ002-037, is a small molecule somatostatin receptor 4 (SSTR4) agonist with potential for treating chronic neuropathic pain, inflammatory pain, and mixed pain. Compared to traditional opioid analgesics and non-steroidal anti-inflammatory drugs like ibuprofen, FZ002-037 offers three differentiated advantages: first, the novel non-addictive target SSTR4, which has already been clinically validated with Proof of Concept (POC); second, its primary action on peripheral nerve tissues, which can effectively enhance drug safety; third, it exhibits ultra-high subtype selectivity, reducing off-target side effects. The Phase I clinical data we have completed shows favorable pharmacokinetic properties and safety.

Our another non-addictive analgesic pipeline, FZ008-145, is the world's second and China's first highly selective second-generation Nav1.8 inhibitor. In addition to its safety and potency advantages, it also exhibits excellent PK properties. Compared with the first-generation Nav1.8 inhibitors, it can provide better analgesic effects at lower doses while avoiding side effects such as addiction. Currently, the most effective acute analgesics are still opioid drugs, but their addictive side effects have led to strict clinical restrictions, resulting in a significant unmet clinical need. Nav1.8 inhibitors are expected to fill this gap.

At the same time, we have laid out more than five pipelines in autoimmune diseases, with target indications including psoriasis, inflammatory bowel disease, rheumatoid arthritis, and more. The TYK2 JH2 inhibitor FZ007-119, approved in February 2024, is the fastest progressing pipeline among them. FZ007-119 exhibits over ten thousand times selectivity for JAK1-3 compared to Deucravacitinib, the world's first drug for this target, showing extremely high potential to translate into clinical benefits such as further improving clinical efficacy and enhancing medication safety. Its ultra-high selectivity gives FZ007-119 the potential to become the best-in-class (BIC). In the future, it can further expand the development of indications such as inflammatory bowel disease (IBD), demonstrating broad market prospects.

VCBeat: What milestone developments can Fermion expect in its R&D pipeline in the coming years?

Dr. Deng Daiguo: The non-addictive chronic analgesic pipeline FZ002-037 is Fermion's leading pipeline and the world’s second, as well as China’s first, SSTR4 agonist to enter clinical trials. The Phase I clinical trial of FZ002-037 has achieved excellent safety data, and it will soon enter Phase II clinical research this year. Combined with the established adaptive Phase II/III clinical protocol, it is believed that it will quickly bring a new treatment option for its lead indication, neuropathic pain. Meanwhile, FZ002-037 is in negotiations with multiple overseas pharmaceutical companies regarding licensing partnerships, with expectations to achieve Guangzhou Fermion Technology Co., Ltd.’s first overseas business development collaboration.

In the CNS fieldAnother non-addictive analgesic pipeline Nav1.8 inhibitor FZ008-145, which we have developed, submitted an IND application to the FDA in March this year, and overseas clinical research will be launched in the second half of the year; at the same time, the domestic rights partner Healthoo has also initiated Phase I clinical research in China.

In the field of autoimmuneThe TYK2 JH2 inhibitor FZ007-119 developed by us has initiated Phase I clinical research in China and will submit an IND application to the FDA in the near future. In addition, the company’s topical formulation product for treating autoimmune skin diseases is planned to submit an IND application to the NMPA in the second half of the year, and another pipeline for treating multiple sclerosis and Parkinson's disease will also commence IND-enabling work.

In the future, we will focus on early innovative targets in the central nervous system (CNS) and autoimmune fields, continuously developing a differentiated portfolio matrix of BIC/FIC small-molecule drugs. At the same time, through pipeline rights transfers or collaborative development, we will actively build a global partnership network to advance our product pipelines into clinical stages, providing more and better treatment options for diseases.

VCBeat: How to Promote the Completion of the AI Pharmaceutical Commercialization Closed Loop?

Dr. Deng Daiguo: Whatever the new technology, the essence of making medicine remains unchanged, which isDevelop high-value products and successfully launch them to the market,Benefiting patients.The chain of new drug research and development is very long, requiring a lot of capabilities, and the commercial landing path of AI pharmaceuticals is particularly important. In recent years, the development of AI technology has been very rapid, but the application of any technology has its boundaries.Focus on specific areas, solve key problems, and closely integrate AI technology with business., becoming more vertically integrated to form a stronger core competitiveness. Small incisions and rapid iterations are the key to the commercial implementation of AI technology.The advantages of applying AI technology can be realized by targeting unmet clinical needs, selecting promising indications and targets, and developing better drugs, which will be more beneficial to the commercial development of AI-driven drug discovery.

VCBeat: How do you view the ups and downs experienced by the AI pharmaceuticals track in the past few years? What are your predictions for the next few years?

Dr. Deng Daiguo:The development of AI technology is propelling AI-driven drug discovery forward, from multinational corporations (MNCs), to biotech companies, and then to traditional pharmaceutical firms. The penetration of AI technology is gradually deepening. However, AI-driven drug discovery is still in the nascent stage of its lifecycle. The advancement of emerging technologies progresses in a spiral manner amidst "bubbles." The fluctuations in the environment over the past few years have accelerated the process of screening and淘汰 within the industry. This has provided greater clarity for AI-driven drug discovery practitioners, pharmaceutical companies, investors, and policymakers alike, helping those truly valuable AI applications secure more resources in future developments. Value and price will not remain mismatched indefinitely; as companies steadily achieve results during their development, they will eventually be recognized.

Innovation is crucial for the biotechnology industry. With the advancement of life science basic research, the iterative upgrading of AI technology, and improvements in economic environment and policies, it is believed that the AI pharmaceuticals industry will usher in a new wave of development. Particularly, new drugs or technologies that can address unmet clinical needs or transform treatment methods—high-value product innovation will always gain market recognition.

From the perspective of company development, due to the long R&D cycle and high investment required for new drugs, it is necessary to balance the balance sheet. We are also in talks with several potential foreign partners for some of our projects, and 2024 will be a year of significant progress for the company's overseas BD.

We will continue to innovate, maintain strategic resolve, and take solid steps to achieve each development milestone, providing more and better treatment options for diseases as soon as possible.