Comparison Between Cox Proportional Hazards and Machine Learning Models for the Prognostication of Recurrence and Survival Following Liver Resection for Hepatocellular Carcinoma
Corresponding Author
Hwee-Leong Tan
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Correspondence:
Hwee-Leong Tan ([email protected])
Brian K. P. Goh ([email protected])
Search for more papers by this authorClaudia Y. T. Liauw
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Search for more papers by this authorTse-Lert Chua
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Search for more papers by this authorAmanda Y. R. Lam
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Department of Endocrinology, Singapore General Hospital, Singapore, Singapore
Search for more papers by this authorCliburn Chan
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina, USA
Search for more papers by this authorYe-Xin Koh
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorJin-Yao Teo
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorPeng-Chung Cheow
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorAlexander Y. F. Chung
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorCorresponding Author
Brian K. P. Goh
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Correspondence:
Hwee-Leong Tan ([email protected])
Brian K. P. Goh ([email protected])
Search for more papers by this authorCorresponding Author
Hwee-Leong Tan
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Correspondence:
Hwee-Leong Tan ([email protected])
Brian K. P. Goh ([email protected])
Search for more papers by this authorClaudia Y. T. Liauw
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Search for more papers by this authorTse-Lert Chua
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Search for more papers by this authorAmanda Y. R. Lam
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Department of Endocrinology, Singapore General Hospital, Singapore, Singapore
Search for more papers by this authorCliburn Chan
Department of Data Science, Singapore General Hospital, Singapore, Singapore
Department of Biostatistics and Bioinformatics, Duke University School of Medicine, Durham, North Carolina, USA
Search for more papers by this authorYe-Xin Koh
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorJin-Yao Teo
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorPeng-Chung Cheow
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorAlexander Y. F. Chung
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Search for more papers by this authorCorresponding Author
Brian K. P. Goh
Department of Hepatopancreatobiliary and Transplant Surgery, Singapore General Hospital and National Cancer Center, Singapore, Singapore
Correspondence:
Hwee-Leong Tan ([email protected])
Brian K. P. Goh ([email protected])
Search for more papers by this authorABSTRACT
Background
A robust prognostication model after liver resection for hepatocellular carcinoma (HCC) can guide clinical management. We aimed to develop a prognostication model for HCC recurrence and survival following liver resection, comparing between Cox proportional hazards (CPH) and supervised machine learning models.
Methods
We studied all patients who underwent liver resection for HCC between January 1, 2000 and October 31, 2022 at our institution. We aimed to predict recurrence-free survival following resection and identify risk categories for HCC recurrence. The CPH model and two supervised machine learning models (random survival forest [RSF] and extreme gradient boosting [XGB]) were used. Model performance was assessed with C-index, time-dependent area under curve (tdAUC) and Brier score.
Results
We studied 1290 patients, with 737 (57.1%) experiencing an event (HCC recurrence or death) over a median follow-up duration of 19.2 months. The CPH model had the overall best performance (C-index: 0.663, tdAUC at 6 months: 0.752; 1 year: 0.740; 2 years: 0.722; 5 years: 0.624). Using this model, patients stratified based on risk score could be discriminated between low, intermediate, and high-risk groups (p < 0.001).
Conclusion
A CPH-derived prognostication model was effective for predicting and risk stratifying recurrence and survival following liver resection for HCC.
Conflicts of Interest
The authors declare no conflicts of interest.
Open Research
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
Supporting Information
| Filename | Description |
|---|---|
| jhbp12186-sup-0001-FigureS1.tifTIFF image, 358.4 KB |
Figure S1. CONSORT diagram of study cohort. |
| jhbp12186-sup-0002-FigureS2.tifTIFF image, 457.5 KB |
Figure S2. Calibration curves of models at (a) 6 months, (b) 1 year, (c) 2 years, and (d) 5 years. |
| jhbp12186-sup-0003-TableS1-S2.docxWord 2007 document , 12.6 KB |
Data S1. |
Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
References
- 1S. Chidambaranathan-Reghupaty, P. B. Fisher, and D. Sarkar, “Hepatocellular Carcinoma (HCC): Epidemiology, Etiology and Molecular Classification,” Advances in Cancer Research 149 (2021): 1–61, https://doi.org/10.1016/bs.acr.2020.10.001.
- 2A. G. Singal, J. M. Llovet, M. Yarchoan, et al., “AASLD Practice Guidance on Prevention, Diagnosis, and Treatment of Hepatocellular Carcinoma,” Hepatology 78, no. 6 (2023): 1922–1965, https://doi.org/10.1097/HEP.0000000000000466.
- 3M. Yilma, X. R. Houhong, V. Saxena, et al., “Survival Outcomes Among Patients With Hepatocellular Carcinoma in a Large Integrated US Health System,” JAMA Network Open 7, no. 9 (2024): e2435066, https://doi.org/10.1001/jamanetworkopen.2024.35066.
- 4M. Reig, A. Forner, J. Rimola, et al., “BCLC Strategy for Prognosis Prediction and Treatment Recommendation: The 2022 Update,” Journal of Hepatology 76, no. 3 (2022): 681–693, https://doi.org/10.1016/j.jhep.2021.11.018.
- 5A. D. Pinna, T. Yang, V. Mazzaferro, et al., “Liver Transplantation and Hepatic Resection Can Achieve Cure for Hepatocellular Carcinoma,” Annals of Surgery 268, no. 5 (2018): 868–875, https://doi.org/10.1097/SLA.0000000000002889.
- 6P. Tabrizian, G. Jibara, B. Shrager, M. Schwartz, and S. Roayaie, “Recurrence of Hepatocellular Cancer After Resection: Patterns, Treatments, and Prognosis,” Annals of Surgery 261, no. 5 (2015): 947–955, https://doi.org/10.1097/SLA.0000000000000710.
- 7R. Nevola, R. Ruocco, L. Criscuolo, et al., “Predictors of Early and Late Hepatocellular Carcinoma Recurrence,” World Journal of Gastroenterology 29, no. 8 (2023): 1243, https://doi.org/10.3748/wjg.v29.i8.1243.
- 8J. N. Vauthey and G. Y. Lauwers, “Prognostic Factors After Resection of Hepatocellular Carcinoma: Are There Landmarks in the Wild Forest?,” Journal of Hepatology 38, no. 2 (2003): 237–239, https://doi.org/10.1016/s0168-8278(02)00407-5.
- 9G. W. Ji, K. Wang, Y. X. Xia, X. C. Li, and X. H. Wang, “Application Value of Machine Learning Algorithms for Predicting Recurrence After Resection of Early-Stage Hepatocellular Carcinoma,” Zhonghua Wai Ke Za Zhi 59, no. 8 (2021): 679–685, https://doi.org/10.3760/cma.j.cn112139-20201026-00768.
- 10H. Nathan, R. D. Schulick, M. A. Choti, and T. M. Pawlik, “Predictors of Survival After Resection of Early Hepatocellular Carcinoma,” Annals of Surgery 249, no. 5 (2009): 799–805, https://doi.org/10.1097/SLA.0b013e3181a38eb5.
- 11W. H. Kang, S. Hwang, M. Kaibori, et al., “Validation of Quantitative Prognostic Prediction Using ADV Score for Resection of Hepatocellular Carcinoma: A Korea-Japan Collaborative Study With 9200 Patients,” Journal of Hepato-Biliary-Pancreatic Sciences 30, no. 8 (2023): 993–1005, https://doi.org/10.1002/jhbp.1319.
- 12K. Furukawa, H. Shiba, T. Horiuchi, et al., “Survival Benefit of Hepatic Resection for Hepatocellular Carcinoma Beyond the Barcelona Clinic Liver Cancer Classification,” Journal of Hepato-Biliary-Pancreatic Sciences 24, no. 4 (2017): 199–205, https://doi.org/10.1002/jhbp.436.
- 13X. Liu, J. Lu, G. Zhang, et al., “A Machine Learning Approach Yields a Multiparameter Prognostic Marker in Liver Cancer,” Cancer Immunology Research 9, no. 3 (2021): 337–347, https://doi.org/10.1158/2326-6066.CIR-20-0616.
- 14 National Comprehensive Cancer Network, “Hepatocellular Carcinoma (Version 3.2024)”.
- 15D. Dindo, N. Demartines, and P. A. Clavien, “Classification of Surgical Complications: A New Proposal With Evaluation in a Cohort of 6336 Patients and Results of a Survey,” Annals of Surgery 240, no. 2 (2004): 205–213, https://doi.org/10.1097/01.sla.0000133083.54934.ae.
- 16N. N. Rahbari, O. J. Garden, R. Padbury, et al., “Posthepatectomy Liver Failure: A Definition and Grading by the International Study Group of Liver Surgery (ISGLS),” Surgery 149, no. 5 (2011): 713–724, https://doi.org/10.1016/j.surg.2010.10.001.
- 17P. R. Galle, A. Forner, J. M. Llovet, V. Mazzaferro, F. Piscaglia, and J. L. Raoul, “European Association for the Study of the Liver. Electronic Address Eee, European Association for the Study of the L. EASL Clinical Practice Guidelines: Management of Hepatocellular Carcinoma,” Journal of Hepatology 69, no. 1 (2018): 182–236, https://doi.org/10.1016/j.jhep.2018.03.019.
- 18H. Ishwaran, U. B. Kogalur, E. H. Blackstone, and M. S. Lauer, “Random Survival Forests,” Annals of Applied Statistics 2, no. 3 (2008): 841–860.
- 19A. Barnwal, H. Cho, and T. Hocking, “Survival Regression With Accelerated Failure Time Model in XGBoost,” Journal of Computational and Graphical Statistics 31, no. 4 (2022): 1292–1302.
- 20P. Royston and D. G. Altman, “External Validation of a Cox Prognostic Model: Principles and Methods,” BMC Medical Research Methodology 13 (2013): 33, https://doi.org/10.1186/1471-2288-13-33.
- 21S. Pölsterl, “Scikit-Survival: A Library for Time-To-Event Analysis Built on Top of Scikit-Learn,” Journal of Machine Learning Research 21, no. 212 (2020): 1–6.
- 22D. Vieira, “XGBoost Survival Embeddings: Improving Statistical Properties of XGBoost Survival Analysis Implementation. 0.2.0 ed., 2020”.
- 23D. Pilon, “Lifelines: Survival Analysis in Python,” Journal of Open Source Software 4, no. 40 (2019): 1317.
10.21105/joss.01317 Google Scholar
- 24P. Virtanen, R. Gommers, T. E. Oliphant, et al., “SciPy 1.0: Fundamental Algorithms for Scientific Computing in Python,” Nature Methods 17, no. 3 (2020): 261–272, https://doi.org/10.1038/s41592-019-0686-2.
- 25Y. L. Linn, M. Y. Chee, Y. X. Koh, et al., “Actual 10-Year Survivors and 10-Year Recurrence Free Survivors After Primary Liver Resection for Hepatocellular Carcinoma in the 21st Century: A Single Institution Contemporary Experience,” Journal of Surgical Oncology 123, no. 1 (2021): 214–221, https://doi.org/10.1002/jso.26259.
- 26S. D. Lu, L. Li, X. M. Liang, et al., “Updates and Advancements in the Management of Hepatocellular Carcinoma Patients After Hepatectomy,” Expert Review of Gastroenterology & Hepatology 13, no. 11 (2019): 1077–1088, https://doi.org/10.1080/17474124.2019.1684898.
- 27S. Qin, M. Chen, A. L. Cheng, et al., “Atezolizumab Plus Bevacizumab Versus Active Surveillance in Patients With Resected or Ablated High-Risk Hepatocellular Carcinoma (IMbrave050): A Randomised, Open-Label, Multicentre, Phase 3 Trial,” Lancet 402, no. 10415 (2023): 1835–1847, https://doi.org/10.1016/S0140-6736(23)01796-8.
- 28T. T. Tran, J. Lee, M. Gunathilake, et al., “A Comparison of Machine Learning Models and Cox Proportional Hazards Models Regarding Their Ability to Predict the Risk of Gastrointestinal Cancer Based on Metabolic Syndrome and Its Components,” Frontiers in Oncology 13 (2023): 1049787, https://doi.org/10.3389/fonc.2023.1049787.
- 29H. A. Lee, Y. S. Lee, B. K. Kim, et al., “Change in the Recurrence Pattern and Predictors Over Time After Complete Cure of Hepatocellular Carcinoma,” Gut Liver 15, no. 3 (2021): 420–429, https://doi.org/10.5009/gnl20101.
- 30K. Iguchi, E. Hatano, K. Yamanaka, S. Tanaka, K. Taura, and S. Uemoto, “The Impact of Posthepatectomy Liver Failure on the Recurrence of Hepatocellular Carcinoma,” World Journal of Surgery 38, no. 1 (2014): 150–158, https://doi.org/10.1007/s00268-013-2247-7.
Online Version of Record before inclusion in an issue
