Wang, XiaoyingDuan, MaotengSu, Po-LanLi, JianyingKrull, JordanJin, JiachengChen, HuSun, YuhanWu, WeidongHe, KaiCarpenter, Richard L.Zhang, ChiCao, ShaXu, DongWang, GuangyuLi, LangXin, GangCarbone, David P.Li, ZihaMa, Qin2025-07-142025-07-142025-05-22Wang X, Duan M, Su PL, et al. Deep-learning-enabled multi-omics analyses for prediction of future metastasis in cancer. Preprint. bioRxiv. 2025;2025.05.16.654579. Published 2025 May 22. doi:10.1101/2025.05.16.654579https://hdl.handle.net/1805/49399Metastasis remains the leading cause of cancer-related mortality, yet predicting future metastasis is a major clinical challenge due to the lack of validated biomarkers and effective assessment methods. Here, we present EmitGCL, a deep-learning framework that accurately predicts future metastasis and its corresponding biomarkers. Based on a comprehensive benchmarking comparison, EmitGCL outperformed other computational tools across six cancer types from seven cohorts of patients with superior sensitivity and specificity. It captured occult metastatic cells in a patient with a lymph node-negative breast cancer, who was declared to have no evidence of disease by conventional imaging methods but was later confirmed to have a metastatic disease. Notably, EmitGCL identified HSP90AA1 and HSP90AB1 as predictable biomarkers for future breast cancer metastasis, which was validated across five independent cohorts of patients (n=420). Furthermore, we demonstrated YY1 transcription factor as a key driver of breast cancer metastasis which was validated through in-silico and CRISPR-based migration assays, suggesting that YY1 is a potential therapeutic target for deterring metastasis.en-USAttribution-NonCommercial 4.0 InternationalCancer-related mortalityMetastasisEmitGCLDeep-learning-enabled multi-omics analyses for prediction of future metastasis in cancerArticle