Journal of Clinical Gastroenterology:
Prognostic Prediction in Hepatocellular Carcinoma: From Art to Science
Strazzabosco, Mario MD, PhD*; Cohen, Eric MD*; Emre, Sukru MD†
*Section of Digestive Diseases, Department of Medicine
†Section of Transplantation and Immunology, Department of Surgery, Yale University School of Medicine, New Haven, CT
The authors declare no conflicts of interest.
Mario Strazzabosco acknowledges the financial help of CeLiveR, Ospedali Riuniti di Bergam. (Yale Liver Center Grant # DK034989-24).
Reprints: Mario Strazzabosco, MD, PhD, Yale University School of Medicine, TAC Building S233, 300 Cedar Street, P.O. Box 208019, New Haven, CT 06520 (e-mail: email@example.com).
Hepatocellular carcinoma (HCC) is the third most common cause of cancer mortality worldwide and is increasing rapidly in incidence.1 Not surprisingly, it is a major clinical factor in chronic liver disease. For example, in a large cohort of hepatitis C virus RNA+ patients with fully compensated cirrhosis followed prospectively for up to 17 years, HCC developed in 38% of cases.2 This significantly outnumbered clinical events like ascites (23%), jaundice (17%), upper gastrointestinal bleeding (6%), and hepatic encephalopathy (1%).2 The main cause of death in this cohort was HCC (44%). These figures underscore the imperative to further characterize this disease for future prognostic and treatment purposes. To do so, one must be familiar with the epidemiology, risk factors, surveillance protocols, and current HCC staging systems.
A number of staging systems are available, though may be outdated, as they were devised when HCC treatment options were more limited. There is no worldwide consensus on the use of any HCC staging system.3,4 In this issue of the Journal of Clinical Gastroenterology, Toyama et al use a large series of HCC cases to address these issues. They raise important concerns about the inadequacies of our current scoring systems. In this editorial, we review some of the information influencing prognosis estimation in HCC.
The incidence of HCC reflects both temporal and geographical variation.5 As an example, one can examine mortality profiles. Mortality from HCC in the US increased from 1.54/100,000 in 1980 to 2.58/100,000 in 1998.6 The impact of immigration on this trend from countries with high incidences of HCC cannot be underestimated. In China, for example, mortality reaches 35/100,000 population.7 The age of peak incidence of HCC is another epidemiologic factor that plays an important role. In high-incidence countries, the peak is reached between 50 and 55 years of age, whereas in low-incidence western countries the age range is 65 to 70 years.1 These variations may reflect differences in the biology of HCC,8 thus limiting the applicability of prognosis estimation from one country to the next.
The etiology of chronic liver disease is also vital in HCC prognostication, and unfortunately, current scoring systems fail to take this variable into account.9 The majority of HCC cases come in the setting of cirrhosis,10 though this is not always the case. For example, hepatitis B virus (HBV), considered an oncogenic virus, dramatically increases HCC risk.11 HCC may in fact develop in noncirrhotic livers with active HBV or even with evidence of previous infection (ie, isolated HbcAb). The risk of developing HCC also increases with the number of risk factors, such as hepatitis C virus-related cirrhosis, active alcohol abuse, previous HBV exposure, and diabetes.12,13
Functional status of the patient is another parameter that is not always captured by current scoring systems, and the widely used tumor, node, metastasis (TNM) system is just one example. With most solid tumors, the estimation of life expectancy is directly related to tumor stage at diagnosis and staging is directly linked to treatment indication3; in HCC, this is not so. The functional impairment of the liver has a significant impact on survival.14,15 This is no small dilemma, and poses a difficult decision regarding choice of treatment. Scoring systems should evaluate the general health status and physical ability, such as the Karnofsky index. These will help identify patients with highly advanced disease, where any therapy may not be advisable. Surveys indicate that the odds ratio for a given patient receiving the “most appropriate treatment” varies significantly from state to state.1
Despite the above variables, our current scoring systems are largely based on the dimension and number of HCC nodules. There exists a correlation between the size of a dysplastic nodule and the degree of differentiation and aggressiveness.16 This is demonstrated by the higher rates of posttransplant recurrence when a given pretransplant tumor burden is exceeded.17 The development of the dysplastic nodule in a cirrhotic liver is likely a multistep process of carcinogenic events with the accrual of genetic damage.18–20 At this point of time, we do not have efficient techniques to detect HCC at this early stage, and as the disease is asymptomatic for most of its natural history, early detection depends on diligent implementation of imaging protocols and serologic tumor markers. The initial diagnosis therefore depends on our ability to image the arterial vascularization of HCC with respect to other regenerating nodules. A proportion of HCC nodules do not show this diagnostic characteristic, whereas others grow in a diffuse, infiltrative pattern. Sheer location of a dysplastic nodule can evade accurate radiographic detection, especially with the use of abdominal ultrasound.21 In these cases, α-fetoprotein, a serologic marker with less sensitivity and specificity, is relied upon. Despite its popularity in clinical practice, an absolute α-fetoprotein level has poor diagnostic value.22
With the myriad of factors involved, our scoring systems have evolved to meet these needs. The first prognostic index specific for HCC was developed by Okuda et al23 and took into account bilirubin, ascites, albumin, and extent of tumor involvement within the liver. It has been abandoned for newer prognostic indexes. Some of the most popular ones include the CLIP (Cancer of the Liver Italian Program), the BCLC (Barcelona Clinic for Liver Cancer), and the modified TNM staging system, each of which have merit and were validated in external studies.24–27 Controversy exists regarding how best to allocate points for patients with HCC and advanced liver disease.
Unpublished data from our group show that the performance of these scoring systems are similar, but the BCLC staging system offers a preferable advantage. Foremost is its ability to employ the health status of the patient, the dimension of the nodules, and the functional state of the liver. Patients are divided into 5 stages: very early, early, intermediate, advanced, and terminal. They are then subdivided according to functional status of the liver and size of tumor. The most important feature of BCLC is that for each stage, a best treatment is suggested.
In our center, patients with liver cirrhosis, HCC, and no metastatic disease follow an algorithm very similar to BCLC. For patients with stage A1 (Child class A, a resectable, single nodule, no portal hypertension) we offer partial hepatectomy and rescue transplantation if the tumor recurs. Patients in stages A2-4 (single nodule <5 cm or 2 to 3 nodules <3 cm) are listed for transplant and concurrently downstaged with radiofrequency ablation (RFA) or transarterial chemoembolization (TACE). Patients with stage B (Child A or B, large, multifocal) are treated with TACE; if vascular invasion is present, they may go on to experimental systemic therapy. Any Child C patient within the Milan criteria is listed fortransplantation, and if a contraindication for transplant exists, these patients undergo palliative measures.
The aforementioned Milan criteria has withstood the test of time and is useful in determining transplantation candidacy.28 Proposed by Mazzaferro and colleagues29 in 1996, the criteria suggests that a patient with a single nodule <5 cm or with no more than 3 nodules <3 cm are expected to fare well posttransplant. Although there is little doubt that the patient with multifocal HCC inside the Milan criteria should be listed for transplantation and downstaged with RFA or TACE to avoid progression of disease while waiting, it is likely that patients meeting the more extensive “UCSF criteria” may turn not to be excellent candidates for transplant.
How then, given the variation in geography, etiology, clinical parameters, and varied staging criteria is a specialist expected to provide optimal care for the patient with HCC? The advances in technology have promoted a chastened optimism within the hepatology community, whereby the art of prognostic prediction will be honed into a science. In the near future, clinical decision making could be based on genetic and biologic features of individual tumors.30 Microarray analysis of resected tumors has shed light on several classes of genes that correlate with prognosis after specific therapies. Studies show that the presence of stem cells within the tumor is associated with a worse prognosis.31,32 Others have identified clusters of genes whose expression allows for the identification of HCC subsets with different prognoses.33–37 Most likely, these novel techniques will overhaul the current staging systems, particularly in the transplant setting.
The study presented by Toyama and colleagues addresses the role of staging systems that decide the treatment strategy for both primary and recurrent tumors, and are the first to include disease recidivism in a prognostic scoring system. This is an important and complex issue. The conclusions they make, although valid, should be regarded cautiously on several grounds. For validation, only the C-index is used; supportive additions could have included statistical models like the Brier score and D-statistic. Furthermore, the discriminatory ability of death at 3 years, according to the ROC curve, is not so different than the Cancer of the Liver Italian Program (CLIP) system. However, the consideration that HCC recurrence carries a very poor prognosis is important.
Unpublished data from our group detail the frequency of tumor recurrence: disease-free survival at 1 and 3 years for TACE (40% and 5%), RFA or resection (65% and 15%), and transplant (80% and 78%). As indicated by Toyama, we need to develop prognostic systems that identify a priori patients with high risk of tumor recidivism to help guide treatment practices. Unfortunately, at the present time, no strategy for secondary HCC prevention has been validated. Clinical trials addressing prevention of recidivism are eagerly anticipated.
In conclusion, although our ability to treat HCC has improved over the years, a new horizon in genomic and proteomic prognostication could redefine our approach to this complex cancer. Until that day arrives, treatment of HCC will remain a formidable task for the multidisciplinary team. We must continue to refine the existing models and develop new ones that incorporate the variables that make HCC such a multidimensional disease.
1. El-Serag HB, Rudolph KL. Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology. 2007;132:2557–2576.
2. Sangiovanni A, Prati GM, Fasani P, et al. The natural history of compensated cirrhosis due to hepatitis C virus: a 17-year cohort study of 214 patients. Hepatology. 2006;43:1303–1310.
3. Sala M, Forner A, Varela M, et al. Prognostic prediction in patients with hepatocellular carcinoma. Semin Liver Dis. 2005;25:171–180.
4. Chung H, Kudo M, Takahashi S, et al. Comparison of three current staging systems for hepatocellular carcinoma: Japan integrated staging score, new Barcelona Clinic Liver Cancer staging classification, and Tokyo score. J Gastroenterol Hepatol. 2007. [Epub ahead of print].
5. Keehn DM, Frank-Stromborg M. A worldwide perspective on the epidemiology and primary prevention of liver cancer. Cancer Nurs. 1991;14:163–174.
6. Kim WR, Gores GJ, Benson JT, et al. Mortality and hospital utilization for hepatocellular carcinoma in the United States. Gastroenterology. 2005;129:486–493.
7. Di Bisceglie AM. Epidemiology and clinical presentation of hepatocellular carcinoma. J Vasc Interv Radiol. 2002;13:S169–S171.
8. Namieno T, Kawata A, Sato N, et al. Age-related, different clinicopathologic features of hepatocellular carcinoma patients. Ann Surg. 1995;221:308–314.
9. Grieco A, Pompili M, Caminiti G, et al. Prognostic factors for survival in patients with early-intermediate hepatocellular carcinoma undergoing non-surgical therapy: comparison of Okuda, CLIP, and BCLC staging systems in a single Italian centre. Gut. 2005;54:411–418.
10. Semela D, Dufour JF. Angiogenesis and hepatocellular carcinoma. J Hepatol. 2004;41:864–880.
11. Dominguez-Malagon H, Gaytan-Graham S. Hepatocellular carcinoma: an update. Ultrastruct Pathol. 2001;25:497–516.
12. Fattovich G, Llovet JM. Risk factors for hepatocellular carcinoma in HCV-cirrhosis: what we know and what is missing. J Hepatol. 2006;44:1013–1016.
13. Donato F, Tagger A, Chiesa R, et al. Hepatitis B and C virus infection, alcohol drinking, and hepatocellular carcinoma: a case-control study in Italy. Brescia HCC Study. Hepatology. 1997;26:579–584.
14. Bialecki ES, Di Bisceglie AM. Clinical presentation and natural course of hepatocellular carcinoma. Eur J Gastroenterol Hepatol. 2005;17:485–489.
15. Lerose R, Molinari R, Rocchi E, et al. Prognostic features and survival of hepatocellular carcinoma in Italy: impact of stage of disease. Eur J Cancer. 2001;37:239–245.
16. Makatsoris T, Petsas T, Tsamandas AC, et al. Hepatocellular carcinoma in hepatectomized patients: biologic and therapeutic implications. Anticancer Res. 2005;25:3067–3073.
17. Parfitt JR, Marotta P, Alghamdi M, et al. Recurrent hepatocellular carcinoma after transplantation: use of a pathological score on explanted livers to predict recurrence. Liver Transpl. 2007;13:543–551.
18. Gish RG. Hepatocellular carcinoma: overcoming challenges in disease management. Clin Gastroenterol Hepatol. 2006;4:252–261.
19. Thorgeirsson SS, Grisham JW. Molecular pathogenesis of human hepatocellular carcinoma. Nat Genet. 2002;31:339–346.
20. Coleman WB. Mechanisms of human hepatocarcinogenesis. Curr Mol Med. 2003;3:573–588.
21. Okazaki N, Yoshida T, Yoshino M, et al. Screening of patients with chronic liver disease for hepatocellular carcinoma by ultrasonography. Clin Oncol. 1984;10:241–246.
22. Gambarin-Gelwan M, Wolf DC, Shapiro R, et al. Sensitivity of commonly available screening tests in detecting hepatocellular carcinoma in cirrhotic patients undergoing liver transplantation. Am J Gastroenterol. 2000;95:1535–1538.
23. Okuda K, Ohtsuki T, Obata H, et al. Natural history of hepatocellular carcinoma and prognosis in relation to treatment. Study of 850 patients. Cancer. 1985;56:918–928.
24. Marrero JA, Fontana RJ, Barrat A, et al. Prognosis of hepatocellular carcinoma: comparison of 7 staging systems in an American cohort. Hepatology. 2005;41:707–716.
25. Ueno S, Tanabe G, Sako K, et al. Discrimination value of the new western prognostic system (CLIP score) for hepatocellular carcinoma in 662 Japanese patients. Cancer of the Liver Italian Program. Hepatology. 2001;34:529–534.
26. Levy I, Sherman M. Staging of hepatocellular carcinoma: assessment of the CLIP, Okuda, and Child-Pugh staging systems in a cohort of 257 patients in Toronto. Gut. 2002;50:881–885.
27. Farinati F, Rinaldi M, Gianni S, et al. How should patients with hepatocellular carcinoma be staged? Validation of a new prognostic system. Cancer. 2000;89:2266–2273.
28. Freeman RB Jr. Transplantation for hepatocellular carcinoma: the Milan criteria and beyond. Liver Transpl. 2006;12:S8–S13.
29. Mazzaferro V, Regalia E, Doci R, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med. 1996;334:693–699.
30. Schwartz M. Liver transplantation for hepatocellular carcinoma. Gastroenterology. 2004;127:S268–S276.
31. Lee JS, Heo J, Libbrecht L, et al. A novel prognostic subtype of human hepatocellular carcinoma derived from hepatic progenitor cells. Nat Med. 2006;12:410–416.
32. Yamamoto T, Uenishi T, Ogawa M, et al. Immunohistologic attempt to find carcinogenesis from hepatic progenitor cell in hepatocellular carcinoma. Dig Surg. 2005;22:364–370.
33. Mitsuhashi N, Shimizu H, Ohtsuka M, et al. Angiopoietins and Tie-2 expression in angiogenesis and proliferation of human hepatocellular carcinoma. Hepatology. 2003;37:1105–1113.
34. Fujii T, Nomoto S, Koshikawa K, et al. Overexpression of pituitary tumor transforming gene 1 in HCC is associated with angiogenesis and poor prognosis. Hepatology. 2006;43:1267–1275.
35. Iizuka N, Oka M, Yamada-Okabe H, et al. Oligonucleotide microarray for prediction of early intrahepatic recurrence of hepatocellular carcinoma after curative resection. Lancet. 2003;361:923–929.
36. Wang W, Yang LY, Huang GW, et al. Genomic analysis reveals RhoC as a potential marker in hepatocellular carcinoma with poor prognosis. Br J Cancer. 2004;90:2349–2355.
37. Fields AC, Cotsonis G, Sexton D, et al. Surviving expression in hepatocellular carcinoma: correlation with proliferation, prognostic parameters, and outcome. Mod Pathol. 2004;17:1378–1385.
© 2008 Lippincott Williams & Wilkins, Inc.
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Data is temporarily unavailable. Please try again soon.