Cytology-based deep learning method for prediction of tumor origin in cancers of unknown primary origin

Cancer of unknown primary site (CUP) poses diagnostic challenges. Many cases of CUP manifest as pleural and peritoneal serous effusions. Researchers at four centers in China developed a deep-learning method for tumor origin differentiation using cytological histology (TORCH) that can identify malignancy and predict tumor origin in both hydrothorax and ascites. In five test sets (n=27,337), TORCH achieved area under the receiver operating curve values ranging from 0.953 to 0.991 for cancer diagnosis and 0.953 to 0.979 for tumor origin localization. TORCH accurately predicted primary tumor origins, with a top-1 accuracy of 82.6% and a top-3 accuracy of 98.9%. Patients with CUP whose initial treatment protocol was concordant with TORCH-predicted origins had better overall survival than those who were administered discordant treatments (median 27 versus 17 months; p=0.006). The authors conclude that TORCH is a valuable ancillary tool in clinical practice, although further validation in randomized trials is warranted.