A consortium of research institutes in South Korea has been formed to promote the utilisation of AI and national supercomputers in discovering novel drugs for treating lung cancer.

According to a press statement, the following organisations will collaborate to develop and test AI for identifying novel drug candidates and predicting its efficacy: 

• DAAN Cancer Research Institute under Yonsei University's College of Medicine

• Daegu Gyeongbuk Institute of Science and Technology (DGIST)

• Korea Research Institute of Chemical Technology (KRICT)

• Korea Advanced Institute of Science and Technology (KAIST)

• Biotechnology company J INTS BIO

WHAT IT'S ABOUT

The consortium's research will be done in four phases: First, the collection of lung cancer tissue and genomic samples, to be led by Yonsei University's DAAN Cancer Research; second, the simulation and analysis of protein-drug interactions using supercomputers and AI at the DGIST Core Protein Resources Center; third, the synthesis of AI-identified drug candidates through pharmacological and toxicological assessments at KRICT; and lastly, the validation through clinical trials of AI's predictive capability and accuracy.

Besides overseeing the entire research, J INTS BIO will also lead the development of patient-specific treatment protocols for administering their potential anti-lung cancer drug. 

The consortium also aims to build "extensive cancer data infrastructures" that will underpin ongoing research. They target to present the initial findings of their research by the first half of next year.

WHY IT MATTERS

This collaboration emphasises the use of AI and supercomputers to complement measures addressing persistent challenges in anti-cancer drug discovery and development, particularly low effectiveness and high toxicity. 

The consortium underscored AI's value in drug discovery by helping hasten the process while helping minimise clinical trial failure rates.

"AI's capacity to process immense biological datasets and apply machine learning for patient-specific predictions can drastically reduce clinical trial failure rates and accelerate the entire drug discovery process," said KAIST professor Joung Ho Kim. 

Utilising AI and supercomputers can also enable the "ultra-precise" analysis of patient tissues and genetic profiles, facilitating the design of personalised treatments, said Byoung Chul Cho, a professor from Yonsei's Severance Hospital. He was also part of the research team that developed a branded oral non-small cell lung cancer drug. 

THE LARGER TREND

Last year, the South Korean government started allowing researchers across the country to access public and clinical data on cancer through the K-CURE service. In August, access to more cancer data types on K-CURE was expanded. 

Meanwhile, the Korean government also recently announced that it intends to develop a system to rapidly develop and verify new drug targets for non-immune solid cancers as part of initiatives under the Korean Advanced Research Projects Agency for Health.

Besides the government, private national efforts are also pursuing AI-enabled drug discovery in South Korea. For one, AI company Standigm has partnered with Institut Pasteur Korea to develop new anti-tuberculosis drug candidates. It tied up with Merck Korea as well. Meanwhile, Pharma giant Daewoong Pharmaceutical unveiled its AI-driven drug development system, powered by an 800 million-compound database.