Shipeng Chen1, Zihan Zhang2,3, Ying Wang1, Meng Fang1, Jun Zhou1, Ya Li4, Erhei Dai5, Zhaolei Feng6, Hao Wang7, Zaixing Yang8, Yongwei Li9, Xianzhang Huang10, Jian'an Jia11, Shuang Li12, Chenjun Huang1, Lin Tong1, Xiao Xiao1, Yutong He1, Yong Duan3, Shanfeng Zhu2, Chunfang Gao1. 1. Department of Laboratory Medicine, Shanghai Eastern Hepatobiliary Surgery Hospital, Shanghai, China. 2. ISTBI and Shanghai Key Lab of Intelligent Information Processing, Fudan University, Shanghai, China. 3. School of Computer Science, Fudan University, Shanghai, China. 4. Department of Laboratory Medicine, the First Affiliated Hospital of Kunming Medical University, Yunnan, China. 5. Department of Laboratory Medicine, Fifth Hospital of Shijiazhuang, Hebei Medical University, Hebei, China. 6. Department of Laboratory Medicine, Jinan Infectious Disease Hospital, Shandong, China. 7. Department of Laboratory Medicine, Shanghai Changzheng Hospital, Shanghai, China. 8. Department of Laboratory Medicine, Taizhou First People's Hospital, Zhejiang, China. 9. Department of Laboratory Medicine, Henan Province Hospital of Traditional Chinese Medicine, Henan, China. 10. Department of Laboratory Medicine, Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangdong, China. 11. Department of Laboratory Medicine, 901 Hospital of the Joint Logistics Support Force of the Chinese People's Liberation Army, Anhui, China. 12. Department of Infectious Diseases, First Affiliated Hospital of Nanjing Medical University, Jiangsu, China.
Abstract
BACKGROUND: Hepatitis B virus (HBV) infection is one of the main leading causes of hepatocellular carcinoma (HCC) worldwide. However, it remains uncertain how the reverse-transcriptase (rt) gene contributes to HCC progression. METHODS: We enrolled a total of 307 patients with chronic hepatitis B (CHB) and 237 with HBV-related HCC from 13 medical centers. Sequence features comprised multidimensional attributes of rt nucleic acid and rt/s amino acid sequences. Machine-learning models were used to establish HCC predictive algorithms. Model performances were tested in the training and independent validation cohorts using receiver operating characteristic curves and calibration plots. RESULTS: A random forest (RF) model based on combined metrics (10 features) demonstrated the best predictive performances in both cross and independent validation (AUC, 0.96; accuracy, 0.90), irrespective of HBV genotypes and sequencing depth. Moreover, HCC risk scores for individuals obtained from the RF model (AUC, 0.966; 95% confidence interval, .922-.989) outperformed α-fetoprotein (0.713; .632-.784) in distinguishing between patients with HCC and those with CHB. CONCLUSIONS: Our study provides evidence for the first time that HBV rt sequences contain vital HBV quasispecies features in predicting HCC. Integrating deep sequencing with feature extraction and machine-learning models benefits the longitudinal surveillance of CHB and HCC risk assessment.
BACKGROUND: Hepatitis B virus (HBV) infection is one of the main leading causes of hepatocellular carcinoma (HCC) worldwide. However, it remains uncertain how the reverse-transcriptase (rt) gene contributes to HCC progression. METHODS: We enrolled a total of 307 patients with chronic hepatitis B (CHB) and 237 with HBV-related HCC from 13 medical centers. Sequence features comprised multidimensional attributes of rt nucleic acid and rt/s amino acid sequences. Machine-learning models were used to establish HCC predictive algorithms. Model performances were tested in the training and independent validation cohorts using receiver operating characteristic curves and calibration plots. RESULTS: A random forest (RF) model based on combined metrics (10 features) demonstrated the best predictive performances in both cross and independent validation (AUC, 0.96; accuracy, 0.90), irrespective of HBV genotypes and sequencing depth. Moreover, HCC risk scores for individuals obtained from the RF model (AUC, 0.966; 95% confidence interval, .922-.989) outperformed α-fetoprotein (0.713; .632-.784) in distinguishing between patients with HCC and those with CHB. CONCLUSIONS: Our study provides evidence for the first time that HBV rt sequences contain vital HBV quasispecies features in predicting HCC. Integrating deep sequencing with feature extraction and machine-learning models benefits the longitudinal surveillance of CHB and HCC risk assessment.