1
|
Li Z, Lan L, Zhou Y, Li R, Chavin KD, Xu H, Li L, Shih DJH, Jim Zheng W. Developing deep learning-based strategies to predict the risk of hepatocellular carcinoma among patients with nonalcoholic fatty liver disease from electronic health records. J Biomed Inform 2024; 152:104626. [PMID: 38521180 DOI: 10.1016/j.jbi.2024.104626] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/20/2023] [Revised: 02/23/2024] [Accepted: 03/20/2024] [Indexed: 03/25/2024]
Abstract
OBJECTIVE The accuracy of deep learning models for many disease prediction problems is affected by time-varying covariates, rare incidence, covariate imbalance and delayed diagnosis when using structured electronic health records data. The situation is further exasperated when predicting the risk of one disease on condition of another disease, such as the hepatocellular carcinoma risk among patients with nonalcoholic fatty liver disease due to slow, chronic progression, the scarce of data with both disease conditions and the sex bias of the diseases. The goal of this study is to investigate the extent to which the aforementioned issues influence deep learning performance, and then devised strategies to tackle these challenges. These strategies were applied to improve hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. METHODS We evaluated two representative deep learning models in the task of predicting the occurrence of hepatocellular carcinoma in a cohort of patients with nonalcoholic fatty liver disease (n = 220,838) from a national EHR database. The disease prediction task was carefully formulated as a classification problem while taking censorship and the length of follow-up into consideration. RESULTS We developed a novel backward masking scheme to deal with the issue of delayed diagnosis which is very common in EHR data analysis and evaluate how the length of longitudinal information after the index date affects disease prediction. We observed that modeling time-varying covariates improved the performance of the algorithms and transfer learning mitigated reduced performance caused by the lack of data. In addition, covariate imbalance, such as sex bias in data impaired performance. Deep learning models trained on one sex and evaluated in the other sex showed reduced performance, indicating the importance of assessing covariate imbalance while preparing data for model training. CONCLUSIONS The strategies developed in this work can significantly improve the performance of hepatocellular carcinoma risk prediction among patients with nonalcoholic fatty liver disease. Furthermore, our novel strategies can be generalized to apply to other disease risk predictions using structured electronic health records, especially for disease risks on condition of another disease.
Collapse
Affiliation(s)
- Zhao Li
- McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, 7000 Fannin Street, Suite 600, Houston, TX 77030, USA
| | - Lan Lan
- McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, 7000 Fannin Street, Suite 600, Houston, TX 77030, USA
| | - Yujia Zhou
- McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, 7000 Fannin Street, Suite 600, Houston, TX 77030, USA
| | - Ruoxing Li
- McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, 7000 Fannin Street, Suite 600, Houston, TX 77030, USA
| | - Kenneth D Chavin
- Department of Surgery, Case Western Reserve University School of Medicine, 11100 Euclid Ave, Cleveland, OH 44106, USA
| | - Hua Xu
- Yale School of Medicine, USA
| | - Liang Li
- Department of Biostatistics, University of Texas MD Anderson Cancer Center, 1400 Pressler Street, FCT4.6008, Houston, TX 77030, USA
| | - David J H Shih
- School of Biomedical Sciences, The University of Hong Kong, Hong Kong Special Administrative Region
| | - W Jim Zheng
- McWilliams School of Biomedical Informatics, University of Texas Health Science Center at Houston, 7000 Fannin Street, Suite 600, Houston, TX 77030, USA.
| |
Collapse
|
2
|
Taru MG, Lupsor-Platon M. Exploring Opportunities to Enhance the Screening and Surveillance of Hepatocellular Carcinoma in Non-Alcoholic Fatty Liver Disease (NAFLD) through Risk Stratification Algorithms Incorporating Ultrasound Elastography. Cancers (Basel) 2023; 15:4097. [PMID: 37627125 PMCID: PMC10452922 DOI: 10.3390/cancers15164097] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/11/2023] [Revised: 08/08/2023] [Accepted: 08/11/2023] [Indexed: 08/27/2023] Open
Abstract
Non-alcoholic fatty liver disease (NAFLD), with its progressive form, non-alcoholic steatohepatitis (NASH), has emerged as a significant public health concern, affecting over 30% of the global population. Hepatocellular carcinoma (HCC), a complication associated with both cirrhotic and non-cirrhotic NAFLD, has shown a significant increase in incidence. A substantial proportion of NAFLD-related HCC occurs in non-cirrhotic livers, highlighting the need for improved risk stratification and surveillance strategies. This comprehensive review explores the potential role of liver ultrasound elastography as a risk assessment tool for HCC development in NAFLD and highlights the importance of effective screening tools for early, cost-effective detection and improved management of NAFLD-related HCC. The integration of non-invasive tools and algorithms into risk stratification strategies could have the capacity to enhance NAFLD-related HCC screening and surveillance effectiveness. Alongside exploring the potential advancement of non-invasive tools and algorithms for effectively stratifying HCC risk in NAFLD, we offer essential perspectives that could enable readers to improve the personalized assessment of NAFLD-related HCC risk through a more methodical screening approach.
Collapse
Affiliation(s)
- Madalina-Gabriela Taru
- Hepatology Department, Regional Institute of Gastroenterology and Hepatology “Octavian Fodor”, 400162 Cluj-Napoca, Romania;
- “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, Romania
| | - Monica Lupsor-Platon
- “Iuliu Hatieganu” University of Medicine and Pharmacy, 400012 Cluj-Napoca, Romania
- Medical Imaging Department, Regional Institute of Gastroenterology and Hepatology “Octavian Fodor”, 400162 Cluj-Napoca, Romania
| |
Collapse
|
3
|
Li H, Liu H, Yan LJ, Ding ZN, Zhang X, Pan GQ, Han CL, Tian BW, Tan SY, Dong ZR, Wang DX, Yan YC, Li T. Performance of GALAD score and serum biomarkers for detecting NAFLD-related HCC: a systematic review and network meta-analysis. Expert Rev Gastroenterol Hepatol 2023; 17:1159-1167. [PMID: 37929312 DOI: 10.1080/17474124.2023.2279175] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/14/2023] [Accepted: 10/31/2023] [Indexed: 11/07/2023]
Abstract
INTRODUCTION The incidence of nonalcoholic fatty liver disease (NAFLD)-related hepatocellular carcinoma (HCC) is increasing globally. We aimed to assess the performance of alpha-fetoprotein (AFP), AFP-L3, des-gamma-carboxy prothrombin (DCP), and GALAD score in detecting NAFLD-related HCC. METHODS We searched the relevant literature in PubMed, Embase and Cochrane. Conventional and network meta-analyses were performed for sensitivity, specificity, Youden index (YI), and the area under the summary receiver operator characteristic curve (AUC). RESULTS Fifteen studies involving 2031 NAFLD participants were included in this meta-analysis. When detecting early-stage NAFLD-related HCC, GALAD score and DCP process excellent performance. The sensitivity and AUC of DCP (0.60, 0.74, respectively) were higher than AFP (0.34, 0.59, respectively). The network meta-analysis showed that DCP and GALAD score had similar performance. In detecting all-stage NAFLD-related HCC, GALAD score (sensitivity = 0.87; YI = 0.77) performed better than AFP (sensitivity = 0.56; YI = 0.50), AFP-L3 (sensitivity = 0.39; YI = 0.36) and DCP (sensitivity = 0.73; YI = 0.62). Network meta-analysis obtained consistent results with conventional meta-analysis. CONCLUSIONS Due to the lower cost-effectiveness, DCP was more suitable for detecting early NAFLD-related HCC. AFP could be used in detecting all-stage NAFLD-related HCC.
Collapse
Affiliation(s)
- Han Li
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Hui Liu
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Lun-Jie Yan
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Zi-Niu Ding
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Xiao Zhang
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Guo-Qiang Pan
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Cheng-Long Han
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Bao-Wen Tian
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Si-Yu Tan
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Zhao-Ru Dong
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Dong-Xu Wang
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Yu-Chuan Yan
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| | - Tao Li
- Department of General Surgery, Qilu Hospital of Shandong University, Jinan, China
| |
Collapse
|
4
|
Kohlhepp MS, Liu H, Tacke F, Guillot A. The contradictory roles of macrophages in non-alcoholic fatty liver disease and primary liver cancer-Challenges and opportunities. Front Mol Biosci 2023; 10:1129831. [PMID: 36845555 PMCID: PMC9950415 DOI: 10.3389/fmolb.2023.1129831] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/22/2022] [Accepted: 01/31/2023] [Indexed: 02/12/2023] Open
Abstract
Chronic liver diseases from varying etiologies generally lead to liver fibrosis and cirrhosis. Among them, non-alcoholic fatty liver disease (NAFLD) affects roughly one-quarter of the world population, thus representing a major and increasing public health burden. Chronic hepatocyte injury, inflammation (non-alcoholic steatohepatitis, NASH) and liver fibrosis are recognized soils for primary liver cancer, particularly hepatocellular carcinoma (HCC), being the third most common cause for cancer-related deaths worldwide. Despite recent advances in liver disease understanding, therapeutic options on pre-malignant and malignant stages remain limited. Thus, there is an urgent need to identify targetable liver disease-driving mechanisms for the development of novel therapeutics. Monocytes and macrophages comprise a central, yet versatile component of the inflammatory response, fueling chronic liver disease initiation and progression. Recent proteomic and transcriptomic studies performed at singular cell levels revealed a previously overlooked diversity of macrophage subpopulations and functions. Indeed, liver macrophages that encompass liver resident macrophages (also named Kupffer cells) and monocyte-derived macrophages, can acquire a variety of phenotypes depending on microenvironmental cues, and thus exert manifold and sometimes contradictory functions. Those functions range from modulating and exacerbating tissue inflammation to promoting and exaggerating tissue repair mechanisms (i.e., parenchymal regeneration, cancer cell proliferation, angiogenesis, fibrosis). Due to these central functions, liver macrophages represent an attractive target for the treatment of liver diseases. In this review, we discuss the multifaceted and contrary roles of macrophages in chronic liver diseases, with a particular focus on NAFLD/NASH and HCC. Moreover, we discuss potential therapeutic approaches targeting liver macrophages.
Collapse
|
5
|
Laface C, Ranieri G, Maselli FM, Ambrogio F, Foti C, Ammendola M, Laterza M, Cazzato G, Memeo R, Mastrandrea G, Lioce M, Fedele P. Immunotherapy and the Combination with Targeted Therapies for Advanced Hepatocellular Carcinoma. Cancers (Basel) 2023; 15. [PMID: 36765612 DOI: 10.3390/cancers15030654] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/12/2022] [Revised: 01/11/2023] [Accepted: 01/17/2023] [Indexed: 01/24/2023] Open
Abstract
One of the most important abilities of a tumor is to establish a state of immunosuppression inside the tumor microenvironment. This is made possible through numerous mechanisms of tumor immune escape that have been identified in experimental studies during the last decades. In addition, the hepatic microenvironment is commonly oriented towards a state of immune tolerance because the liver receives blood from the hepatic arteries and portal veins containing a variety of endogenous antigens. Therefore, the hepatic microenvironment establishes an autoimmune tolerance, preventing an autoimmune reaction in the liver. On this basis, hepatic tumor cells may escape the immune system, avoiding being recognized and destroyed by immune cells. Moreover, since the etiology of Hepatocellular Carcinoma (HCC) is often related to cirrhosis, and hepatitis B or C, this tumor develops in the context of chronic inflammation. Thus, the HCC microenvironment is characterized by important immune cell infiltration. Given these data and the poor prognosis of advanced HCC, different immunotherapeutic strategies have been developed and evaluated for these patients. In this review, we describe all the clinical applications of immunotherapy for advanced HCC, from the drugs that have already been approved to the ongoing clinical trials.
Collapse
|
6
|
Zhang K, Yu M, Liu H, Hui Z, Yang N, Bi X, Sun L, Lin R, Lü G. Upregulated TUBG1 expression is correlated with poor prognosis in hepatocellular carcinoma. PeerJ 2022; 10:e14415. [PMID: 36523478 PMCID: PMC9745943 DOI: 10.7717/peerj.14415] [Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/22/2022] [Accepted: 10/28/2022] [Indexed: 12/11/2022] Open
Abstract
Background Hepatocellular carcinoma (HCC) development is a complex pathological process. Tubulin gamma 1 (TUBG1) plays an oncogenic role in several human cancers; however, its functional role in HCC tumorigenesis remains unknown. Methods Herein we first evaluated the gene expression levels of TUBG1 in HCC using data from The Cancer Genome Atlas and Gene Expression Profiling Interactive Analysis databases. We then elucidated the association between TUBG1 gene expression levels and survival rates of patients with HCC. Cell cycle, proliferation, transwell migration, and matrigel invasion assays were used to study the effects of TUBG1 on the malignant phenotypes of HCC cells. Results Based on the data obtained from the aforementioned databases and our in vitro experiments, TUBG1 was found to be overexpressed in HCC and patients with high TUBG1 expression levels showed a remarkably poor overall survival rate. In addition, the expression of TUBG1 significantly promoted the malignant phenotypes of HCC cells in vitro. Gene ontology term enrichment analysis revealed that co-regulated genes were enriched in biological processes mainly involved in chromosome segregation, chromosomal region, and chromatin binding; moreover, Kyoto Encyclopedia of Genes and Genome pathway analysis showed that they were mainly involved in cell cycle, oocyte meiosis, platinum drug resistance, and the p53 signaling pathway. Conclusions We report that TUBG1 is an important oncogene in HCC. It promotes HCC progression and may serve as a potential prognostic biomarker for HCC. Future studies are warranted to unveil molecular biological mechanisms underlying TUBG1 carcinogenesis.
Collapse
Affiliation(s)
- Kainan Zhang
- Xinjiang Medical University, Urumqi, Xinjiang, China,State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Mengsi Yu
- Department of Clinical Laboratory, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Hui Liu
- State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Zhao Hui
- Department of Clinical Laboratory, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Ning Yang
- State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Xiaojuan Bi
- State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Li Sun
- State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - RenYong Lin
- State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China
| | - Guodong Lü
- State Key Laboratory of Pathogenesis, Prevention, and Treatment of Central Asian High Incidence Diseases, Clinical Medical Research Institute, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, Xinjiang, China,College of Pharmacy, Xinjiang Medical University, Urumqi, Xinjiang, China
| |
Collapse
|