Abstract This manuscript provides a concise surgical review of hepatic epitheliod hemangioendothelioma. A detailed review of diagnosis, pre-surgical radiologic evaluation, surgical techniques, including liver transplantation, and post-surgical care of the patient with hepatic epitheliod hemangioendothelioma is presented. Keywords Hepatic epitheliod hemangioendothelioma • Liver surgery • Liver transplantation • Hepatobiliary surgery • Liver tumor Introduction Hepatic Epitheliod Hemangioendothelioma (HEHE) remains a diagnostic and therapeutic challenge to the practicing hepatobiliary surgeon. With less than 1000 reported cases since its initial description by Weiss and Enzinger in 1982 [ 1 ] and a widely variable clinical course, HEHE remains a diagnosis that affords the clinician a unique opportunity to tailor therapy to the patient. Infantile hemangioendothelioma, a rare neonatal vascular tumor associated with congestive heart failure, thrombocytopenia, and consumptive coagulopathy, is a unique clinical entity that will not be addressed in this manuscript [ 2 ]. Presentation HEHE is a vascular tumor of endothelial cell origin with an incidence of approximately 1/1,000,000 population [ 3 ]. Since the initial series of 32 patients reported by Ishak in 1984 [ 4 ], our understanding of this rare disease has evolved through collective case reports, database surveys, and meta analyses. HEHE expresses a slight female J. F. Renz (*) University of Chicago Medicine , 5841 S. Maryland, Room J517 MC5027 , Chicago , IL 60637 , USA e-mail: [email protected] 86 preponderance (3:2) and is most often diagnosed in the fourth decade of life [ 5 ]. Presentation can vary widely from an incidental fi nding on routine imaging described in approximately 25 % of new cases to overt liver failure . Extra-hepatic involvement is present in over a third of patients at the time of diagnosis [ 5 , 6 ]. The most frequent presentation includes a history of intermittent right upper quadrant pain , malaise, and weight loss. As the indolent tumor replaces more hepatic volume, late fi ndings of hepatomegaly, jaundice , hepatic outfl ow obstruction (Budd-Chiari syndrome), Kasabach-Merritt syndrome, hemorrhage secondary to tumor rupture, and acute liver failure emerge [ 7 – 10 ]. The presence of symptoms at diagnosis has been validated as a poor prognostic indicator by MVA [ 11 ]. To date, no clear risk factors predisposing to HEHE have emerged; however, oral contraceptives, vinyl chloride, viral hepatitis, and trauma to the liver have been implicated in its development [ 5 , 12 ]. Notably, HEHE is not associated with chronic liver disease [ 5 ]. This affords the physician typically normal hepatic parenchyma to accommodate medical, radiologic , or surgical therapy. Diagnosis As HEHE lies variably within the spectrum between hemangioma and angiosacroma, diagnosis requires integration of radiologic , histologic, and immunologic data. For the diagnosis of HEHE, magnetic resonance imaging (MRI) is emerging as the preferred therapy over ultrasound and computed tomography ( CT ) [ 13 ]. HEHE is described radiographically as two types: nodular and diffuse. The nodular type is an early manifestation of HEHE characterized by independent peripheral lesions, ranging from <1 cm to several centimeters in diameter, within the liver. Presentation typically involves both hepatic lobes with a preponderance of tumor in the right hepatic lobe. As the disease progresses, the multifocal tumors coalesce into bulky subcapsular disease throughout the liver defi ning the advanced diffuse form of HEHE. Capsular retraction develops secondary to scarring and fi brosis [ 14 ]. When evaluating a CT , the bulk of disease is best appreciated on unenhanced imaging where intra-tumoral calcifi cation and capsular retraction can be appreciated. Contrast CT fi ndings include arterial phase marginal enhancement that may appear target-like and is often described as a “halo.” The concentric zonal or target- like appearance refl ects the histology of an avascular, central stomal region with fi ngerlike tumor projections extending peripherally along hepatic sinusoids. These areas become isodense to hepatic parenchyma on post-contrast imaging [ 14 , 15 ]. On MRI, the central, hypocellular regions may demonstrate previous hemorrhage, thrombus, necrosis, or calcifi ciation with low signal T-1 weighting with T-2 hyperintesity. Gadolinium administration optimally demonstrates the peripheral halo with progressive centripetal fi lling on subsequent images. The key fi ndings for any cross-sectional imaging modality are: multiple heterogeneous lesions, subcapsular location, capsular fl attening or retraction, and peripheral delayed contrast enhancement with centripetal fi lling [ 16 ]. The utility of FDG-PET is variable. J.F. Renz 87 FDG-PET has not proven sensitive in screening or diagnosis : however, when it is positive in approximately 40 % of cases, it can be useful in monitoring response to therapy [ 17 ]. Suggestive radiologic fi ndings must be followed by histologic and immunologic analysis to secure the diagnosis of HEHE. Adequate tissue can be obtained by percutaneous , ultra-sound-guided liver biopsy or diagnostic laparoscopy . HEHE is an endothelial cell origin tumor with an appearance of spindle-shaped endothelial cells multiplying along vascular planes. The histology is variable within the spectrum of hemangioma to angiosarcoma but the tumor characteristically expresses Factor VIII-related antigen, CD34 (human hematopoietic progenitor cell antigen), and CD31 (platelet endothelial cell adhesion molecular 1). Immunoanalysis for at least two of these three antigens is necessary to secure the diagnosis. Therefore, precise pathologic interpretation is integral to identifying malignant features of HEHE and predicting clinical behavior [ 5 ]. Potential genetic translocations associated with HEHE have been postulated [ 18 ]; however, the rarity of the disease has impeded linkage analysis. Serum chemistries and standard tumor markers are non-diagnostic at presentation with one exception: an elevated CA19-9 is a negative prognostic factor for HEHE and should guide the clinician toward biliary origin malignancies [ 11 ]. Treatment The wide clinical spectrum of disease at presentation and its variable biologic behavior afford the clinician the opportunity to utilize a variety of therapeutic modalities in “tailoring” therapy to the HEHE patient. The incidence of HEHE has prevented the establishment of guidelines and resulted in the application of a multitude of successful therapeutic endeavors ranging from chemotherapy to liver transplant ation . At the moment, the benchmark therapies remain surgical and, whenever possible, resection is preferred [ 5 , 6 ]. Historically, the bulk of disease at diagnosis has favored liver transplant ation ; however, recent advances in surgical technique coupled with the fact that HEHE typically occurs in the setting of otherwise normal hepatic parenchyma have opened the possibility of good outcomes in the setting of repetitive surgical resection versus liver transplantation. Grotz et al. reported a retrospective series of 30 HEHE patients treated by surgical resection (SR), liver transplantation (LTX), medical therapy, or no therapy at the Mayo Clinic between 1984 and 2007 [ 6 ]. While patients were not randomized to SR or LTX, the group maintained a very aggressive protocol toward SR whenever possible. At a median follow- up of >41 months, the SR group, which contained approximately the same number of patients as the LTX group, demonstrated comparable disease-free survival and overall survival as LTX with a lower incidence of post-operative complications and period of hospitalization. The 1-, 3-, and 5-year overall survival for SR was 100 %, 86 %, and 86 % versus 91 %, 73 %, and 73 % for LTX, respectively. The 1-, 3-, and 5-year 8 Hepatic Epithelioid Hemangioendothelioma 88 disease-free survival for the SR group was 78 %, 62 %, and 62 % versus 64 %, 46 %, and 46 % for LTX, respectively. Hospital stay and the occurrence of Clavien ≥ stage IV complications were lower in SR but did not achieve statistical signifi cance. Clinicopathologic predictors of prolonged disease-free survival have been proposed by Grotz et al. based upon their retrospective series data but have not been prospectively validated. These include: largest tumor size ≤10 cm, total tumor number ≤10, and hepatic involvement ≤4 segments [ 6 ]. This led the authors to advocate for SR as the surgically preferred option for patients with HEHE “regardless of bilobar distribution provided the hepatic disease can be resected [ 6 ].” The recent description of liver partition with portal vein ligation for staged hepatectomy described by Schlitt and others offers a new opportunity to dramatically extend the realm of hepatic resection and thereby avoid liver transplant ation [ 19 ]. However, one must remember that HEHE is a widely variable disease entity and Grotz et al. concede the biologic behavior of each presentation factored largely into their decision to recommend curative surgical therapy [ 6 ]. An alternative strategy of hepatectomy followed by carbon-ion radiotherapy has also been advocated [ 20 ]. Intent to cure must remain the goal as palliative surgical debulking has been demonstrated to enhance progression [ 21 ]. Liver transplantation has proven a durable therapy for the treatment of HEHE. Initially described by Marino et al. in 1988 [ 22 ], the application of LTX to patients with extensive bilateral disease has yielded excellent results on three continents [ 11 , 23 , 24 ]. Mehrabi et al. performed a meta analysis from 1984 through 2005 identifying 402 cases [ 5 ]. Of this group, 45 % were treated by LTX, 25 % received no treatment, 21 % received chemotherapy and/or radiation therapy, and only 9 % SR. Within this group, the 1- and 5-year survival for LTX were 96 % and 55 % respectively. These results were bested only by the SR group that demonstrated 1- and 5-year survival of 100 % and 75 %, respectively. However, it is impossible to determine through meta analysis the extent of disease approached through SR. Notably, the authors identifi ed extra-hepatic disease in 37 % of patients at the time of diagnosis but the presence of extra-hepatic disease did not portend a poor prognosis. The unique fi nding of extra-hepatic disease not impacting long-term survival was confi rmed by Lerut et al. who reported the results of the European Liver Transplant Registry in 2007 [ 23 ]. In their analysis of 59 patients followed for a median of greater than 6 years, the disease-free survival at 1-, 5-, and 10-years postLTX were 90 %, 82 %, and 64 %, respectively. Overall recurrence in the cohort was 24 % with a median time to recurrence of 49 months. The extent of disease reported in referring to LTX included bilobar tumor 96 %, >15 tumor nodules 86 %, pre-LTX therapy 30 %, lymph node invasion 30 %, and extra-hepatic disease 17 %. In this context, the overall results obtained with LTX were excellent and led the authors to conclude pre-existing extrahepatic disease as well as lymph node localization are not contraindications to LTX. Vascular invasion upon histologic examination reduced overall patient survival but not disease-free survival. Thus, the pattern of continual treatment of a low grade malignant tumor with a slowly progressive phenotype re-emerged as the authors’ inclusion of extra-hepatic disease was limited J.F. Renz 89 to that amenable to surgical resection with or without radiation therapy. The fi nding of carcinomatosis excluded LTX [ 23 ]. Data from the United Network for Sharing on 110 transplanted patients between 1987 and 2005 were reported by Rodriguez et al. in 2008 [ 24 ]. Their analysis was limited through inclusion of children transplanted for the infantile variant of HEHE and a relatively short median follow-up of only 24 months. The authors reported patient and allograft survival on a cohort including adults and children with an overall mortality related to HEHE recurrence of 16 %, presumably all in adults as the pediatric form is thought to be benign. Unfortunately, their study was not powered to determine the effect of extra-hepatic disease at LTX [ 24 ]. When considering LTX, it is imperative to exclude angiosarcoma as its biologic behavior is an absolute contraindication [ 25 ]. Disease recurrence has been widely reported as distant as 12 years following LTX and is best approached with surgery and radiation therapy where applicable [ 5 , 26 ]. A role for adjuvant chemotherapy in the management of post-LTX recurrence is theoretically attractive but unproven. Alternative Therapies The epithelial-cell origin of HEHE and its consistent over-expression of vascular endothelial growth factor (VEGF) have made it a natural target for anti-angiogenic therapy [ 27 ]. To date, medical therapy alone has delivered inferior results to surgical therapy [ 5 , 6 ]; however, a variety of chemotherapeutics have been reported to affect HEHE in individual cases. These include thalidomide, doxorubicin, 5-fl uorouracil, vincristine, cyclophosphamide, interferon-alpha 2B, bevacizumab, sunitinib, and lenalidomide [ 28 – 33 ]. Chevreau reported results of a European multicenter, phase II trial of15 patients utilizing sorafenib [ 34 ]. Their early results were indeterminant, but as more information is elicited on the genetic composition of HEHE, the promise of medical therapy, particularly in highly aggressive disease prompting acute liver failure as well as very slowly progressing indolent disease is promising. Summary HEHE is a rare disease with a widely variable presentation and clinical course. Accurate diagnosis through a combination of radiology, histology, and immunochemistry is challenging but essential for anticipating the tumor’s biologic behavior. Ultimately, the biologic behavior guides the practitioner to the most appropriate course of therapy with surgery , either resection or transplantation, the preferred avenue for cure. However, further scientifi c understanding of this unique biologic entity may yield superior outcomes through anti-angiogenic therapy. 8 Hepatic Epithelioid Hemangioendothelioma 90 References 1. Weiss S, Enzinger F. Epitheloid hemangioendothelioma: a vascular tumor often mistaken for a carcinoma. Cancer. 1982;50:970–81. 2. Dasgupta M, Das S, Patra C, Sarker S. Symptomatic infantile hepatic hemangioendothelioma succesfully treated with steroid. J Clin Neonatol. 2013;2:187–9. 3. Bioulac-Sage P, Laumonier H, Laurent C, Blanc J, Balabaud C. Benign and malignant vascular tumors of the liver in adults. Semin Liver Dis. 2008;28:302–14. 4. Ishak K, Sesterhenn I, Goodman M, Rabin L, Stromeyer F. Epithelioid hemangioendothelioma of the liver: a clinicopathologic and follow-up study of 32 cases. Hum Pathol. 1984;15:839–52. 5. Mehrabi A, Kashfi A, Fonouni H, et al. Primary malignant hepatic epithelioid hemangioendothelioma. Cancer. 2006;107:2108–21. 6. Grotz T, Nagorney D, Donohue J, et al. Hepatic epithelioid haemangioendothelioma: is transplantation the only treatment option? HPB. 2010;12:546–53. 7. Makhlouf H, Ishak K, Goodman Z. Epithelioid hemangioendothelioma of the liver. Cancer. 1999;85:562–82. 8. Frider B, Bruno A, Selser J, Vanessa R, Pascual P, Bistoletti R. Kasabach-Merritt syndrome and adult hepatic epithelioid hemangioendothelioma an unusual association. J Hepatol. 2005;42:282–3. 9. Hayashi Y, Inagaki K, Hirota S, Yoshikawa Y, Ikawa H. Epithelioid hemangioendothelioma with marked liver deformity and secondary Budd-Chiari syndrome: pathologic and radiologic correlation. Pathol Int. 1999;49:547–52. 10. Kim G, Kim Y, Kim H, et al. A case of primary hepatic hemangioendothelioma with spontaneous rupture. Korean J Hepatol. 2009;15:510–6. 11. Wang L, Zhou J, Zhou Y, et al. Clinical experience with primary hepatic epithelioid hemangioendothelioma: retrospective study of 33 patients. World J Surg. 2012;36:2677–83. 12. Dean P, Haggitt R, O’Hare C. Malignant epithelioid hemangioendothelioma of the liver in young women. Relationship to oral contraceptive use. Am J Surg. 1985;9:695–704. 13. Chen Y, Yu R-S, Qiu L-L, Jiang D-Y, Tan Y-B, Fu Y-B. Contrast-enhanced multiple-phase imaging features in hepatic epithelioid hemangioendothelioma. World J Gastroenterol. 2011;17:3544–53. 14. Amin S, Chung H, Jha R. Hepatic epithelioid hemangioendothelioma: MR imaging fi ndings. Abdom Imaging. 2010;36:407–14. 15. Radin D, Craig J, Colletti P, Ralls P, Halls J. Hepatic epithelioid hemangioendothelioma. Radiology. 1988;169:145–8. 16. Roth C, Mitchell D. Hepatocellular carcinoma and other hepatic malignancies: MRI imaging. Radiol Clin N Am. 2014;52:683–707. 17. Dong A, Dong H, Wang Y, Gong J, Lu Z, Zou C. MRI and FDG PET/CT fi ndings of hepatic epithelioid hemangioendothelioma. Clin Nucl Med. 2013;38:66–73. 18. Woelfel C, Liehr T, Weise A, et al. Molecular cytogenetic characterization of epithelioid hemangioendothelioma. Cancer Genet. 2011;204:671–6. 19. Schnitzbauer A, Lang S, Goessmann H, et al. Right portal vein ligation combined with in situ splitting induces rapid left lateral liver lobe hypertrophy enabling 2-staged extended right hepatic resection in small-for-size settings. Ann Surg. 2012;255:405–14. 20. Komatsu S, Iwasaki T, Demizu Y, et al. Two-stage treatment with hepatectomy and carbon-ion radiotherapy for multiple hepatic epithelioid hemangioendotheliomas. World J Gastroenterol. 2014;20:8729–35. 21. Ben-Haim M, Roayaie S, Ye M, et al. Hepatic epithelioid hemangioendothelioma: resection or transplantation, which and when? Liver Transplant Surg. 1999;5:526–31. 22. Marino I, Todo S, Tzakis A, et al. Treatment of hepatic epithelioid hemangioendothelioma with liver transplantation. Cancer. 1988;62:2079–84. J.F. Renz 91 23. Lerut J, Orlando G, Adam R, et al. The place of liver transplantation in the treatment of hepatic epithelioid hemangioendothelioma: report of the European transplant registry. Ann Surg. 2007;246:949–57. 24. Rodriguez J, Becker N, O’Mahony C, Goss J, Aloia T. Long-term outcomes following liver transplantation for hepatic hemangioendothelioma: the UNOS experience from 1987 to 2005. J Gastrointest Surg. 2008;12:110–6. 25. Orlando G, Adam R, Mirza D, et al. Hepatic hemangiosarcoma: an absolute contraindication to liver transplantation-the European Liver Transplant Registry experience. Transplantation. 2013;95:872–7. 26. Rude M, Watson R, Crippin J. Recurrent hepatic epithelioid hemanioendothelioma after orthotopic liver transplantation. Hepatology. 2014;59:2050–2. 27. Emamaullee J, Edgar R, Toso C, et al. Vascular endothelial growth factor expression in hepatic epithelioid hemangioendothelioma: implications for treatment and surgical management. Liver Transpl. 2010;16:191–7. 28. Raphael C, Hudson E, Williams L, Lester J, Savage P. Successful treatment of metastatic hepatic epithelioid hemangioendothelioma with thalidomide: a case report. J Med Case Rep. 2010;22:1186–9. 29. Salech F, Valderrama S, Nervi B, et al. Thalidomide for the treatment of metastatic hepatic epithelioid hemangioendothelioma: a case report with a long term follow-up. Ann Hepatol. 2011;10:99–102. 30. Saada E, Saint Paul M-C, Gugenheim J, Follana P, Francois E. Metastatic hepatic epithelioid hemangioendothelioma: long-term response to sunitinib malate. Oncol Res Treat. 2014;37:124–6. 31. Pallotti M, Nannini M, Agostinelli C, et al. Long-term durable response to lenalidomide in a patient with hepatic epithelioid hemangioendothelioma. World J Gastroenterol. 2014;20:7049–54. 32. Lakkis Z, Kim S, Delabrousse E, et al. Metronomic cyclophosphamide: an alternative treatment for hepatic epithelioid hemangioendothelioma. J Hepatol. 2013;58(6):1254–7. 33. Sangro B, Inarrairaegui M, Fernandez-Ros N. Malignant epithelioid hemangioendothelioma of the liver successfully treated with Sorafenib. Rare Tumor. 2012;4:34–9. 34. Chevreau C, Cesne A, Ray-Coquard I, et al. Sorafenib in patients with progressive epithelioid hemangioendothelioma. Cancer. 2013;15:2639–44.

Archita P. Desai and Helen S. Te Abstract Hepatocellular carcinoma (HCC) continues to be a signifi cant cause of mortality in the United States. However, HCC is curable if detected early in its course. Cirrhosis is a well-established risk factor for HCC, but direct evidence demonstrating the benefi t of screening for HCC in this population remains under contention today. Ultrasound (US) every 6 months is currently the proposed screening methodology. Serum alpha-feto protein (AFP) has been dropped from screening guidelines, yet recent prospective data reported an added effi cacy with the combination of serum AFP and US. Technological advances in cross-sectional imaging have dramatically impacted the fi eld of hepatobiliary imaging, making them attractive alternatives for HCC screening in selected populations. While computed tomography (CT) does not appear to confer any signifi cant advantage to US performed by trained personnel, magnetic resonance imaging (MRI) with hepatobiliary phase (HBP) and diffuse weighted imaging (DWI) offers the best sensitivity and specifi city for HCC largely due to its superiority in detecting and characterizing lesions <2 cm. Its costeffectiveness as a screening tool, however, remains to be seen. Keywords Alpha-feto protein • Ultrasound • Computed tomography • Magnetic resonance imaging • Hepatocellular cancer • Screening • Surveillance • Liver transplantation A. P. Desai University of Arizona , 1501 N. Campbell Avenue, Rm 6309A , 245136 , Tucson , AZ 85724 , USA e-mail: [email protected] H. S. Te (*) University of Chicago Medical Center , 5841 S. Maryland Ave., MC 7120 , Chicago , IL 60615 , USA e-mail: [email protected] 94 Introduction Despite the continuing medical advances in the management of chronic liver disease, the incidence of hepatocellular carcinoma (HCC) has steadily risen in the past two decades. Globally, HCC has become the fi fth leading cause of cancer and the second leading cause of cancer-related death in adult men [ 1 ]. In the United States, the age-adjusted incidence rates have doubled since the mid 1980s [ 2 ], causing similar increases in HCC-related mortality and hospitalization rates [ 3 , 4 ]. Although the incidence of HCC appears to have plateaued in the past decade, HCC-related deaths remain on the rise [ 5 , 6 ]. Hepatocellular carcinoma is curable if detected early in its course. Liver transplantation for HCC cases that fall within the Milan criteria has demonstrated excellent results with 5-year survival rates exceeding 70 %. Hepatic resection in non-cirrhotic patients or in well-compensated cirrhotic patients with no portal hypertension and no signifi cant liver functional impairment has led to 5-year survival rates exceeding 70 % as well [ 7 ]. However, to achieve a cure, the diagnosis must be made early, and early diagnosis is only possible if screening is performed. Evidence demonstrating the benefi t of screening for HCC remains under contention today. A meta-analysis found that evidence supporting the benefi t of HCC screening in at-risk patients (cirrhotics and noncirrhotics) were of very low-strength [ 8 ]. While cirrhosis is a well-established risk factor for HCC, there has been no randomized controlled trial (RCT) performed in the US to validate the benefi t of HCC screening in this population, partially due to ethical reasons and patient refusal [ 9 ]. Investigators have resorted to modeling techniques to demonstrate the cost- effectiveness of HCC surveillance in cirrhosis, and screening has been found to provide a survival benefi t in targeted patients who are viable candidates for interventions at acceptable costs [ 10 – 15 ]. In fact, the American Association for the Study of Liver Diseases (AASLD) [ 7 ] and the European Association for the Study of the Liver (EASL) guidelines [ 16 ] recommend HCC surveillance with an ultrasound every 6 months for patients with cirrhosis of any cause, wherein the incidence of HCC is estimated to be 1.5 % per year or greater [ 7 ]. However, the question remains, does biannual ultrasound provide the best benefi t in screening cirrhotic patients for HCC in the United States in 2014? Search Strategy A literature search of English language publications from 2000 to 2014 was used to identify published data on screening for HCC in cirrhotic patients using the PICO outline (Table 9.1 ). Databases searched were PubMed, Medline and Cochrane Evidence Based Medicine . Terms used in the search were “hepatocellular carcinoma/ screening/ cirrhosis ,” “liver cancer /screening/cirrhosis.” Manual searches of reference lists from applicable studies were performed to identify any studies that may have been missed by the computer-assisted search. As the quality of studies for each A.P. Desai and H.S. Te 95 screening modality varied, different inclusion criteria were used for the studies included in this review. For the performance of serum AFP, studies were excluded if they specifi cally addressed diagnosis of hepatocellular carcinoma after a lesion has been detected in the liver rather than screening, or if noncirrhotic patients were the only study subjects. For the performance of ultrasound as a screening test, only prospective studies of cirrhotic cohorts were included. For cross-sectional imaging, studies reviewed refl ect the latest progress in the technology of computed tomography ( CT ) and magnetic resonance imaging (MRI). For CT, only studies assessing the performance of 16-slice or more multidectector CT (MDCT) with triple or quadruple phase imaging and explant pathology used as the reference standard were included. For MRI, studies using dynamic MRI with both hepatobiliary phase and diffusion weighted imaging MRI were included. The data was classifi ed using the GRADE system. Results Clinical Relevance and Risk Factors of Hepatocellular Carcinoma Screening is the administration of diagnostic tests to subjects who have a defi ned risk for developing HCC, but in whom there is no suspicion for HCC to be present prior to the screening . Surveillance is the repeated administration of screening tests. An intervention is considered effective if it provides an increase in longevity of about 100 days or 3 months [ 17 ], and interventions that can be achieved at a cost of <$50,000/year of life gained is considered to be cost-effective in 1992 [ 18 ], although this cost is likely higher in today’s market rates. Major advances in medicine have increased the ability to cure HCC when diagnosed early, although medical and locoregional interventions may still extend Table 9.1 PICO table for screening for hepatocellular carcinoma in cirrhotic patients P (Patients) I (Intervention) C (Comparator group) O (Outcomes measured) Patients with cirrhosis Serum AFP assay No screening Diagnosis of hepatocellular carcinoma within Milan criteria, mortality Patients with cirrhosis Ultrasound examination No screening or screening via another modality Diagnosis of hepatocellular carcinoma, mortality Patients with cirrhosis CT examination No screening or screening via another modality Diagnosis of hepatocellular carcinoma, mortality Patients with cirrhosis MRI examination No screening or screening via another modality Diagnosis of hepatocellular carcinoma, mortality 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 96 survival in later stages. Despite the absence of high quality data from RCT of HCC surveillance versus no HCC surveillance, the ability to change patient outcomes with appropriate interventions at acceptable costs is compelling reason to prompt the AASLD to recommend HCC surveillance in high-risk population. Cirrhosis is a well-established risk factor for HCC, wherein surveillance for an HCC incidence of 1.5 %/year is expected to increase survival by about 3 months [ 10 ]. Hepatitis B carriers who are Asian males aged over 40 years and Asian females aged over 50 years have an HCC incidence of 0.4–0.6 %/year, and those with a family history of HCC are known to have a higher HCC incidence than those without. African and North American Blacks with hepatitis B also are known to develop HCC at a younger age. Cirrhotic hepatitis B carriers have an HCC incidence of 3–8 %/year, while patients with hepatitis C cirrhosis and stage 4 primary biliary cirrhosis have an HCC incidence of 3–5 %/year. The HCC incidence in cirrhotics with genetic hemochromatosis, alpha 1- antitrypsin defi ciency and other causes are not established but may approach >1.5 %/year for most cases [ 7 ]. Another growing group at risk for HCC is the population affl icted with non-alcoholic steatohepatitis, where the incidence is 2.6 %/year in one study [ 19 , 20 ]. Screening Strategies Serum Alpha-Feto Protein (AFP) Serum AFP is the most widely tested biomarker in the diagnosis of HCC, but its performance as a surveillance tool has been suboptimal. The only RCT evaluating serum AFP alone as a screening tool for HCC was conducted in hepatitis B infected patients who were mostly non-cirrhotic; the fi ndings included an earlier diagnosis of HCC [ 21 ]. Data in cirrhotic patients have been limited to case control studies, which have consistently shown a low sensitivity of a serum AFP cut-off of 20 ng/ml at about 60 % for detection of HCC (Table 9.2 ) [ 22 – 27 ]. Trevisani et al. reported the positive predictive value of a serum AFP cut-off value of >20 ng/ml to be dismal at 25.1 % in a population with an HCC prevalence of 5 % [ 25 ]. Currently, serum AFP is not part of the recommended screening process for HCC by both the AASLD [ 28 ] and the EASL guidelines [ 16 , 28 ]. The potential value of measuring serial serum AFPs to survey for HCC has also been investigated in another case-control study involving hepatitis C-infected patients with advanced fi brosis or cirrhosis . Lee et al. found that both the standard deviation and the rate of increase of serum AFP were independently associated with HCC. Incorporation of these metrics along with patient-specifi c risk factors resulted in improved accuracy for HCC prediction to an area under the receiver-operating characteristic curve of 0.81 when compared with 0.76 when only the most recent serum AFP value was used [ 29 ]. However, these fi ndings need to be validated further to determine its true value in HCC screening . A.P. Desai and H.S. Te 97 Ultrasonography (US) with or Without Serum AFP While US technology has been used for diagnostic purposes since the 1940s, its use for screening for liver masses was fi rst reported in the early 1980s [ 30 ]. Since then, many studies evaluating the performance of US with or without serum AFP in surveillance programs aimed to detect early stages of HCC have been done [ 8 , 31 – 33 ]. Despite a variety of study methodologies involved, a recent meta-analysis calculated the pooled odds of early detection as 2.08 (95 % CI 1.80–2.37). More importantly, screening increased the odds of receipt of curative therapy (OR 2.24, 95 % CI 1.99– 2.52) and the odds of 3-year survival (OR 1.90, 95 % CI 1.90–2.17) [ 33 ]. It is important to note that this data has limited accuracy and applicability, as most of the studies included in the meta-analysis were observational studies susceptible to lead-time and selection bias, and these also included a wide variety of patient populations, utilized different screening programs and had largely different screening uptakes as well. Three randomized trials have been done in Europe and Asia to study the effi cacy of screening with ultrasound with or without serum AFP [ 34 – 36 ], but cirrhotic patients were not included or defi ned [ 36 ] so these studies are not included in this discussion. More recent data on HCC screening in patients with cirrhosis published after 2000 included prospective studies that assessed the performance of US with or without serum AFP in the real world. Overall, the sensitivity of US for detecting HCC has a wide range of 43–90 % and a specifi city of 83–97 % (Table 9.3 ) [ 26 , 34 , 37 – 41 ]. While current guidelines do not recommend use of serum AFP, most studies included serum AFP testing in their surveillance program. In fact, one study attempted to assess the role of serum AFP by creating randomization arms that consisted of imaging surveillance without serum AFP, but discovered high usage rates of serum AFP assays in the imaging alone group, necessitating the fi nal analysis to include imaging in combination with serum AFP [ 34 ]. The optimal interval Table 9.2 Sensitivity and specifi city of serum AFP in detecting early stage HCCs Study AFP cut-off (ng/ml) Sensitivity (%) Specifi city (%) Study type ( quality of evidence) Gamberin-Gelwan et al. [ 22 ] 20 58 91 Case-control (low) Trevisani et al. [ 25 ] a >20 60 91 Case-control (low) Nguyen et al. [ 24 , 64 ] >20 63 80 Case-control (low) Snowberger et al. [ 27 ] b 8.9 62 80 Case-control (low) 50 31 96 Marrero et al. [ 24 ] >20 59 90 Case-control (low) Lok et al. [ 23 ] >20 61 81 Case-control (low) Singal et al. [ 26 ] >20 66 90 Prospective cohort (moderate) a In a population with 50 % HCC prevalence b Used explants as gold standard 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 98 Table 9.3 Summary of prospective studies of hepatocellular carcinoma (HCC) screening in cirrhotics using ultrasonography (US) with or without serum alpha feto protein (AFP) Study Screening method Cohort Sensitivity Specifi city Clinical effectiveness Study type ( quality of evidence) Henrion et al. [ 37 ] US and serum AFP assay every 6 months Belgian cirrhotics 83 % Insuffi cient data Compliance to surveillance program overall was 66 % and was signifi cantly higher in those with cirrhosis due to non-alcoholic causes Prospective cohort (low) Bolondi et al. [ 35 ] US and serum AFP assay every 6 months Italian Child- Pugh A or B cirrhotics 82 % Insuffi cient data Higher rate of unifocal and smaller HCC for screened vs control group (80 % vs. 53 %, p < 0.001) Prospective cohort (low) Longer 3-year survival for screened vs control group (45 % vs 31.7 %, p = 0.02) Similar rate of treatment for HCC in screened vs control group (58.6 % vs 68.8 %, p = NS) Cost per year of life saved = $112,996 Sangiovanni et al. [ 39 ] Annual US and serum AFP assay a Italian Child A and B cirrhotics, aged 36–72 Insuffi cient data Insuffi cient data Smaller nodules detected in latest study period with mean diameter of 2.2 cm (1997–2001) vs. 3.0 cm (1992–1996) vs. 3.7 cm (1987–1991) Prospective cohort (low) Improved survival in those screened in the latest study period, linked to earlier diagnosis and improved linkage to treatment A.P. Desai and H.S. Te 99 (continued) Study Screening method Cohort Sensitivity Specifi city Clinical effectiveness Study type ( quality of evidence) Trinchet et al. [ 32 ] US every 3 or 6 months French or Belgian Child-Pugh A or B cirrhotics 86.6 % Insuffi cient data No difference in diagnosis of early HCC or overall survival Randomized control trial (high) Increased incidence of nonmalignant focal lesions (especially those ≤10 mm) when US was completed every 3 months. If lesions were followed per EASL recommendations, the time of HCC diagnosis and treatment would not have changed Singal et al. [ 26 ] US and serum AFP assay every 6–12 months b American Child-Pugh A or B cirrhotics US: 43.9 % US: 91.5 % 38.7 % of study population received inconsistent or no surveillance Prospective cohort (moderate) US + AFP: 90.2 % US + AFP: 83.3 % US sensitivity was lower in non-Caucasian race and lower MELD score Majority of participants only had one positive surveillance study (AFP or US) prior to HCC diagnosis Di Martino et al. [ 36 ] Doppler US followed by cross-sectional imaging Italian OLT candidates over 3 year period who underwent all three imaging tests in 1 month e 71 % 62 % US sensitivity decreased with decreasing nodule size, with only 22 % sensitivity for nodules under 10 mm Prospective cohort (moderate) Lower specifi city due to signifi cant numbers of false positive examinations when state of the art US technology was used 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 100Table 9.3 (continued) Study Screening method Cohort Sensitivity Specifi city Clinical effectiveness Study type ( quality of evidence) Pocha et al. [ 38 ] US every 6 months vs. annual CT c,d American Child A cirrhotics f US: 71.4 % US: 97.5 % Signifi cant number of participants went off-protocol Randomized control trial (moderate) CT: 66.7 % CT: 94.4 % Ultrasound comparable to CT in detecting early stage HCC AFP added little value to overall HCC detection Cost of one HCC detected was signifi cantly higher with CT ($35,383) vs US ($17,041) a If AFP >20 ng/dL, US and serum AFP assay frequency were increased to every 6 months b Screening was obtained at the discretion of the treating hepatologist, no reminders were sent to patient or medical care providers c computed tomography d ±AFP every 6 months in both arms e Excluded three cholangiocarcinoma nodules (in two patients) and three hepatocholangiocarcinoma nodules (in one patient) f Participants had to be potential candidates for treatment of HCC; advanced medical condition and those who were unable to receive IV contrast due to renal insuffi ciency or allergy were excluded A.P. Desai and H.S. Te 101 between screenings is not clear with most studies using 6 month or 12 month intervals. Shorter intervals appears to only increase detection of non-malignant focal lesions in the 3-month group, particularly those <10 mm in diameter, which did not impact clinical care or outcomes in most cases [ 34 ]. Although one study in a Veterans Administration hospital setting found no incremental benefi t of the addition of serum AFP to imaging studies [ 40 ], one prospective cohort study involving 446 patients in real world practice reported substantial advantage of using serum AFP in conjunction with US than with US alone, with increased sensitivity to 90.2 % and specifi city to 83.3 % [ 26 ]. Cross Sectional Imaging The clinical effectiveness of US in screening for HCC is largely limited by low adherence rates, variability in operator experience, diffi culty in visualization of the liver in patients with morbid obesity or very nodular livers, and poor sensitivity when identifying early HCC’s (i.e., those <20 mm). These limitations have fueled the search for improved modalities in screening those at high risk for HCC. Both computed tomography ( CT ) and magnetic resonance imaging (MRI) of the abdomen have been used to further evaluate and diagnose liver lesions found during ultrasound examination. Therefore, both offer attractive alternatives for screening of those at risk for HCC. Computed Tomography Beginning in the 1990s, use of arterial-phase imaging during intravenous contrast medium-enhanced CT studies improved the sensitivity of small nodule detection [ 42 ]. In the 2000s, multidetector-row helical CT (MDCT) technology allowed for faster acquisition, thinner slices (0.5 mm for 64-slice MDCT vs. 5–10 mm for helical CT) in a single breath-hold and repetitive imaging during multiple perfusion phases after contrast material injection [ 42 , 43 ]. In combination with four-phase CT protocols, which provide images during the pre-contrast, arterial, portal venous and delayed phases, 16-, 64- and even 128-slice MDCT has allowed for the detection of a signifi cantly higher number of cases of HCC [ 42 , 44 ]. Current UNOS policy outlines minimum criteria for the use of CT to accurately diagnose HCC radiographically [ 45 ]. Many of the studies assessing the sensitivity and specifi city of triple phase 16- or 64-slice MDCT have been performed in individuals awaiting liver transplantation . Most are retrospective; however, many have used the explant pathology as the gold standard, thus increasing their validity with sensitivities of 77–89 % and specifi cities of 44–93 % (Table 9.4 ) [ 44 , 46 – 48 ]. Due to the lack of prospective data, the screening interval for CT has yet to be established. Overall, even studies of the most advanced CT technology only show marginal superiority over US as a screening tool, largely due to better sensitivity for lesions <20 mm in size. 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 102 Table 9.4 Summary of studies using triple or four-phase protocols with 16- or greater slice multidetector-row helical computed tomography (MDCT) for the diagnosis of hepatocellular carcinoma (HCC) with explant pathology as gold standard Study Imaging method Cohort Sensitivity Specifi city Clinical performance Study type ( quality of evidence) Ronzoni et al. [ 48 ] Triple-phase MDCT original reports vs. images reviewed retrospectively a OLT recipients with CT within 6 months of OLT Original report: 64 % by lesion, 77 % by patient Original report: 75 % by patient f 32 % of false negative lesions were due to prior locoregional therapy and 70 % were <10 mm in size Retrospective (low) Retrospective review: Retrospective review: 77.5 % by patient c 14 % of the study population would have been denied or referred to OLT due to false-negative or false-positive results, respectively 73.3 % by lesion, 83.4 % by patient Denecke et al. [ 47 ] Triple-phase MDCT images reviewed retrospectively by two radiologists OLT recipients with HCC with CT within 100 days of OLT c Observer 1: 78 % Observer 1: 45 % Sensitivity for both observers decreased with decreasing size of the nodule (43 % and 53 % observers 1 and 2 for nodules <10 mm, respectively, vs. 94 % for both observers for 11–20 mm nodules vs. 89 % and 95 %, respectively, for nodules >20 mm) Retrospective (low) Observer 2: 83 % Observer 2: 44 % Poor specifi city with high false positive rate for both observers noted A.P. Desai and H.S. Te 103 Study Imaging method Cohort Sensitivity Specifi city Clinical performance Study type ( quality of evidence) Luca et al. [ 44 ] Triple-phase MDCT images retrospectively reviewed by three blinded radiologists b OLT recipients imaged with CT pre-transplant Hypervascular lesion with washout on portal venous and/or delayed phase images: 43 % Hypervascular lesion with washout: 93 % Pattern of hypervascular nodule with washout has poor sensitivity but excellent specifi city for the diagnosis of HCC Retrospective (low) Overall: 89 % e Limitation to hypervascular lesions with delayed phase washout only improved diagnostic accuracy Pattern of hypervascular nodule with washout established an accurate staging of disease in 46 % of cases, underestimated in 52 % and overestimated in only 2 % Hypervascular lesions >10 mm without washout and hypovascular lesions >20 mm have signifi cant risk of HCC Addley et al. [ 46 ] Triple-phase MDCT reviewed by three radiologist OLT recipients with CT within 5 months of OLT d Observer 1:78 % Observer 1:47 % Sensitivity for each radiologist was signifi cantly lower if lesion <20 mm Retrospective (low) Observer 2: 72 % Observer 2: 69 % Observer expertise and years of experience improved accuracy of HCC detection but decreased sensitivity Observer 3: 65 % Observer 3: 88 % a Diagnosis reached by consensus between two unblinded radiologists b Diagnosis reached by consensus amongst three blinded radiologists c Patients with incomplete histopathologic report and those who had intermittent tumor treatment were excluded d Patients who had previously undergone treatment of HCC, those with malignancy other than HCC and those with multifocal HCC were excluded e MDCT fi ndings of hypervascular nodule with washout, hypervascular nodule without washout, and hypovascular nodules were included f Specifi city by lesion not reported 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 104 Magnetic Resonance Imaging As with CT , MRI is evolving with new techniques allowing for improved diagnostic accuracy. With MR, however, there are a variety of techniques for imaging acquisition, image sequences and optimization, as well as image processing that can impact the ultimate performance of MR in liver lesion detection and accurate characterization. Current UNOS policy outlines specifi c minimum technical requirements for MRI in diagnosing HCC radiographically [ 49 ]. Of the recent advances in MR technology, dynamic imaging, hepatobiliary phase (HBP) imaging with hepatocyte- specifi c contrast agents, and use of diffusion weight image (DWI) characteristics have allowed MRI to evaluate tumor vascularity, increased tissue cellularity and absence of normal hepatocytes with greater detail and accuracy [ 50 – 54 ]. The reported sensitivity of dynamic contrast enhanced MR imaging (with an extracellular fl uid contrast agent) is noted to be from 70 to 100 % in various studies with a pooled sensitivity of 81 % [ 31 ]. Even with dynamic, contrast-enhanced MRI, the detection of early HCC remains a challenge, particularly with the background of a cirrhotic liver where arterial enhancement of small <1 cm lesions is less diagnostically accurate for HCC. For these lesions, MRI offers several advantages over CT despite some inherent limitations [ 55 – 57 ]. Hepatobiliary phase (HPB) imaging may improve the ability to detect and diagnose early HCC. Use of gadoxetic acid (Eovist ® ) or gadobenate dimeglumine (MultiHance ® ) allows for both dynamic imaging during the extracellular phase and delayed imaging in the HBP phase, when the contrast is taken up by the hepatocytes [ 52 , 53 ]. These agents can help differentiate arterial-enhancing pseudolesions. One study documented 95.5 % of HCC’s displayed hypointensity during the HBP while 94.3 % of pseudolesions were isointense during the HBP. In this study, the sensitivity of MRI for diagnosing HCC was 93.9 % [ 58 ]. Another study showed lesions without arterial phase hyperenhancement, but with both venous phase hypoenhancement and HBP hypointensity, carry a higher probability of being welldifferentiated HCC in cirrhotic livers [ 55 ]. A meta-analysis of gadoxetic acid-based MRI reported a pooled sensitivity of 91 % and specifi city of 95 %. For studies focused on lesions <2 cm, performance continued to be excellent with a reported sensitivity range of 87–99 % and specifi city range of 92–96 % [ 52 ]. Diffusion weighted imaging (DWI) in MRI is a functional MRI technique that offers its own advantages in the detection and diagnosis of HCC in those with cirrhosis . DWI capitalizes on the changes that accompany the development of HCC, such as changes in cellularity and extracellular space structure, to detect HCC lesions. A reported 70–80 % of HCC’s, including those <1–2 cm in size, appear hyperintense on DWI [ 51 ]. The sensitivity and specifi city of DWI were better at 91.2 % and 82.9 %, respectively, when compared to dynamic, gadolinium-based MRI at 67.6 % and 61 %, respectively [ 59 ]. A study also highlighted the utility of MRI enhanced with DWI as a screening tool for HCC for those in whom intravenous iodine contrast administration is contraindicated [ 54 ]. In another study, use of A.P. Desai and H.S. Te 105 DWI characteristics improved the sensitivity of conventional MRI from 83–85 % to 98 % in detecting HCC lesions <2 cm [ 60 ]. Furthermore, several studies that assessed the impact of combining HBP imaging with DWI to detect HCC concluded that DWI can incrementally improve the performance of MRI in the detection of HCC (Table 9.5 ) [ 61 – 63 ]. While these techniques are promising, the optimal interval of imaging and the cost of such a screening tool have not been examined. Therefore, evidence supporting the use of MRI for the screening and surveillance of cirrhotics for HCC is lacking, and their use has not been incorporated in the current practice guidelines. Recommendations While the clinical effectiveness of screening individuals with cirrhosis has yet to be determined, indirect evidence supports a survival benefi t with screening of targeted individuals who are viable candidates for interventions. Biannual ultrasound with or without serum AFP is the most validated tool, offers good performance in the general population, and in experienced hands likely has similar performance to triple phase MDCT for the detection of tumors >2 cm. However, as locoregional therapy becomes more widely available, the detection of early HCC may offer a survival benefi t. In this context, dynamic MRI with hepatobiliary phase and diffusion weighted imaging may be the best-performing screening test. Screening, regardless of modality, should be done in expert hands to optimize effectiveness of the test. The cost of such a surveillance program will have to be compared to its clinical effectiveness, which is largely dependent on uptake of screening and linkage to treatment. A Personal View of the Data Many advances have been made in the diagnosis of HCC over the past decade. Improvements in CT and MRI technology form the basis of this progress; however, translation into clinical practice and guidelines is limited by the quality of data supporting their use in screening and surveillance programs. Current data have created a strong platform for imaging-based screening of HCC, whereas the importance of tumor markers and invasive method such as biopsy has declined. In an era where the morbidity and mortality associated with HCC is rising, efforts to improve early diagnosis must be made in order to impact patient outcomes and reduce the healthcare burden associated with HCC. In this context, MRI-based imaging has the most promise for accuracy, although its cost is a major deterrence in its use as the fi rst line tool. 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 106 Table 9.5 Summary of studies of the performance of gadoxetic acid-enhanced MRI with or without DWI in the detection of HCC Study Imaging method Cohort Sensitivity Specifi city Clinical performance Study type (Quality of evidence) Park et al. [ 63 ] Gadoxetic acidenhanced MRI with or without DWI at 3.0-T as reviewed by three independent, blinded observers HCC lesion <2 cm proven by surgical resection in those who had undergone MRI a Gadoxetic acid alone: 81.4 % Gadoxetic acid alone: 98.4 % Adding DWI analysis to gadoxetic-acid enhanced MRI improved the sensitivity and specifi city Retrospective (very low) DWI alone: 78.8 % DWI alone: 96.8 % Combined: 92.4 % Combined: 97.5 % Sensitivity of combined image sets in the detection of lesions <1 cm is lower than for those lesions >1 cm (84.8 % vs. 95.7 %) Specifi city of combined image sets in the detection of lesions <1 cm is slightly lower than for those lesions >1 cm (94.5 % vs. 99.6 %) Park et al. [ 62 ] Gadoxetic acidenhanced MRI with or without DWI at 3.0-T as reviewed by two independent, blinded observers Those with suspected lesion on MDCT or US with lesion <2.0 cm b Combined: 98.5 % Combined: 90.9 % Arterial hyperintensity, hypointensity on HBP and hyperintensity on DWI was present in 65 % of <1 cm lesions Retrospective (very low) The majority of lesions that did not have typical characteristics of HCC on MDCT were able to be characterized as HCC based on HBP and DWI characteristics A.P. Desai and H.S. Te 107 Study Imaging method Cohort Sensitivity Specifi city Clinical performance Study type (Quality of evidence) Hwang et al. [ 61 ] Gadoxetic acidenhanced MRI with or without DWI at 3.0-T as reviewed by two independent, blinded observers OLT recipients who underwent MRI within 90 days of OLT c Gadoxetic acid alone: 72 % Gadoxetic acid alone: 96 % Adding DWI analysis to gadoxetic-acid enhanced MRI improved the sensitivity and specifi city Retrospective (low) Combined: 79 % Combined: 93 % Sensitivity of gadoxetic acid-enhanced MRI with and without DWI decreased with lesion size (for combined imaging, 93 % for lesions >2.0 cm vs. 61 % for lesions <1.0 cm) Sensitivity of gadoxetic acid-enhanced MRI with and without DWI decreased with increasing severity of liver disease (for combined imaging, 97 % in Child-Pugh class A vs 56 % in Child-Pugh class C) a Control group included those with suspected HCC on initial imaging but with negative diagnostic work up b Included individuals without cirrhosis . Individuals excluded if they had not had MDCT as well as MRI. Lesion deemed either HCC or benign hepatocellular nodule based on imaging criteria, biopsy, surgical resection or explant pathology c Included 8/63 non-cirrhotics 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 108 Recommendations 1. MRI with HBP and DWI offers the best sensitivity and specifi city of HCC largely due to its superiority in detecting and characterizing lesions <2 cm. While there are no data on screening interval for MRI, annual imaging in those with no worrisome lesions can be inferred based on tumor doubling time. The costeffectiveness of this approach, however, remains to be studied. 2. In the general population, ultrasound with or without serum AFP every 6 months offers acceptable performance in the screening of HCC and should be used when cross-sectional imaging is not available or tolerated or is contraindicated. The combination of US with serum AFP has demonstrated increased accuracy in a larger prospective study than US alone. Furthermore, the interval change in serum AFP may offer more value in the detection of HCC than a single serum AFP assay alone. 3. In those awaiting liver transplant ation , where accurate assessment of the burden of HCC can signifi cantly alter management , MRI with HPB phase and DWI should be used, with the best performance noted in those with Child-Pugh class A and B cirrhosis . 4. Survival benefi t of screening for HCC has yet to be established in randomized controlled trials, but it is unlikely for such trials to come to fruition due to diffi – culty with patient enrollment. Limiting screening to those individuals who are eligible for treatment will improve clinical effectiveness of surveillance program. References 1. Jemal A, et al. Global cancer statistics. CA Cancer J Clin. 2011;61(2):69–90. 2. El-Serag HB, et al. The continuing increase in the incidence of hepatocellular carcinoma in the United States: an update. Ann Intern Med. 2003;139(10):817–23. 3. El-Serag HB. Hepatocellular carcinoma: recent trends in the United States. Gastroenterology. 2004;127(5 Suppl 1):S27–34. 4. Kim WR, et al. Mortality and hospital utilization for hepatocellular carcinoma in the United States. Gastroenterology. 2005;129(2):486–93. 5. Altekruse SF, et al. Changing hepatocellular carcinoma incidence and liver cancer mortality rates in the United States. Am J Gastroenterol. 2014;109(4):542–53. 6. Rahib L, et al. Projecting cancer incidence and deaths to 2030: the unexpected burden of thyroid, liver, and pancreas cancers in the United States. Cancer Res. 2014;74(11):2913–21. 7. Bruix J, Morris S. Management of hepatocellular carcinoma: an update. AASLD Practice Guidelines 2010 [cited 2014 9/17/14]; Available from: http://www.aasld.org/sites/default/fi les/ guideline_documents/HCCUpdate2010.pdf. 8. Kansagara D, et al. Screening for hepatocellular carcinoma in chronic liver disease: a systematic review. Ann Intern Med. 2014;161(4):261–9. 9. Poustchi H, et al. Feasibility of conducting a randomized control trial for liver cancer screening: is a randomized controlled trial for liver cancer screening feasible or still needed? Hepatology. 2011;54(6):1998–2004. A.P. Desai and H.S. Te 109 10. Sarasin FP, Giostra E, Hadengue A. Cost-effectiveness of screening for detection of small hepatocellular carcinoma in western patients with Child-Pugh class A cirrhosis. Am J Med. 1996;101(4):422–34. 11. Andersson KL, et al. Cost effectiveness of alternative surveillance strategies for hepatocellular carcinoma in patients with cirrhosis. Clin Gastroenterol Hepatol. 2008;6(12):1418–24. 12. Arguedas MR, et al. Screening for hepatocellular carcinoma in patients with hepatitis C cirrhosis: a cost-utility analysis. Am J Gastroenterol. 2003;98(3):679–90. 13. Lin OS, et al. Cost-effectiveness of screening for hepatocellular carcinoma in patients with cirrhosis due to chronic hepatitis C. Aliment Pharmacol Ther. 2004;19(11):1159–72. 14. Mourad A, et al. Hepatocellular carcinoma screening in patients with compensated hepatitis C virus (HCV)-related cirrhosis aware of their HCV status improves survival: a modeling approach. Hepatology. 2014;59(4):1471–81. 15. Patel D, et al. Cost-effectiveness of hepatocellular carcinoma surveillance in patients with hepatitis C virus-related cirrhosis. Clin Gastroenterol Hepatol. 2005;3(1):75–84. 16. European Association For The Study Of The, L, R. European Organisation For, C. Treatment Of. EASL-EORTC clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol. 2012;56(4):908–43. 17. Naimark D, Naglie G, Detsky AS. The meaning of life expectancy: what is a clinically signifi – cant gain? J Gen Intern Med. 1994;9(12):702–7. 18. Laupacis A, et al. How attractive does a new technology have to be to warrant adoption and utilization? Tentative guidelines for using clinical and economic evaluations. CMAJ. 1992;146(4):473–81. 19. Ascha MS, et al. The incidence and risk factors of hepatocellular carcinoma in patients with nonalcoholic steatohepatitis. Hepatology. 2010;51(6):1972–8. 20. Kim Y, et al. Temporal trends in population-based death rates associated with chronic liver disease and liver cancer in the United States over the last 30 years. Cancer. 2014;120(19):3058–65. 21. Chen JG, et al. Screening for liver cancer: results of a randomised controlled trial in Qidong, China. J Med Screen. 2003;10(4):204–9. 22. Gambarin-Gelwan M, et al. Sensitivity of commonly available screening tests in detecting hepatocellular carcinoma in cirrhotic patients undergoing liver transplantation. Am J Gastroenterol. 2000;95(6):1535–8. 23. Lok AS, et al. Des-gamma-carboxy prothrombin and alpha-fetoprotein as biomarkers for the early detection of hepatocellular carcinoma. Gastroenterology. 2010;138(2):493–502. 24. Marrero JA, et al. Alpha-fetoprotein, des-gamma carboxyprothrombin, and lectin-bound alpha-fetoprotein in early hepatocellular carcinoma. Gastroenterology. 2009;137(1):110–8. 25. Trevisani F, et al. Serum alpha-fetoprotein for diagnosis of hepatocellular carcinoma in patients with chronic liver disease: infl uence of HBsAg and anti-HCV status. J Hepatol. 2001;34(4):570–5. 26. Singal AG, et al. Effectiveness of hepatocellular carcinoma surveillance in patients with cirrhosis. Cancer Epidemiol Biomark Prev. 2012;21(5):793–9. 27. Snowberger N, et al. Alpha fetoprotein, ultrasound, computerized tomography and magnetic resonance imaging for detection of hepatocellular carcinoma in patients with advanced cirrhosis. Aliment Pharmacol Ther. 2007;26(9):1187–94. 28. Bruix J, Sherman M, D. American Association for the Study of Liver. Management of hepatocellular carcinoma: an update. Hepatology. 2011;53(3):1020–2. 29. Lee E, et al. Improving screening for hepatocellular carcinoma by incorporating data on levels of alpha-fetoprotein, over time. Clin Gastroenterol Hepatol. 2013;11(4):437–40. 30. Takashima T, et al. Diagnosis and screening of small hepatocellular carcinomas. Comparison of radionuclide imaging, ultrasound, computed tomography, hepatic angiography, and alpha 1-fetoprotein assay. Radiology. 1982;145(3):635–8. 31. Colli A, et al. Accuracy of ultrasonography, spiral CT, magnetic resonance, and alphafetoprotein in diagnosing hepatocellular carcinoma: a systematic review. Am J Gastroenterol. 2006;101(3):513–23. 9 What Is the Best Way to Screen Cirrhotic Patients for Hepatocellular Carcinoma… 110 32. Singal A, et al. Meta-analysis: surveillance with ultrasound for early-stage hepatocellular carcinoma in patients with cirrhosis. Aliment Pharmacol Ther. 2009;30(1):37–47. 33. Singal AG, Pillai A, Tiro J. Early detection, curative treatment, and survival rates for hepatocellular carcinoma surveillance in patients with cirrhosis: a meta-analysis. PLoS Med. 2014;11(4):e1001624–e1001624. 34. Trinchet J-C, et al. Ultrasonographic surveillance of hepatocellular carcinoma in cirrhosis: a randomized trial comparing 3- and 6-month periodicities. Hepatology. 2011;54(6):1987–97. 35. Wang J-H, et al. Hepatocellular carcinoma surveillance at 4- vs. 12-month intervals for patients with chronic viral hepatitis: a randomized study in community. Am J Gastroenterol. 2013;108(3):416–24. 36. Zhang B-H, Yang B-H, Tang Z-Y. Randomized controlled trial of screening for hepatocellular carcinoma. J Cancer Res Clin Oncol. 2004;130(7):417–22. 37. Bolondi L, et al. Surveillance programme of cirrhotic patients for early diagnosis and treatment of hepatocellular carcinoma: a cost effectiveness analysis. Gut. 2001;48(2):251–9. 38. Di Martino M, et al. Hepatocellular carcinoma in cirrhotic patients: prospective comparison of US, CT and MR imaging. Eur Radiol. 2013;23(4):887–96. 39. Henrion J, et al. Surveillance for hepatocellular carcinoma: compliance and results according to the aetiology of cirrhosis in a cohort of 141 patients. Acta Gastroenterol Belg. 2000;63(1):5–9. 40. Pocha C, et al. Surveillance for hepatocellular cancer with ultrasonography vs. computed tomography – a randomised study. Aliment Pharmacol Ther. 2013;38(3):303–12. 41. Sangiovanni A, et al. Increased survival of cirrhotic patients with a hepatocellular carcinoma detected during surveillance☆. Gastroenterology. 2004;126(4):1005–14. 42. Choi BI. The current status of imaging diagnosis of hepatocellular carcinoma. Liver Transpl. 2004;10(2 Suppl 1):S20–5. 43. Boone JM. Multidetector CT: opportunities, challenges, and concerns associated with scanners with 64 or more detector rows. Radiology. 2006;241(2):334–7. 44. Luca A, et al. Multidetector-row computed tomography (MDCT) for the diagnosis of hepatocellular carcinoma in cirrhotic candidates for liver transplantation: prevalence of radiological vascular patterns and histological correlation with liver explants. Eur Radiol. 2010;20(4):898–907. 45. United Network for Organ Sharing. HRSA/OPTN Policy 3.6 organ distribution: allocation of livers. Table 9–3: recommendations for dynamic contrast-enhanced CT of liver. Available at: http://optn.transplant.hrsa.gov/PoliciesandBylaws2/policies/pdfs/policy_8.pdf. Accessed on 10 Oct 2014. 46. Addley HC, et al. Accuracy of hepatocellular carcinoma detection on multidetector CT in a transplant liver population with explant liver correlation. Clin Radiol. 2011;66(4):349–56. 47. Denecke T, et al. Multislice computed tomography using a triple-phase contrast protocol for preoperative assessment of hepatic tumor load in patients with hepatocellular carcinoma before liver transplantation. Transplant Int. 2009;22(4):395–402. 48. Ronzoni A, et al. Role of MDCT in the diagnosis of hepatocellular carcinoma in patients with cirrhosis undergoing orthotopic liver transplantation. AJR Am J Roentgenol. 2007;189(4):792–8. 49. United Network for Organ Sharing. HRSA/OPTN Policy 3.6 organ distribution: allocation of livers. Table 1. Available at: http://optn.transplant.hrsa.gov/PoliciesandBylaws2/policies/pdfs/ policy_8.pdf. Accessed 10 Oct 2014. 50. Barr DC, Hussain HK. MR imaging in cirrhosis and hepatocellular carcinoma. Magn Reson Imaging Clin N Am. 2014;22(3):315–35. 51. Lim KS. Diffusion-weighted MRI of hepatocellular carcinoma in cirrhosis. Clin Radiol. 2014;69(1):1–10. 52. Liu X, et al. Gadoxetic acid disodium-enhanced magnetic resonance imaging for the detection of hepatocellular carcinoma: a meta-analysis. PLoS ONE. 2013;8(8):e70896–e70896. 53. Seale MK, et al. Hepatobiliary-specifi c MR contrast agents: role in imaging the liver and biliary tree. Radiographics. 2009;29(6):1725–48. A.P. Desai and H.S. Te 111 54. Taouli B, Koh D-M. Diffusion-weighted MR imaging of the liver. Radiology. 2010;254(1):47–66. 55. Bartolozzi C, et al. Contrast-enhanced magnetic resonance imaging of 102 nodules in cirrhosis: correlation with histological fi ndings on explanted livers. Abdom Imaging. 2013;38(2):290–6. 56. Bolondi L, et al. Characterization of small nodules in cirrhosis by assessment of vascularity: the problem of hypovascular hepatocellular carcinoma. Hepatology. 2005;42(1):27–34. 57. Hanna RF, et al. Cirrhosis-associated hepatocellular nodules: correlation of histopathologic and MR imaging features. Radiographics. 2008;28(3):747–69. 58. Sun HY, et al. Gadoxetic acid-enhanced magnetic resonance imaging for differentiating small hepatocellular carcinomas (< or =2 cm in diameter) from arterial enhancing pseudolesions: special emphasis on hepatobiliary phase imaging. Investig Radiol. 2010;45(2):96–103. 59. Vandecaveye V, et al. Diffusion-weighted MRI provides additional value to conventional dynamic contrast-enhanced MRI for detection of hepatocellular carcinoma. Eur Radiol. 2009;19(10):2456–66. 60. Xu P-J, et al. Added value of breath hold diffusion-weighted MRI in detection of small hepatocellular carcinoma lesions compared with dynamic contrast-enhanced MRI alone using receiver operating characteristic curve analysis. J Magn Reson Imaging. 2009;29(2):341–9. 61. Hwang J, et al. Pre-transplant diagnosis of hepatocellular carcinoma by gadoxetic acidenhanced and diffusion-weighted magnetic resonance imaging. Liver Transpl. 2014;20(12):1436–46. 62. Park MJ, et al. Validation of diagnostic criteria using gadoxetic acid-enhanced and diffusionweighted MR imaging for small hepatocellular carcinoma (<= 2.0 cm) in patients with hepatitis- induced liver cirrhosis. Acta Radiol. 2013;54(2):127–36. 63. Park MJ, et al. Small hepatocellular carcinomas: improved sensitivity by combining gadoxetic acid-enhanced and diffusion-weighted MR imaging patterns. Radiology. 2012;264(3):761–70. 64. Nguyen MH, et al. Racial differences in effectiveness of alpha-fetoprotein for diagnosis of hepatocellular carcinoma in hepatitis C virus cirrhosis. Hepatology. 2002;36(2):410–7.