Hepatocellular carcinoma prevention: A worldwide emergence between the opulence of developed countries and the economic constraints of developing nations

Abstract

Hepatocellular carcinoma (HCC) is the fifth most common neoplasm, the major cause of death in patients with liver cirrhosis, and the third most common cause of cancer-related death in the world. The geographic distribution of HCC varies significantly and 80% of cases occur in developing countries (Far East and South Asia) where the prevalence of viral hepatitis is higher. The treatment of HCC is difficult because most patients are diagnosed when the tumour is in an advanced stage and is not amenable to potential curative therapy, thus prevention is the key to reducing HCC and its related morbidity and mortality. HCC is unique among cancers, occurring mostly in patients with a known risk factor. Ninety percent of HCCs develop in the context of chronic liver diseases and mainly in patients with cirrhosis. Viral hepatitis is the most common cause of HCC worldwide, followed by alcoholic liver disease (ALD) and other causes such as non-alcoholic fatty liver disease (NAFLD), genetic haemocromatosis (GH) and primary biliary cirrhosis in an advanced stage (III-V). In certain areas of the People’s Republic of China, exposure to aflatoxin and HBV infection are thought to be responsible for the extraordinary high risk of HCC. Substantial progresses in the prevention of virusl-related hepatitis (screening of blood units, use of disposable sanitary tools, HBV vaccination) have been achieved in developed countries, but in the same areas, alcohol- and dysmetabolism-related HCCs are emerging problems which require specific interventions in terms of public health measures. In developing countries, economic constraints limit the development of any program for the prevention of viral hepatitis transmission (including health education campaigns, healthcare politics, primary prevention and the improvement of hygienic and sanitary conditions). When viral liver disease is established, only a minority of patients are treated worldwide and benefit a possible preventive effect of medical treatment on HCC development. Thus the real contribution of medical treatment to HCC prevention in patients with chronic viral hepatitis is small. Great efforts are needed to identify more effective medical measures for primary and secondary prevention of HCC.

INTRODUCTION

Hepatocellular carcinoma (HCC) is the fifth most common neoplasm, the major cause of death in patients with liver cirrhosis, and the third most common cause of cancer-related death in the world[1,2]. Every year, 560 000 people in the world develop HCC and about the same number die of liver cancer. It is reported to be the leading cause of death in patients with cirrhosis in Europe[3] and Asia[4]. In Italy, its incidence varies from 2% to 4% each year in patients with cirrhosis[5-7].

The geographic distribution of HCC varies significantly and 80% of cases occur in developing countries (Far East and South Asia) where the prevalence of viral hepatitis is higher[8]. Epidemiological data show that its incidence is changing around the world according to the etiology since it is increasing in many developed countries, whereas HCC is declining in developing countries[9]. Studies on the epidemiology and natural history of chronic hepatitis C virus (HCV) infection suggest that HCC frequency has been increasing over the past 20-30 years in USA[10,11] and Europe[12]. HCC is one of the few types of cancer increasing in frequency and mortality in USA[11]. In Japan there has been a three-fold increase in HCC incidence since 1970[13]. Moreover, it has been shown that 80% of newly diagnosed HCCs in Japan are HCV-positive patients[14]. It has been hypothesized that HCC incidence and mortality will indeed increase over the next 10-20 years because of the increased incidence of HCV-related cirrhosis[10]. With regard to the decline of hepatocellular cancer in developing countries, the introduction of mass hepatitis B virus (HBV) vaccination programs seems to play an important role. In Taiwan the carrier rate of HBsAg in children and adolescents has been significantly reduced[15], as has the incidence of childhood HCC[16]. Nevertheless, the decrease in the incidence of HCC among adults is still a long way off and is not expected for another 3 or 4 decades later[17]. Thus, the role of HBV vaccination in the current decline of HCC incidence, at least in Taiwan, is still to be clarified. In addition, other factors may have a role in changing the epidemiology of HCC incidence around the world: the migration of people from endemic areas and better detection of HCC in Western countries.

The treatment of HCC is difficult because most patients are diagnosed when the tumour is in an advanced stage and is not amenable to potential curative therapy[18], thus prevention is the key to reducing HCC and its related morbidity and mortality.

Notably, HCC is unique among cancers, occurring mostly in patients with a known risk factor. Ninety percent of HCCs develop in the context of chronic liver diseases and mainly in patients with cirrhosis[19,20]. Viral hepatitis is the most common cause of HCC worldwide, accounting for more than 80% of cases, followed by alcoholic liver disease (ALD) and other causes such as non-alcoholic fatty liver disease (NAFLD), genetic haemocromatosis (GH) and primary biliary cirrhosis in an advanced stage (III-V)[20-23]. In certain areas of the People’s Republic of China where HCC is the leading cause of cancer death, exposure to aflatoxin, a potent hepatocarcinogen produced by food contaminant, and HBV infection are thought to be responsible for the extraordinary high risk of HCC[24].

Therefore, a prevention strategy should be addressed to each single risk factor ranging from food conservation to alcohol abuse, metabolic disease (NAFLD) and viral infection. Then, HCC prevention can focus on different levels: prevention of the underlying disease (primary prevention), prevention of liver disease progression to cirrhosis and cancer (secondary prevention), prevention of cancer relapse in successfully treated HCC (tertiary prevention).

HCC RELATED TO VIRUSES AND OTHER FACTORS

HBV

HBV-related chronic hepatitis is the most common cause of HCC in the world[25]. The incidence and ways of transmission of this virus vary substantially in different endemic areas throughout the world. HBV is usually acquired in adulthood and its incidence is low in Western countries, while its incidence is high in Asia and in most African countries and its transmission is vertical from mother to child[26].

The relationship between HBV and HCC development is not completely understood. Nevertheless, it seems that two main mechanisms play a role in the induction of liver cancer: one involves repeated liver damage causing necroinflammation and consequent regeneration of hepatocytes, the other is the direct oncogenic property of HBV which integrates itself into the hepatocyte DNA, producing cis- or transactivation of the cellular oncogenes[27,28]. It is important to note that HBV and HCV have been classified among human carcinogens by the International Agency for Research on Cancer[1].

There is a relationship between viral replication (defined as the presence of detectable HBV-DNA by non-PCR assay or HBeAg positivity) and HCC development. A recent study showed that the cumulative risk of HCC development is higher (87%) in patients at the age of 30-70 years persistently positive for HBsAg and HBeAg than in those (12%) who are positive for HBsAg only and in those (1%) who are HBsAg-negative[29]. HBeAg prevalence, a marker of infectivity and viral proliferation, decreases with increasing age. HBeAg seroconversion (both spontaneous and IFN-induced) usually confers a favourable outcome. However, the selection of the e minus mutant (HBV-DNA positive) leads to the development of cirrhosis (not only in the Mediterranean area but also worldwide) which is the prerequisite for HCC development[30].

Therefore, two strategies can be used to prevent HBV-related HCC: one is the prevention of HBV infection, the other which is more questionable is the treatment of HBV-related liver disease and cirrhosis. Only a minority of patients treated with interferon-α (IFN-α) achieve HBs clearance while the majority of HBsAg-positive and HBeAg-negative patients have a persistent viral replication. These patients are still at risk of developing cancer. About 35% of patients[31] who respond to IFN-α treatment achieve HBs clearance and therefore can be considered cured. In the remaining HBsAg-positive patients who can clear HBeAg, a persistent viral replication is still detectable[31]. These patients, especially those with liver cirrhosis remain at risk of developing cancer. On the other hand, the occult HBV infection (HBsAg-negative patients) is related to an increasing or a persistant risk of HCC development in both anti-HCV-related and unrelated cirrhosis[32].

Therefore, occult HBV infection should be investigated specially in patients with chronic HCV infection and considered as an additional factor related to HCC development.

HCV

HCV-related carcinogenesis is still not clearly understood. Hepatitis C virus infection increases the risk of developing HCC and its prevalence among patients with liver cancer varies between different geographic areas from 20% to 90%[10,33]. HCC occurs mostly in patients with end stage liver disease and advanced fibrosis[34]. In this population, the risk of developing liver cancer is 2%-8% per year[35-37] based on clinical studies. The only prospective study evaluating HCC risk in HCV-positive patients, has attributed a 20-fold increased risk to viral infection, but the presence of cirrhosis in the study population was not evaluated[38]. Other studies have not evaluated the presence of co-factors, such as alcohol intake or HBV co-infection and the eventual role of antiviral therapy. Therefore, a bias may exist in the literature, which may explain the variation in the incidence estimates of HCC among patients with HCV infection[11].

HCV itself probably has a direct role in the induction of HCC. Some studies showed that the HCV replicon and both non-structural and core viral proteins are capable of inducing a complex series of intracellular events leading to apoptosis suppression and cellular growth through the down-regulation of PKR and the up-regulation of transcription factors including NF-κB, STAT-3 and ATF-6[39-41].

Other causes of HCC

Alcohol consumption is a risk factor for the development of HCC[42,43]. HCC develops mostly in patients with cirrhosis although a direct carcinogenetic role of alcohol has been documented and is a known risk factor also for cancers other than HCC[44]. Ethanol is mainly metabolized in the liver, producing acetaldehyde and free radicals which produce liver injury and DNA damage through the increase of oxidant stress[45]. Moreover, alcohol abuse produces an accumulation of iron which in turn, contributes to oxidative stress[46] with an additive and possibly synergistic mechanism causing hepatocellular damage.

Hereditary hemochromatosis is a common genetic defect in the Caucasian population causing iron accumulation in different organs as well as in the liver[47]. As has already been stated, iron itself may have a direct role in the development of HCC. The risk of developing HCC in these patients is reported to be 20-200 times greater than that in healthy controls[48-51]. Male sex, older age, cirrhosis and the presence of co-factors such as alcohol, viral hepatitis and tobacco smoking are known risk factors[52,53].

Non-alcoholic fatty liver disease (NAFLD) has recently been described in cohorts of patients as a cause of HCC[54-56], but larger prospective studies are needed to define the real incidence and risk factors for HCC development in this setting of patients

Aflatoxins are hepatocarcinogens produced by Aspergillus Flavus and contaminate the food supply in South-east Asia and Africa[57,58], increasing the risk of HCC development, especially in patients with HBV infection, by causing a mutation of the p53 gene[59]. This kind of contamination occurs mostly in developing countries with a hot-humid climate where a variety of oilseeds and cereal crops are produced[60].

PUBLIC HEALTH MEASURES FOR PRIMARY PREVENTION OF LIVER DISEASE

As already mentioned, HCC occurs in the context of known risk factors and its treatment is disappointing. Therefore the best way to prevent its development is preventing the onset of acquired liver diseases and applying surveillance programs for patients with estabilished liver diseases.

In developed countries, viral hepatitis infection has decreased since the mid-1980s when blood donor screening for viral hepatitis became available. Moreover, the use of medical disposables has contributed to this phenomenon[61] and nowadays, the risk of HBV infection is limited to sexual risk relations, intravenous drug users and in a few cases of patients undergoing dental therapy, acupuncture, piercing and tattooing[62-64]. In developing countries, this field is still a matter of public health because almost half of the blood units are not screened for viral hepatitis[65], sanitary conditions are poor and disposable sanitary tools are often not available.

With regard to HBV, the development of a specific vaccine has dramatically changed the virus epidemiology. This is very important in developing countries where HBV transmission is mostly vertical. HBV vaccination has been shown to be able to prevent cancer[16]. Effective vaccines have been available since 1982 and the World Health Organization has recommended the implementation of mass immunization programs since 1991. This has led to a significant decrease in HCC incidence among children and adolescents in Taiwan and it is expected to occur in the next few decades, also in those countries where the vaccination program was started later[16,66-68]. Different HBV vaccines are available, with recombinant HBs antigen (Table 1). HBV vaccination is one of the fundamental and effective forms of public health prevention measures. Unfortunately, immunization programs have a cost and a large proportion of children do not receive basic vaccines because of economic problems in many developing countries[65]. A plant-derived HBV vaccination could be a cheaper approach, this vaccine is under study with the aim of incorporating the vaccine into the alimentary chain[69].

Table 1.

Recommended doses and schedules for HBV vaccine

Vaccine	Patients	Dose (mcg)	Volume (mL)	Schedule (mo)
Engerix-B	< 11 yr	10	0.5	0, 1, 6
(GlaxoSmithKline, Research Triangle Park, NC)	11-19 yr	10	0.5	0, 1, 6
	> 20 yr	20	1.0	0, 1, 6
	Dialysis	40	2.0	0, 1, 2, 6
Recombivax HB (Merk & Co. Inc., Whitehouse Station, NJ)	< 11 yr	5	0.5	0, 1, 6
	11-19 yr	5	0.5	0, 1, 6
	> 20 yr	10	1.0	0, 1, 6
	Predialysis/ dialysis	40	1.0	0, 1, 6

Especially in developing countries where HBV transmission is mainly vertical with a high prevalence of infection, the screening for HBV is advisable at least in late pregnancy. Mothers positive for HBV-DNA should be given lamivudine or continue lamivudine treatment in order to reduce the viral load and the risk of vertical transmission[70]. Passive immune prophylaxis and immediate vaccination of newborns are required immediately within the perinatal period[70].

Unfortunately, an effective vaccine against HCV is still unavailable. The lack of a small animal model, genomic HCV diversity and the difficulties in establishing an in vitro culture of large quantities of HCV are some of the problems encountered by researchers in this field[71]. Therefore, the uncertainty of obtaining an effective vaccine imposes the call for the hygienic strategies already applied in the Western countries in order to reduce the incidence of new cases from about 50 × 10^5[72,73] to < 1 × 10⁵ inhabitants[74].

The first step in the prevention of alcohol and NAFLD-related HCC is the fight against chronic abuse and the education for a healthier lifestyle in the general population. Similarly, warnings against alcohol abuse or unsuitable diets and a campaign for the control of body weight through moderate physical activity could play a major role. These concepts are also good for the primary prevention of colon, breast and pancreatic cancer which recognizes a risky lifestyle as their primum movements.

Iron depletion is reported to have a protective effect[75] against the development of hepatocellular carcinoma in genetic hemochromatosis, therefore phlebotomy should be performed when indicated in patients with iron accumulation.

With regard to aflatoxin, the best preventive approach would be pre-harvesting crop management, avoiding infection of the crop with Aspergillus[60]. Irrigation and use of fungicides or pesticides may be helpful, but this approach is expensive, especially in poor countries. Post-harvesting technologies are available to limit fungus growth and crop contamination. There are drugs capable of modulating the aflatoxin metabolism once ingested. Oltipraz, an antischistosomal drug, has been shown to be effective in detoxifying patients with serum aflatoxin-albumin adducts[76]. Chlorophyllin is a cheaper drug which has been tested in a perspective, randomised study in China with good results[77].

PATIENTS WITH ESTABLISHED LIVER DISEASE

HBV

Although it varies according to different geographic areas, HCC incidence among subjects chronically infected with HBV is higher in cirrhotics than in patients with chronic viral hepatitis[6,7,18,30,78-100] (Table 2, adapted from[101]), suggesting that other factors play a relevant role in cancerogenesis. In general, endemic East Asian countries as compared to Western countries where HBV prevalence is low or intermediate have a higher incidence of HCC among all groups considered: asymptomatic carriers, inactive carriers, chronic hepatitis and cirrhosis. It was reported that the five-year cumulative incidence in cirrhotic patients is 15% in East Asian studies and 10% in European studies with a 3-fold higher risk of developing liver cancer in endemic areas than in those with a lower prevalence[101].

Table 2.

HCC incidence rates according to clinical setting and geographic areas

Clinical setting	Geographic area	Studies (n)	Patients (n)	Mean follow- up (yr)	HCC incidence	95% CI
Asympto- matic carrier	North America	2	1804	16	0.1	0.07-0.14
	Taiwan and China	4	18869	8	0.7	0.61-0.70
	Japan	1	513	7.3	0.2	0.08-0.39
Inactive carrier	Europe	3	410	16	0.02	0-0.04
	Taiwan	1	189	8	0.2	0-0.42
Chronic hepatitis	Europe	6	471	5.9	0.1	0-0.27
	Taiwan	2	461	4.0	1.0	0.36-1.56
	Japan	2	737	5.1	0.8	0.46-1.06
Compensated cirrhosis	Europe	6	401	5.8	2.2	1.62-2.80
	Taiwan/ Singapore	3	278	4.3	3.2	1.94-4.55
	Japan	2	306	5.8	4.3	3.40-5.25

Adapted from ref. 101 with permission from American Gastroenterological Association.

This phenomenon is due to several factors associated with an increased risk of HCC development such as patient’s age at onset of infection, core promoter variants, the presence of HBe antigen, and probably HBV genotypes. However, the role of HBe antigen and HBV DNA replication in the development of HCC is still being debated. A large study from Taiwan on 11 893 men with chronic hepatitis B, found that the risk of developing HCC is 10-fold higher in patients with HBsAg alone at diagnosis and 60-fold higher in those with both HBs and HBeAg than in HBsAg negative patients[102]. The EUROHEP cohort, with a smaller sample size and a low incidence of HCC showed different results, suggesting that probably in different geographical areas, the same co-factors may produce different results[78]. However, two additional European cohort studies showed that cirrhotic patients, who cleared HBeAg and HBV-DNA (and eventually HBsAg), and achieved ALT normalization, are at low risk of developing HCC[3,103]. Therefore, suppressing HBV replication which represents “per se” a major risk factor may prevent HCC development. Only one randomised controlled trial has reported a decreased incidence of HCC in 67 Taiwanese men treated with interferon as compared to 34 untreated men followed up for 1-12 years. HCC occurred in 1.5% of patients in the treated group as compared to 12% in the untreated group (P = 0.04). It is important to note that the only patient who developed HCC in the treated group initially cleared HBeAg, but relapsed later, again becoming HBeAg-positive with elevated transaminases[96]. A meta-analysis of 7 studies comparing treated patients versus untreated controls with HBV-related compensated cirrhosis has shown a very weak protective effect of interferon with a 6.4% difference in risk (95% CI: 2.8% to 10%)[104].

In conclusion, interferon treatment may have a beneficial effect on the development of HCC in HBV-infected patients, particularly when a virological response is achieved.

Nowadays, different nucleotide and nucleoside analogues are available for the treatment of HBV. The oldest in this family is lamivudine, a cytidine analogue, which is able to inhibit HBV replication, improve liver enzymes and inflammatory score and arrest progression to fibrosis[105]. A retrospective study evaluated the efficacy of lamivudine in terms of HCC prevention in 377 Japanese patients, compared to the same number of untreated HBV infected controls. HCC occurred in 1.1% of patients with an annual incidence of 0.4% (patient/year) in the treated group and in 13.3% of patients with an annual incidence of 2.5% (patient/year) in the untreated group (P < 0.001)[106]. Only one prospective randomized controlled study has been conducted using HCC development as an endpoint. In this study 651 patients with HBV-related cirrhosis or advanced fibrosis were randomly assigned with a 2:1 ratio, to receive lamivudine or a placebo. The study was stopped after a median duration of treatment of 32.4 mo based on the recommendations of an independent data and safety monitoring board because of a significant difference between the treated groups in the number of endpoints reached. HCC occurred in 3.9% in the lamivudine group and 7.4% in the placebo group (hazard ratio, 0.49; P = 0.047)[25]. This study showed that lamivudine plays a role in preventing HCC but a longer follow-up is needed to confirm this result in consideration of the high rate of lamivudine resistance which could reactivate HBV DNA replication.

A large Italian retrospective analysis of 656 patients with chronic HBV infection, with or without cirrhosis, showed that the likelihood of developing HCC is significantly less in cirrhotic patients having a virological response than in those having a virological breakthrough[84]. Further studies are also needed to evaluate the effect of the newest antiviral drugs (adefovir, entecavir, tenofovir, emtricitabine, etc) in terms of HCC prevention.

In conclusion, different factors contribute to the risk of HCC development in patients with HBV infection: HBV replication, HBV direct oncogenetic effect through the integration in host genoma and cirrhosis itself. Until now, the only mechanism we could control pharmacologically is viral replication and consequently liver disease evolution to cirrhosis. Therefore, a strict imaging follow-up is recommended in all patients even those responding to therapy.

HCV

HCC incidence is higher in patients with cirrhosis due to chronic hepatitis C infection with a variable incidence in cirrhotics of 2%-8% per year[35-37,107]. As for HBV its incidence varies between geographic areas. Studies from Japan [79,108-112] have documented a summary HCC incidence of 1.8 per 100 subjects per year in patients with chronic hepatitis as compared to 7.1 in patients with cirrhosis. Therefore there is a 4-fold risk of developing HCC when cirrhosis is present[101]. In Europe and the United States[5-7,23,36,37,78,80,113-117], the summary incidence is 3.7 in cirrhotic patients whereas it is impossible to calculate it in non-cirrhotic patients due to the lack of HCC in the only study available[36]. Analyzing these data showed that the 5-year cumulative risk of HCC development in cirrhotics is 17% in Europe and 30% in Japan[101]. Thus the best way of preventing HCC development in HCV-infected patients is preventing cirrhosis itself. This is a challenging issue but difficult to carry out, as a large proportion of patients are unaware of their status and the disease is mostly asymptomatic. Therapy is offered to a small proportion of patients worldwide, and moreover, the infection can be eradicated in only about half of the treatable patients[118-120].

IFN may prevent HCC by preventing liver damage evolution by the eradication of viral hepatitis, this is generally the case of sustained responders[121]. Eradicating HCV infection may prevent HCC development, at least in patients without cirrhosis. With regard to cirrhosis, a different incidence of HCC between treated and untreated cirrhotics (Table 3) and a decline in HCC incidence in patients achieving a sustained virological response have been reported[104,108]. The residual risk is related to the cirrhosis itself. Moreover, if some dysplastic or neoplastic cells are present, IFN alone is not capable of eradicating them. As a consequence, we should follow up patients by ultrasonography even after eradication of the infection. In fact, the cirrhotic background probably persists and HCC may develop years after clearance of the virus[122]. A study recently published by our group has shown the best results in terms of HCC prevention in cirrhotic patients treated with IFN plus ribavirin as compared to those treated with IFN alone, and the reduction in HCC incidence after re-treatment of non-responders[122]. The tight relationship between a high Ag-NOR proliferative index [silver stained (Ag)-nucleolar organizing region (NOR)] and HCC development in this study confirms that clinical utility of cellular proliferation markers can predict HCC development as reported in previous studies[5,122]. Since IFN is an anti-proliferative cytokine, it may have a beneficial role in preventing HCC development in cirrhotic paptiens through the reduction of hepatocyte proliferation, even when a virological response is not achieved[122]. Moreover, a recent retrospective study has shown a lower incidence of HCC in patients with HCV chronic hepatitis who were non-responders to long-term IFN treatment as compared to those treated for less than 24 mo[123].

Table 3.

Effect of IFN on HCC incidence among HCV cirrhotic patients

Author, year	n	Follow-up month (range)	HCC rate
			Treated (n/n)	Controls (n/n)
Nishiguchi 1995[150]	90	54 (24-86)	2/45	17/45
Mazzella 1996[7]	284	32 (12-71)	5/193	9/91
Bruno 1997[5]	163	68 (60-84)	6/82	14/81
Fattovich 1997[35]	329	60 (1-153)	7/193	16/136
IIHCSG 1998[151]	491	n.r.	21/232	48/259
Imai 1998[111]	52	48 (3-65)	8/32	7/20
Gramenzi 1998[115]	144	72 (n.r.)	6/72	19/72
Serfaty 1998[116]	103	40 (6-42)	2/59	9/44
Sofia 1998[152]	162	43 (n.r.)	11/103	4/59
Benvegnù 1999[6]	152	72 (n.r.)	4/75	20/77
Mura 1999[153]	57	76 (n.r.)	0/28	5/29
Shioda 1999[154]	646	55 (n.r.)	22/588	18/58
Yoshida 1999[108]	337	52 (n.r.)	33/230	29/107
Valla 1999[117]	99	37 (37-53)	5/47	9/52
Overall	3109		132/1979	224/1130

Adapted from ref. 104 with permission from EASL. Meta-analysis of the three randomised controlled trials and 11 non randomised trials. Risk difference: -12.8; 95% CI: -8.3-17.2; P ≤ 0.0001 (chi square test for heterogeneity).

Data on non-cirrhotic patients derive from retro-spective analysis. The largest from Japan, included 2890 patients, 490 of whom were untreated. HCC incidence was higher in the untreated group (3.1% per year) than in the treated group (1.1% per year), but the difference reached a statistical significance only among patients with stage 2 or 3 fibrosis. Patients were also stratified according to virological response showing the greater benefit in responder patients with fibrosis F2-F3[108]. These data have been confirmed by a recent update of the study[124].

This study has also confirmed that HCC development seems related to the transaminase level, which is basically the rationale of treatment, the aim of which is to reduce liver inflammation. The long term use of glycyrrhizin, an aqueous extract from licorice root capable of reducing transaminases when administered for a short period of time, can significantly reduce HCC incidence in HCV patients[125]. Long-term glycyrrhizin injections have recently been shown to reduce the incidence of HCC in patients with HCV-related chronic hepatitis C with or without cirrhosis and in non-responders to IFN-based treatment[126]. Similarly, ursodeoxycholic acid (UDCA) administration has been shown to reduce transaminases in patients with HCV-related chronic hepatitis although the data are insufficient to support this effect on viral markers, mortality, incidence of cirrhosis, or liver histology[127,128]. The mechanisms involved in ALT normalisation are unknown. However, the reduction of HLA antigen expression and cholestasis, changes in membrane plasticity and bile acid pool hydrophobicity may reduce liver inflammation[129]. All these mechanisms are probably involved in the ability of UDCA to lower the incidence of HCC both in rats and in humans[130,131]. Vitamin K₂ may also have a protective effect against HCC development in women with cirrhosis[132].

The preventive role of these drugs needs to be further investigated and their use in the treatment of HCV is not recommended as the only effective drug is interferon.

Finally, epidemiological studies have shown that coffee consumption lowers the incidence of HCC in patients with HCV-, HBV- or alcohol-related cirrhosis[133-136], and that the preventive effect is cup-dependent (3 cups decrease the risk of HCC by 75%).

HCC RECURRENCE PREVENTION (TERTIARY PREVENTION)

HCC recurrence is about 80% after 5 years of curative treatment[137,138] with a local tumour progression of 10% after 5 years of radiofrequency ablation[138]. Recurrence may originate in intrahepatic metastasis from the treated tumour or from multicentric occurrence.

Several approaches to the prevention of HCC recurrence have been attempted. The most convincing seems to be polyprenoic acid administration. Polyprenoic acid is an acyclic derivate of retinoic acid which has been shown to have chemopreventive activity[139-142] by interfering with cellular regulation and differentiation. It was reported that oral administration of polyprenoic acid could significantly reduce the incidence of new HCCs after a median follow-up of 38 mo[143,144]. Its effect does not seem to be mediated by the reduction of liver inflammation (the transaminase level did not differ between the treatment and control groups)[145], but is related to the clearance of dysplastic/HCC clones by apoptotic mechanisms. The mechanism of polyprenoic acid seems to be different from that of interferon which seems to be related to multiple mechanisms such as increased immune-surveillance, S0 arrest of the cell cycle and restoration of apoptosis. A randomised controlled study on patients with HCV-related HCC showed that administration of IFN-α for 48 wk after the effective ethanol injection treatment of HCC, could improve the prognosis by reduction of the second and third recurrence with an improvement of survival in treated patients[146]. Thus, IFN seems to play a more significant role in cirrhosis as a pre-cancer condition than in liver cancer itself. IFN-α lymphoblastoids also seem to be capable of reducing HCC recurrence and improving survival in HBsAg-positive patients with their HCC resected[147]. Another small randomised controlled study showed that the recurrence rate is lower in patients treated with IFN-β after surgery or curative alcohol injection than in untreated patients[148]. The problem in these studies is that they included a small number of patients who were treated with different kinds of IFN and different treatment schedules.

Another approach is irradiation of the liver using intra-arterial iodine¹³¹-labeled lipiodol after HCC resection with the aim of eradicating neoplastic foci. This strategy has been used in a small randomised controlled trial, showing a reduction in the recurrence rate and an increase in disease-free survival when compared to controls[149].

There is evidence that IFN and polyprenoic acid have a modest effects at least in small studies. In practice, when HCC develops in a cirrhotic liver, medical treatment has a very limited efficacy on the prevention of recurrence.

CONCLUSION

HCC prevention is a public health challenge. Since HCC mostly develops in a sick liver and the secondary prevention is of little effect, the best strategy is to prevent the onset of liver diseases.

Substantial progresses have been achieved in developed countries in the prevention of virusl-related hepatitis (screening of blood units, use of disposable sanitary tools, HBV vaccination) , but alcohol- and dysmetabolism-related HCCs are emerging problems in the same areas, which require specific interventions in terms of public health measures.

In developing countries, there is still a long way to go. In fact, economic constraints limit the development of any program for the prevention of viral hepatitis transmission (including health education campaigns, healthcare politicies, primary prevention and improvement of hygienic and sanitary conditions). In this setting, the first step should be a political approach aimed to increase people awareness, extensive vaccination programs, blood units screening, sanitary conditions and use of disposables in poor countries. It is obvious that pharmacological approach should follow. In Western countries our efforts should be directed towards educational approach and counceling activities in order to reduce alcohol abuse and obesity.

As a rule, when the diagnosis of viral liver disease is established, only a minority of patients are treated worldwide and benefit from a possible preventive effect of medical treatment for HCC. Once again economic support in developed countries should be given to treat those patients who could benefit from antiviral treatment.

With regard to HCC recurrence, the real contribution of medical treatment to prevention of chronic viral hepatitis is small. However, more effective medical measures against secondary and tertiary prevention of HCC should be taken.

Finally, the role of co-factors in the development of HCC is still unknown and should be further investigated. Co-factors may also clarify the large differences in the geographic incidence of HCC and give further targets of intervention.