GC Genome Unveils External Cohort Validation Results for "ai-CANCERCH" at Japanese Conference

Analysis of Clinical Data from 8,436 Individuals
Specificity 95.5%, Sensitivity 79.7%
"Addressing Unmet Needs in Hard-to-Diagnose Cancers"

GC Genome's artificial intelligence (AI)-based multi-cancer early screening test "ai-CANCERCH" demonstrated high specificity and sensitivity in an external cohort-based performance validation.

GC Genome announced on the 9th that it disclosed the performance validation results of ai-CANCERCH at the "Japanese Society for the Study of Liquid Biopsy" conference held in Kyoto, Japan on the 6th and 7th.

GC Jineom disclosed the performance validation results of Aikaenseochi at the liquid biopsy conference 'Japanese Society for the Study of Liquid Biopsy' held in Kyoto, Japan on the 6th and 7th. GC Jineom

ai-CANCERCH is a multi-cancer early detection (MCED) test that can detect multi-cancer signals using only a single blood tube, based on GC Genome's proprietary AI algorithm and whole genome sequencing (WGS) technology. In this study, GC Genome evaluated the test performance for various cancer types by utilizing a large-scale external validation cohort.

The model was trained on a cohort of 4,063 individuals, including 1,117 cancer patients. Performance was then validated in an external cohort of 4,373 individuals, including 846 cancer patients and 3,527 non-cancer controls, so that clinical data from a total of 8,436 individuals were used in the analysis.

According to the study results, ai-CANCERCH achieved a specificity of 95.5%. Overall sensitivity and stage-weighted sensitivity were 79.7% and 80.2%, respectively. By cancer type, it showed sensitivity of up to more than 80% for pancreatic cancer and hepatobiliary cancer, for which standard screening protocols are insufficient. This indicates relatively high sensitivity for cancers that are difficult to diagnose at an early stage. For cancers such as lung, colorectal, and breast cancer, where existing screening systems are already established, it showed sensitivity levels in the range of 61% to 76%.

GC Genome stated, "It is meaningful in that the study was conducted based on large-scale clinical data, thereby securing both statistical significance and technical reliability," and added, "By combining an AI analysis method that goes beyond simple genome analysis to incorporate the fragmentomic characteristics of cell-free DNA (cfDNA), together with a low-cost, high-efficiency low-coverage whole genome sequencing (lcWGS)-based technology, we have also increased the potential for commercialization."

GC Genome plans to use this external cohort-based performance validation as a starting point to accelerate clinical collaboration with Japanese medical institutions and discussions on the introduction of the test. Japan is a market where demand for early diagnosis is rapidly increasing due to population aging and a rising cancer incidence rate. The company sees this study as highly significant in that it has demonstrated the practical applicability of ai-CANCERCH in the Japanese clinical setting.

Eunhae Cho, Head of Research at GC Genome, said, "We will address the unmet medical needs for cancers that have been in the blind spots of existing screening systems and strengthen our competitiveness in the global multi-cancer early detection market."

This content was produced with the assistance of AI translation services.