CATCHERS Inc. is a pioneering biotechnology company dedicated to the early detection and auxiliary diagnosis of cancer. Our core technology platform utilizes validated salivary biomarkers for esophageal cancer and powerful bioinformatics to bridge the gap between genomic sequencing and accessible real-world clinical solutions.

We processed the raw saliva microbiome sequencing data from 844 individuals using a bioinformatics pipeline to identify specific microbial features that differentiate esophageal cancer patients from healthy controls and periodontitis patients. By analyzing these extracted features with a machine learning model—such as a random forest classifier, we developed a proprietary algorithm. The algorithm was trained on extensive, curated datasets from esophageal cancer patients, healthy individuals, and periodontitis patients. The resulting adaptive model generates the core of the "EarlyDetect for Esophageal Cancer" Kit/prototype.

In the discovery cohort (N=235), the prototype demonstrated strong key performance metrics (KPMs), with a sensitivity of 0.839—reflecting its ability to correctly identify those with esophageal cancer (true positive rate), and a specificity of 0.911—indicating its ability to correctly identify non-cancer individuals (true negative rate). These results yield an area under the curve (AUC) of 0.932 (95%CI: 0.900-0.965), reflecting high discriminative power between esophageal cancer and non-cancer cases, where a value of 1.0 represents perfect discrimination.

In two independent validation cohorts, the model also exhibited robust performance, achieving an AUC of 0.856 (95% CI: 0.806-0.906) with a sensitivity of 0.768 and a specificity of 0.820 (N=236), and an AUC of 0.800 (95% CI: 0.749-0.851) with a sensitivity of 0.726 and a specificity of 0.832 (N=373).