Traditional histopathology performed by pathologists through naked eyes is insufficient for accurate and efficient diagnosis of bladder cancer (BCa). We collected a total of 643 hematoxylin-eosin strained BCa images from Shanghai General Hospital and The Cancer Genome Atlas (TCGA). We constructed and cross-verified automatic diagnosis and prognosis models by performing machine learning algorithm based on patho-mics data. Our study indicated that high diagnostic efficiency of the machine learning-based diagnosis model was observed in BCa patients, with area under the curve (AUC) values of 96.3%, 89.2% and 94.1% in the training cohort, test cohort and external validation cohort, respectively. Our diagnosis model also performed well in distinguishing BCa patients from patients with glandular cystitis, with AUC value of 93.4% in General cohort. Significant differences were found in overall survival in TCGA cohort (HR = 2.09, 95% CI: 1.56-2.81, p < 0.0001) and General cohort (HR = 5.32, 95% CI: 2.95-9.59, p < 0.0001) comparing BCa patients with high versus low risk stratified by risk score, which was proved to be an independent prognostic factor for BCa. The integration nomogram based on our risk score and clinicopathologic characters displayed higher prediction accuracy than current tumor stage/grade systems, with AUC values of 77.7%, 83.8% and 81.3% for 1-, 3- and 5-year overall survival prediction of BCa patients. However, prospective studies are still wanted for further verifications.
Cancer science. 2021 May 01 [Epub ahead of print]
Siteng Chen, Liren Jiang, Xinyi Zheng, Jialiang Shao, Tao Wang, Encheng Zhang, Feng Gao, Xiang Wang, Junhua Zheng
Department of Urology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China., Department of Pathology, Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China., Department of Pharmacy, Huashan Hospital, Fudan University, Shanghai, China.