EUS-guided FNA could be another important tool for the early diagnosis of hepatocellular carcinoma

Abstract

CalloutIt is unlikely that an invasive and costly test like EUS could be used for routine screening/surveillance of hepatocellular carcinoma unless it is used as a part of screening/surveillance for esophageal varices.Hepatocellular carcinoma (HCC) is the 5th most common cancer in the world, with a very wide geographic variation.1Parkin D.M. Bray F.I. Devesa S.S. Cancer burden in the year 2000. The global picture.Eur J Cancer. 2001; 37: S4-S66Abstract Full Text Full Text PDF PubMed Google Scholar Although the incidence is high in many parts of Asia, Africa, and Southern Europe, it has been low in North America and western Europe until recently. However, the incidence of HCC has recently increased in the United States and Europe, perhaps because of the silent epidemic hepatitis C virus (HCV).2El-Serag H.B. Mason A.C. Rising incidence of hepatocellular carcinoma in the United States.N Engl J Med. 1999; 340: 745-750Crossref PubMed Scopus (2702) Google Scholar, 3Nair S. Shiv Kumar K. Thuluvath P.J. Mortality from hepatocellular and biliary cancers: changing epidemiological trends.Am J Gastroenterol. 2002; 97: 167-171Crossref PubMed Google Scholar, 4Davila J.A. Morgan R.O. Shaib Y. et al.Hepatitis C infection and the increasing incidence of hepatoceullular carcinoma: a population based study.Gastroenterology. 2004; 127: 1372-1380Abstract Full Text Full Text PDF PubMed Scopus (419) Google Scholar In the United States, for example, there has been a 3-fold increase in the incidence of HCC between 1981 and 1998. Although HCC may develop in the absence of cirrhosis in hepatitis B virus (HBV), it is almost always seen in the presence of cirrhosis in other chronic liver diseases, including HCV.4Davila J.A. Morgan R.O. Shaib Y. et al.Hepatitis C infection and the increasing incidence of hepatoceullular carcinoma: a population based study.Gastroenterology. 2004; 127: 1372-1380Abstract Full Text Full Text PDF PubMed Scopus (419) Google Scholar, 5Davila J.A. Morgan R.O. Shaib Y. et al.Diabetes increases the risk of hepatocellular carcinoma in the United States: a population based case control study.Gut. 2005; 54: 533-539Crossref PubMed Scopus (545) Google Scholar, 6Beasley R.P. Hwang L.Y. Lin C.C. et al.Hepatocellular carcinoma and hepatitis B virus. A prospective study of 22,707 men in Taiwan.Lancet. 1981; II: 1129-1133Abstract Scopus (402) Google Scholar, 7Fattovich G. Stroffolini T. Zagni I. et al.Hepatocellular carcinoma in cirrhosis: incidence and risk factors.Gastroenterology. 2004; 127: S35-S50Abstract Full Text Full Text PDF PubMed Scopus (1872) Google Scholar, 8Bruix J. Sherman M. Management of hepatocellular carcinoma. AASLD Practice Guideline.Hepatology. 2005; 42: 1208-1233Crossref PubMed Scopus (5017) Google Scholar The 5-year cumulative incidence of HCC in patients with HCV cirrhosis is around 15%, and the risk increases 2- to 6-fold in the presence of other concomitant risk factors, such as alcohol, obesity, diabetes mellitus, and HBV.5Davila J.A. Morgan R.O. Shaib Y. et al.Diabetes increases the risk of hepatocellular carcinoma in the United States: a population based case control study.Gut. 2005; 54: 533-539Crossref PubMed Scopus (545) Google Scholar It is unlikely that an invasive and costly test like EUS could be used for routine screening/surveillance of hepatocellular carcinoma unless it is used as a part of screening/surveillance for esophageal varices. The prognosis of patients with symptomatic HCC is dismal, with a 5-year survival between 0% to 10%. If the diagnosis is made at an early stage, before vascular invasion or before the tumor becomes very large, then the prognosis is much better, with a 5-year survival ranging from 50% to 80% when using treatment modalities such as percutaneous alcohol injection, radiofrequency ablation, or liver transplantation.9Mazzaferro 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: 693-699Crossref PubMed Scopus (5529) Google Scholar, 10Yao F.Y. Ferrell L. Bass N.M. et al.Liver transplantation for hepatocellular carcinoma: expansion of the tumor size does not adversely impact survival.Hepatology. 2001; 33: 1394-1403Crossref PubMed Scopus (1714) Google Scholar, 11Yoo H.Y. Patt C.H. Geschwind J.F. et al.The outcome of liver transplantation in patients with hepatocellular carcinoma in the US between 1988 and 2001: 5-year survival has improved significantly with time.J Clin Oncol. 2003; 21: 4329-4335Crossref PubMed Scopus (214) Google Scholar Unfortunately, the majority of patients present at a very late stage, when most treatment options, except for transarterial chemoembolization, are not applicable because of advanced tumor size, metastases, or vascular invasion. Early diagnosis could be achieved by screening and surveillance of high-risk patients if sensitive and specific blood tests or imaging modalities are easily available and if those tests are cost effective. The risk of HCC is higher than 1.5% in patients with cirrhosis (of any cause) and higher than 0.2% in HBV carrier per year.8Bruix J. Sherman M. Management of hepatocellular carcinoma. AASLD Practice Guideline.Hepatology. 2005; 42: 1208-1233Crossref PubMed Scopus (5017) Google Scholar By using theoretical models, it has been suggested that screening is cost effective in these patients. Blood tests that are currently available for clinical use, including alpha-fetoprotein (AFP), are either insensitive or nonspecific as screening/surveillance tests for HCC. Although higher cutoff values may improve specificity (AFP >400 ng/mL), they reduce sensitivity to less than 20%, making the test of borderline utility. In contrast, a lower cutoff (>20 ng/mL) may improve the sensitivity (∼60%) while reducing specificity to less than 20%. However, elevated AFP (>200 ng/mL) in the presence of a mass on imaging has a very high positive predictive value, and, similarly, a rising AFP also has a very high positive predictive value of making a diagnosis of HCC. Other blood tests, such as des-gamma-carboxy-prothrombin, AFP-L3 fraction, alpha fucosidase, and glypican 3, have not been adequately tested as screening/surveillance or diagnostic tests for HCC.8Bruix J. Sherman M. Management of hepatocellular carcinoma. AASLD Practice Guideline.Hepatology. 2005; 42: 1208-1233Crossref PubMed Scopus (5017) Google Scholar Therefore, most centers depend on imaging modalities, such as abdominal US, CT, or magnetic resonance imaging (MRI) for screening/surveillance and diagnosis of HCC. Of these, US performed every 6 to 12 months has been evaluated extensively in many countries and has been found to have adequate sensitivity (60%-80%) and specificity (around 80%), and, thus, it is currently recommended as a screening/surveillance test for HCC in high-risk patients as per American Association for the Study of Liver Disease (AASLD) guidelines.8Bruix J. Sherman M. Management of hepatocellular carcinoma. AASLD Practice Guideline.Hepatology. 2005; 42: 1208-1233Crossref PubMed Scopus (5017) Google Scholar In addition, it has been suggested that abdominal US every 6 to 12 months is also cost effective as a screening/surveillance tool. However, this experience is not uniformly shared by many hepatologists in the United States, including this author. Abdominal US is operator dependent, with a variable sensitivity and specificity in the presence of advanced cirrhosis and especially in patients with obesity. HCC tumors may appear as hyper-, hypo-, or isoechoic lesions on US (which will also apply to EUS), making the diagnosis difficult in the presence of underlying cirrhosis, especially in the presence of regenerating or dysplastic nodules. Although not well studied as a screening/surveillance test, based on their diagnostic superiority, many centers use triple-phase high-resolution CT, or contrast-enhanced MRI for screening/surveillance of HCC, despite its costs. With improvements in technology, sensitivity of both CT and MRI has increased significantly, while significantly decreasing the specificity. A firm diagnosis of HCC, without histologic confirmation, could be made if a mass lesion shows “early wash-out” on contrast-enhanced triple-phase CT, 2 separate imaging modalities are suggestive of HCC, or a combination of 1 imaging test plus AFP ≥200 ng/mL.8Bruix J. Sherman M. Management of hepatocellular carcinoma. AASLD Practice Guideline.Hepatology. 2005; 42: 1208-1233Crossref PubMed Scopus (5017) Google Scholar In the absence of these, histology is the only firm method of establishing the diagnosis of HCC. The biopsy becomes problematic when small lesions identified on CT or MRI are not visualized readily on abdominal US, which occurs quite frequently in the author's experience. Although FNA or biopsy of these lesions could be performed under CT, this requires considerable expertise. Therefore, investigators have been searching for better blood tests (biomarkers) and imaging modalities for the screening and diagnosis of HCC. In this issue of Gastrointestinal Endoscopy, Singh et al12Singh P. Erickson R.A. Mukhopadhyay P. et al.EUS for detection of the hepatocellular carcinoma: results of a prospective study.Gastrointest Endosc. 2007; 66: 265-273Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar report on the utility of EUS for the diagnosis of HCC in a prospective study that involved 17 patients. All patients had EUS, and the majority had a CT (n = 15), an MRI (n = 14), and a US (n = 13). EUS was performed before a CT and MRI in 53% and 60%, respectively. Eight of these patients were found to have HCC, and, based on this study, the investigators report that the diagnostic accuracy of US, CT, MRI, EUS, and EUS-guided FNA (EUS-FNA) for the diagnosis of HCC were 38%, 69%, 65%, 92%, and 94%, respectively. In this study, the specificity of EUS was only 25%, and, when combined with FNA, specificity rose to 100%, suggesting that the real advantage of EUS is the ability to perform FNA during the same procedure. There are few concerns with this study. Although this is reported as a “comparative” study, more than half of these patients had EUS or EUS-FNA before either CT or MRI. In addition, half of the patients who were enrolled in the study were found to have HCC, a remarkably high prevalence, unless these patients were “selected” based on “highly” suspicious findings. Moreover, this high prevalence may skew the diagnostic accuracy of the tests because diagnostic accuracy is dependent on the prevalence. In patients with cirrhosis, even those with suspicious findings, the reported prevalence of HCC found with any diagnostic modality is not anywhere near 50%, unless these patients had significantly elevated AFP or highly suggestive findings on other imaging modalities, such as US, CT, or MRI. Because AFP was normal in 5 patients with HCC and only moderately elevated in another 3, it is probable that EUS was performed based on suspicious findings on abdominal US. Hence, one may have to conclude that EUS or EUS-FNA was used as a diagnostic test in this study rather than a screening or surveillance test. Previous case series also reported the utility of EUS and FNA for the diagnosis of primary liver cancer, hilar cholangiocarcinoma, and left lobe liver metastases, with sensitivity in the range of 82% to 94% and positive predictive value reaching 100%.13Nguyen P. Feng J.C. Chang K.J. Endoscopic ultrasound (EUS) and EUS-guided fine-needle aspiration (FNA) of liver lesions.Gastrointest Endosc. 1999; 50: 357-361Abstract Full Text Full Text PDF PubMed Scopus (153) Google Scholar, 14Awad S.S. Fagan S. Abudayyeh S. et al.Preoperative evaluation of hepatic lesions for the staging of hepatocellular and metastatic liver carcinoma using endoscopic ultrasonography.Am J Surg. 2002; 184: 601-604Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar, 15DeWitt J. LeBlanc J. McHenry L. et al.Endoscopic ultrasound-guided fine needle aspiration cytology of solid liver lesions: a large single-center experience.Am J Gastroenterol. 2003; 98: 1976-1981PubMed Google Scholar, 16Hollerbach S. Willert J. Topalidis T. et al.Endoscopic ultrasound-guided fine-needle aspiration biopsy of liver lesions: histological and cytological assessment.Endoscopy. 2003; 35: 743-749Crossref PubMed Scopus (43) Google Scholar, 17McGrath K. Brody D. Luketich J. et al.Detection of unsuspected left hepatic lobe metastases during EUS staging of cancer of the esophagus and cardia.Am J Gastroenterol. 2006; 101: 1742-1746Crossref PubMed Scopus (35) Google Scholar, 18Fritscher-Ravens A. Broering D.C. Sriram P.V. et al.EUS-guided fine-needle aspiration cytodiagnosis of hilar cholangiocarcinoma: a case series.Gastrointest Endosc. 2000; 52: 534-540Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar These studies suggest that EUS-FNA has an impact in the management of patients with undiagnosed small liver lesions, especially when they are seen in the left hepatic lobe or the hilum. The current study suggests that lesions in the right lobe also could be visualized on EUS; however, the investigators give no information on the exact locations of these lesions in the right lobe. In other studies, similar to the current study, new or additional malignant liver lesions that were not seen on dynamic CT were identified on EUS, suggesting that EUS may be more sensitive for identifying smaller liver lesions.13Nguyen P. Feng J.C. Chang K.J. Endoscopic ultrasound (EUS) and EUS-guided fine-needle aspiration (FNA) of liver lesions.Gastrointest Endosc. 1999; 50: 357-361Abstract Full Text Full Text PDF PubMed Scopus (153) Google Scholar, 14Awad S.S. Fagan S. Abudayyeh S. et al.Preoperative evaluation of hepatic lesions for the staging of hepatocellular and metastatic liver carcinoma using endoscopic ultrasonography.Am J Surg. 2002; 184: 601-604Abstract Full Text Full Text PDF PubMed Scopus (68) Google Scholar, 15DeWitt J. LeBlanc J. McHenry L. et al.Endoscopic ultrasound-guided fine needle aspiration cytology of solid liver lesions: a large single-center experience.Am J Gastroenterol. 2003; 98: 1976-1981PubMed Google Scholar, 16Hollerbach S. Willert J. Topalidis T. et al.Endoscopic ultrasound-guided fine-needle aspiration biopsy of liver lesions: histological and cytological assessment.Endoscopy. 2003; 35: 743-749Crossref PubMed Scopus (43) Google Scholar, 17McGrath K. Brody D. Luketich J. et al.Detection of unsuspected left hepatic lobe metastases during EUS staging of cancer of the esophagus and cardia.Am J Gastroenterol. 2006; 101: 1742-1746Crossref PubMed Scopus (35) Google Scholar, 18Fritscher-Ravens A. Broering D.C. Sriram P.V. et al.EUS-guided fine-needle aspiration cytodiagnosis of hilar cholangiocarcinoma: a case series.Gastrointest Endosc. 2000; 52: 534-540Abstract Full Text Full Text PDF PubMed Scopus (147) Google Scholar The current study and previous studies showed that adequate tissue for cytopathologic assessment of liver lesions visualized on EUS could be obtained without any major complication by EUS-FNA. The retrospective nature, heterogeneity of patient population, and patient-selection bias make it difficult to draw any firm conclusions from these studies. Nevertheless, it is reasonable to conclude that EUS and FNA may be helpful in the management of a subset of patients with small liver lesions that are difficult to sample by percutaneous US-guided techniques. What else can we learn from these case series? It is unlikely that an invasive and costly test like EUS could be used for routine screening/surveillance of HCC unless it is used as a part of screening/surveillance for esophageal varices. Moreover, EUS may not be able to visualize all 8 segments of the liver, thereby reducing the sensitivity of the test considerably. The real importance of EUS may be to make a firm tissue diagnosis of suspicious lesions in the segments of the liver (left lobe and caudate lobe, and, possibly, part of the right lobe) that could be easily visualized by EUS. This case series and previous studies suggest that EUS could be a valuable tool for this purpose, because suspicious lesions could be sampled by FNA without another additional procedure. However, many questions remain unanswered from this small series and similar previous studies. What is the sensitivity, specificity, positive and negative predictive values of EUS and FNA for the diagnosis of HCC in a large group of unselected patients with cirrhosis and suspicious lesions? Is it going to be more accurate and safer than percutaneous US-guided biopsy or aspiration? The risk of needle-track spread of HCC is estimated to be less than 1% with percutaneous US-guided biopsy based on large series, but the risk of spread from EUS + FNA remains undefined. Although the risk of EUS-FNA is thought to be low for many other cancers, including pancreatic cancer and lymph-node aspiration, liver cancers may be different for many reasons, including increased tumor vascularity and longer distance that has to be traversed to enter the liver capsule, increasing the risk of possible spillage of tumor cells into the peritoneal cavity. Because patients with small HCC have excellent outcomes, with 5-year tumor-free survival reaching close to 80% with liver transplantation, any additional risk of recurrence from peritoneal spillage may have an important negative impact.9Mazzaferro 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: 693-699Crossref PubMed Scopus (5529) Google Scholar, 10Yao F.Y. Ferrell L. Bass N.M. et al.Liver transplantation for hepatocellular carcinoma: expansion of the tumor size does not adversely impact survival.Hepatology. 2001; 33: 1394-1403Crossref PubMed Scopus (1714) Google Scholar, 11Yoo H.Y. Patt C.H. Geschwind J.F. et al.The outcome of liver transplantation in patients with hepatocellular carcinoma in the US between 1988 and 2001: 5-year survival has improved significantly with time.J Clin Oncol. 2003; 21: 4329-4335Crossref PubMed Scopus (214) Google Scholar Careful observational studies may answer many of these important questions. Based on their observations, the investigators proposed an algorithm for the diagnosis of suspicious lesions in the liver. It is important to define what investigators mean by “suspected” HCC. It is also perhaps premature to recommend an algorithm based on this small study and previous case series. If a suspicious lesion is found on US and not found on a 3-phase, rapid sequence CT, one could repeat a CT in 6 to 12 weeks or do an MRI if the index of suspicion is high, before attempting EUS or EUS-FNA. In patients where the suspected lesion is in the left lobe or the caudate lobe, EUS ± FNA may be an alternate option. FNA should be used judiciously in patients with suspected HCC, and it should be done only if the diagnosis will have an impact on the management, because the real risk of recurrence from peritoneal spillage is unknown, especially in those patients who are candidates for liver transplantation. However, EUS-FNA would be extremely useful in patients with HCC and suspicious lymph nodes, because these patients could be excluded from transplantation, if they were found to have extrahepatic spread. Future studies should focus on carefully mapping out the segments that could be visualized consistently on EUS, defining HCC imaging characteristics and its variations in the presence of cirrhosis; staging of tumors, including vascular invasion; establishing the risks associated with EUS-FNA; and, finally, exploring the potential and efficacy of treatment modalities such as EUS-guided ethanol injection of small HCC. Comparative studies would also be useful to determine whether EUS is more sensitive or specific than other modalities, such as US, CT, or MRI, for the diagnosis of HCC, at least in those segments that could be clearly visualized by EUS. It is very unlikely that EUS could be used as a primary screening or surveillance test because of the incomplete imaging, its invasiveness, and costs. The exploratory study published in this issue of Gastrointestinal Endoscopy clearly suggests that EUS may be another useful tool in the diagnosis of small HCC. The author has no conflict of interest. EUS for detection of the hepatocellular carcinoma: results of a prospective studyGastrointestinal EndoscopyVol. 66Issue 2PreviewEarly detection of hepatocellular carcinoma (HCC) and accurate determination of the number of lesions are critical in determining eligibility for liver transplantation or resection. Current diagnostic modalities (CT and magnetic resonance imaging [MRI]) often miss small lesions. Full-Text PDF