Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide and ranks in the top three causes of cancer deaths in the Asia-Pacific (AP) region.1,2 Hepatitis B and C virus (HBV and HCV) infections are the most common causes of HCC worldwide. Due to the high prevalence of HBV in the AP region, 75% of HCC patients are seen in this region. The incidence of HCC has been static over the years in the AP region; however, it is rising in the western world, Japan and Australia due to an epidemic of HCV infections.1,2 The number of patients with HCC is expected to increase by two times over the next two decades.3 Eighty percent of HCCs develop in patients with liver cirrhosis. The annual incidence of HCC in HBV-related cirrhosis varies from 2% to 6%, while in HCV-related cirrhosis it is 3–5%.4 The majority of HCCs are detected at a late stage with high mortality. Thus, the yearly fatality ratio is close to one indicating almost all patients with HCC die within one year.
There have been significant advances in diagnostic and therapeutic modalities for early HCC. During 1980–1990, detection of early HCC and curative treatment was possible in only 5–10% patients, while this number increased to 30–40% in 1990–2010.1 In a Japanese study,5 it has been shown that in the last three decades there has been an increasing incidence of early stage HCCs, which has led to the potentially curative treatment of these patients. Clinic-based studies from Italy have also shown that there is a decreasing trend in mortality in liver cirrhosis patients with HCC in the last three decades.6 Looking at these data it seems reasonable to have a surveillance program for early detection of HCC.
Hepatocellular carcinoma is a suitable target for a cancer screening program as it qualifies the following criteria: (i) It is a significant health problem. (ii) It occurs in a well-defined at-risk population. (iii) Cirrhosis is the primary risk factor. (iv) HCC has a protracted subclinical phase. (v) Treatment of sub-clinical disease offers advantage over treatment of symptomatic disease. (vi) During the sub-clinical phase there are no distinctive symptoms. (vii) More than 80% of the tumors detected in the symptomatic stage are unresectable. (viii) Prognosis of early HCC has improved significantly. The routinely used screen tests for HCC are ultrasound and/or alpha fetoprotein (AFP), which are affordable and acceptable to the population and these screening tests have moderate accuracy.1,7,8
Surveillance for HCC has been recommended by the various guidelines published by the hepatology and gastroenterology organization around the world, as well as a recent AP Working party.8 HCC surveillance has been recommended for patients with chronic HBV-related cirrhosis and for certain categories of chronic HBV-related non-cirrhotic patients (males above the age of 40 and females above the age of 50 years, patients with family history of HCC, and patients with high serum HBV DNA (> 2000 IU/mL). All patients with chronic HCV-related cirrhosis should be screened (especially patients with age more than 40 years, patients with concomitant alcoholism, chronic HBV or HIV co-infection or metabolic risk factors (obesity, diabetes). All other patients with liver cirrhosis are recommended to undergo surveillance. However, the benefits of an HCC surveillance program in this population are uncertain.7,8
The outcomes of a HCC surveillance program depends on data from clinical trials converted into clinical practice. The main issues of outcome of HCC screening are: (i) Is it used? (ii) How is it used? (iii) Frequency and type of patient population. (iv) Recall strategy. (v) Is appropriate therapy given?
It has been estimated that surveillance practices are followed by more than 60% of physicians worldwide who consult on patients with cirrhosis.9 The benefits and harms associated with screening are unknown. There is no randomized controlled trial to study the effect of surveillance. Indeed, one study published in abstract form showed that in an attempted randomized controlled trial of surveillance for HCC more than 80% of the informed patients declined to participate and preferred to undergo ultrasound surveillance versus no surveillance.10 In a recently published study from the USA, it was shown that for patients who were on a standard of care surveillance program, nearly 70% with HCC were eligible for liver transplantation, as compared with 35% of HCCs diagnosed outside a formal surveillance program.11 Only 61% of the HCCs referred had received surveillance,11 and 32% of the 70% patients eligible for liver transplantation received a donor organ.
In a recently conducted meta-analysis for surveillance with ultrasound for early stage HCC in patients with cirrhosis, the authors found limited sensitivity of 63%, and AFP provided no additional benefit to ultrasound. Although 6-monthly intervals were better than yearly interval,12 AFP has limited efficacy and is not recommended for surveillance except when ultrasound is not available. However, in spite of widespread practice of HCC surveillance programs and an increasing array of treatment options, fewer than half of the candidates for potentially curative treatment of HCC actually receive it.
Cost effective and cost utility analysis of HCC surveillance was studied in a systemic review which included 29 study reports.13 The overall conclusion from these studies was that an HCC surveillance program increases the diagnosis of small HCCs which are amenable to potential curative treatment. Incremental cost effective ratio for 6-monthly AFP and ultrasound varies between $US24 500 to $46 000 per quality-adjusted life-year. The impact on quality of life in cirrhotic patients undergoing surveillance was highest in younger patients. Impact on quality of life in HCC patients was seen in those who underwent liver transplantation. Cost effective analysis based on a computerized decision analytical model from seven studies showed ultrasound plus AFP 6-monthly in a mixed etiology cohort is the most effective surveillance strategy. Cost effectiveness of surveillance strategies was highest in HBV-related cirrhosis and lowest in alcoholic cirrhosis. Factors that affect the cost effectiveness are the rate of incidentally detected small HCCs and annual incidence of HCC in the risk group. Adoption of liver transplantation as a treatment strategy and younger age of screen population are also relevant.8
In this issue of JGH, Qian et al.14 report their results on a retrospective review of all patients who underwent HCC screening in their hospital for 6 years. This analysis showed the benefits of a HCC screening program. Ultrasonography and AFP were used for HCC screening. Out of 22 detected HCCs, 17 were potentially curable, but at the end of follow up, only 10 patients were alive. Of these 10 patients, six had received liver transplantation and three had received locoregional ability therapy. The cost per potentially curable HCC was $A17 680. Although this study is a retrospective single tertiary care centre, it addresses important issues of HCC surveillance. The surveillance technique and treatments offered were the best standard of care for the present situation. This study highlights the benefits of liver transplantation as an important modality for treatment of HCC. Liver transplantation offers a cure for underlying liver cirrhosis and HCC, and hence becomes a more effective modality than locoregional therapies.
Surveillance of HCC is appropriate and effective, but we need to do much better. A surveillance program should be widely and effectively applied not only by hepatologists and gastroenterologists, but internists and surgeons taking care of these patients. We need better ultrasound techniques and serum markers that are more sensitive and specific for the detection of early HCC. Finally, liver transplantation needs to be more widely available as a treatment modality for patients with HCC.