The international quest for the imaging diagnosis of liver cancer

The systematic review by Cannella et al assessed adherence to a high-risk population criteria for the imaging diagnosis of HCC, specifically focusing on the Liver Imaging Reporting and Data System (LI-RADS) and the European Association for the Study of the Liver (EASL) definitions in published research studies.¹ The accuracy of diagnostic imaging tests, such as CT, MRI, and contrast-enhanced ultrasound, is affected by the pretest probability of HCC.² Thus, to minimize the probability of a false-positive diagnosis of lesions that mimic HCC on imaging studies, the imaging criteria for noninvasive diagnosis of HCC are only applicable to patients with a high pretest probability of having HCC. Therefore, adherence to high-risk population criteria is crucial for validating the diagnostic accuracy of imaging features in research and for improving the early and accurate diagnosis of HCC in clinical practice. This systematic review found that adherence to high-risk population criteria in the literature was optimal only in half of the LI-RADS studies and less than one-third of EASL studies. In addition, adherence to high-risk population criteria improved according to CT/MRI LI-RADS versions and publication year, but no significant differences were observed in the versions of contrast-enhanced ultrasound LI-RADS or EASL.

There are some major contributing factors to the heterogeneity in the adherence to high-risk population criteria for the imaging diagnosis of HCC among published studies. Research conducted in the Asian countries where chronic hepatitis C without cirrhosis is included in high-risk populations in their guidelines, such as the Asian Pacific Association for the Study of the Liver³ or the Korean Liver Cancer Association-National Cancer Center⁴ seems to have contributed to the clear violation of CT/MRI LI-RADS. Similarly, in recent studies of the EASL guidelines, the inclusion of patients with chronic HBV or HCV infection without cirrhosis was a major reason for inadequate adherence. Another factor that may contribute to the violation is the discrepancy between target populations for surveillance and those for HCC imaging diagnosis in the LI-RADS and EASL⁵ (Table 1). For example, according to the EASL guidelines,⁶ patients with non-cirrhotic chronic HBV with an intermediate or high risk of HCC according to the PAGE-B score, and patients with non-cirrhotic chronic HCV with bridging fibrosis (F3), are included as a high-risk population for HCC surveillance, but they cannot be diagnosed with HCC using the EASL’s imaging criteria. According to the LI-RADS, HCC surveillance is recommended for patients with vascular disorder-related cirrhosis, but the LI-RADS diagnostic algorithm should not be applied to these patients. Therefore, the international community should strive for more consistency in the high-risk population criteria. However, it is still a matter of debate as to which patients with chronic HBV or HCV infection should be included in the noninvasive diagnosis of HCC, and whether it is appropriate to exclude some patients with cirrhosis depending on the cause.

TABLE 1 - Target populations for HCC surveillance and imaging diagnosis

Target population		LI-RADS 2018a	EASL 2018 6	KLCA-NCC 2022 4	APASL 2017 3
Surveillance	HBV cirrhosis	+	+	+	+
	HCV cirrhosis	+	+	+	+
	Other causes of cirrhosis	+	+	+	+
	Chronic HBV without cirrhosis	+	Intermediate or high risk of HCC	+	+
	Chronic HCV without cirrhosis	—	≥F3 fibrosis	+	—
	Other causes of chronic liver disease without cirrhosis	—	≥F3 fibrosis	—	—
Imaging diagnosis	HBV cirrhosis	+	+	+	+
	HCV cirrhosis	+	+	+	+
	Other causes of cirrhosis	+	+	+	+
	Chronic HBV without cirrhosis	+	—	+	+
	Chronic HCV without cirrhosis	—	—	+	+
	Current or prior HCC without cirrhosis	+	—	—	—
	Exclusion	Cirrhosis due to congenital hepatic fibrosis or vascular disorders	—	Cirrhosis due to vascular disorders	—

^aAccording to LI-RADS Manual (https://www.acr.org/Clinical-Resources/Reporting-and-Data-Systems/LI-RADS/LI-RADS-CT-MRI-v2018).

Abbreviations: APASL, Asian Pacific Association for the Study of the Liver; EASL, European Association for the Study of the Liver; KLCA-NCC, Korean Liver Cancer Association and National Cancer Center; LI-RADS, Liver Imaging Reporting and Data System.

To date, in addition to the LI-RADS and EASL guidelines, many imaging-based diagnostic systems for HCC exist worldwide (Table 1). Although these systems are similar, they have differences resulting from different target populations, reflecting differences in the prevalence of chronic liver disease and regional priorities for HCC treatment.⁷ In North America and Europe, where liver transplants are frequently used for managing early-stage HCC, high specificity in the imaging diagnosis of HCC is desired to ensure fair allocation of liver allografts. In contrast, in many Asian countries where hepatic resection and locoregional ablation therapies are major options for treating early-stage HCC and chronic HBV infection is prevalent, maximal sensitivity is preferred.⁸ In addition to the differences in the high-risk population criteria, HCC imaging diagnosis criteria also differ according to the guidelines (Table 2), which can lead to confusion in the data collection and interpretation of HCC research. Many studies have compared the accuracy of imaging-based HCC diagnoses based on “major HCC features.” In addition, there is a large variation between the guidelines to date for lesions with inconclusive imaging findings, which require appropriate verification and further study with respect to the best additional diagnostic tasks or management. For inconclusive findings, the principles of biopsy application differ according to the guidelines. The LI-RADS and the American Association for the Study of Liver Diseases guidelines recommend biopsy as an option for LR-4 (probably HCC) lesions which have a higher risk of HCC among inconclusive nodules.² On the other hand, the EASL guidelines suggest that biopsy could be considered for all inconclusive nodules.⁶ While biopsy can be challenging due to sampling errors, potential complications such as bleeding, and the risk of tumor seeding, it remains an essential diagnostic tool for reaching a definitive diagnosis and deciding on an appropriate course of treatment. Recently, more active biopsy strategies have been introduced that are applied even when HCC confirmation is possible by imaging tests to get information on the tumor’s molecular profile. This approach can help in directing targeted therapy and immunotherapy, as well as contribute to clinical trials and research studies.

TABLE 2 - Imaging diagnostic criteria for definite HCC according to the latest LI-RADS and EASL guidelines

Imaging modality	Size	LI-RADS criteria (CT/MRI LI-RADS 2018, CEUS LI-RADS 2017)	EASL 2018 criteria
CT/MRI	10–19 mm	Nonrim APHE AND one or more of the following:  Nonperipheral washouta  Threshold growth	APHE AND washouta
	≥20 mm	Nonrim APHE AND one or more of following:  Enhancing capsule  Nonperipheral washouta  Threshold growth	—
CEUS	≥10 mm	Nonrim APHE AND late (≥60 sec) and mild washout (as a first-line imaging technique)	APHE AND late (≥60 sec) and mild washout (as a second-line imaging technique)

^aWashout is determined on portal venous or delayed phases on CT and MRI using extracellular contrast agents or gadobenate dimeglumine; on portal venous phase only on MRI using gadoxetate disodium.

Abbreviations: APHE, arterial phase hyperenhancement, EASL, European Association for the Study of the Liver; LI-RADS, Liver Imaging Reporting and Data System.

Given that unifying HCC imaging systems globally is the ultimate goal, the adoption of standardized terms is a fundamental element in achieving this goal. Inconsistent terminology and definitions used in the literature interfere with the process of data extraction from published studies, the conduct of structured reviews and meta-analyses, and the translation of research findings into clinical practice.⁹ The implementation of standardized terminology for imaging findings could drive the growth of registries and facilitate data sharing and adoption of scientific advancements from other systems. This would also enhance communication and collaboration among health care professionals, ultimately leading to better HCC imaging systems.⁹

In conclusion, the global unification of existing imaging-based HCC diagnostic systems, which requires further research and vigorous validation of current imaging systems, is expected to improve the consistent diagnosis and proper management of HCC. Adopting standardized terminology and complying with the criteria set out in each guideline could be a basic step toward harmonizing HCC imaging systems.