Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
American Association for the Study of Liver Diseases
alpha-fetoprotein
Asian Pacific Association for Studies of the Liver
computed tomography
clathrin heavy chain
dysplastic foci
dysplastic nodule
European Association for the Study of the Liver
early hepatocellular carcinoma
early, well-differentiated hepatocellular carcinoma
enhancer of zeste homolog 2
hemotoxylin and eosin
hepatitis B virus
hepatocellular carcinoma
high-grade dysplastic nodule
heat shock protein 70
Japan Society of Hepatology
a keratin
a Although the prefix CK is widely used in surgical pathology to designate human cytokeratins, consensus nomenclature recommends the replacement of “cytokeratin” with “keratin” and the prefix “CK” with “K.” (Schweizer J, Bowden PE, Coulombe PA, et al. New consensus nomenclature for mammalian keratins. J Cell Biol 2006;174:169–174.)
liver imaging reporting and data system
large cell change
low-grade dysplastic nodule
large regenerative nodule
magnetic resonance imaging
progressed hepatocellular carcinoma
proliferating cell nuclear antigen
organic anion transporting polypeptide 8
small cell change
telomerase reverse transcriptase
Most patients with compensated cirrhosis with “de novo” hepatocellular carcinoma (HCC) at an early stage (single nodule <5 cm, or 3 nodules each ≤3 cm) may expect 5-year survival rates exceeding 50%. This significant improvement in prognosis of HCC over recent decades is a reflection of three major factors:
Protocols for regular screening of cirrhotic patients to detect early HCC (eHCC)
Availability of successful modalities of therapy such as orthotopic liver transplantation, resection, and percutaneous ablation for HCC
Internationally recognized criteria for accurate selection of patients based on precise definitions of number and size of HCC nodules
Regular ultrasound surveillance of patients with cirrhosis leads to frequent detection of small, 0.5- to 3-cm lesions as hypoechoic or, less frequently, hyperechoic nodules arising in the cirrhotic parenchyma. The diagnostic approach to these small nodular lesions is a challenge. Difficulties arise from the fact that 7% to 40% of these small lesions are represented by nonmalignant or premalignant nodules ( Table 31.1 ). This wide range of nonmalignant hepatocellular nodules distribution may depend on the different selection criteria used in the different studies being mostly ultrasonographic earlier, and based on dynamic imaging later. Moreover, until 2008, standardized histologic criteria were lacking, in particular those outlining the features of eHCC.
Reference | No. of Lesions | HCC | NM-HN | Other |
---|---|---|---|---|
Kondo F et al (1990) | 161 | 132 (82%) | 28 (17.5%) | 1 (0.5%) |
Borzio M et al (2003) | 317 | 184 (58%) | 125 (39.5%) | 8 (2.5%) |
Caturelli E et al (2004) | 294 | 258 (87%) | 31 (11%) | 5 (2%) |
Bolondi L et al (2005) | 72 | 60 (83%) | 12 (17%) | 0 |
Forner A et al (2008) | 89 | 60 (67%) | 24 (27%) | 5 (6%) |
Leoni S et al (2010) | 75 | 55 (73%) | 20 (27%) | 0 |
Sangiovanni A et al (2010) | 67 | 44 (67%) | 21 (31%) | 2 (2%) |
Manini MA et al (2014) | 119 | 84 (71%) | 32 (27%) | 3 (2%) |
The size of a radiologically detected nodule is highly predictive of malignancy ( Table 31.2 ). The great majority of lesions less than 1 cm are nonmalignant, whereas lesions exceeding 2 cm are mostly HCC; this association is so strong that the diagnosis of nonmalignancy in a nodule greater than 2 cm should raise the possibility of a diagnostic error.
Reference | Total | NM-HN/HCC | NM-HN/HCC | ||
---|---|---|---|---|---|
<10 mm | 10–20 mm | >20 mm | |||
Borzio M et al (2003) | 317 | 89/228 | 12/1 | 58/29 | 19/198 |
Bolondi L et al (2005) | 72 | 12/60 | — | 12/29 | 0/31 |
Forner A et al (2008) | 89 | 29/60 | 11/2 | 18/58 | — |
Leoni S et al (2010) | 75 | 20/55 | — | 16/27 | 4/28 |
Sangiovanni A et al (2010) | 65 | 21/44 | 0/2 | 20/34 | 1/8 |
Manini MA et al (2014) | 116 | 32/84 | 6/1 | 23/43 | 3/40 |
It is well established that, in cirrhosis, progression from nonmalignant nodules to dysplastic nodules (DNs) and on to early and progressed HCC is characterized by progressive loss of portal venous blood and development of “de novo” arterial vascularization (neoangiogenesis). This arterial vascularization produces a characteristic radiologic pattern, known as “wash in” during arterial phases and “wash out” during early-late portal phases, respectively, on contrast-enhanced imaging techniques. This pattern is distinctive enough to establish in lesions of any size, a firm diagnosis of malignancy on radiology alone and serves as the basis of the “noninvasive” criteria for diagnosis of HCC, according to practice guidelines. A biopsy is required in all cases, irrespective of size, when imaging features are inconclusive or atypical.
This chapter discusses the pathobiology, natural history, and diagnosis of premalignant nodules and early HCC, the study and accurate diagnosis of which have great clinical relevance ( Box 31.1 ). In addition, the wide evidence linking DNs to HCC provides the potential to offer insights into the process of hepatocarcinogenesis; knowledge that will be invaluable for the diagnosis, prognosis, and treatment of HCC.
To select patients harboring low-grade dysplastic nodules for optimal follow-up
To select patients at higher risk of malignant transformation (those with high-grade dysplastic nodule) for optimal management
To direct patients with malignant nodules to appropriate therapy: surgical resection/waiting list for transplantation/ablative therapy according to international guidelines
To decrease the likelihood of prematurely transplanting patients who do not have hepatocellular carcinoma and, conversely, to avoid the exclusion of patients with multiple nodules, some of which may be nonmalignant
To accurately stage patients with malignant nodules, particularly those amenable to hepatic resection
To offer individualized prognosis based on an interdisciplinary approach (histology, radiology, clinical parameters) and not on the false belief that every nodule in cirrhosis is (or is likely to become) malignant, thus avoiding overtreatment
To obtain accurate correlative data for molecular/genetic expression studies to characterize individual lesions for individual therapy
As mentioned previously, premalignant and early malignant hepatocellular lesions are most commonly detected in cirrhosis. Consequently, eastern and western scientific societies identify cirrhotic patients, regardless of etiology, as the ideal target population for surveillance programs. The Asian Pacific Association for Studies of the Liver (APASL), the Japan Society of Hepatology (JSH), and the European Association for the Study of the Liver (EASL) extend their recommendation to noncirrhotic patients with chronic hepatitis C and advanced fibrosis (F3, according to METAVIR classification). It is agreed that patients with the hepatitis B virus (HBV) under antiviral treatment should be monitored whereas surveillance of patients with hepatitis C virus (HCV) achieving a sustained complete response is questioned. The American Association for the Study of Liver Diseases (AASLD) guidelines for the management of HCC stated that surveillance is deemed cost effective if the expected HCC risk exceeds 1.5% per year in patients with hepatitis C and 0.2% per year in patients with hepatitis B. Under surveillance several hepatic nodules other than progressed HCC are detected in the target population. All these nodules need characterization and follow up because several of them have the features of premalignant lesions and the question of their clinical management is raised.
The vast majority of HCC (up to 90%) develops in the background of chronic hepatitis/cirrhosis following a multistep sequence, which includes a number of lesions considered premalignant. The currently accepted nomenclature for these lesions was proposed in 1995, by an international consensus, namely the International Working Party, and further updated in East-West consensus meeting in 2008. Premalignant lesions include dysplastic foci (DF) and DNs, and small cancerous lesions (up to 2 cm) can be both early (eHCC) and progressed (pHCC) HCC. DF are microscopic abnormalities (<1 mm) consisting of an expansile group of hepatocytes with cytologic changes suggestive of premalignancy (in particular small cell changes and iron-free foci). A DN (from a few millimeters up to about 2 cm) possibly originates from a DF and it shows regenerative features and/or dysplasia of low (low-grade DN [LGDN]) or high (high-grade DN [HGDN]) degree. DNs are usually macroscopically recognizable, and those with low-grade dysplasia can evolve into those with high-grade dysplasia. In practice, hepatocellular dysplastic lesions can be distinguished into those detectable only as microscopic foci and those forming grossly recognizable nodules in cirrhosis (see Fig. 31.1 , eSlide 31.1, eSlide 31.2 , and Box 31.2 ).
Cluster of hepatocytes, <1 mm in size, characterized by presence of either small cell change or large cell change
A distinctly nodular lesion that differs from the surrounding liver parenchyma with regard to size, color, and texture and that bulges from the surrounding liver on cut surface (see Fig. 31.1A ). It can be detected but not accurately diagnosed at imaging. Histologically it is distinguished into two categories:
Low-grade: lesions with features suggestive of a regenerative growth, with or without clonal cell populations, lacking major architectural alterations (see Fig. 31.1B–D ) and with an overall nonmalignant morphology. The old category of regenerative nodules has been incorporated into this group of lesions.
High-grade: lesions showing a certain degree of cytologic atypia and architectural alterations but which are insufficient for a diagnosis of malignancy (see Fig. 31.1E–G; also see eSlide 31.1 ).
These are small foci, less than 1 mm in size, of dysplastic hepatocytes recognized within the liver parenchyma in chronic liver disease, particularly at advanced fibrotic stages (cirrhosis). Dysplastic foci are named according to the main cell type as large cell foci , small cell foci , or iron-free foci ; the latter occur in hemochromatosis or any condition with significant parenchymal iron overload. In most cases, DF are recognized incidentally in a liver biopsy; dysplastic hepatocytes showing large cell changes (LCC) or small cell changes (SCC) are also part of the spectrum of cytologic abnormalities occurring in DNs. Several studies conducted have been conducted in the past to address the precancerous role played by these foci in the human liver, particularly as far as LCC and SCC, previously called large cell dysplasia and small cell dysplasia , respectively, are concerned. An extensive and detailed analysis of these lesions is beyond the scope of the present chapter; however, the main features are summarized in Box 31.3 and illustrated in Fig. 31.2 . From a practical point of view, in routine clinical practice, look for and mention dysplastic foci in the pathology report when they are present.
Abnormal hepatocytes, increased proliferation/apoptosis rate ; significant telomere shortening and loss of cell cycle checkpoint
Common detection of morphophenotypic features of liver progenitor cells such as K19 21
Rare (5%) in cirrhotics under surveillance; frequent in high-grade dysplastic nodules and noncancerous liver adjacent to hepatocellular carcinoma (HCC)
Less robust predictor of HCC development
Accepted microscopic precursor of HCC
Abnormal hepatocytes, possibly senescent or slowly proliferating ; slight telomere shortening, altered cell cycle checkpoint control
Morphophenotypic features of more differentiated hepatocytes
Frequent (20% to 30%) in cirrhotics under surveillance, dysplastic nodules (low- and high-grade), and noncancerous liver adjacent to HCC
Robust predictor of HCC development (ie, liver more susceptible to malignant transformation)
Heterogeneous lesion likely including two populations: one reactive and terminally differentiated, the other as a very early precursor of HCC
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here