Keywords
hepatocellular carcinoma - staging - locoregional therapy
Hepatocellular carcinoma (HCC) is the fifth leading cause of cancer and the second
leading cause of cancer mortality worldwide.[1]
[2] According to the Centers for Disease Control and Prevention, liver cancer is the
only cancer for which the incidence and mortality are rising: between 2003 and 2012,
the incidence of liver cancer increased by 38% and the mortality increased by 56%.[3] The incidence has almost tripled in the last three decades in the United States
and is expected to continue to rise through 2030 with the rise in obesity and nonalcoholic
steatohepatitis (NASH).[4] The African American and Hispanic communities are expected to experience the highest
growth in the United States.[5] In Asia and Africa, the leading cause of HCC is chronic hepatitis B virus (HBV)
infection and aflatoxin exposure, while in the United States hepatitis C virus (HCV)
infection is the leading cause, with alcoholic cirrhosis and NASH/obesity as growing
factors.[6] Without treatment, the median survival of HCC is 6 to 20 months, and the 5-year
survival is approximately 10 to 14%. Treatment options vary significantly by disease
stage, and outcomes are better with early diagnosis and treatments that can cure or
curb the disease. Thus, accurate diagnosis and staging remain critical for providing
the appropriate treatment for patients.[6] This review will describe current accepted diagnosis paradigms, discuss treatment
options with a focus on locoregional therapies, and describe the ideal candidates
for locoregional therapies vis-a-vis the currently accepted and proposed staging models,
which will be critically evaluated. Lastly, we will review methods for assessing response
after locoregional therapy to determine when repeat treatment is indicated.
Surveillance and Diagnosis
Surveillance and Diagnosis
HCC develops in a setting of chronic inflammation due to hepatitis B or C viral infection,
ethanol damage, NASH, and other etiologies.[7] Given the long latency period for the development of HCC, various screening protocols
have been established to allow for early detection, when patients may still be eligible
for curative treatments. The American Association for the Study of Liver Diseases
(AASLD) and the European Association for the Study of the Liver (EASL) recommend surveillance
for adult patients at risk for developing HCC (Level 1A evidence).[8]
[9] The risk of developing HCC is dependent on the underlying etiology of liver inflammation
and the presence of cirrhosis. For instance, the incidence of HCC in Asian men with
HBV is 0.3 to 0.6% per year if there is no cirrhosis but 3 to 5% per year if cirrhosis
is present.[8] Therefore, the recommendation is to screen patients with cirrhosis of any cause
or HBV without cirrhosis in Asian and African American men older than 40 years and
Asian women older than 50 years.[10]
In most countries, surveillance is performed with ultrasound, with or without the
addition of serum α-fetoprotein (AFP) levels (which is not always elevated in HCC).[6] One Chinese randomized controlled trial (RCT) demonstrated that screening with ultrasound
and AFP can lead to a 37% reduction in mortality.[11] In the United States, initial surveillance is also often performed with multiphase
computed tomography (CT) or magnetic resonance imaging (MRI), with the advantage of
being less operator dependent but for much greater expense. Cost effectiveness studies
have been done which confirm a benefit of HCC screening and surveillance with ultrasound.[12] No RCT has been done to evaluate the sensitivity or specificity of surveillance
in patients with HCV or other causes of cirrhosis, although given the current clinical
standard of surveillance in this population, the feasibility of being able to conduct
an RCT in this cohort is logistically and possibly ethically questionable.[13] There has been growing acceptance of contrast-enhanced ultrasound (CEUS) to diagnose
HCC. A recent meta-analysis concluded that CEUS has a sensitivity of 85% and a specificity
of 91% for detecting HCC.[14] However, the AASLD recommends larger scale studies to confirm these findings before
widely recommending CEUS for HCC detection.[8]
Once surveillance ultrasound identifies a lesion suspicious for liver cancer, both
AASLD and EASL recommend further imaging with multiphasic CT or MRI.[8]
[9] The AASLD recommends diagnosis based on the American College of Radiology's (ACR's)
Liver Imaging Reporting and Data System (LI-RADS; [Fig. 1]), whereas EASL recommends basing diagnosis on their own criteria. Arterial phase
hyperenhancement and portal venous/delayed phase washout remain the hallmarks of HCC
for both LI-RADS and EASL ([Fig. 2]). If a lesion does not demonstrate classic features of HCC, then biopsy is recommended
to guide treatment. Likewise, if a lesion is suspicious for HCC on imaging in a patient
without cirrhosis or underlying liver disease (e.g., HCV, HBV, or NAFLD) then biopsy
is also indicated to establish the diagnosis.
Fig. 1 Flow chart demonstrating the different categories of findings on multiphasic liver
computed tomography or magnetic resonance imaging with next management steps.
Fig. 2 Common findings of HCC on multiphasic MRI after administration of a hepatobiliary
contrast agent (gadoxetic acid). (a) Arterial phase axial T1-weighted fat-saturated image demonstrating an arterially
hyperenhancing lesion in segment 8 which washes out on delayed and hepatobiliary sequences
(b, c) (white arrows). (d) The lesion (white arrow) demonstrates restricted diffusion on a DWI sequence. DWI,
diffusion-weighted imaging; HCC, hepatocellular carcinoma; MRI, magnetic resonance
imaging.
Staging
Treatments for HCC vary by stage and degree of hepatic decompensation from cirrhosis,
and treatment in the early stage can be potentially curative. In the intermediate
and advanced stages, locoregional and systemic therapies can be offered with the goal
of downstaging to transplantation, bridging to transplantation, controlling disease,
or palliation. Categorizing HCC patients into stages helps guide the treatment choice
([Fig. 3]) to maximize tumor response while minimizing adverse outcomes and potentially futile
or morbid interventions.
Fig. 3 Flow chart describing a staging approach to evaluate patients for locoregional therapy
with corresponding staging according to the Barcelona Liver Clinic system (BCLC) and
Hong Kong Liver Clinic system (HKLC). CP, Child–Pugh; PS, performance status; OS,
overall survival; TACE, transarterial chemoembolization; TARE, transarterial radioembolization,
TKI/CPI, tyrosine kinase inhibitor/check point inhibitor; SBRT, stereotactic body
radiotherapy.
The TNM staging system is not regularly used to guide liver cancer treatment decisions
because it does not take into account factors which impact prognosis and treatment
response, namely, liver function and patient's performance status (PS; Eastern Cooperative
Oncology Group 0–5). HCC-specific staging systems that incorporate these factors include
the Barcelona Clinic Liver Cancer (BCLC) staging system and Hong Kong Liver Cancer
(HKLC) staging system; less commonly used are the Okuda staging system, Japanese integrated
score, and Italian Liver Cancer staging system.[15] The BCLC and HKLC systems both take into account the tumor burden, PS, and liver
function as measured by the Child–Pugh (CP) score A–C.
The AASLD and EASL recommend staging HCC patients using the BCLC system, which is
composed of stages 0–D.[8]
[9] Very early and early stages 0–A include patients with low tumor burden (≤3 tumors,
<3 cm) and preserved liver function (CP A) and PS (0), with a corresponding life expectancy
of >5 years. The intermediate stage includes multifocal disease, maintained liver
function (CP A–B7), and preserved PS (0), with a life expectancy of approximately
2.5 years. Advanced disease includes symptomatic, invasive, or metastatic disease
with preserved liver function but worsening PS (1–2); life expectancy is greater than
10 months. End-stage disease includes any tumor volume with poor liver function (CP
C) or PS >2; life expectancy is only 3 months.[9] The widespread adoption of the BCLC system is due to its simplicity and reproducibility
with respect to prognosis for each stage. However, it has been critiqued as being
overly conservative particularly in the intermediate and advanced stage settings.
To better predict prognosis and capture patients who would benefit from locoregional
treatments and other more aggressive therapies, the HKLC system was developed. The
HKLC system consists of nine stages and offers more patient outcome stratification
at the cost of increased complexity.[9] The BCLC and HKLC methods can be integrated into conceptual early, intermediate/advanced,
and end-stage disease ([Fig. 3]; patient examples in [Fig. 4]).
Fig. 4 Two representative patients presenting with HCC. Patient 1 has a 20.6-mm arterial
phase hyperenhancing (a) lesion with washout (b) on MRI. Given their CP A and PS 0 status, they were treated with percutaneous thermal
ablation. MRI 2 years later (c) demonstrates no residual arterial phase hyperenhancement with intrinsic T1-weighted
hyperintense consistent with blood products in a nonviable ablation cavity (circle).
Patient 2 has a 37-mm arterial hyperenhancing lesion with washout (white arrow) on
delayed phases on contrast-enhanced CT (d, e). Given the CP B score, they were treated with TACE. (f) Representative digitally subtracted angiography of the right hepatic artery during
the TACE procedure demonstrating tumor blush with a component that is tubular (arrowhead).
(g) Non–contrast-enhanced CT after TACE demonstrating dense lipiodol uptake in the right
posterior portal vein consistent with tumor invasion, which was not noted prior to
the procedure. Given the progression to advanced disease, the patient was transitioned
to systemic therapy. CP, Child–Pugh; CT, computed tomography; HCC, hepatocellular
carcinoma; MRI, magnetic resonance imaging; PS, performance status; TACE, transarterial
chemoembolization.
A common weakness of both the BCLC and HKLC systems is their dependence on the CP
system for cirrhosis, which relies on the subjective assessments of encephalopathy
and ascites. To develop a more objective measure of liver function, the ALBI score
was developed based on serum albumin and bilirubin levels. The ALBI score has been
validated in patients after liver resection, TARE, and treatment with sorafenib.[16]
[17]
[18] Currently, the AASLD/EASL makes no recommendation for the usage of the ALBI score,
but ongoing studies are aimed at improving staging with the use of objective criteria.
Hepatocellular Carcinoma Treatment by Stage
Hepatocellular Carcinoma Treatment by Stage
If a patient has early-stage HCC (BCLC 0/A, HKLC I/II), a curative approach may be
pursued through transplantation, resection, or ablation. Liver transplantation remains
the gold standard curative approach, but the scarcity of transplantable organs and
stringent pretransplant health and social requirements make it an untenable option
for most patients. In the nontransplant setting, both the AASLD and EASL recommend
resection and thermal ablation (RFA, MWA, cryoablation) as first-line therapies for
the treatment of early-stage disease, with the goal of a 75% 5-year survival.[8]
[9] RCTs have demonstrated an advantage of thermal ablation over percutaneous ethanol
ablation, particularly for lesions greater than 2 cm.[19]
[20] A recent meta-analysis showed similar outcomes between resection and thermal ablation,
with the latter being less expensive and less morbidity.[21]
[22] When thermal ablation is not feasible due to potential thermal damage to surrounding
critical structures (biliary ducts, gallbladder, bowel, etc.), irreversible electroporation,
a nonthermal ablative technique, has been shown to be effective and safe.[23] TARE has recently been proposed as a potentially curative option due to retrospective
studies showing median overall survival of approximately 6.7 years and a 5-year survival
of 75%.[24] Additional prospective trials are needed to validate these findings. Patients with
early-stage disease usually spend longer than 6 months on the transplant waiting list;
therefore, it is recommended to perform locoregional therapies for listed patients
as a bridge to transplantation.[8] Although surgery and ablation represent potentially curative treatments for early-stage
disease, most patients present with intermediate-stage disease, for which locoregional
therapies represent the best options for disease control.
Locoregional interventions for intermediate-stage disease (BCLC B/C, HKLC IIb/IIIa/IIIb/IVa/IVb)
are aimed at prolonging survival. Currently, treated intermediate and advanced stage
patients have an estimated survival of up to 26 and 10 months, respectively.[9]
[25] For patients with BCLC stage B disease, RCTs and meta-analyses have demonstrated
a survival benefit of locoregional therapies, making them the current recommendation
of AASLD/EASL.[8]
[9]
[25]
[26] Catheter-based locoregional therapies include TAE with bland embolic particles;
conventional TACE (cTACE) with lipiodol mixed with chemotherapeutic agents, plus or
minus an embolic material; drug-eluting bead TACE (DEB-TACE), where particles are
conjugated to a chemotherapeutic agent; or TARE with Yttrium-90 eluting microparticles.
RCTs and meta-analyses debate the efficacy of cTACE versus TAE versus DEB-TACE.[27]
[28] The largest RCT comparing cTACE to DEB-TACE, the PRECISION V trial, did not demonstrate
a difference between the two treatments, although subgroup analysis showed potentially
increased response rates for DEB-TACE with decreased liver toxicity in CP B, ECOG
1, and patients with bilobar disease or recurrent disease.[29] A RCT comparing DEB-TACE to TAE showed similar outcomes, suggesting that the main
mechanism for cellular death is ischemia rather than by the chemotherapeutic agent.[27] This finding is further supported by noting the outcomes from cTACE are independent
of the type or number of chemotherapeutic agents used, which suggests that these agents
confer little to no additional tumoricidal benefit.[30] Given the conflicting evidence surrounding which version of TACE is superior, the
choice usually depends on operator and institutional preference. However, if future
percutaneous ablation or SBRT is planned, cTACE or DEB-TACE plus lipiodol can be done
to render the tumor radiopaque for use as a target or fiducial. The addition of lipiodol
can be particularly useful for tumors between 3 and 7 cm, where the combination of
TACE and MWA has been shown to confer additional disease control rates and overall
survival.[31]
[32]
[33]
While not currently recommended by the societies for BCLC B patients, a meta-analysis
of studies comparing TACE to TARE demonstrated similar 1-year overall survival (OS)
but improved OS at 2 and 3 years with TARE.[34] Additionally, a phase II RCT study evaluating cTACE versus TARE in early to intermediate
stage patients showed a significantly increased time to progression, with a difference
of at least 22 months, suggesting a potential benefit for TARE in this patient population,
although overall survival was similar between the groups.[35] More data are needed to validate these findings before TARE can be recommended over
TACE for this population. Another limitation for TARE is the need for a total bilirubin
less than 2 mg/dL when performing lobar treatment where TACE can usually be done safely
up to less than 3 mg/dL. In select intermediate-stage patients, locoregional therapies
can also be considered to downstage patients back into transplantation criteria. In
a prospective case series involving 118 patients, 77 were able to be downstaged to
transplantation criteria and 54% were eventually transplanted.[36]
For advanced (BCLC C) disease, recent trials have demonstrated modest increased survival
with the use of systemic agents, including tyrosine kinase inhibitors (sorafenib,
lenvatinib, regorafenib).[37] New research shows modest responses in HCC patients with check point inhibitors
targeting PD-1/PD-L1 receptors, such as pembrolizumab, nivolumab, cemiplimab, and
atezolizumab. The recent RCT of atezolizumab and bevacizumab versus sorafenib showed
improved 1-year survival and overall response rate (67.2 vs. 54.6% and 27 vs. 12%,
respectively).[38] Two RCTs evaluated TACE with the addition of systemic agents (brivanib or sorafenib)
for intermediate-stage disease and found no additional OS benefit with combination
therapy.[39]
[40]
Locoregional therapy is not routinely recommended for patients with advanced-stage
disease. These patients may have portal vein thrombosis or other tumoral vascular
invasion which may render embolizing procedures like TACE, TAE, and DEB-TACE more
dangerous, given the greater risk of infarction. Nonetheless, retrospective studies
combing TACE with systemic therapies in advanced-stage disease show promising results
in select patients, with longer time to progression and overall survival.[41]
[42]
[43]
[44]
[45] The SARAH and SIRveNIB trials studied the role of TARE and did not demonstrate an
overall survival difference between TARE and sorafenib for locally advanced disease,
although quality of life and tumor response rates were improved with TARE.[46]
[47] Ongoing trials are evaluating the combination of TARE and sorafenib for HCC (e.g.,
STOP-HCC). A recent RCT assessing boosted doses of TARE with ablative intent (>205
Gy) in intermediate- and advanced-staged patients with tumors greater than 10 cm,
many with portal vein thrombosis, has also shown promising overall survival improvements
of approximately 20 months.[48]
Currently, there is no role for surgery (other than transplantation) or minimally
invasive, image-guided procedures for the treatment of end-stage HCC (BCLC D, HKLC
V). Management is mostly directed at symptom control rather than prolonging life.
Interventional radiologists offer supportive procedures such as paracentesis, peritoneal
tunneled drainage catheter placement, and port placement to assist with easier venous
access for medication administration.
Given the different staging systems and variety of treatment options alone and in
combination, HCC treatment recommendations are best made by a multidisciplinary tumor
board. This team typically includes members from interventional radiology, hepatology,
transplant surgery, medical oncology, radiation oncology, and diagnostic radiology.
This team approach facilitates consideration of the full range of options and limitations
within the context of a patient's overall treatment goals. Before and after the tumor
board makes a recommendation, the patient should be seen in an outpatient consultation.
This interaction offers an opportunity to educate the patient regarding their options
and to exercise their autonomy in eliciting their preferences. This visit may exclude
some treatment options due to patient preference, lack of availability close to the
patient, or cost/lack of insurance coverage. Some systemic agents, SBRT, and TARE
can be cost prohibitive for some patients. In these cases, a different treatment modality
(e.g., TACE) may be favored. The clinic visit also serves to open a line of dialogue
with the patients and their family about realistic expectations of potential outcomes.
Moreover, it paves the way for future end-of-life discussions if the patient becomes
ineligible for additional treatment.
Treatment Evaluation
Follow-up with imaging is critical to assess the treatment efficacy and determine
when further treatment is indicated. Follow-up is usually obtained 1 to 3 months after
the first treatment and then every 3 months after locoregional therapies or up to
every 6 months after successful ablation or resection. LI-RADS defines criteria for
the assessment of residual/recurrent disease. Viable disease is described by nodular,
mass-like, or thick irregular tissue in or along the treated lesion with arterial
phase hyperenhancement, washout, or a similar enhancement pattern prior to treatment.[49] The degree of tumor response is measured with the AASLD-recommended modified Response
Evaluation Criteria in Solid Tumors (mRECIST) or the EASL classification.[8]
[9] For recurrent or residual disease, the patient should be presented again at the
multidisciplinary conference for determination for the next best treatment option
(e.g., repeat same intervention after TACE, switch modalities to TARE or SBRT, or
continue surveillance). Practice patterns vary widely with regard to the number of
times a modality will be attempted before it is deemed to be ineffectual or insufficient
for a certain patient, and the data do not support one approach over another.
Conclusions
The successful management of patients with HCC depends on accurate and early diagnosis,
the use of staging systems that predict both prognosis and the appropriate treatment,
and multidisciplinary discussions that incorporate patient preference. Locoregional
therapies remain the mainstay for intermediate-stage disease, both at diagnosis and
for recurrent disease.
