Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Primary malignant tumours of the liver
Introduction
Primary liver malignancies comprise a group of tumours, some very rare, that have traditionally been divided into three main categories based on the cell type of origin—hepatocellular (hepatocellular carcinoma [HCC]), biliary (intrahepatic cholangiocarcinoma [ICC]), and mesenchymal (angiosarcoma, epithelioid haemangioendothelioma). Yet, as our understanding of the molecular genetics and tumorigenesis of primary liver malignancy evolves, the once clear distinction between and uniformity within categories continue to blur. Subtypes, such as fibrolamellar HCC, demonstrate morphomolecular and clinical features distinct from typical HCC, and mixed hepatocellular-cholangiocellular carcinoma represents the merging of two traditionally different tumour types from different cell lineages. , As such, primary malignant tumours of the liver often exist as part of a continuum, with some falling somewhere within a spectrum of cellular and genomic differentiation.
Understanding the underlying pathogenesis of primary malignant liver tumours is imperative for appropriate diagnosis, and has implications for management, treatment response and prognosis.
Hepatocellular carcinoma
HCC is the most common primary liver tumour, accounting for 75–85% of primary liver malignancies. It is the sixth most commonly diagnosed cancer overall and in 2020, was third leading cause of cancer mortality, with an estimated 830,000 deaths reported worldwide. The rates of HCC incidence and mortality are two to three times higher among men than women, and vary widely between geographical regions. In many countries in Eastern Asia, South-Eastern Asia, and Northern and Western Africa, HCC is the most commonly diagnosed malignancy and leading cause of cancer-related deaths. ,
Less common risk factors include α 1 -antitrypsin deficiency, hereditary hemochromatosis, primary biliary cirrhosis and autoimmune hepatitis. Due in large part to widespread public health efforts to improve vaccination against HBV, limit hepatitis B and C viral transmission and reduce aflatoxin exposure, the incidence and mortality rates of HCC have mostly decreased over the last several decades, particularly in high-risk countries of Eastern and South-Eastern Asia, and Eastern Sub-Saharan Africa. , In patients with chronic viral hepatitis who have already developed cirrhosis, successful antiviral therapy for both HBV and HCV has been demonstrated to significantly reduce, although not entirely eliminate, the risk of developing HCC across both low- and high-risk regions. As such, antiviral treatment for patients with chronic infections from HBV and HCV should follow the recommendations of existing European Association for the Study of the Liver (EASL) guidelines. , In other areas, the major risk factors for HCC appear to be in transition. For example, in low-risk regions, such as Europe, North America and South America, there has been an increase in the incidence of HCC in recent years, which may reflect increases in non-alcoholic fatty liver disease (NAFLD) related to the increased prevalence of obesity, diabetes and metabolic syndrome in these populations. , The effect of risk modification on the development of HCC in patients with NAFLD is unclear.
The main risk factors for the development of HCC include chronic infection from HBV and HCV, exposure to foods contaminated with aflatoxins, heavy alcohol consumption, obesity, type 2 diabetes and smoking.
Shifting patterns in the incidence of HCC worldwide reflect geographic differences in risk factors, vaccination and screening protocols. While the incidence of HCC is decreasing in many high-risk countries, it is increasing in previously low-risk regions, such as Europe, North America and South America.
Treatment with antiviral therapies have demonstrated efficacy in preventing progression to cirrhosis and decreasing the incidence of HCC and are recommended in all patients with chronic hepatitis due to HBV and HCV.
Pathogenesis of HCC
The association between HCC and cirrhosis is well established—while only an estimated 1–3% of patients with cirrhosis develop HCC annually, approximately 90% of patients with HCC have underlying cirrhosis. Repeated rounds of cellular injury followed by regeneration in the setting of chronic inflammation contributes to the development of liver fibrosis, cirrhosis and eventually HCC in a multi-step process estimated to take approximately 20 to 40 years. These sustained cycles of hepatocyte injury and repair, compounded by the host inflammatory response to viral infection, increase the risk of genomic alterations and contribute to oxidative stress, DNA damage and chromosome instability. The role of the host immune system in the development of cirrhosis and eventual HCC in patients with NAFLD is less well understood, but is thought to be mediated by metabolic pathways, insulin resistance and lipotoxicity.
The accumulation of genetic aberrations within cirrhotic nodules from repeated injury and repair promotes progressive atypia and eventual malignant transformation to HCC.
Screening for HCC
Because HCC develops over decades, often silently, screening and surveillance in high-risk individuals is one of the most important components in a comprehensive HCC program. The primary goal of screening is to detect HCC in early stages when effective treatment and long-term cure are still possible. There is currently only a single, randomised controlled trial (RCT) examining the impact of screening compared with no screening on survival. Performed in China, 18,816 patients with HBV were randomly assigned to either a screening program, which consisted of serum α-fetoprotein levels (AFP) and liver ultrasound (US) every 6 months, or no screening. Despite only a 68% adherence, their results still demonstrated a 37% reduction in HCC-related mortality with biannual screening protocol. Additional population cohort studies and cost-effectiveness analyses have largely reinforced the benefits of biannual screening in high-risk patients, leading to this practice being universally accepted. How high-risk is defined and what the true cost-effectiveness of screening is in certain patient populations, such as those with advanced liver disease who are not candidates for transplantation, high-risk non-cirrhotics or those with NAFLD, is less clear.
Several tests may be used for surveillance, including both imaging and serological studies. The most widely used exam for HCC surveillance is liver US, as it is non-invasive, poses little risk to the patient, has a relatively moderate cost and can detect additional related conditions, such as ascites or portal vein thrombosis, which may need prompt treatment. A meta-analysis of 19 studies demonstrated that US surveillance detected the majority of HCC tumours before clinical presentation, with a pooled sensitivity of 94%. However, the sensitivity of US to detect early-stage HCC was only 63%. The use of contrast-enhanced US has been proposed in an effort to improve the detection of small, early-stage HCC tumours; however, this has not been proven and its use has not been widely adopted. While multidetector computed tomography (CT) and dynamic magnetic resonance imaging (MRI) can be valuable studies in circumstances where US is less effective, such as in fatty liver disease, their associated high-cost, radiation exposure, need for contrast agents and considerably high false-positive rates have limited their role in long-term surveillance schemes. Several serological tests have been investigated to aid early diagnosis of HCC, including AFP, prothrombin induced by vitamin K absence II (PIVKA II), the ratio of glycosylated AFP (L3 fraction) to total AFP, α-fucosidase, and glypican. Although AFP is the most widely used biomarker for HCC, it has largely been a tool of diagnosis rather than surveillance. Indeed, studies examining the utility of AFP for surveillance have been disappointing, and current guidelines recommend against its routine use in HCC screening. Other serum markers have primarily been studied as diagnostic or prognostic aids, and insufficient evidence exists to recommend their use for screening purposes.
Screening for HCC should be performed by experienced personnel in all high-risk individuals using abdominal US every 6 months.
Diagnosis of HCC
Non-invasive imaging forms the cornerstone of diagnosis for HCC in the cirrhotic liver. Image-based diagnosis relies on a high pre-test probability in the setting of cirrhosis, combined with the presence of hallmark radiographic findings that are typical of HCC—arterial phase hyperenhancement and non-peripheral wash-out on portal venous and/or delayed phases ( Fig. 6.1a –d). Adopted in 2011 by the American College of Radiology, the Liver Imaging Reporting and Data System (LI-RADS) provides a set of standardised terminology and establishes a framework for the classification of liver lesions in the setting of cirrhosis, ranging from LR-1 (definitely benign) to LR-5 (definitely HCC) ( Table 6.1 ). The accuracy of the radiographic diagnosis of HCC relies on two major factors—tumour size and imaging modality. Because small tumours are more challenging to diagnose, the EASL and European Organisation for Research and Treatment of Cancer (EORTC) developed a sequential algorithm for the work-up of tumours < 10 mm and 10–20 mm in size ( Fig. 6.2 ). ,
Table 6.1
American College of Radiology Liver Reporting & Data System (LI-RADS) for CT and MRI imaging for hepatocellular carcinoma (HCC)
| LI-RADS | Description | Management |
|---|---|---|
| Negative | No observations detected | Return to surveillance in 6 months |
| LR-NC | Not categorizable due to image degradation or omission | Repeat or alternative imaging in < 3 months |
| LR-1 | Definitely benign observation | Return to surveillance in 6 months |
| LR-2 | Probably benign | Consider repeat diagnostic imaging in 6 months |
| LR-3 | Intermediate probability of malignancy | Repeat or alternative imaging in 3–6 months |
| LR-4 | Probably HCC | Multidisciplinary discussion for further work-up |
| LR-5 | Definitely HCC | Multidisciplinary discussion for management consensus |
| LR-M | Probably/definite malignancy not HCC specific | Multidisciplinary discussion. Consider biopsy |
| LR-TIV | Definite tumour in vein | Multidisciplinary discussion. May include biopsy |
While several imaging modalities may be used to diagnose HCC, multiphasic contrast-enhanced CT and MRI are the most widely used and studied. In general, MRI appears to have a higher sensitivity than CT, although this depends on tumour size. , A recent prospective, multicentre study of 544 nodules in 381 patients demonstrated an increased sensitivity and specificity with contrast-enhanced MRI (72.3% and 89.4%, respectively) compared with CT (67.9% and 76.8%, respectively) for lesions 10–20 mm in size; for lesions 20–30mm in size, the sensitivity and specificity of contrast-enhanced MRI and CT were comparable (70.6% and 83.2% vs 71.6% and 93.6%, respectively). Importantly, this study also found that the combination of CT and MRI resulted in a specificity of 100%, but a sensitivity of only 55%, supporting the recommendation against routine dual imaging for small lesions.
The use of contrast-enhanced ultrasound (CEUS) for the diagnosis of HCC is controversial. Although refinements in the definition of the typical hallmark features of HCC on CEUS have led to improved diagnostic capacity, its sensitivity remains significantly lower than CT or MRI, missing roughly 13% of HCCs seen on cross-sectional imaging. Furthermore, because CEUS does not provide a panoramic view of the liver, does not carry the added benefit of staging and is not easily reviewed across imaging software platforms, it is not recommended as a first-line imaging technique for the diagnosis of HCC. Still, it may be beneficial when used to characterise one or a few nodules seen on conventional US, or when both CT and MRI are either contraindicated or inconclusive.
The diagnosis of HCC in cirrhotic patients with tumours larger than 1 cm is largely based on non-invasive criteria using four-phase multidetector CT or dynamic contrast-enhanced MRI. The typical hallmark features of HCC include a hypervascular tumour in the arterial phase with washout in the portal venous or delayed phases.
In situations where CT or MRI are inconclusive, imaging suggests a malignancy other than HCC, or in non-cirrhotic patients, pathologic confirmation should be obtained.
HCC staging systems
Many staging systems for HCC exist, each utilising different combinations of clinicopathologic factors and displaying varying degrees of applicability, tumour-specific focus and prognostic value. The American Joint Committee on Cancer (AJCC) TNM classification is the standard staging system for cancer worldwide. In HCC, the TNM classification focuses on four major clinicopathologic factors—tumour size, multifocality, vascular invasion and extrahepatic spread. Although it is widely referenced for HCC, TNM system relies on pathologic characteristics of resected specimens only, and fails to incorporate underlying liver function and patient-specific factors. The Barcelona Clinic Liver Cancer (BCLC) staging system is clinically-based, applicable to all stages of disease and has been externally validated in different clinical settings. , Based on tumour stage, liver function and performance status, it stratifies patients with HCC into four categories—early, intermediate, advanced and terminal—and makes treatment recommendations for each ( Fig. 6.3 ). ,
Due to the unique nature of the disease, any clinical staging system for HCC should include tumour-specific factors, underlying liver function, and patient performance status.
The BCLC staging system is the most widely used staging system for HCC and is the recommended system for prognosis and treatment allocation in Europe.
Surgery
Surgery is considered first-line therapy for HCC and is the only potentially curative treatment option, with 5-year survival ranging from 60–80% in well-selected patients. The choice of resection or transplantation is generally dictated by the degree of underlying liver dysfunction and the burden of disease.
Liver resection
Liver resection for HCC is unique, in that it requires consideration of not just technical and biologic resectability, but also underlying liver dysfunction. In liver surgery, the technical resectability of a tumour is defined not only by what can be removed, but more importantly, by what will be left behind. Hepatic artery and portal vein inflow, hepatic vein outflow, biliary drainage, and the quantity and quality of the future liver remnant (FLR) must be adequate in order to maintain hepatic function and allow for regeneration following resection. Inflow, outflow and biliary drainage can be determined using high-quality cross-sectional imaging.
The Child-Pugh score is the most commonly used measurement of liver quality and function. Liver resection is still considered safe and appropriate in patients with a Child-Pugh A score without evidence of portal hypertension. Other measurements that may be helpful in improving patient selection for resection include the model for end-stage liver disease (MELD), MELD-Na, indocyanine green kinetics, liver stiffness measurement and cholinesterase/bilirubin ratio. The assessment of liver quantity can be performed by evaluating the expected postoperative FLR as a ratio of the total liver volume (minus the tumour volume). Ratios of at least 25% are considered safe in healthy, non-cirrhotic livers. An FLR of at least 30–40% is recommended in patients with chronic liver disease and a Child-Pugh A score. In patients whose FLR may be inadequate, but who are otherwise candidates for liver resection, pre-operative portal vein embolisation (PVE) should be considered.
The biologic principles of resection for HCC largely focus on four factors—tumour size, multifocality, vascular invasion and extrahepatic spread. As previously mentioned, these factors form the basis of the AJCC TNM staging system, and are incorporated into various other clinical staging systems, such as the BCLC. , , However, unlike transplantation for HCC, specific oncologic criteria are not as well defined for resection and in practice, the determination of resectability from a tumour biology standpoint can vary widely. For example, resection may still be considered feasible in select patients with large tumours (size > 5 cm) or limited multifocal disease. The benefit of resection in patients with major vascular invasion, however, is questionable and should be pursued only in highly selected patients. With few exceptions, the presence of extrahepatic disease, including regional lymph node metastasis, is a contraindication to resection in patients with HCC.
With regards to underlying liver dysfunction, liver resection is the treatment of choice in non-cirrhotic patients with HCC. Challenges arise in patients with HCC in the setting NAFLD and metabolic syndrome, but without significant fibrosis or cirrhosis, as postoperative morbidity following resection is increased compared to those with normal livers. Still, postoperative mortality remains less than 2% and patients demonstrate prolonged long-term survival compared to patients with cirrhosis undergoing resection. , In patients with cirrhosis, the decision to proceed with liver resection should rely on a combination of tumour-related factors, the degree of underlying liver dysfunction, and the extent of resection required to achieve complete tumour extirpation. In 2012, the EASL and EORTC developed a set of guidelines for resection of HCC in patients with cirrhosis—solitary tumours and well-preserved liver function, platelet count > 100,000/mL or the absence of portal hypertension.
Technical resectability, tumour biology, underlying hepatic dysfunction and patient comorbidities must all be considered prior to resection in patients with HCC.
Resection is the treatment of choice for HCC in the non-cirrhotic liver.
Liver transplantation
HCC is the only malignancy for which solid organ transplantation is widely accepted. The Milan criteria (solitary tumour ≤ 5 cm or ≤ 3 tumours each ≤3 cm in size, without vascular invasion) is the most widely accepted and reliable tool for determining transplantability. , Patients within Milan criteria have an expected 5-year survival rate that ranges from 65% to 80%, and have a longer survival than patients who fall outside the criteria. In centres where the conventional Milan criteria is used, patients beyond Milan may still be offered transplantation if they can be successfully down-staged into the accepted criteria. In order to limit dropout and improve the success rate of downstaging strategies, restricted eligibility for downstaging protocols using tumour number, size and AFP is recommended. ,
Several expanded criteria have been developed and externally validated, and demonstrate similar post-transplant survival to the Milan criteria ( Table 6.2 ). While there is no consensus on the applicability of any single expanded criteria, surrogate markers of tumour biology, such as AFP, and response to neoadjuvant therapy are likely to replace conventional, morphology-based criteria to determine transplant eligibility. Regardless of the criteria used, major vascular invasion and extrahepatic disease are absolute contraindications to liver transplantation for patients with HCC. ,
Table 6.2
Summary of validated selection criteria for liver transplantation in patients with hepatocellular carcinoma
| System | Criteria | Survival a |
|---|---|---|
| Milan , |
|
4-yr OS: 85% |
| UCSF , |
|
5-yr OS: 75% |
| Up-to-7 |
|
5-yr OS: 71% |
| TTV + AFP |
|
4-yr OS: 75% |
| AFP-French model |
|
5-yr OS: 68% (low-risk score) 5-yr OS: 48% (high-risk score) |
| Toronto |
|
5-yr OS: 68% |
AFP, alpha-fetoprotein; OS, overall survival; TTV, total tumour volume; UCSF, University of California, San Francisco.
a All studies reported no statistically significant difference between patients within Milan and those outside Milan, but within the proposed criteria, among the respective study cohorts.
Due to challenges with organ availability and allocation, wait times can be significant, and patients with HCC who are candidates for liver transplantation are at risk of tumour progression while waiting. Neoadjuvant ‘bridging’ therapy using various combinations of ablation and transarterial techniques has been shown to reduce dropout due to progression of disease and has become nearly universally accepted. This is particularly true in regions where the expected wait time is greater than 6 months.
Liver transplantation is the recommended first-line treatment for patients who fall within Milan criteria and are not eligible for resection.
There is no consensus on the use of expanded criteria for transplantation in HCC.
Patients beyond Milan criteria may be considered candidates for liver transplantation after successful downstaging.
Major vascular invasion and the presence of extrahepatic disease are absolute contraindications to liver transplantation in HCC.
In patients with HCC who are transplant candidates, the use of locoregional and liver-directed therapies to bridge the patient while on the wait list is recommended.
Locoregional therapies
Locoregional therapy is the recommended treatment approach for patients with HCC who are not candidates for resection or transplantation.
Ablation
Tumour necrosis using ablative techniques can be achieved using either chemical ablation (percutaneous ethanol injection [PEI]), thermal ablation (radiofrequency [RFA] or microwave ablation [MWA]) or cryoablation. All ablative procedures may be done using a percutaneous, laparoscopic or open approach. In order to improve the efficacy of ablation, image guidance using US, CT, or cone-beam CT should be used. Caution should be used, however, when lesions are close to important structures as thermal spread can lead to injury to major bile ducts and surrounding organs.
PEI is one of the most common techniques for treating small, well-circumscribed HCC, and can achieve complete necrosis in 90% of tumours < 2 cm in size. However, PEI is less successful in tumours larger than 2 cm, may demonstrate inconsistent distribution within the tumour, and is associated with a high recurrence rate. Studies comparing PEI with RFA have consistently demonstrated improved overall survival (OS), disease-free survival and recurrence-free survival with RFA.
RFA uses high-frequency alternating currents to generate frictional heat, inducing cell death. The heat produced by RFA creates a coagulative necrosis, which can be applied beyond the boundaries of the tumour, treating a margin potentially containing microscopic tumour cells. Outcomes with RFA vary widely but are generally favourable for early HCC, with an estimated 3-year OS of 76% and RFS of 46% for tumours < 3 cm in size. Larger tumours may still be successfully treated with RFA by using multipolar approaches or by combining RFA with transarterial chemoembolisation (TACE). , Multiple studies have been performed comparing RFA with surgery for HCC. A 2017 Cochran review comparing RFA with surgery for small, solitary HCC found no difference in survival between the two. Surgery was associated with fewer HCC recurrences, while RFA was associated with fewer adverse events and shorter hospital length of stay. For tumours > 5 cm or multifocal tumours > 3 cm, surgery is associated with improved survival compared with RFA, and should be the treatment of choice in resectable patients.
MWA uses electromagnetic energy to heat tissue. One of the benefits of this technique is that it is less affected by nearby vessels, and thus, less prone to heat sink than RFA. Although data are still emerging, MWA has been shown in several studies to be at least equivalent to RFA in terms of survival, recurrence and adverse events, and may be considered an alternative to RFA in patients with small or unresectable HCC.
Irreversible electroporation (IRE) is an emerging ablative technique that delivers non-thermal, high-voltage electric pulses to tumour cells, resulting in increased cell membrane permeability and subsequent apoptosis. Similar to MWA, IRE is less likely to demonstrate heat sink and is safe to use near blood vessels, bile ducts and other critical structures. Still, high-quality data examining the long-term effectiveness of IRE for HCC are lacking.
RFA is the standard of care for patients with early-stage HCC tumours that are not otherwise suitable for surgical resection.
In patients with small (<3 cm), solitary tumours in favourable locations, RFA may be considered an alternative to surgery.
Ethanol injection is an option in tumours < 2 cm when thermal ablation is not technically feasible.
MWA is associated with acceptable local control and survival, comparable to RFA, but long-term data are still needed.
You're Reading a Preview
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here