Hepatocellular Carcinoma

Educational Objectives

The goal of this program is to improve the management of hepatocellular carcinoma (HCC). After hearing and assim­ilating this program, the clinician will be better able to:

1.   Recommend mandatory screening for those patients considered to be at high risk.

2.   Utilize an algorithm for the diagnosis and treatment of HCC.

3.   Optimize patient selection for each of the treatment strategies.

4.   Elaborate on the use of sorafenib as a treatment option.

5.   Discuss the indications, contraindications, and complications of radiofrequency ablation, chemoembolization, and radioembolization.

Faculty Disclosure

In adherence to ACCME Standards for Commercial Support, Audio-Digest requires all faculty and members of the plan­ning committee to disclose relevant financial relationships within the past 12 months that might create any personal con­flicts of interest. Any identified conflicts were resolved to ensure that this educational activity promotes quality in health care and not a proprietary business or commercial interest. For this program, the following has been disclosed: Dr. Sherman is a speaker for Bayer AG. Dr. Wallace and the planning committee reported nothing to disclose. In his lecture, Dr. Wallace presents information that is related to the off-label or investigational use of a therapy, product, or device.

Current Management of Hepatocellular Carcinoma

Morris Sherman, MD, PhD, Associate Professor of Medicine, University of Toronto Faculty of Medicine, To­ronto, ON

Survival: majority of hepatocellular carcinoma (HCC) now curable; requires aggressive high-quality screening and management of small screening-detected lesions

Mandatory screening: those with hepatitis B; Asian men >40 yr of age; Asian women >50 yr of age; individuals with hepatitis B and family history of first-degree relative with hepatoma; screen earlier in blacks of North Ameri­can, African, and Caribbean origin, since disease tends to occur at younger age; those with cirrhosis; in hepatitis B, cirrhosis not precondition to screening

Diagnostic algorithm: lesion <1 cm on ultrasonography (US)  —chances of malignancy small (no further investiga­tion); repeat US at 3- to 4-mo intervals; if lesion stable, repeat US at 3- to 4-mo intervals for 2 yr; if unchanged at 2 yr, unlikely cancerous; resume surveillance; if lesion enlarging, proceed according to lesion size; tumors >2 cm  —  typical vascular pattern (hypervascularity in arterial phase and hypovascularity in venous phase) on computed to­mography (CT) or magnetic resonance imaging (MRI) highly specific; treat as cancer (no biopsy); tumors 1 to 2 cm  —  gray area; use 2 dynamic imaging studies (CT and MRI); if both show typical pattern, treat as cancer (no bi­opsy); if one or neither shows typical features, biopsy required; if biopsy diagnostic, treat as cancer; negative biopsy  —  cannot rule out cancer; follow closely with repeat biopsy or imaging; if change in size or profile of lesion seen on imaging follow-up, repeat imaging and/or biopsy; rationale — most nodules <1 cm not cancer (probably cirrhotic); biopsy difficult and unreliable; biopsy appearance of very small lesions most often shows features of early HCC; radiology  —highly specific; 3 phases of CT (arterial, portal, and delayed); in lesions <2 cm, adding de­layed phase increases sensitivity; if lesion >2 cm, sensitivity »100%; speaker’s study  —  89 patients with 107 nod­ules 1 to 2 cm in diameter; all patients had CT, MRI, and contrast US; found that sequential imaging markedly increased sensitivity, with some loss of specificity but increased overall accuracy, resulting in reduction in number of biopsies

Recent modifications of algorithm: if lesion <1 cm, guidelines unchanged; guidelines now address all lesions >1 cm; require 4-phase dynamic contrast-enhanced CT or dynamic contrast-enhanced MRI (start with either modal­ity); 4 phases (no contrast, arterial, venous, and delayed); if arterial hypervascularity and venous washout found with first modality, treat as cancer (no biopsy); if features not seen, perform second study using other modality; if features indicative of cancer seen, treat as such; if typical features not seen, perform biopsy; if biopsy negative, can­not rule out cancer; follow patient closely

Very early HCC: first identifiable morphologic changes of cancer; called “vaguely nodular lesions” (VNL; no dis­tinct outlines); characteristics of VNL  —  smaller than small HCC, which often have well-differentiated morphol­ogy; only 15% of VNL exclusively well differentiated, although most have some foci of moderate differentiation; hypovascularity most important diagnostic feature of VNL; small HCC often contain fat whereas VNL do not; both VNL and small HCC exhibit stromal invasion; VNL never have capsule (50% of small HCC have capsule); VNL tend to evolve into small HCC

Early HCC: composed of well-differentiated cancer tissue; initially, various numbers of portal tracts still present and may show replacing growth pattern at tumor boundary, ie, cancer cells growing into cords; at later stage, cancer cells expand liver cords to give typical appearance; CD34  —  stains vascular endothelium; normally, only positive vascular endothelium within portal tracts; in cancer, spaces between cords arterialized, resulting in positive stain; CK7  —  biliary stain; portal tracts not usually found in early HCC, although possible in very early HCC; recommended that all 3 stains, plus glutamine synthetase, be used in all small cancers where diagnosis uncertain; stromal invasion  —  hepatocyte-looking cells found against vein, invading portal tract (not present in normal portal tract); in­dication of malignancy; first identifiable invasive behavior; nomenclature of small lesions  —  low-grade dysplastic nodule (LGDN); high-grade dysplastic nodule (HGDN); well-differentiated or very early cancer; more advanced cancer; LGDN and HGDN probably not malignant

Management: divided into early, intermediate, advanced, and end stage; majority of patients (50%-60%) present in intermediate or advanced stage; »30% present earlier and »10% present late; in countries where US screening rig­orously performed, »60% present in early stage

Strategy for treatment (Barcelona Clinic Liver Cancer [BCLC] staging system): performance status (PS); Child-Pugh (CP) score; best patients have no abnormalities on PS, CP class A (CP-A), with tumor <2 cm; if portal pressure and bilirubin normal, resection indicated; if portal pressure and bilirubin elevated, resection not indicated; if no associated diseases present, patient candidate for transplantation, even though tumor very small; if associated diseases present, radiofrequency ablation (RFA) indicated; transplantation preferred in all categories if enough or­gans available; advanced disease  —  PS possibly good and CP score possibly normal or slightly prolonged; if <3 le­sions <3 cm present, and PS good, or if 3 lesions £3 cm present, transplantation indicated (utilize Milan criteria); multiple lesions >3 cm  —  chemoembolization (CE) indicated; if portal invasion or metastatic nodes present, sorafenib; if patient presents with jaundice and ascites, symptomatic care; prognosis for untreated HCC in interme­diate and advanced stage  —  overall 5-yr survival 7%; intermediate stage, 16%; advanced stage, 0%

Resection: no level 1 evidence demonstrating advantage over other forms of therapy; however, consensus exists on some aspects of patient selection, although few patients candidates; contraindications  —  most CP-B and all CP-C classes; evidence of extrahepatic cancer; major portal or hepatic vein invasion; in minor portal vein invasion, resec­tion performed, but recurrence rate significantly higher; portal hypertension (HTN)  —  measure portal pressure gra­dient or determine presence of esophageal varices, splenomegaly, or platelet count <100,000/μL; inadequate liver function  —  elevated serum bilirubin; CP-B or CP-C class; indocyanine green retention >15% at 15 min; remaining liver ³450 mL and 1 cm clear margin required; centrally situated tumors particularly difficult; survival  —  decreases with multiple lesions, compared to single lesions in CP-A and in CP-B; presence of portal HTN and multiple le­sions not contraindication but worsens outcome; outcome  —  overall 5-yr survival after resection »70%, and 10-yr survival »30%; 5-yr recurrence rate 70% to 90%

Liver transplantation (LT): no level 1 evidence showing that LT better than other forms of therapy, although post-LT survival data highly satisfactory; study showed that tumor-free survival at 3 yr significantly better with LT; al­though 5-yr survival with LT similar to that of resection, recurrence rate significantly lower with LT; study compar­ing resection to LT found that patients with LT had better outcomes; even better outcomes if no nodes and metastases present and margins negative; current policy  —  Milan criteria for single lesion <5 cm or £3 lesions, none >3 cm; studies show that once Milan criteria exceeded, outcomes worsen; factors predicting recurrence post-LT  —  vascular invasion and poorly differentiated histology strong predictors; tumor diameter not factor unless >5 cm; effect of adding patients with HCC to list of patients awaiting LT  —  no effect on patients on LT list without HCC (overall survival same); once Milan criteria exceeded and survival decreased after resection, absolute risk and number needed to harm increase; if mortality 50% in hepatoma group, 1 extra death for every 6.7 LTs performed; if mortality 80%, 1 extra death for every 3.3 LTs performed; consider in formulating LT policy; management of pa­tients with HCC while awaiting LT  —  no recommendations

Local ablation: size of lesion that can be treated depends on size of probe; need 0.5- to 1-cm margin; probes £7 cm available; study of lesions £8.5 cm using alcohol able to achieve 58% complete ablation; study  —  independent pre­dictors of survival CP score and whether complete initial response obtained, ie, able to ablate lesion to extent that no visible tumor seen on arterial-phase contrast imaging; if tumor <2 cm, complete response 96% (2.1-3.0 cm, 78%); overall survival at 3 yr 27%; if patient CP-A class with complete initial response, 5-yr survival 42%; if lesion <2 cm, 5-yr survival 63%

Radiofrequency ablation (RFA): randomized controlled trial (RCT) of RFA vs resection  —  tumors <5 cm; outcome similar in overall survival and disease-free survival; those who had local ablation had better survival, while those who had surgery had fewer recurrences at 5 yr; overall, no statistically significant difference; RFA vs alcohol injec­tion — slight advantage in overall survival and greater advantage in recurrence-free survival with RFA; RFA for HCC  —  Italian study looked at tumors <2 cm treated with RFA; complete response rate (no detectable tumor on ra­diography) 97%; 5-yr survival 69%; local recurrence rate over 5 yr <1%; crucial to find small lesions because of curability with outpatient procedure

Resection for small HCC: local ablation with RFA can be first-line treatment for lesions <2 cm; resection better for lesions >2 cm; validation still required

LT for small lesions: 5-yr survival for lesions <2 cm and CP-A 63% with RFA; 5-yr survival in LT for HCC 65% (in­tention-to-treat analysis); therefore, LT should not be offered to patients with lesions <2 cm and possibly reserved for recurrences or second primary lesion

Chemoembolization: RCTs show survival advantage; meta-analysis clearly showed advantage over control group; standard of care for those with multinodular disease and CP-A

Sorafenib: for extrahepatic disease or portal vein invasion; multikinase inhibitor; antiangiogenic agent; shown in RCTs to prolong survival, compared to placebo; first chemical agent approved by Food and Drug Administration to manage HCC; average increase in survival 10.7 mo (vs 7.9 mo); conventional chemotherapy not shown to increase survival

BCLC staging system: provides blueprint for treatment strategies in different stages of disease; de facto staging sys­tem used; only one with defined population survival for different stages of disease

A Radiologist’s Perspective on Treatment of HCC

Michael J. Wallace, MD, Associate Professor and Section Head, Interventional Radiology, University of Texas MD Anderson Cancer Center, Houston

Locoregional treatment options: percutaneous ablation (chemical and thermal); transvascular approaches (arterial infusion, embolization, CE, and radioembolization); chemical percutaneous ablation (percutaneous ethanol injec­tion [PEI]); thermal ablation  —  RFA; cryoablation; microwave ablation; laser ablation; high-intensity focused ul­trasound

Radiofrequency ablation: probes produce local ionic agitation and create frictional heat at tip to ablate tissue; goal to achieve temperature of 60°C to 100°C to cause instantaneous coagulation necrosis; several types of devices available; typically requires electrical generator, applicator, and dispersive electrode; probe shapes variable; aside from ablating tumor, goal is to obtain margin around tumor; 3 cm rough threshold used; reasonably acceptable out­comes with lesions <3 cm; complete ablation rates (on imaging) upper 80% to 90% (lower for PEI), but decrease with lesions ³3 cm; looking at pathologic necrosis, even lesions £3 cm have lower success rates with complete ab­lation; tract seeding  —complication (12% in small series; £1% in larger series); if subcapsular lesions present, use indirect approach, going through normal liver, and ablate tract; risk increased with previous biopsy or poorly differ­entiated HCC; several RCTs showed benefit over PEI, with better local recurrence-free survival, fewer treatments, and lower tumor-free progression; several series show 5-yr survival rates in mid to upper 50%; criteria for treatment  —  absence of extrahepatic disease; »3 tumors; tumor size £3 cm; with larger tumors, combined therapy usual

Chemoembolization

Mechanism of action: takes advantage of hypervascularity of tumor; treating through arterial system; combination of embolization, causing ischemia or hypoxia, and delivering chemotherapeutic agent to tumor, reducing flow and in­creasing time chemotherapeutic agent dwells in tumor; current regimens involve chemotherapeutic agent with ethiodized oil (Ethiodol), which acts as emulsifying agent with drug and as contrast agent; dual embolization effect achieved by aiding emulsified product to treat portal venules through peribiliary plexus; embolic material added af­terwards to reduce flow to tumor; CT or MRI used to guide follow-up studies

Contraindications: hepatic encephalopathy; elevated bilirubin (>3 mg/dL); relative contraindications include exten­sive hepatic tumor and impaired liver function; portal vein thrombosis or compromise often viewed as contraindica­tion, but able to perform CE safely, with some survival benefit; those with complex biliary abnormalities (with stents or biliary bypasses) at higher risk for liver abscess

Complications: »10%; liver failure and abscess formation; relatively well tolerated; aggressive antibiotic use and bowel preparation reduce risk for liver abscess; goal cytoreduction or palliation; well-circumscribed lesions that ac­cumulate ethiodized oil homogeneously have better outcomes, compared to ill-defined or multinodular lesions; also used for downstaging tumors for resection or LT; use previously controversial, with several studies showing no ben­efit, until 2 RCTs in 2002 showed evidence of survival benefit

Drug-eluting beads: comparison to conventional CE showed peak of drug in systemic circulation after 1 hr, com­pared to blunted response with drug-eluting bead product; European study  —  showed no substantial difference with conventional transarterial CE (TACE), but significant difference in higher-risk patients; fewer side effects in drug-eluting bead cohort

Multitherapy approaches: for larger (£7 cm) solitary lesions, combination of CE and ablation showed results com­parable to resection; in 12-mo follow-up study, no recurrence found in smaller lesions (<5 cm) compared to lesions 5 to 12 cm; using CE with RFA, better results seen with smaller lesions (£5 cm)

Portal vein (PV) tumor thrombus: relative contraindication; negative prognostic factor; median survival 3 mo if un­treated; survival benefit may be seen with CE, depending on severity of PV involvement; some studies show sur­vival benefit £15 mo, depending on level of PV occlusion

Radioembolization

Mechanism of action: delivers b-emitting particle through vascular system, exploiting hypervascular nature of tu­mors; patients must have good PS and preserved hepatic and renal function

Contraindications: previous liver irradiation; pulmonary insufficiency; biliary bypass or biliary enteric anastomosis (higher risk for liver abscess); portal HTN

Complications: postembolization syndrome  —  not as severe as in CE; abdominal pain does not usually require nar­cotics; fatigue usually lasts 2 wk; gastrointestinal  —  avoidable, if meticulous search for vessels performed; biliary, hepatic, and pulmonary complications rare; median survival relatively comparable to CE, although no head-to-head comparison performed; dose dependence  —  those who received dose >104 Gy had better survival than those who received smaller doses

Acknowledgements

Dr. Sherman was recorded at 25th Annual New Treatments in Chronic Liver Disease, held March 27-28, 2010, in San Diego, CA, and sponsored by Scripps Clinic. Dr. Wallace was recorded at 34th Annual Texas Program, held September 11-13, 2009, in Hous­ton, TX, and sponsored by the American College of Gastroenterology, Texas Society for Gastroenterology and Endoscopy, and SGNA Texas Regional Societies. The Audio-Digest Foundation thanks the speakers and the sponsors for their cooperation in the production of this program.

The 26th Annual New Treatments in Chronic Liver Disease will be held in San Diego, CA, on March 26-27, 2011. To attend this meeting, call 1-800-SCRIPPS or visit www.scripps.org, click on CME, select Annual Meetings, and click Gastroenter­ology.

Suggested Reading

Cheung TT et al: Combined resection and radiofrequency ablation for multifocal hepatocellular carcinoma: prognosis and outcomes. World J Gastroenterol. 2010 Jun 28;16(24):3056-62; Cillo U et al: Intention-to-treat analysis of liver transplan­tation in selected, aggressively treated HCC patients exceeding the Milan criteria. Am J Transplant. 2007 Apr;7(4):972-81; Dhanasekaran R et al: The effectiveness of locoregional therapies versus supportive care in maintaining survival within the Milan criteria in patients with hepatocellular carcinoma. J Vasc Interv Radiol. 2010 Aug;21(8):1197-204; quiz 204; Ma­hady SE et al: Locoregional therapies for hepatocellular carcinoma: which patients are most likely to gain a survival advan­tage? J Gastroenterol Hepatol. 2010 Jul;25(7):1299-305; Miyayama S et al: Chemoembolization for the treatment of large hepatocellular carcinoma. J Vasc Interv Radiol. 2010 Aug;21(8):1226-34; Orsi F et al: High-intensity focused ultrasound ablation: effective and safe therapy for solid tumors in difficult locations. AJR Am J Roentgenol. 2010 Sep;195(3):W245-52; Peng ZW et al: Radiofrequency ablation as first-line treatment for small solitary hepatocellular carcinoma: long-term re­sults. Eur J Surg Oncol. 2010 Nov;36(11):1054-60; Seo YS et al: Preliminary result of stereotactic body radiotherapy as a local salvage treatment for inoperable hepatocellular carcinoma. J Surg Oncol. 2010 Sep 1;102(3):209-14; Strigari L et al: Efficacy and toxicity related to treatment of hepatocellular carcinoma with 90Y-SIR spheres: radiobiologic considerations. J Nucl Med. 2010 Sep;51(9):1377-85; Tseng PL et al: Optimal treatment increased survival of hepatocellular carcinoma pa­tients detected with community-based screening. J Gastroenterol Hepatol. 2010 Aug;25(8):1426-34; Wakui N et al: Re­currence incidence of small HCC in cirrhosis patients by ablation versus injection. Hepatogastroenterology. 2010 Mar-Apr;57(98):195-201; Welker MW et al: Efficacy and safety of sorafenib in advanced hepatocellular carcinoma under daily practice conditions. J Chemother. 2010 Jun;22(3):205-11; Yoshizumi T et al: Living donor liver transplantation in patients older than 60 years. Transplantation. 2010 Aug 27;90(4):433-7.