FRONTEO, extracted by AI drug discovery support service 'Drug Discovery AI Factory'...

　FRONTEO Inc. (Headquarters: Minato-ku, Tokyo; Representative Board of Director and President: Masahiro Morimoto; hereinafter, FRONTEO) has launched the AI drug discovery support service "Drug Discovery AI Factory (hereinafter, DDAIF)".^*1Drug discovery target molecules for pancreatic cancer using^*2Candidates were extracted and in vitro (test tube) tests were conducted to confirm a certain degree of efficacy in inhibiting the proliferation of cancer cells.^*3This experiment was conducted by our company with the aim of proving the effectiveness of DDAIF in extracting associations between target molecules and diseases that have not been described in the literature.

■ Reduced target molecule search time from approximately two years to two days, and searched for numerous highly novel target molecule candidates
In this experiment, DDAIF was used to successfully extract 2 candidate target molecule genes from a total of approximately 2 human genes in just two days. In conventional drug discovery approaches, the "target discovery" process to extract such target molecules often takes more than two years, and it is said to be extremely difficult to find highly novel target molecules that have not been described in the literature, even if you spend a lot of time doing so. The use of DDAIF technology to extract candidate target molecules in a short period of time makes it possible to dramatically accelerate the "target discovery" process, which is the most important step in drug discovery.

Furthermore, of these 17 genes, six genes were confirmed to inhibit the proliferation of pancreatic cancer cells in experiments, four of which have no published papers reporting their association with pancreatic cancer, and the remaining two genes have only been reported in one paper (as of April 6, 4).^*4This was an extremely novel candidate target molecule.
Traditionally, target discovery has been carried out by thoroughly examining the literature to find promising target candidates, but it is said to be extremely difficult to find highly novel target molecules that have not been described in the literature, even if one spends a lot of time on this.

The results of this validation prove that FRONTEO's in-house developed AI "KIBIT" is capable of discovering associations between unknown drug discovery target molecules and diseases from known literature, and we believe that the technology that can extract candidate target molecules in a short period of time will dramatically accelerate the "target discovery" process, which is the most important step in drug discovery.
FRONTEO will continue to accelerate the further promotion and social implementation of DDAIF, which will be a new approach to treating difficult-to-treat diseases including pancreatic cancer.

Pancreatic cancer: an area with extremely high unmet medical needs
　Pancreatic cancer has a 5-year survival rate of less than 10%^*5In addition, there are limited treatment options, and patients have to rely on highly toxic chemotherapy drugs. In addition, there are few treatment options when chemotherapy drugs become ineffective.^*6It is one of the most common diseases.

■ Extraction of many highly novel target molecule candidates
　Of the six genes listed above for which cell proliferation inhibition was confirmed, no papers had reported a relationship between four genes and pancreatic cancer. There was only one paper on two genes, so we were able to extract a large number of highly novel target molecule candidates.

■ Cell proliferation inhibition effect
　The effects of the 17 extracted genes on cell proliferation were examined in pancreatic cancer cell lines. It was found that when the activity of six of these genes was inhibited, cancer cell proliferation was suppressed by approximately 6 to 4 percent compared to when nothing was done.
　This ratio is comparable to that of KRAS, a gene known to cause pancreatic cancer and have a particularly important effect on promoting cell carcinogenesis, and may have great potential as a candidate target molecule for drug discovery in pancreatic cancer.

Going forward, we plan to work with co-creation partners, including Tokyo University of Science, to clarify the mechanism of action of the target molecule candidates that have shown efficacy and to verify them through animal testing. In the future, we plan to license them out to pharmaceutical companies.^*7We will also consider this.

Comment from Masayuki Murata, Specially Appointed Professor at the Graduate School of Science, Tokyo University of Science
　"I think that the ability to narrow down unreported drug discovery targets to multiple candidates in a short period of time - a matter of days depending on the subject of analysis - is an innovative technology that would not be possible with conventional approaches. Although the results of this validation study are at the cellular level, I believe that this is an example that shows that KIBIT can greatly contribute to the advancement and acceleration of drug discovery research and to improving the success rate. Our research group at Tokyo University of Science is developing a variety of cell function analysis and evaluation methods with the aim of gaining a deeper understanding of the life phenomena that are the foundation of drug discovery and cell therapy.
In the future, we expect that by systematically verifying the functions of the target molecule candidates narrowed down by FRONTEO through cell line experiments through our ongoing collaborative research, we will be able to accelerate drug discovery through the identification of functionally effective molecules.

Challenges in drug discovery
　In June 2025, the Cabinet Office established the "Public-Private Council for Improving Drug Discovery Capacity"^*8As symbolized by the establishment of the Center for Drug Discovery and Development, strengthening drug discovery capabilities is an urgent national issue for Japan. Japan has had a long-standing trade deficit with pharmaceutical companies.^*9It is urgent to establish a system that is not dependent on overseas sources for the supply of pharmaceuticals.

[Source] Cabinet Office: Initiative meeting to rapidly deliver the latest medicines to the public by improving drug discovery capabilities, Document 5 Current state of drug discovery in Japan and the world*9

　However, drug discovery research has already reached maturity, and many drugs that are relatively easy to develop have already been launched on the market. However, as diseases become more complex and diverse, the search for new drug targets is becoming more difficult every year, and the "exhaustion" of target molecules is becoming a serious issue.
　Furthermore, overseas, there are moves to limit the number of generic drugs that can be approved, and the number of “first-in-class” drugs is on the rise.^{* 10}The development of new drugs with a strong first-mover advantage, known as "potential first-mover advantage," has become an essential theme for corporate competitiveness and, ultimately, for the nation's pharmaceutical strategy. To achieve this, it is essential to discover "unknown target molecules" that are not described in papers or databases, but identifying such information has been extremely difficult using conventional research methods or AI.

■Significance of this project using DDAIF
　FRONTEO's "Drug Discovery AI Factory (DDAIF)" addresses these structural challenges by utilizing its unique natural language processing AI technology, enabling innovative drug discovery support that can extract "unknown target molecules" that have not been reported in papers.
　FRONTEO will continue to contribute to solving structural issues in the field of life sciences and the sustainable development of the medical industry through research and development of AI specialized in natural language processing and its social implementation.

Notes:*1 An AI drug discovery support service in which FRONTEO's drug discovery experts, who are well versed in AI and drug discovery, utilize the natural language processing technology and unique analysis methods of the company's specialized AI "KIBIT" to search for target molecules and indications and provide supporting hypotheses.*2 The molecule (gene) on which a drug acts.*3 It was confirmed that temporarily suppressing the activity of candidate drug discovery target molecules extracted by DDAIF significantly inhibited the proliferation of pancreatic cancer cells.*4 PubMed (A database of biomedical papers operated by the National Center for Biomedical Information at the U.S. National Library of Medicine.https://pubmed.ncbi.nlm.nih.gov/) Search the title and abstract of the article using the gene name and "pancreatic cancer" (as of April 2025, 4)*5 National Cancer Center: Latest Cancer Statistics, https://ganjoho.jp/reg_stat/statistics/stat/summary.html*6 Need for new drugs and treatments for diseases for which no effective treatment has been found*7 In addition to conducting research and development in-house, pharmaceutical companies may also acquire development and sales rights for drug candidate substances from other companies or organizations. When other companies or organizations grant such rights to pharmaceutical companies for the latter purpose, this is called licensing out.* 8　https://www8.cao.go.jp/iryou/kanmin_kyogikai.html*9 Cabinet Office: Initiative to Rapidly Deliver the Latest Pharmaceuticals to the Public by Improving Drug Discovery Capabilities, Document 5: Current Status of Drug Discovery in Japan and the World, https://www.cas.go.jp/jp/seisaku/souyakuryoku/dai1/siryou5.pdf*10 The first new drug approved in its category.

About FRONTEO Drug Discovery AI Factory (DDAIF) URL:https://lifescience.fronteo.com/products/drug-discovery-ai-factory/

"FRONTEO Drug Discovery AI Factory (DDAIF)" is an AI drug discovery support service that combines the knowledge of FRONTEO's drug discovery researchers and AI engineers with "KIBIT" (patented in Japan and the US), an AI specialized in natural language processing. It strongly supports researchers' decision-making in drug development by analyzing disease-related gene networks and building hypotheses about target candidates.
This service has already been introduced at several major pharmaceutical companies and is accumulating a proven track record.

[Reference: Co-creation project with pharmaceutical companies]・ FRONTEO andMaruhoLaunches co-creation project to discover drug discovery targets in the dermatology field using Drug Discovery AI Factory, https://www.fronteo.com/pr/20250710- Fuji Pharmaceuticaland FRONTEO launch joint project to evaluate drug discovery seeds in the field of women's health, https://www.fronteo.com/pr/20250709- Metagene Therapeuticsand FRONTEO begin joint research into microbiome drug discovery, which is attracting worldwide attention, https://www.fronteo.com/pr/20250630・ FRONTEO andChugai PharmaceuticalLaunched a co-creation project on target discovery using Drug Discovery AI Factory, https://www.fronteo.com/pr/20250515 - EA Pharmaand FRONTEO launch joint project to explore drug target discovery using AI, https://www.fronteo.com/pr/20250512・ FRONTEO andエーザイLaunched a co-creation project on target discovery using Drug Discovery AI Factory, https://www.fronteo.com/pr/20250128 ・ FRONTEO andMaruishi PharmaceuticalLaunched a co-creation project on biomarker discovery using Drug Discovery AI Factory, https://www.fronteo.com/pr/20250109・ FRONTEO andUBE, Started a co-creation project on drug repositioning using Drug Discovery AI Factory, https://www.fronteo.com/pr/20241114

[Reference: Joint research projects with academia]・ FRONTEO andTokyo University of ScienceStarted joint research on discovery of new drug discovery targets using "Drug Discovery AI Factory", https://www.fronteo.com/pr/20250513・ FRONTEO andKumamoto UniversityStarted joint research to discover new cancer treatments using Drug Discovery AI Factory, https://www.fronteo.com/pr/20250508

■About FRONTEO Co., Ltd.　URL: https://www.fronteo.com/FRONTEO supports the judgment of experts in various fields who face social issues day and night through the provision of its proprietary specialized AI "KIBIT," and creates the starting point for innovation. Unlike general-purpose AI, our proprietary natural language processing technology (patented in Japan, the United States, and Europe) enables high-speed and high-precision analysis without relying on the amount of training data or computing power. In addition, by utilizing patented technology that maps (visualizes the structure) the analyzed information, "KIBIT" can directly influence the insights of experts, and in recent years, KIBIT's technology has also been used in hypothesis generation and target discovery in drug discovery.

Through KIBIT's unique technology and approach, we aim to realize our philosophy of "providing solutions that do not overlook risks and opportunities hidden in records, and realizing fairness in the information society."Life science AI, Risk Management (Compliance Support Area,economic security field,Legal Tech AI field), DX (Professional Support Areas) and promotes social implementation in each of our businesses.

Founded in August 2003, listed on the Tokyo Stock Exchange Mothers (now Tokyo Stock Exchange Growth) on June 8, 2007. Has operations in Japan, the United States, South Korea, and Taiwan. Obtained a Type 6 medical device manufacturing and sales license and filed a controlled medical device sales notification. Capital: 26 yen (as of March 899,176, 2025).

*FRONTEO has obtained a total of 21 patents in Japan, South Korea, the United States, and Europe for the technology used in Drug Discovery AI Factory.
*FRONTEO, KIBIT, and Drug Discovery AI Factory are trademarks or registered trademarks of FRONTEO in Japan, Europe, the United States, and Korea.