Hepatocellular carcinoma (HCC) or hepatoma is the most common primary malignancy of the liver. It ranks as the fifth most common tumor in the world and the third most common cause of cancer-related death. The incidence rates of HCC in the United States have historically been lower than in many countries. However, in recent decades, HCC age-adjusted incidence rates have doubled. This is largely due to an increasing prevalence of ailments that predispose to hepatic cirrhosis, such as chronic viral hepatitis, obesity and alcohol abuse.1
At present, curative treatment options for HCC include liver transplantation, partial hepatic resection and, in some reported instances, percutaneous or surgical ablation.2 In 1996, Dr. Vincenzo Mazzaferro and colleagues defined the subset of patients with unresectable HCC for whom liver transplantation was the optimal treatment. Termed the “Milan criteria,” patients with a single HCC tumor < 5 cm in diameter or 2-3 tumors all < 3 cm in diameter, demonstrated overall and recurrence-free survival rates of 75% and 83%, respectively, at 4 years.3 Unfortunately, wait time for liver transplantation in patients with HCC has risen, increasing waitlist dropout rates and associated mortality. Dropout rates after 1 year of waiting for a liver transplant have been reported at 20% to 60%.4 Thus, to control tumor burden of those meeting Milan criteria, reduce the waitlist dropout for patients awaiting orthotopic liver transplantation(OLT), and provide downstaging of disease for those exceeding criteria, many centers have relied on pretransplant transarterial locoregional therapy (LRT).
The increased use of LRT emphasizes the diagnostic radiologist’s need to be familiar with common therapies for treating and managing HCC. The radiologic hallmark of HCC, contrast uptake in the arterial phase and washout in the venous/late phase (Figure 1), remains fundamental to the diagnosis and post-treatment evaluation of lesions. This focused review will introduce the basic principles and forms of transarterial LRTs, briefly discuss the role of LRT in the management of HCC, examine assessment of imaging response to LRT procedures, and attempt to illuminate the nuances of interpreting post-treatment imaging.
Principles of Transarterial Locoregional Therapy
The normal liver receives a dual blood supply from the hepatic artery (25%) and the portal vein (75%). As the hypervascular HCC tumor grows, it increasingly depends on the hepatic artery for blood supply. Once an HCC tumor nodule reaches a diameter of ≥ 2 cm, most of the blood supply derives from the hepatic artery (Figure 2).5 This unique dual blood supply property of the liver provides the rationale for exploitation via transarterial embolization, or purposeful blocking of the hepatic tumoral arterial supply, as the nutrient portal venous flow to the liver remains intact.
Transarterial LRTs play an important role in the management of HCC. The varying types of LRTs exert their effect by delivering ischemic, chemical or radioactive doses to targeted liver lesions. These common therapies have been shown to achieve reproducible downstaging for patients outside of Milan criteria attempting to be listed for transplantation (Figure 3), and extend bridging of patients already listed and awaiting liver transplantation.5 Some authors have also shown increased survival for transarterial chemoembolization (TACE) in 2 randomized controlled studies.6,7 Similarly, longer time to progression has been suggested with transarterial radioembolization (TARE) compared to TACE.8 The practice of treating HCC patients with transarterial LRT is common in transplant centers for the purpose of downstaging or bridging to transplantation, and palliation in those with preserved liver function and adequate performance status. The form of transarterial LRT utilized derives from a comprehensive understanding of the patient’s performance status, oncologic tumor stage, and laboratory results, and is usually discussed at a multidisciplinary conference involving hepatologists, radiologists, and transplant surgeons.
Transarterial embolization (TAE), also known as bland embolization, and TACE consist of the selective angiographic occlusion of the arterial blood supply to a targeted lesion with a variety of embolic agents, with or without use of a chemotherapeutic agent. The bland occlusion by embolic particles only, as seen in TAE, results in tumor hypoxia and necrosis. In theory, the addition of chemotherapy produces an additive cytotoxic anti-tumor effect, as seen in TACE. However, controversy exists regarding the efficacy of TACE over TAE and a recent meta-analysis suggests that TAE is as equally effective as TACE at a lower cost point and with potentially fewer side effects.5 For both TAE and TACE, portal venous thrombosis is a relative contraindication.
Multiple forms of TACE are used in clinical practice. Conventional TACE (cTACE) is a single agent or cocktail of chemotherapeutic agents coupled with either gelfoam or iodinated poppy seed oil as the embolic agent. The presence of iodinated oil mandates 4-phase imaging, and inclusion of a noncontrast sequence, when evaluating with computed tomography (CT) post treatment (Figure 4). There is significant heterogeneity to the cocktail regimen and embolic material utilized for cTACE in clinical practice, making it difficult to study. Similarly, drug-eluting bead TACE (DEB-TACE) is a common form of LRT used at many centers. DEB-TACE is typically a single drug, most commonly doxorubin, loaded on polyvinyl alcohol or acrylic polymeric beads (Figure 5). The potential benefit of DEB-TACE is delivery of a local, controlled, sustained dose of chemotherapeutic agent to the tumor with the potential for less systemic toxicity and post embolization syndrome (PES). A summary of PES is seen in Table 1.9 A randomized controlled trial of 212 patients with HCC showed greater complete response, objective response, and disease control at 6 months with DEB-TACE vs. cTACE with a suggestion of improved tolerability, reduced liver toxicity and doxorubicin-related side effects.10
Transarterial radioembolization (TARE), also referred to in the literature as selective internal radiation (SIRT) or radioembolization, is a catheter-based therapy delivering microspheres loaded with yttrium-90 (Y90) to a hepatic malignancy. TARE provides a microembolic effect while delivering high-energy radiation over a limited range (mean penetration of radiation into tissues is 2.6 mm), thereby minimizing the radiation exposure to normal liver parenchyma.11 Given its microembolic effect, Y90 can be used in patients with compromised portal venous flow (ie, bland thrombus, tumor thrombus, or hepatofugal flow), a clear practical advantage when compared to TAE or TACE. Following radioembolization, arterial embolization is much less extensive compared to TAE or TACE. The goal of TARE is not to produce an extensive embolization, but rather to administer the highest radiation dose possible to the tumor. This concept spares the nontumorous liver from radiation-induced liver damage and maintains the patency of the artery for future transarterial treatments. TARE has the unique benefit of inducing hypertrophy of the future liver remnant (Figure 6), augmenting surgical resection.12
The formal assessment of HCC response to LRT has experienced a natural progression and evolution as our knowledge and understanding of both HCC and LRT has grown. The World Health Organization (WHO) criteria13 followed by the Response Criteria in Solid Tumors (RECIST) guidelines, were early standardized approaches to report degrees of tumor response, recurrence and disease-free interval. WHO and RECIST provide guidelines in which radiologic observations can be quantified for the measurement of tumor response to therapy and have provided the basis of response assessment in clinical trials.14,15 Both techniques rely on lesion size as the indicator of response, summarized in Tables 2 and 3. Hallmark to WHO is the bilinear product approach, which is calculated by multiplying the maximal diameter of a lesion by its longest perpendicular diameter, while the RECIST group proposed a model of single linear summation for target lesions (Figure 7).
The limitations of WHO and RECIST guidelines are due to their reliance upon changes in tumor dimension as the measure of radiologic response, ie, tumor shrinkage. The WHO criteria and RECIST do not take into account tumor necrosis that occurs without a reduction in overall tumor size. Furthermore, neither criteria mandate arterial phase imaging to demonstrate the radiologic hallmark finding of HCC. Recent literature has shown that assessments of radiologic response based only on size criteria do not correlate well with clinical benefit from LRT.15,16
In 2000, the European Association for the Study of Liver Disease (EASL) panel of experts put forth a new model for evaluating response in solid tumors. The EASL guidelines take into account tumor necrosis induced by treatments17 as shown in Figure 8. The EASL panel mandated evaluation with contrast-enhanced radiologic imaging for assessing treatment response. Viable tumor was defined as enhancement in the arterial phase on dynamic CT or MRI. This concept was further supported in 2005 by the American Association for the Study of Liver Disease.18
Given the complexities and unique properties of HCC, a more practical model for tumor assessment was required to augment trial design and facilitate clinical management. As a result, the modified RECIST (mRECIST) criteria were proposed and are based on measurements of tumor viability as indicated by arterial phase enhancement19,20 as depicted in Figure 8. mRECIST also include guidelines for assessing pleural effusions and ascites, porta hepatis lymph nodes, portal vein thrombosis, and the development of new lesions. The mRECIST is now the most widely used criteria for assessing tumor response to LRTs. Table 4 depicts the definitions and response categories for HCC according to mRECIST.
Modified RECIST standardizes image acquisition, interpretation and assessment. Variables that affect imaging quality can produce inconsistencies in the practical application of this criteria. Thus, mRECIST mandates high-quality dual-phase (arterial and portal venous) imaging (CT or MRI) after injecting intravenous contrast so as not to introduce pseudo-changes related to the method rather than pathology. According to mRECIST, for lesions with atypical enhancement dynamics (without arterial enhancement), conventional RECIST criteria must be applied. Perilesional arterial and portovenous phase enhancement is a common benign finding that results from arterioportal shunting after LRT (Figure 9).
For an HCC target lesion to be assessed by mRECIST, the following must exist: > 1 cm per RECIST, suitability for repeat measurement, and classic intratumoral arterial enhancement on CT or MRI. Therefore, infiltrative forms of HCC should be excluded from mRECIST evaluation. Even with these limitations, the mRECIST criteria allows for reproducible, objective, quantitative measures needed to define a common language between radiologists and clinicians facilitating informed treatment decisions.
Functional MRI techniques, such as diffusion-weighted imaging (DWI) have been recently investigated, aiding specificity to locoregional therapy response assessment and outcome prediction.21 Additionally, the degree of fluorodeoxyglucose (FDG) uptake on positron emission tomography (PET)/CT has been widely used as an indicator of metabolic activity in many malignancies and may show promise in evaluating post LRT HCC.22 Future developments using PET/CT and DWI are on the horizon but remain in the research state.
In 2011, the American College of Radiology (ACR) developed LI-RADS to address inconsistencies in imaging technique and reporting standards. It has since been updated and evolved into a widely accepted standardization of the performance, interpretation, and reporting of lesions on CT or MR in those at high risk for HCC. LI-RADS is a structured diagnostic algorithmic system that categorizes lesions from LR-1 (definitely benign) to LR-5 (definitely HCC) (Table 5). LI-RADS clearly mandates imaging technique performance standards as seen in Figure 10, and includes protocols for MRI hepatobiliary imaging agents not mandated by the United Network for Organ Sharing and Organ Procurement and Transplantation Network (UNOS-OPTN). LI-RADS LR-5 lesions, those which demonstrate the radiologic hallmark of HCC, do not require biopsy for diagnosis. LR-5 lesions are essentially equivalent to UNOS-OPTN Class 5 lesions, are diagnostic of HCC, and are a classification upon which transplant candidacy is determined. The use of LI-RADS after LRT is really quite simple. A lesion after locoregional therapy is simply categorized as LR-Treated (LR-T) as shown in Figure 11, and either has residual/recurrent disease (Figure 12) or is without residual/recurrent disease. A more comprehensive algorithmic approach to deal with the variable appearances and complexities of the LR-T category is expected from the ACR, to be termed ACR TR-LIRADS. For a more comprehensive review of LI-RADS visit http://www.acr.org/quality-safety/resources/LIRADS.23
Radiologic assessment of HCC after LRT has substantially improved over time and has become increasingly important to clinicians for patient management. With continually evolving assessment criteria, it is critical that diagnostic radiologists remain abreast of new guidelines and familiarize themselves with the bevy of HCC treatment regimens. At present TAE, TACE and TARE have become cornerstone therapies for bridging to transplant, downstaging to transplant, and palliation of HCC. mRECIST has emerged as the standard image-driven means of assessing response to LRTs, serves as the basis for response in clinical trials, and demonstrates significant improvement over prior guidelines. LI-RADS is a structured algorithmic system that standardizes the performance, interpretation, and reporting of lesions on CT or MR in those at high risk for HCC. LI-RADS LR-T lesions can be difficult to evaluate in the presence of benign post-procedural changes and in tumors presenting with a heterogeneous pattern of enhancement. ACR TR-LIRADS guidelines are expected and will likely elucidate the many challenges of interpretation and standardize the evaluation of hepatoma after LRT.
Montgomery J, Trebelev AE, Bianco BA. Imaging of Hepatocellular Carcinoma after Transarterial Locoregional Therapy: A Practical Review and Discussion of Treatment Res. J Am Osteopath Coll Radiol. 2017;6(3):5-12.