Approach Considerations
The diagnosis of hepatocellular carcinoma (HCC) can often be established on the basis of noninvasive imaging, without biopsy confirmation. Even when biopsy is needed, imaging is usually required for guidance. [4]
Because the outcome in patients with advanced HCC is uniformly dismal, early diagnosis is crucial in order to provide effective treatment. Consequently, routine screening for HCC is recommended in patients with cirrhosis from any cause; some guidelines also recommend testing in other patients at high risk (see Guidelines). Screening is typically performed using ultrasonography (US), with or without serum alpha-fetoprotein (AFP) measurement, generally every 6 months.
AFP is elevated in 75% of cases. The level of elevation correlates inversely with prognosis. An elevation of greater than 400 ng/mL predicts for HCC with specificity greater than 95%. In the setting of a growing mass, cirrhosis, and the absence of acute hepatitis, many centers use a level greater than 1000 ng/mL as presumptive evidence of HCC (without biopsy). AFP alone is inadequate for screening purposes because of the high rate of false positives in active hepatitis; it has only 40-64% sensitivity because many tumors do not produce AFP at all or do so only at a very advanced stage. [35]
US as a screening method is reported to have 60% sensitivity and 97% specificity in the cirrhotic population, and it has been demonstrated to be cost-effective. [36, 37] Findings on US should then be confirmed with further imaging studies—multiphase computed tomography (CT) or magnetic resonance imaging (MRI)—and potentially biopsy.
With aggressive screening, the rate of resectable HCC diagnosed in patients who are at high risk reaches 30-50%, which is nearly twice the rate of unscreened populations. [38] Despite the significant risk of recurrence, even in treated patients, the screening protocols appear to be cost effective in this population. [39]
On CT, HCC generally appears as a focal nodule with early enhancement on the arterial phase with rapid washout of contrast on the portal venous phase of a three-phase contrast scan. MRI of HCC generally demonstrates high signal intensity on T2 imaging. Biopsy is indicated in patients with HCCs that are larger than 2 cm with low AFP or in whom ablative treatment or transplant is contraindicated. In patients with elevated AFP and consistent imaging characteristics, patients can be treated presumptively for HCC without a biopsy. Patients should also undergo evaluation for extrahepatic disease (primarily pulmonary metastasis) with cross-sectional imaging, as the presence of extrahepatic disease would preclude curative locoregional therapy. For more information, see Hepatocellular Carcinoma Imaging.
Laboratory evaluation of patients with newly diagnosed HCC should include testing to determine the severity of the underlying liver disease, such as the following:
-
Complete blood cell count (CBC)
-
Electrolyte levels
-
Liver function tests (LFTs)
-
Coagulation studies (eg, international normalized ratio [INR], partial thromboplastin time [PTT])
-
AFP determination
Laboratory Studies
Laboratory results suggestive or indicative of disease severity include the following:
-
Anemia - Low hemoglobin may be related to bleeding from varices or other sources
-
Thrombocytopenia - A platelet count below 100,000/μL is highly suggestive of significant portal hypertension/splenomegaly
-
Hyponatremia is commonly found in patients with cirrhosis and ascites and may be a marker of advanced liver disease
-
Increased serum creatinine level may reflect intrinsic renal disease or hepatorenal syndrome
-
Prolonged prothrombin time (PT)/INR reflects significant impairment of hepatic function that may preclude resection
-
Elevated liver enzymes reflect active hepatitis due to viral infection, current alcohol use, or other causes
-
Increased bilirubin level usually indicates advanced liver disease
-
Hypoglycemia may represent end-stage liver disease (no glycogen stores)
Laboratory findings associated with particular disease etiologies include the following:
-
Hepatitis B surface antigen (HBsAg)/hepatitis B core antibody (anti-HBc), anti-HCV - Viral hepatitis (current/past)
-
Increased iron saturation (> 50%) - Underlying hemochromatosis
-
Low α1-antitrypsin levels - α1-Antitrypsin deficiency
-
Increased AFP - Levels higher than 400 ng/mL are considered diagnostic with appropriate imaging studies
-
Hypercalcemia - Ectopic parathyroid hormone production is possible in 5-10% of patients with HCC
Alpha fetoprotein
AFP levels may be elevated because of production by the tumor or by regenerating hepatocytes. Therefore, AFP levels are also frequently elevated in chronic active hepatitis C (levels of 200-300 ng/mL are not uncommon), but in those patients the levels tend to fluctuate and do not progressively increase. AFP levels can also be elevated because of other conditions, such as following liver resection (transient until regeneration complete), recovery following toxic injury, or seroconversion following hepatitis B infection (typically inducing transient exacerbation of inflammation).
When elevated, the AFP is 75-91% specific, and values greater than 400 ng/mL are generally considered diagnostic of HCC in the proper clinical context, including appropriate radiologic findings. [40] (See Table 2 below.) Better biologic markers, including AFP variants, are being investigated. [41, 42]
Table 2. Serum Alpha-Fetoprotein (AFP) Determination in Liver Disease [40] (Open Table in a new window)
Alpha-Fetoprotein (ng/mL) |
Interpretation |
> 400-500 |
- HCC likely if accompanied by space-occupying solid lesion(s) in cirrhotic liver or levels are rapidly increasing. - Diffusely growing HCC, may be difficult to detect on imaging. - Occasionally in patients with active liver disease (particularly HBV or HCV infection) reflecting inflammation, regeneration, or seroconversion |
Normal value to < 400 |
- Frequent: Regeneration/inflammation (usually in patients with elevated transaminases and HCV) - Regeneration after partial hepatectomy - If a space-occupying lesion and transaminases are normal, suspicious for HCC |
Normal value |
Does not exclude HCC (cirrhotic and noncirrhotic liver) |
Ultrasonography
Accurate diagnosis and surgical planning require adequate cross-sectional imaging studies. Although US is commonly used for screening, it does not provide sufficient anatomic detail for planning surgical resection or ablation. Correlation between ultrasonographic findings and explant liver pathology has revealed that a significant number of small lesions may not be detected with ultrasound screening. Pooled estimates from one meta-analysis suggested that US is only 60% sensitive. [36]
US identification of HCC can be difficult in the background of regenerative nodules in the cirrhotic liver. In general, HCC appears as a round or oval mass with sharp, smooth boundaries. The lesions have a range of echogenicity, from hypoechoic to hyperechoic, depending on the surrounding parenchyma and the degree of fatty infiltration. The border between the HCC and the liver can become indistinct with nodular HCC. The use of Doppler analysis to characterize the lesion can be helpful, in that HCC is more likely to have a significant arterial blood supply and neovascularization as compared to regenerative nodules. (See the image below.)
Computed Tomography
Triple-phase CT (including an arterial phase, a portal venous phase, and a late washout phase) has been found to be highly accurate in the diagnosis and characterization of HCCs but, like US, may miss smaller lesions. Pooled estimates reveal a sensitivity of 68% and a specificity of 93%. [36] Disadvantages of CT include cost, radiation exposure, and the need for iodinated contrast.
Classic CT findings of HCC include a hypervascular pattern with arterial enhancement and rapid washout during the portal venous phase. [43] (See the images below.) In contrast, regenerative nodules generally appear isoattenuating or hypoattenuating when compared with the remaining parenchyma. Other characteristics that support the diagnosis of HCC include visualization of a tumor capsule, demonstration of an internal mosaic resulting from variable attenuation within the tumor, and portal vein branch invasion. Unfortunately, all of these characteristics are more easily demonstrated in large lesions. Consequently, small lesions are frequently missed on CT examination.
Magnetic Resonance Imaging
MRI provides an excellent method for characterizing HCC without radiation and the need for iodinated contrast. Technologic improvements have reduced scanning time and improved the specificity of the study. Pooled analysis demonstrated a sensitivity of 81% and a specificity of 85%. [36]
HCC demonstrates a variety of features on MRI, depending on the tumor architecture, grade, and amount of intratumoral fat and glycogen. [43] (See the image below.) The lesion ranges from isointense to hyperintense (bright) on T1-weighted images. Similarly, T2 images may vary from isointense to hyperintense. Well-differentiated tumors are more commonly hyperintense on T1 images and isointense on T2 images, whereas moderately or poorly differentiated tumors tend to be hyperintense on T2 images and isointense on T1 images. Although imaging characteristics may be suggestive, a significant overlap may occur between the tumor and regenerative nodules.
The benefits of contrast-enhanced studies must be balanced against the risks if any anatomic or functional renal impairment is possible. Iodinated contrast for CT may worsen renal failure, and gadolinium enhancement on MRI has been linked to a syndrome of severe systemic fibrosis in a patient with renal failure. [44]
Biopsy
The decision to biopsy a lesion suspected of being hepatocellular carcinoma is the subject of ongoing controversy. In patients with large tumors who are not candidates for resection or transplantation, biopsy is frequently not indicated to confirm the diagnosis prior to initiating palliative procedures, because clinical and imaging evidence is convincing and biopsy is potentially risky.
In patients with lesions smaller than 1 cm, fewer than 50% of the lesions will be malignant, and the false-negative result rate is high. Thus, conservative management with close follow-up and no biopsy is recommended. [38]
In patients with 1- to 2-cm lesions, a biopsy should be performed; these patients have a significant risk of malignancy. If the result is positive, they are candidates for resection, transplantation, or ablative therapy. As in the smaller lesions, there is a significant false-negative result rate, and close follow-up is indicated in patients with a negative biopsy result.
Patients with lesions larger than 2 cm, cirrhosis, characteristic imaging studies, and elevated AFP values can be managed without biopsy. In these patients, the risk of tumor seeding must be taken into account. Whereas some groups require biopsy before transplantation, [38] others are willing to proceed on clinical characteristics alone. [45] In patients with more atypical findings on imaging studies, the value of AFP should not be overemphasized, because an excessive number of patients submitted to transplantation did not have HCC. [41]
In patients with cirrhosis who are being considered for resection, survival following resection has been previously correlated with the degree of portal hypertension. In some centers, determination of the wedged hepatic vein pressure is advocated to then determine the safety of resection. Resection can, in general, be safely undertaken in patients with a wedged hepatic venous pressure gradient of less than 10. [38] Patients should also have a platelet count lower than 100,000/μL and a normal bilirubin level. In patients with small tumors but significant hepatic dysfunction, transplantation is the preferred option.
Histologic Findings
Histology is quite variable: tumors range from well differentiated to anaplastic. The fibrolamellar subtype is associated with a better prognosis, possibly because it is not associated with cirrhosis and is more likely to be resectable. The presence of intracellular bile or staining for AFP may be helpful in distinguishing HCC from other hepatic malignancies (eg, cholangiocarcinoma). Various other immunohistochemical markers are available to establish the diagnosis of HCC. Nguyen et al reported that arginase-1 and hepatocyte paraffin antigen 1 (Hep Par 1) had the highest sensitivity for well-differentiated HCC, whereas arginase-1 and glypican-3 had the highest sensitivity for poorly differentiated HCC. [46]
Aberrations of chromosome 1 and 8 are common features of HCC that can be detected by fluorescent in situ hybridization (FISH) technique. The role of FISH in the diagnosis of hepatocellular carcinoma is still under investigation.
Staging
The prognosis of HCC is a reflection of both tumor characteristics (ie, size, location, tumor biology) and the degree of underlying liver disease. The traditional pathologic TNM (tumor-node-metastasis) staging system, while helpful in determining a prognosis in patients undergoing resection, is not as useful in planning treatment, because it fails to include measures of the severity of the liver disease. However, the tumor size is predictive of outcome, as it predicts the likelihood of major venous involvement. [47]
Likewise, the Child-Pugh-Turcotte score predicts perioperative survival after resection, but it does not incorporate tumor size, number, and location, which have important implications for respectability and treatment. Among the scales that integrate the tumor and liver disease characteristics, the Barcelona Clinic Liver Cancer (BCLC) system, [38] the Japan Integrated Staging System, and the Cancer of the Liver Italian Program (CLIP) are the most widely used staging systems.
BCLC algorithm
The Barcelona-Clinic Liver Cancer (BCLC) approach to hepatocellular carcinoma management. Adapted from Llovet JM, Fuster J, Bruix J, Barcelona-Clinic Liver Cancer Group. The Barcelona approach: diagnosis, staging, and treatment of hepatocellular carcinoma. Liver Transpl. Feb 2004;10(2 Suppl 1):S115-20.
The BCLC system is very useful in deciding among potential treatment options and correlates best with patient outcome among the major staging systems. [48]
In the BCLC system, stage 0 patients have lesions smaller than 2 cm, normal bilirubin levels, and normal portal pressure measurements. These patients can often undergo resection safely with excellent long-term survival.
Patients with larger tumors (ie, single tumors < 5 cm or multiple [≤ 3] tumors < 3 cm) are considered for resection if they have preserved liver function or for transplantation if they have decompensated cirrhosis.
In patients whose tumor exceeds these measurements, palliative therapy can be offered depending upon hepatic reserve. Fewer than 10% of these patients survive longer than 3 years.
CLIP scoring system
A score of 0-2 is assigned for each of the 4 features listed below; a cumulative score ranging from 0-6 is the CLIP score.
Child-Pugh class:
-
Class A = 0
-
Class B = 1
-
Class C = 2
Tumor morphology:
-
Uninodular and extension less than 50% = 0
-
Multinodular and extension less than 50% = 1
-
Massive and extension greater than 50% = 2
Alpha-fetoprotein:
-
Less than 400 = 0
-
Greater than 400 = 1
Portal vein thrombosis:
-
Absent = 0
-
Present = 1
Estimated survival based on CLIP score
Patients with a total CLIP score of 0 have an estimated survival of 31 months; those with score of 1, about 27 months; score of 2, 13 months; score of 3, 8 months; and scores 4-6, approximately 2 months.
For more information, see Hepatocellular Carcinoma Staging.
-
Large hepatocellular carcinoma. Image courtesy of Arief Suriawinata, MD, Department of Pathology, Dartmouth Medical School.
-
Photomicrograph of a liver demonstrating hepatocellular carcinoma. Image courtesy of Arief Suriawinata, MD, Department of Pathology, Dartmouth Medical School.
-
MRI of a liver with hepatocellular carcinoma.
-
Ultrasonographic image of hepatocellular carcinoma.
-
Arterial phase CT scan demonstrating enhancement of hepatocellular carcinoma.
-
Portal venous phase CT scan demonstrating washout of hepatocellular carcinoma.
-
The Barcelona-Clinic Liver Cancer (BCLC) approach to hepatocellular carcinoma management. Adapted from Llovet JM, Fuster J, Bruix J, Barcelona-Clinic Liver Cancer Group. The Barcelona approach: diagnosis, staging, and treatment of hepatocellular carcinoma. Liver Transpl. Feb 2004;10(2 Suppl 1):S115-20.
-
Hepatocellular carcinoma: pathobiology.
-
Right hepatectomy. Part 1: Dissection of portal vein. Courtesy of Memorial Sloan-Kettering Cancer Center, featuring Leslie H. Blumgart, MD. (From Blumgart LH. Video Atlas: Liver, Biliary & Pancreatic Surgery. Philadelphia, PA: Saunders; 2010.)
-
Right hepatectomy. Part 2: Devascularization of right liver. Courtesy of Memorial Sloan-Kettering Cancer Center, featuring Leslie H. Blumgart, MD. (From Blumgart LH. Video Atlas: Liver, Biliary & Pancreatic Surgery. Philadelphia, PA: Saunders; 2010.)
-
Right hepatectomy. Part 3: Suturing and dividing. Courtesy of Memorial Sloan-Kettering Cancer Center, featuring Leslie H. Blumgart, MD. (From Blumgart LH. Video Atlas: Liver, Biliary & Pancreatic Surgery. Philadelphia, PA: Saunders; 2010.)
-
Hepatocellular carcinoma (HCC). Dilated collateral superficial abdominal veins in a 67-year-old man with cirrhosis, HCC, and portal vein occlusion.
-
Hepatocellular carcinoma. Plain radiograph immediately following chemoembolization, demonstrating catheter placement and Ethiodol enhancement of tumors.
Tables
Alpha-Fetoprotein (ng/mL) |
Interpretation |
> 400-500 |
- HCC likely if accompanied by space-occupying solid lesion(s) in cirrhotic liver or levels are rapidly increasing. - Diffusely growing HCC, may be difficult to detect on imaging. - Occasionally in patients with active liver disease (particularly HBV or HCV infection) reflecting inflammation, regeneration, or seroconversion |
Normal value to < 400 |
- Frequent: Regeneration/inflammation (usually in patients with elevated transaminases and HCV) - Regeneration after partial hepatectomy - If a space-occupying lesion and transaminases are normal, suspicious for HCC |
Normal value |
Does not exclude HCC (cirrhotic and noncirrhotic liver) |
Author (Year) |
N |
Survival Rate |
|
1 year |
5 years |
||
Mazzefero (1996) |
48 |
84% |
74% |
Bismuth (1999) |
45 |
82% |
74% |
Llovet (1999) |
79 |
86% |
75% |
Jonas (2001) |
120 |
90% |
71% |
Stage |
Criteria |
0 – Very Early |
Child-Pugh class A Single < 2 cm nodule ECOG PS 0-1 |
A – Early |
Child-Pugh class A-B Single or 2-3 nodules < 3 cm ECOG PS 0-1 |
B – Intermediate |
Child-Pugh class A-B Multinodular ECOG PS 0-1 |
C - Advanced |
Child-Pugh class A-B Portal vein invasion, N1, M1 ECOG PS 0-2 |
D - Terminal |
Child-Pugh class C Any T, N, or M ECOG PS > 2 |
ECOG PS = Eastern Cooperative Oncology Group Performance Status |
|
