Implementation of a novel microarray‐based diagnostic test for cancer of unknown primary

摘要: Patients with carcinoma of unknown primary (CUP) present metastatic disease for which the site cannot be found, despite extensive standard investigation. Here, we describe development and implementation first clinically available microarray-based test this cancer type (CUPPrint), based on 633 individual tumors representing 30 17 noncarcinoma classes. Tissue origin prediction either fresh frozen or paraffin-embedded tumor samples is achieved use a custom 8-pack 1.9k microarray robust classification algorithm. An expression profile 495 genes was used to predict by applying k-nearest neighbor Internal cross-validation analysis an independent, previously published, 229-sample dataset revealed that informative predictions were made up 94% analyzed. Analysis 13 published CUP specimens yielded predicted origins supported clinical suspicion in 12 cases (92%). Microarray profiling presents promising tool assist identification might direct more tailored treatment patients.