Primary liver cancer represents approximately 4% of all new cancers diagnosed worldwide, and of all the neoplasias originating in the liver, approximately 90% correspond to hepatocellular carcinoma.1

There has been a global increase in hepatocellular carcinoma incidence over the last few decades; it is the fifth most frequent neoplasia and the third cause of cancer-related death. The greatest prevalence is in Asia and Africa, whereas prevalence is much lower in America and Europe.2 In the United States, the Hispanic population has experienced the greatest annual percentage increase in the last decade. Compared with other ethnic groups, this population presented an important increase and according to a study by El-Serag et al. encompassing the decade of 1992 to 2002, women predominated at 63%, compared with 31% for men.3 This was confirmed by Alterkruse et al.4 in their analysis of death certificates in the U.S. Hispanic population carried out from 1975 to 2005. They also reported a triplication of hepatocellular carcinoma in this population, finding an annual percentage increase of 4% within their study time frame. It is striking that 40% of the Hispanics included in the analysis were born outside of the United States and that the increase in this disease was greater in native Hispanics than in their immigrant peers.

A rise in incidence of this pathology has been observed in Mexico for several decades. Cortes-Espinosa et al.5 demonstrated that incidence doubled within a 25-year period (1965-1990). They obtained this datum through a necropsy analysis (n = 12,556) at a referral hospital in Mexico City (0.35% for 1965-1969 vs 0.69% in 1985-1989). More recently, Méndez-Sánchez et al.6 analyzed official death certificates in Mexico from the year 2000 to 2006 and reported a national increase of 14% in the death rate from HHC (4.16 deaths per 100,000 inhabitants in the year 2000 vs 4.74 in 2005). Women experienced the greatest change, with a 15% increase, compared with the 12.5% increase in men. And finally, the Mexican National Health Information System (available at www.sinais.salud.gob.mx) analyzed causes of death from 1979 to 2008 in the population and corroborated an increase in mortality from this disease from 0.4% in the 1980s to 1.3% for 2008. According to the available information, the age groups above 55 years are the most affected, with an equal mortality rate for men and women.

Table 1 presents the data from the published series in both the texts and abstracts on patients with HCC and its primary epidemiologic characteristics. In relation to sex, the international literature generally describes a predominance in men; according to the national information presented by different authors6–10 and recent international epidemiologic reports,3,4 predominance has been equaling out over the last decade and may possibly increase for women in the future. According to the international literature, the most affected age groups are persons in the seventh decade of life in regions with low prevalence and from the sixth decade of life in regions with high prevalence. In Mexico, as shown in table 1, hepatocellular carcinoma presents mainly in the sixth decade of life. The main risk factor for 70 to 90% of the patients with HCC is cirrhosis of the liver. Currently, the two main causes of cirrhosis of the liver in Mexico are alcoholic liver disease (39.5%) and hepatitis C virus infection (36.6%); to a lesser degree are cryptogenic cirrhosis (10.4%), primary biliary cirrhosis (5.7%), hepatitis B virus (5%), and other reported causes (2.8%).11 It is noteworthy that cryptogenic cirrhosis has been catalogued as the third cause in relation to cirrhosis frequency, which could correspond to nonalcoholic fatty liver disease, given its strong association with obesity, diabetes mellitus, and dyslipidemia.

Before the discovery of the hepatitis C virus, 2 case series published in Mexico described alcoholic liver disease as the main risk factor for HCC.7,12 After its identification, different case series described this virus as the main risk factor associated with hepatocellular carcinoma. Table 1 describes these and other epidemiologic characteristics related to this pathology. The presence of underlying hepatopathy in the cases of HCC varies from 55%7 to 92%,12 a figure in accordance with internationally published figures.

The presence of advanced fibrosis without cirrhosis associated with hepatitis B and C virus infections has been reported to be a possible etiologic factor for developing HCC.13 It should be underlined that there is a higher risk associated with inflammatory activity and a greater viral burden in the presence of chronic virus B infection. However, this does not appear to be a frequent cause of HCC in Mexico, where the prevalence of this infection is considered low (HBsAg < 2% in the general population) according to the World Health Organization. In addition, universal hepatitis B vaccination was implemented in Mexico in 1998, and infants born in that year and those that followed have been vaccinated, resulting in coverage above 80% in the infant population for the year 2007, according to the official website of that organization.

Obesity is another recognized factor that predisposes to an increase in the incidence of different cancers, including HCC. This tumor is 4.5% more frequent in men with a BMI above 35 and 1.7% more frequent in women with a BMI above 35, compared with normal-weight subjects.1 The presence of diabetes mellitus is a frequent finding, particularly in cases of cryptogenic cirrhosis, which in turn, can be secondary to nonalcoholic fatty liver disease, and therefore its causal role in HCC is not clearly defined. Other risk factors described for the development of HCC include iron toxicity, aflatoxin exposure, Wilson disease, alpha-1-antitrypsin deficiency, and other metabolic disorders whose prevalence in Mexico is unknown.

Recommendations:

• -

  HCC incidence in Mexico has increased over the last few decades (23/24 = 95.8%).

• -

  The mean age for HCC presentation in Mexico is in the sixth decade of life and the woman:man ratio appears to be equal (22/24 = 91.6%).

• -

  HCC is frequently associated with underlying hepatopathy (23/24 = 95.8%).

• -

  Chronic hepatitis C virus infection is the main cause of HCC reported in Mexico (22/24 = 91.6%).

• -

  Chronic hepatitis B virus infection does not appear to be a frequent cause of HCC in Mexico, whereas cirrhosis due to nonalcoholic fatty liver disease may emerge in the near future as an important cause of HCC in Mexico (23/24 = 95.8%).

• -

  Hereditary metabolic diseases are a very infrequent cause of HCC in Mexico (23/24 = 95.8%).

• -

  There is no available data on aflatoxin exposure in the Mexican population (23/24= 95.8%).

Screening consists of the application of minimally invasive tests that enable early-stage disease detection, allowing curative treatment to be offered and thus improving the survival rate. In the high-risk population for HCC, lesions < 2cm are considered potentially curable. Surveillance is the programmed application of these screening tests to this risk group.

Surveillance is carried out by the attending physician of patients with chronic liver disease (cirrhosis), a high-risk group for developing HCC.14,15

HCC surveillance requires a multidisciplinary team that includes hepatologists, radiologists, and pathologists specifically trained in the detection and study of space-occupying lesions of the liver.

The most widely used screening study is transabdominal ultrasound. It is a simple, relatively inexpensive, and minimally invasive study with approximately 65-80% sensitivity and over 90% specificity. Despite its being operator-dependent, it is the screening study recommended by the current international guidelines,16–18 in the absence of a more effective biomarker,19,20 and approach algorithms are available in the event of an abnormal study. According to a recent meta-analysis carried out by Singal et al.,21 the addition of alpha-fetoprotein to this strategy is of minimal benefit and raises costs. Abdominal ultrasound is recommended every 6 months, even though the ideal interval is unknown. It is believed that a period of 4 to 12 months is needed for an undetectable lesion to become identifiable (2cm) by this method and undergo curative treatment. In contrast, performing the study at shorter intervals (every three months) increases the frequency of abnormal findings, without modifying the survival rate of those patients with confirmed and treated tumors.22

A frequent ultrasound finding is an echogenic nodule surrounded by a hypoechogenic halo, and when Doppler color ultrasound is applied, aberrant intratumor vascularity can be observed.19,23,24 However, it should be emphasized that any recently appearing or not previously identified nodule merits evaluation. Routine AFP determination is not recommended as an isolated screening and surveillance test because it is not cost-effective.19,25–27

Surveillance is indicated in high-risk groups, specifically in patients with cirrhosis of the liver of any etiology, with sufficient liver reserve (Child-Pugh stage A and B) and adequate functional capacity. Contrastingly, screening for HCC is not recommended for patients with Child-Pugh C cirrhosis of the liver, except in patients considered for placement on a waiting list for liver transplantation.28

When a nodule smaller than one centimeter is identified through abdominal ultrasound In risk groups under surveillance, an ultrasound study every 3-4 months is recommended for a period of 18-24 months. If there is growth in the morphologic behavior of the image during this follow-up period, a dynamic study such as multiphase computed tomography or nuclear magnetic resonance should be carried out.16,29

To be considered technically adequate, dynamic studies should include the following phases: non-contrast phase, hepatic artery phase, portal vein phase, and late phase.16,30 It is very important to standardize the multiphase dynamic study to avoid interpretation variation and errors.

The choice between computed axial tomography (CAT) or magnetic resonance depends on experience and availability in relation to each hospital center. The typical dynamic study image of an HCC tumor is a hypo/isodense lesion in the noncontrast phase that is stronger in the early arterial phase and has an early venous washout.16

In a high-risk population, a dynamic study with an image typical of an HCC lesion is sufficient for making HCC diagnosis. If the image is atypical, another dynamic study should be carried out; if the studies are not concordant, then the recommendation is to perform a liver biopsy.16,17,31 Biopsy is not justified in high-risk groups with nodules regarded as typical lesions, and likewise, biopsy is not indicated in patients that are not candidates for any specific type of treatment.16,17

If a patient without cirrhosis has a lesion that is impossible to diagnose through imaging studies, biopsy of the lesion is always recommended.31 Ultrasound-guided or CT-guided Tru-Cut biopsy is suggested and should always be reviewed by a pathologist with training and/or experience in hepatic lesions. When the liver biopsy is difficult to interpret, histochemistry studies can be used, such as reticulum staining and immunohistochemistry with CD 34, glypican-3, heat shock protein 70, and glutamine synthetase. Unfortunately, the majority of these are not available for routine use; their foremost value is in distinguishing between regeneration nodule, dysplastic nodule, and HCC.31,32

According to current management recommendations,16,17,22 the determination of alpha-fetoprotein levels has no diagnostic usefulness, even though it could be useful in evaluating treatment response if levels were initially high.

There are 7 proposals in regard to staging and they include the TNM classification system created by the American Joint Committee on Cancer (AJCC), the Cancer of the Liver Italian Program (CLIP), the Japanese Integrated System (JIS), the Grupe d’Etude de Traitment du Carcinoma Hepatocellulaire (GRETCH), the Chinese University Prognostic Index (CUPI), the Okuda system, and the Barcelona Clinic Liver Cancer (BCLC) system. Traditionally, tumor prognosis is directly related to its size and extension, but in the case of HCC, liver function reserve also affects outcome and limits treatment. Therefore, the ideal classification should include variables related to the tumor, as well as liver function classification and functional capacity of the individual presenting with 2 diseases, cirrhosis and cancer, and finally, it should include the efficacy of the treatment that was administered. The Okuda, TNM, CLIP, and JIS classifications do not include variables that evaluate the functional status of the individual, limiting their prognostic capacity. Both the CUPI and the GRETCH consider indirect liver function reserve data. Of these 7 systems, the only scale that includes the abovementioned points is the BCLC, originally published in 1999.33 Marrero et al.34 compared the 7 different existing scales in a consecutive cohort of 239 individuals with HCC and found that the main independent survival predictors were tumor diameter, the presence of portal vein thrombosis, the Model of End Stage Liver Diseases (MELD) score, the functional capacity of the individual, and the administered treatment. Of the 7 scales, the BCLC had the greatest independent survival predictive power, because all these parameters are included in its variables, and therefore it is the staging classification recommended by the American and European guidelines, as well as by our group.16,17 Even though recent information suggests that stage B (intermediate) may be sufficiently heterogeneous to merit being a subclassification, this remains to be defined in the future.35

Recommendations:

• -

  Surveillance is recommended for groups at high risk for developing HCC in Child-Pugh stages A or B, and also stage C, but only if the patient is on a waiting list or considered a candidate for liver transplantation (22/24 = 91.6%).

• -

  Surveillance for risk groups should be carried out with abdominal ultrasound every six months (23/24 = 95.8%).

• -

  All new lesions or those not previously identified in high-risk patients merit an initial evaluation with either a multiphase computed tomography or nuclear magnetic resonance dynamic study (23/24 = 95.8%).

• -

  A dynamic study with a characteristic lesion > 1cm in a high-risk population is sufficient for making diagnosis (23/24 = 95.8%).

• -

  A dynamic study with an undetermined lesion in this population makes another dynamic study or guided liver biopsy obligatory (23/24 = 95.8%).

• -

  Any recently appearing lesion in a healthy liver that cannot be diagnosed through an imaging study merits a liver biopsy (20/24 = 83.3%).