Page 76 - Read Online

Osho et al. Hepatoma Res 2020;6:55 I http://dx.doi.org/10.20517/2394-5079.2020.42 Page 5 of 15

ultrasound and serum biomarkers for most patients, with multiphase CT or MRI considered in the highest
[7]
risk patients .

Unfortunately, a systematic review found that less than 20% of patients with cirrhosis in the U.S. undergo
HCC surveillance, with even lower rates of guideline-concordant semi-annual surveillance [48,49] . Patients
and providers have reported potential barriers to surveillance including knowledge deficits, time
constraints, and financial costs of tests that need to be addressed to increase surveillance utilization [50,51] .
Studies have demonstrated promise for inreach efforts such as electronic medical record reminders or
outreach strategies including mailed invitations to complete ultrasound surveillance [52-54] .

While awaiting ongoing trial data for both novel biomarkers and cross-sectional imaging techniques,
ultrasound with or without AFP remains the gold standard surveillance strategy.

ROLE OF IMAGING IN HCC DIAGNOSIS
For surveillance to be effective, recall procedures must be followed for patients with abnormal surveillance
[55]
tests . In patients with an ultrasound nodule < 1 cm in maximum diameter, the risk of HCC is low and
professional society guidelines recommend repeat short-interval ultrasound in ~3 months. If the lesion is
stable in size, it can be followed on ultrasound; however, diagnostic evaluation with cross-sectional imaging
[2]
(i.e., contrast-enhanced MRI or 4-phase CT) is recommended once a lesion is ≥ 1 cm in size [Figure 1].
HCC is unique compared to other cancers, in that the diagnosis can often be made radiographically without
histological confirmation. Historically, HCC diagnosis has been made on the basis of the presence of “arterial
enhancement and delayed washout”, i.e., hypervascularity during the arterial phase and hypointensity on
the portal venous or delayed phases of imaging. This classic appearance is related to the liver’s dual blood
supply, where the background liver receives most of its blood supply from the portal vein and HCC lesions
obtain most of their blood supply from hepatic artery branches. In the setting of cirrhosis, this appearance
was demonstrated to have a specificity of 95% for the diagnosis of HCC [56,57] .

LI-RADS criteria
More recently, the American Association for Cancer Research (AACR) and AASLD have adopted the
LI-RADS criteria for HCC diagnosis and classification, and have chosen specific populations for which
[57]
these criteria should be applied, namely patients with cirrhosis and chronic hepatitis B infection . The
LI-RADS criteria do not apply to pediatric patients or patients with cirrhosis secondary to vascular
[58]
disorders (e.g., Budd-Chiari syndrome, sinusoidal obstruction syndrome) . The LI-RADS criteria include
a combination of major and minor imaging criteria, and classifies lesions on a scale ranging from LR-1
(definitely benign) to LR-5 (definitely HCC) or LR-M (malignant but not definite for HCC) [Table 1].
Major LI-RADS criteria include arterial phase hyperenhancement (APHE), delayed washout, enhancing
capsule, and threshold growth. Patients with LR-1 and LR-2 lesions are definitely and likely benign,
respectively, so most of these patients can return to ultrasound-based surveillance. Patients with LR-3
and LR-4 lesions have an intermediate risk of HCC, so these patients can be considered for continued
observation versus biopsy after multidisciplinary discussion. A recent systematic review found 38 and 74%
of LR-3 and LR-4 lesions were HCC, respectively, highlighting that these lesions should not be ignored and
must be followed clinically . In this systematic review, LR-5 lesions had a positive predictive value of 94%
[59]
for being HCC, and therefore do not warrant biopsy for histological confirmation prior to treatment. The
LR-M classification is reserved for lesions that are suspicious for malignancy but have features that are not
definite for HCC, e.g., peripheral enhancement, and can be seen in other malignancies such as intrahepatic
cholangiocarcinoma. Therefore, biopsy is typically recommended in these cases to make a definitive
diagnosis. It should also be noted that the LI-RADS criteria do not apply to patients without cirrhosis
and/or chronic HBV infection, as the positive predictive value of the aforementioned classic imaging