[Edaily Reporter Kim Jinsoo ] Cancer treatment is increasingly shifting toward a model where the patient’s tumor genes are analyzed to select the most appropriate therapy. Consequently, the key factor determining the competitiveness of AI-driven precision medicine companies is shifting from the “sophistication of AI algorithms” to “how long and how extensively they have worked with actual cancer patient samples.” In response, domestic companies such as Macrogen, Inc.(038290) and NGeneBio Co., Ltd.(354200)are advancing AI-driven precision medicine based on data.

According to the pharmaceutical and biotech industry on the 15th, the next-generation sequencing (NGS) market is growing rapidly as the development of targeted anticancer drugs gains momentum worldwide.

The National Center for Biotechnology Policy Research projected that, as of 2024, the clinical NGS market will grow at an average annual rate of 11.6%, reaching approximately $12 billion (about 18.5 trillion won) by 2030. Given that NGS accounted for about 10% of the total clinical diagnostics market in 2022, its share is estimated to have increased further.

A key driver behind the growth of the NGS market is regulation by the U.S. Food and Drug Administration (FDA). When approving anticancer drugs, the FDA requires that corresponding genetic diagnostic tests (companion diagnostics, or CDx) be approved alongside them. This creates a structure where demand for diagnostics follows the release of each targeted anticancer drug.

As the market grows, the criteria for competition among AI-driven precision medicine companies are also shifting. While the focus has traditionally been on how sophisticated AI algorithms are, it is gradually shifting toward “how much data a company possesses.”

The Key Is “Genomic Data”

The human body contains approximately 25,000 genes. In cancer patients, mutations occur in 400 to 500 of these genes. NGS identifies these mutated genes and analyzes which treatment will be most effective.

Genetic analysis requires the ability to accurately and quickly translate vast amounts of genomic data into “clinical language.” If the sample quality is poor or the tissue volume is insufficient, the test itself becomes impossible; it is also common for the analysis to take several days, causing patients to miss the optimal window for treatment.

The key point is that this data cannot be accumulated in a short period of time. This is because data cannot be purchased with capital, nor can it be generated by AI. Since mutation patterns vary depending on various factors, such as ethnicity, the process of handling actual patient samples and validating the results in clinical settings must be repeated over several years before the data becomes “real data” that AI can utilize.

Temperus AI, a U.S.-based AI precision medicine company, was recognized with one of the industry’s highest valuations during its 2024 Nasdaq listing. The market’s high valuation of Temperus AI stems from the company’s asset base of millions of clinical and genomic data points accumulated over approximately 10 years in collaboration with major cancer centers.

Guardant Health, a liquid biopsy company, has also expanded its insurance coverage in the U.S. based on data accumulated through its blood-based diagnostic kits. Foundation Medicine started with NGS panels for solid tumors and, by gradually adding blood-based recurrence monitoring tests and AI-powered mutation interpretation capabilities, has established itself as a key partner for pharmaceutical companies.

Companies leading the way in “data acquisition” in South Korea

are... In South Korea as well, as AI-driven precision medicine companies ramp up their operations, securing data has become directly linked to a company’s competitiveness.

Macrogen, Inc. held a completion ceremony last month for the “Songdo Global Genome Center.” With a total floor area of 18,500 square meters (approximately 5,600 pyeong) and eight stories above ground, it serves as a precision medicine hub that handles everything from genomic big data production to AI interpretation and clinical analysis under one roof. It is also equipped with automated and standardized facilities capable of performing next-generation multi-omics analyses, such as single-cell analysis and spatial transcriptomics.

A key figure to note is “300,000 samples per year”—the current sequencing volume handled by Macrogen, Inc. The company has also secured the infrastructure necessary to handle large-scale national projects, such as the National Integrated Bio-Big Data Construction Project (K-DNA). Macrogen, Inc. is looking ahead to a genomic big data project covering 10 million people. The genomic analysis infrastructure and data accumulated over the 29 years since its founding in 1997 serve as the foundation for its transition to AI.

NGeneBio Co., Ltd. has entered the stage of directly combining AI and NGS and then validating the results using clinical data.

At the end of last month, NGeneBio Co., Ltd. signed a joint research agreement with Professor Kim Tae-jeong’s team in the Department of Pathology at Catholic University of Korea’s Yeouido St. Mary’s Hospital to develop an “AI-NGS Integrated Companion Diagnostic Solution.” The system works as follows: the AI pathology analysis software “NPAS (NPathAnalySys)” first screens tissue pathology images to identify patients with a high likelihood of genetic mutations, after which the NGS precision diagnostics software “NGAS (NGeneAnalySys)” performs the final genetic analysis and clinical interpretation.

The “NPAS-EGFR” model for predicting EGFR mutations in lung cancer incorporates “EXAONE Path 2.0,” for which NGeneBio Co., Ltd. secured an exclusive license from LG AI Research earlier this year.

Validation is conducted by directly comparing the system with actual prescription data from Thermo Fisher’s “Oncomine”—a global standard companion diagnostic product—to confirm equivalence. Through this process, the company plans to secure real-world evidence (RWE) that goes beyond merely aiding interpretation and can be used in actual prescribing decisions.

The research is being led by Professor Kim Tae-jeong of Yeouido St. Mary’s Hospital at The Catholic University of Korea, who serves on the Pathology Committee of the International Association for the Study of Lung Cancer (IASLC). NGeneBio Co., Ltd. plans to complete the validation by the end of the year and apply for digital medical device approval.

NGeneBio Co., Ltd.’s confidence is backed by concrete data. In 2017, NGeneBio Co., Ltd. became the first company in Korea to receive product approval from the Ministry of Food and Drug Safety (MFDS) for “BracaAccuTest,” an NGS test for breast and ovarian cancer. Since then, the company has expanded its portfolio to include “HimAcuteTest,” a panel for five types of blood cancers, and “OncoAcutePanel,” a solid tumor panel jointly developed with Asan Medical Center in Seoul and the Harvard Dana-Farber Cancer Institute. Through ongoing joint research and patent collaborations with numerous hospitals, the company has built up its assets by accumulating clinical and genomic data from the medical field.

NGeneBio Co., Ltd. plans to establish a recurring revenue model by introducing a fee structure based on the number of AI-powered diagnostic tests performed. Starting with lung cancer, the company intends to expand its scope to solid tumors—such as breast and colorectal cancers—where securing specimens is difficult and rapid diagnosis is essential.

An industry insider stated, “Companies that have accumulated real patient data through NGS panels or diagnostic kit businesses are now in the most advantageous position during this AI transition,” adding, “Ultimately, the strength of precision medicine companies stems not from algorithms but from genomic data.”