Introduction
Hepatocellular carcinoma (HCC) staging guides patient prognosis and treatment allocation; however, there is no universally accepted staging system for HCC. The most widely endorsed staging system is the Barcelona Clinic Liver Cancer (BCLC) system, which incorporates tumor burden, functional status, and liver function.1 We aimed to compare the discriminant ability of several staging systems for HCC in a geographically diverse multicenter United States (US) cohort.
Methods
We conducted a retrospective cohort study of patients newly diagnosed with HCC at four US health systems between June 1, 2012 and May 31, 2013.
Patients were identified by ICD-9 codes for HCC (155.0 or 155.2), tumor conference lists, and prospectively maintained databases. Authors adjudicated HCC cases to confirm they met diagnostic criteria, based on published criteria.2 We excluded patients with missing data for any of the included staging systems.
Patient clinical course, including dates of treatment, follow-up imaging, and death was recorded at each site. Outcomes were recorded from the clinical notes and in the case of death, all attempts were made to verify the date of death including search of local obituary listings. Patients were followed from enrollment to death, referral to hospice care, or the end of the follow-up period. This study was approved by the Institutional Review Boards at each study site.
Statistical Analysis
We assessed the prognostic performance of each system (the ITA.LI.CA, Hong Kong Liver Cancer (HKLC), Cancer of the Liver Italian Program (CLIP), and the Model to estimate survival in ambulatory patients with hepatocellular carcinoma (MESIAH) systems.3–6) based on the Akaike information criterion (AIC)7, the Discriminatory Ability Linear Trend χ28, and the concordance index (C-index)9. We calculated AIC and the linear trend χ2 based on separate Cox models for each staging system.
We resampled the dataset with replacement (bootstrap) 1000 times to calculate an empirical distribution for the C-index and difference between C-indices for each staging system and pair of staging systems. We then used these empirical distributions to estimate a central 95% confidence interval for each C-index and difference between C-indices. All analyses were completed in SAS v 9.3 (SAS Institute, Cary, NC).
Results
In total, 320 patients met inclusion criteria. Baseline characteristics were notable for being predominantly male (75.3%), white (48.8%) with a mean age of 61.0 ± 9.4. The most common liver disease etiology was hepatitis C (48.8%) and the majority of patients had Child Turcotte Pugh (CTP) class A liver disease (49.4%; CTP B 33.7%; CTP C 16.9%.) The median ECOG score was 1 (IQR: 0–1). The median number of tumors was 1 (IQR: 1–3), median tumor diameter was 3.8 cm (IQR: 2.2–7.5). Vascular invasion and metastases were present in 22.5% and 8.1%, respectively. Mean follow-up was 382 ± 302 days and 1-year survival was 58.8%.
The MESIAH system had the lowest AIC and highest Discriminatory Ability Linear Trend χ2, while the HKLC showed the highest C-index at 0.769. Notably, the BCLC system had the worst discriminant ability in all measures, including the highest AIC, lowest Discriminatory Ability Linear Trend χ2, and lowest C-statistic. The CLIP and the ITA.LI.CA systems were intermediate in their discriminant ability.
Bootstrapped comparison of the C-indices of the staging systems is shown in Table 1. While no staging system showed superiority over the others, both the HKLC and MESIAH had significantly higher C-indices than the BCLC (p=0.004 and 0.026, respectively).
Table 1.
Comparison of the C-indices for the staging systems. ITA.LI.CA – Italian Liver Cancer; BCLC – Barcelona Clinic Liver Cancer; HKLC – Hong Kong Liver Cancer; CLIP - Cancer of the Liver Italian Program; MESIAH - Model to estimate survival in ambulatory patients with hepatocellular carcinoma
| A | B | Mean(A–B) | 95% CI | Two –side P- value |
|---|---|---|---|---|
| ITA.LI.CA | BCLC | 0.018 | −0.010, 0.044 | 0.17 |
| HKLC | −0.019 | −0.044, 0.006 | 0.12 | |
| CLIP | 0.005 | −0.019, 0.030 | 0.69 | |
| MESIAH | −0.016 | −0.039, 0.006 | 0.15 | |
| BCLC | HKLC | −0.037 | −0.065, −0.011 | 0.004 |
| CLIP | −0.012 | −0.043, 0.017 | 0.43 | |
| MESIAH | −0.034 | −0.064, −0.005 | 0.026 | |
| HKLC | CLIP | 0.024 | −0.007, 0.058 | 0.16 |
| MESIAH | 0.003 | −0.027, 0.035 | 0.91 | |
| CLIP | MESIAH | −0.021 | −0.049, 0.006 | 0.1 |
Discussion
Accurate staging is central to prognosticating and allocating treatment in patients with HCC. We report the comparative ability of several HCC staging systems in a multicenter, geographically and demographically diverse cohort in the US. We found that the HKLC and MESIAH staging systems outperformed the BCLC in discriminant ability.
Our study had some notable strengths and weaknesses. First, we captured receipt of all HCC treatments at our institutions; however, there could be ascertainment bias for medical care received at outside institutions. In addition, our study was conducted at four academic centers and our results may not be generalized to all practice settings.
In conclusion we have shown that the HKLC and MESIAH staging systems are superior to the widely accepted BCLC in a multicenter US cohort. Notably, the maximum C index in our cohort was 0.769 which is in the “very good” range. To improve on this, future staging systems may rely on features such as imaging characteristics or tumor genetic alterations, rather than solely on laboratory and patient-level variables. While better systems are being developed, we have shown that some of the more contemporary systems for HCC staging deserve stronger consideration for more widespread adoption.
Acknowledgments
Grant Support:
This work was conducted with support from the Agency for Health Research and Quality Center for Patient-Centered Outcomes Research (R24 HS022418).
Abbreviations
- AASLD
American Association for the Study of Liver Disease
- AIC
Akaike information criterion
- BCLC
Barcelona Clinic Liver Cancer
- CLIP
Cancer of the Liver Italian Program
- CTP
Child Turcotte Pugh
- EASL
European Association for Study of the Liver
- HCC
Hepatocellular Carcinoma
- HKLC
Hong Kong Liver Cancer
- ITA.LI.CA
Italian Liver Cancer
- MESIAH
Model to estimate survival in ambulatory patients with hepatocellular carcinoma
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures: The authors of this manuscript have nothing to disclose
Author Contributions:
Neehar D Parikh - study concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis; study supervision
Steve Scaglione- acquisition of data; analysis and interpretation of data; critical revision of the manuscript for important intellectual content; study supervision
Yumeng Li - drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis
Corey Powell - drafting of the manuscript; critical revision of the manuscript for important intellectual content; statistical analysis
Olutola A Yerokun - acquisition of data; administrative, technical, or material support
Paulina Devlin - acquisition of data; administrative, technical, or material support
Allyce Cains - acquisition of data; administrative, technical, or material support
Sahil Mittal - acquisition of data; analysis and interpretation of data; critical revision of the manuscript for important intellectual content; study supervision
Amit G Singal - study concept and design; acquisition of data; analysis and interpretation of data; critical revision of the manuscript for important intellectual content; statistical analysis; study supervision
References
- 1.Llovet JM, Bru C, Bruix J. Prognosis of hepatocellular carcinoma: the BCLC staging classification. Seminars in liver disease. 1999;19(3):329–38. doi: 10.1055/s-2007-1007122. Epub 1999/10/13. [DOI] [PubMed] [Google Scholar]
- 2.Bruix J, Sherman M. Management of hepatocellular carcinoma: an update. Hepatology. 2011;53(3):1020–2. doi: 10.1002/hep.24199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Yau T, Tang VY, Yao TJ, et al. Development of Hong Kong Liver Cancer staging system with treatment stratification for patients with hepatocellular carcinoma. Gastroenterology. 2014;146(7):1691–700 e3. doi: 10.1053/j.gastro.2014.02.032. [DOI] [PubMed] [Google Scholar]
- 4.A new prognostic system for hepatocellular carcinoma: a retrospective study of 435 patients: the Cancer of the Liver Italian Program (CLIP) investigators. Hepatology. 1998;28(3):751–5. doi: 10.1002/hep.510280322. [DOI] [PubMed] [Google Scholar]
- 5.Yang JD, Kim WR, Park KW, et al. Model to estimate survival in ambulatory patients with hepatocellular carcinoma. Hepatology. 2012;56(2):614–21. doi: 10.1002/hep.25680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Farinati F, Vitale A, Spolverato G, et al. Development and Validation of a New Prognostic System for Patients with Hepatocellular Carcinoma. PLoS Med. 2016;13(4):e1002006. doi: 10.1371/journal.pmed.1002006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Stone M. Encyclopedia of Biostatistics. John Wiley & Sons, Ltd; 2005. Akaike's Criteria. [Google Scholar]
- 8.Pourhoseingholi MA, Hajizadeh E, Moghimi Dehkordi B, et al. Comparing Cox regression and parametric models for survival of patients with gastric carcinoma. Asian Pac J Cancer Prev. 2007;8(3):412–6. [PubMed] [Google Scholar]
- 9.Steyerberg EW, Vickers AJ, Cook NR, et al. Assessing the performance of prediction models: a framework for traditional and novel measures. Epidemiology. 2010;21(1):128–38. doi: 10.1097/EDE.0b013e3181c30fb2. [DOI] [PMC free article] [PubMed] [Google Scholar]