Version May 2024
INTRODUCTION — Hepatocellular cancer (HCC) is an aggressive tumor that frequently occurs in the setting of underlying liver disease, particularly cirrhosis. The two factors that are most important in determining a patient's prognosis and potential treatment options are tumor mass/location and the patient's hepatic reserve. Patients with unresectable tumors who do not fulfill criteria for liver transplantation are typically managed with nonsurgical locoregional therapy (percutaneous needle-based ablation techniques, transarterial therapies, and/or radiation therapy [RT]) or systemic therapy (immune checkpoint inhibitors or molecularly targeted agents, such as atezolizumab with bevacizumab and sorafenib). (See "Surgical resection of hepatocellular carcinoma", section on 'Preoperative assessment for resectability' and "Liver transplantation for hepatocellular carcinoma", section on 'Requirements for listing and management while on the wait list'.)
The antitumor effect of many nonsurgical locoregional treatment modalities and many systemic therapies is not accurately reflected by conventional bidimensional tumor measurements on radiographic studies. Accurate assessment of response requires evaluation of tumor size and other features, including the ablative margin, tumor viability and vascularity (as reflected by contrast enhancement), and early detection of residual/recurrent tumor and new areas of tumor involvement.
This topic review will cover the alternative methods of tumor assessment used for HCC that is treated either by nonsurgical local means or systemic therapy. Antitumor efficacy of specific nonsurgical locoregional treatments and systemic therapy is covered elsewhere. (See "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates not eligible for local thermal ablation" and "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates who are eligible for local ablation" and "Systemic treatment for advanced hepatocellular carcinoma".)
TREATMENT ALGORITHMS AND GENERAL APPROACH TO THE PATIENT WITH HCC — Treatments for HCC include potentially curative surgery (resection, liver transplantation), liver-directed therapies for unresectable liver-isolated disease in those ineligible for transplant, and systemic therapies:
●In general, the optimal treatment for early-stage HCC is surgical resection or liver transplantation.
●Patients with unresectable tumors that are ineligible for liver transplantation are typically managed with nonsurgical locoregional therapy or systemic therapy. Nonsurgical locoregional therapy may also be used as a bridge to liver transplantation or to downstage tumors that initially exceed usual transplantation criteria in order to downstage them to possible candidacy for liver transplantation. (See "Surgical resection of hepatocellular carcinoma" and "Liver transplantation for hepatocellular carcinoma", section on 'Requirements for listing and management while on the wait list' and "Liver transplantation for hepatocellular carcinoma", section on 'Bridging therapy'.)
•Nonsurgical locoregional therapies used for treatment of HCC include percutaneous needle-based ablation techniques, transarterial therapies, and radiation therapy (RT):
-Image-guided percutaneous ablation is most appropriate for one or two tumors that are no larger than 4 cm. Methods include radiofrequency ablation (RFA), microwave ablation, cryoablation, percutaneous ethanol injection (PEI), irreversible electroporation, and high-intensity focused ultrasound (US) ablation. (See "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates who are eligible for local ablation".)
-Arterially therapies can be used for patients who are not appropriate candidates for ablation; they include transarterial chemoembolization (TACE; includes both conventional and drug-eluting bead TACE), radioembolization (also termed selective internal RT), and bland embolization. (See "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates not eligible for local thermal ablation".)
-External beam RT approaches include photon irradiation, stereotactic body RT (SBRT), and charged particle (eg, proton beam) irradiation. (See "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates not eligible for local thermal ablation", section on 'Stereotactic body radiation therapy'.)
•Indications for systemic therapy are evolving. In general, systemic therapy is appropriate for patients with an acceptable performance status and underlying liver function who have extrahepatic spread, or have liver-limited disease that is refractory to nonsurgical locoregional therapies, or unsuitable for such therapies because of extensive vascular invasion or large intrahepatic tumor extent. (See "Systemic treatment for advanced hepatocellular carcinoma".)
An algorithmic approach to the treatment of patients with HCC is shown in the figure (algorithm 1). An alternative treatment algorithm is available from Barcelona Clinic Liver Cancer group, which was updated in 2022 (figure 1) [1]. However, attempts to generate algorithmic approaches to the treatment of HCC are difficult since new treatments and indications for various treatments are evolving rapidly. Furthermore, therapeutic approaches tend to vary based upon the available expertise, as well as variability in the criteria for hepatic resection and liver transplantation. These issues are discussed in detail elsewhere. (See "Overview of treatment approaches for hepatocellular carcinoma", section on 'Treatment algorithms'.)
TECHNIQUES FOR ASSESSMENT OF TREATMENT RESPONSE — Response assessment of HCC is a requirement for clinical evaluation of the effect of locoregional and systemic therapies, and for use as endpoints in clinical trials as a surrogate for survival. The gold standard for response to therapy is pathological evaluation of tissue. However, this is impractical as it is invasive and can have both false positive and false negative results, depending upon site of biopsy. Radiographic imaging offers a noninvasive, robust, and very accessible method of evaluation of tumor response.
●Typically, the efficacy of nonsurgical local ablation therapy is assessed by dynamic computed tomography (CT) or contrast-enhanced magnetic resonance imaging (MRI) one month after therapy and every three to six months thereafter. In general, MRI is preferred over CT due to its inherent superior tissue contrast resolution and sensitivity for detection of both the lesion and associated post-treatment changes. MRI is also beneficial for patients who have an allergy to iodinated contrast and for those who have undergone ethiodized oil (Lipiodol)-based therapies. (See 'MRI' below and 'Response after locoregional therapy' below.)
●For patients receiving systemic therapy, reimaging is typically carried out using cross-sectional imaging after the initial six to eight weeks of therapy, and then every two to three months thereafter. (See 'Response to systemic therapy' below.)
In either situation, for the subset of patients with an elevated alpha-fetoprotein (AFP), this marker can be useful to follow response; elevations may precede radiographic evidence of disease progression by several weeks or months. (See 'Tumor markers' below.)
Imaging studies — Contrast-enhanced, cross-sectional imaging with MRI or CT is typically used for diagnosis of HCC and follow-up after therapy. There is no clear evidence of superiority of MRI over CT for assessing HCC response to therapy. MRI may be preferred for individuals with acute or chronic kidney injury, those who are allergic to intravenous iodinated contrast, and for follow-up of patients treated with Lipiodol-based conventional transarterial chemoembolization (TACE).
Computed tomography — Contrast-enhanced CT is the most common radiographic modality used for diagnosis of HCC and follow-up after therapy. The ability of CT to detect HCC has improved with the development of helical CT technology, which involves the rapid administration of intravenous contrast in combination with extremely fast imaging. Multiphase contrast-enhanced thin-slice (multidetector row) helical CT (MDCT) is now the most common scanner type for CT evaluation. Thin section images allow multiplanar reconstruction, and the liver can be imaged in a single breath-hold, providing good quality images.
Attention must be paid to the type of scan that is ordered for treatment evaluation in patients with HCC. There is institutional variation in the scanning acquisitions for CT of post-treatment HCC. Some choose to add an unenhanced phase to characterize residual post-treatment enhancement. Routine imaging of HCC post treatment includes a single late arterial phase, a portal venous phase, and a late venous phase after iodinated contrast injection [2]. Standard single-phase CT with scanning during the portal venous phase is not generally sufficient to assess vascular characteristics associated with determining response.
The CT appearance of treated HCC is discussed in detail below. (See 'Imaging appearance of treated HCC' below.)
MRI — Magnetic resonance imaging (MRI) is generally preferred over CT for assessing treatment response due to its inherent superior tissue contrast resolution and sensitivity for detection of both the lesion and associated post-treatment changes. Compared with CT, MRI has the advantage of achieving high-resolution images of the liver without the use of nephrotoxic contrast agents or ionizing radiation. On contrast-enhanced MRI, HCC appears as a moderately high-intensity pattern on T2-weighted images and a low intensity pattern on T1-weighted images precontrast, often with an enhancing capsule (image 1). The characteristic appearance of HCC on contrast-enhanced MRI is arterial phase hyperenhancement, with subsequent washout on portal or delayed phases, lower signal intensity in the hepatobiliary phase (when gadoxetic acid is used), and capsular (pseudocapsule) enhancement.
Our institution uses the following protocol for liver evaluation on MRI: coronal fast spin echo T2-weighted images (which enhance T2 differences between tissues), axial fast spin echo T2-weighted images with and without chemical presaturation to reduce signal from fat (FAT-SAT), axial gradient recalled echo (GRE; which allows images to be acquired rapidly in a single breath-hold), in-phase and opposed-phase T1-weighted image, and axial three-dimensional FAT-SAT GRE T1-weighted imaging before and after dynamic injection of an extracellular gadolinium-based contrast agent (GBCA). (See "Principles of magnetic resonance imaging".)
The volumetric T1-weighted imaging sequence is the most important sequence for HCC detection and characterization. Acquisitions are performed in the arterial, portal venous, late venous or equilibrium phase (three minutes after injection) and hepatobiliary phase at 10 or 20 minutes (for gadoxetic acid) [3-5].
The added value of liver-specific gadoxetic acid (Eovist) over extracellular GBCAs to assess response after locoregional therapy of HCC is unclear. One study suggested no benefit for the detection of local recurrence after RFA [6], whereas another study comparing CT with gadoxetic acid-enhanced MRI after RFA showed higher interobserver agreement and a better performance in general for MRI [7]. The added potential value of gadoxetic acid may be in the differentiation of small areas of viable tumor from "pseudolesions" seen during the arterial phase by assessing washout on the late venous phase or hypointensity in the hepatobiliary phase (which reflects the lack of functioning hepatocytes) [8]. However, the contrast kinetics of gadoxetic acid are such that the arterial phase lacks the robustness of gadolinium, and this may compromise the identification of new lesions.
For patients being assessed after nonsurgical ablation methods such as RFA or TACE, image subtraction techniques are used in every patient. This is helpful because without subtraction, post-treatment coagulative hemorrhagic necrosis tends to be bright on precontrast (unenhanced) T1-weighted imaging, leading to difficulty in evaluation of postcontrast tumor enhancement [9]. In addition, unlike CT, in which the beam-hardening effects of high-attenuation Lipiodol may obscure small enhancing tumors, Lipiodol does not adversely affect MRI signal-intensity characteristics, so residual enhancement can be detected, especially when image subtraction is used (image 2) [10,11]. Contrast-enhanced dynamic T1-weighted imaging with diffusion-weighted MRI is a type of functional imaging that can also be advantageous in assessing treatment-related changes in HCC and is superior to contrast-enhanced CT after Lipiodol-based TACE therapies [10,12,13].
The MRI appearance of treated HCC and the Liver Reporting and Data System (LI-RADS) treatment response algorithm to assess response after locoregional therapies for HCC are discussed in detail below. (See 'Response after locoregional therapy' below.)
Diffusion-weighted imaging with MRI is a technique that derives its imaging contrast on the basis of differences in the mobility of extracellular protons in tissues. Diffusion is a physical process that results from the thermally driven random motion of water molecules. The signal of a diffusion-weighted MRI is proportional to the compactness of the cells, which in turn means a lack of extracellular space. This is exploited in interrogation of tumors to evaluate cellularity and cell death. This in turn can be an indirect measure of treatment effect. The diffusion characteristics are measured by the apparent diffusion coefficient. Most institutions do not routinely use diffusion-weighted imaging during MRI scans; however, it is increasingly being incorporated in routine protocols, especially for patients with renal insufficiency who cannot get intravenous contrast. There is some potential for diffusion-weighted imaging for monitoring the response to anti-HCC treatment, but this is not yet a standard approach [14]. (See 'Functional imaging' below.)
Other imaging studies
Ultrasound — In general, ultrasound (US) is used for HCC screening and is not used for follow-up imaging after locoregional or systemic treatment of HCC. US is limited by lower sensitivity, inability to distinguish viable versus nonviable tumors, and its operator dependency.
Dual energy/dual source CT — The use of dual energy/dual source CT (DECT) is not established, and further studies are necessary prior to incorporating this approach into routine clinical practice. Dual energy CT scanners are also not universally available across imaging practices.
DECT is a technology in which images are acquired by using two different energy beams that can be from two different sources in the same gantry, or by alternating high- and low-energy radiation from the same source at high speed. DECT provides a higher temporal resolution and can provide an iodine map with virtual nonenhanced images, which decreases radiation dose as it eliminates the need for nonenhanced image acquisition.
DECT increases detection of hypervascular lesions, such as HCC, and it may improve characterization of indeterminate subcentimeter lesions [15,16]. DECT may also have utility in evaluating HCC response to locoregional therapy, such as radiofrequency ablation (RFA), because of a higher lesion-to-liver contrast-to-noise ratio on iodine maps, aiding in the detection of residual tumor [17].
Tumor markers
AFP — In the subset of patients with an elevated alpha-fetoprotein (AFP), this marker can be useful to follow response to nonsurgical locoregional therapies and provide an early screening test for progression, which should later be confirmed by imaging [18-20]. In most, but not all [21], studies, serial measurements of AFP appear to be a valid surrogate endpoint for benefit from systemic chemotherapy, including molecularly targeted agents [22-29]. In one study, AFP responders (arbitrarily defined as patients whose serum AFP decreased by more than 20 percent after a minimum of two cycles of chemotherapy) had better median survival than did nonresponders (13.5 versus 5.6 months), and AFP response was strongly associated with radiographic response [23]. Furthermore, an AFP response was frequently observed in patients with radiographically stable disease and served to identify a subgroup of these patients with better survival.
Unfortunately, AFP levels are not elevated in up to 40 percent of patients with HCC <2 cm in diameter and in 28 percent of those with tumors between 2 and 5 cm. (See "Clinical features and diagnosis of hepatocellular carcinoma", section on 'Alpha-fetoprotein'.)
DCP — Use of an alternative tumor marker, des-gamma-carboxy prothrombin (DCP), has been suggested in patients with advanced HCC, but the available data suggest that DCP levels do not always correlate with survival after treatment with either locoregional or systemic therapies [30-34], and it is not clear that DCP adds value to monitoring serum levels of AFP when both markers are elevated. Furthermore, serum assay for DCP is not widely available. The utility of DCP for screening and diagnosis of HCC is discussed in detail elsewhere. (See "Clinical features and diagnosis of hepatocellular carcinoma", section on 'Other markers'.)
Plasma VEGF levels — Changes in circulating plasma vascular endothelial growth factor (VEGF) may represent a novel prognostic marker for HCC; a staging system has been proposed that includes plasma VEGF concentration along with the Cancer of the Liver Italian Program (CLIP) score (V-CLIP score). (See "Staging and prognostic factors in hepatocellular carcinoma", section on 'New prognostic markers and methods under investigation'.)
Plasma VEGF levels may represent a surrogate marker for benefit from sorafenib, an inhibitor of multiple receptor kinases including the VEGF receptor, but the available data are conflicting:
●In a retrospective review of 63 patients with advanced HCC, a decrease in plasma VEGF levels of >5 percent from the pretreatment baseline at eight (but not four) weeks after initiating sorafenib was associated with significantly longer median overall survival compared with patients without a 5 percent decrease (30.9 versus 14.4 months) [35]. While intriguing, this evaluation is based upon a very small number of patients (n = 14) with a >5 percent decrease in serum VEGF at eight weeks, and additional data are needed to confirm and validate these findings.
●On the other hand, an analysis of data from a pivotal phase III trial that established the survival benefit of sorafenib in advanced HCC failed to find an association between plasma biomarkers, including VEGF, and outcomes [36].
Liquid biopsy analysis — Liquid biopsies, or blood samples, can be analyzed to detect circulating tumor cells and plasma cell-free circulating tumor DNA (ctDNA), but whether these assays aid in prognostication or response classification is unclear:
●A systematic review of 112 studies, including 25 that focused on treatment monitoring or selection, concluded that changes in numbers of circulating tumor cells could more accurately identify patients with an HCC recurrence after resection or locoregional therapy than pretreatment counts alone [37].
●Few studies have examined the effect of post-treatment ctDNA on prognosis, and most have generally failed to show an association with progression-free survival (PFS) or overall survival [38,39]. One notable exception is that among patients treated with first-line atezolizumab plus bevacizumab, at least one preliminary report noted that longer PFS was observed in patients whose ctDNA became undetectable post-treatment, although the confidence intervals were wide [40]. The median PFS in patients with ctDNA present versus cleared at C4D1 was 6.5 months and not reached, respectively (HR for progression 12.0 [95% CI 1.7-93], log-rank p <0.00029).
●In addition, of the several studies evaluating changes in ctDNA after therapy, at least three found that changes in ctDNA preceded radiographic documentation of tumor progression by up to eight weeks [40-42]. However, the lack of assay standardization and high risk of bias complicates interpretation of these studies. The clinical utility of liquid biopsy analysis in management of patients with HCC remains uncertain.
IMAGING APPEARANCE OF TREATED HCC
Response after locoregional therapy — Imaging performed immediately after locoregional therapy can aid in evaluation of the treatment and act as a guide to expected outcomes. Although not entirely reliable, the absence of contrast uptake within the tumor is thought to reflect tumor necrosis, while the persistence of contrast uptake generally indicates persistent disease. Recurrence of tumor in the treated area (or elsewhere) is signaled by the reappearance of vascular enhancement.
●When treating HCC with transarterial chemoembolization (TACE) using ethiodized oil (Lipiodol), unenhanced computed tomography (CT) can be performed immediately to assess Lipiodol distribution. In general, complete retention of iodized oil is highly correlated with complete lesion necrosis, while incomplete retention may represent complete necrosis or residual viable tumor [43,44]. If unenhanced CT shows lack of Lipiodol uptake within the tumor, this likely represents arteriovenous shunting through the tumor or a variant tumor vascular supply (including one that may be extrahepatic), and that tumor will likely require retreatment [45].
The iodized oil is not apparent on magnetic resonance imaging (MRI), and thus, MRI is generally preferred to assess for recurrent/viable tumor over time.
●To assess for technical success after selective internal radiation therapy (SIRT), the patient is immediately transferred to the nuclear medicine department for a single-photon emission computed tomography (SPECT) scan, which shows the distribution of yttrium-90 (Y90) in the liver. Some institutions perform Y90 positron emission tomography/CT and Bremsstrahlung SPECT/CT to evaluate the distribution of radiation dose with proprietary software showing the distribution of treatment across the tumor and assessing need of retreatment [46].
●After ablation (radiofrequency ablation [RFA], microwave, cryoablation), immediate success is defined as a nonenhancing ablation zone encompassing the entire tumor and an ablative safety margin of at least 5 mm of the surrounding nontumor hepatic parenchyma [47]. Following RFA, gas bubbles can form as a result of treatment and should not be mistaken for infection or infarction [48].
A thin, uniform rim of enhancement around the treated zone that reflects transient hyperemia and represents a physiologic response to thermal injury, and embolization of the hepatic parenchyma can be seen after RFA, TACE, or SIRT [49]. This usually resolves within several months of follow-up imaging, a fact that permits it to be differentiated from residual malignancy [50]. However, small foci of residual tumor may be obscured by transient hyperemia; cases showing persistent arterial enhancement with washout on delayed phase images at short-term follow-up should be evaluated further to determine whether additional directed therapy is needed [49].
Analysis of enhancement patterns on the arterial phase of the first follow-up CT is helpful for predicting progression of treated HCC. Regardless of the mode of therapy, nodular or thick areas of arterial enhancement along the margin of a treated HCC are consistent with residual tumor, especially when there is associated contrast washout. This treatment response is categorized as LI-RADS TR-viable (algorithm 2); in general, these are areas that require additional treatment (image 3) [48,51].
Radiologic patterns after treatment vary according to treatment modality:
●Ablative therapy – At short-term follow-up, coagulative necrosis, as induced by locoregional ablative therapies, appears as hyperdensity on CT and high signal intensity on T1-weighted MRI. Both of these changes limit the evaluation of contrast-enhanced images, which are of paramount importance to evaluate the efficacy of treatment, given that residual enhancement implies residual tumor. This can be circumvented by the use of image subtraction on MRI, which is helpful in differentiating coagulative necrosis, hemorrhage, and residual enhancing tumor (image 2). (See 'MRI' above.)
This problem is most often seen on the one-month post-treatment images. The hyperdensity on CT and increased T1 signal in the treated area resulting from coagulative necrosis frequently resolve on further follow-up imaging [50]. For this reason (and to reduce exposure to ionizing radiation), we prefer MRI over CT at the one-month time point.
Following ablation therapy, residual viable tumor or tumor regrowth is most often detected at the periphery of the treated lesion, either as irregular thickening at the margin or as a new tumor nodule (image 4) [52]. By MRI, increased signal intensity on T2-weighted imaging or restriction on diffusion-weighted MRI with associated nodular enhancement is characteristic of an incompletely treated tumor. If HCC is ablated using RFA, arterioportal shunts can be seen due to thermal injury to small vessels in the hepatic parenchyma but will resolve on follow-up evaluation [53].
●TACE – For patients treated with TACE using drug-eluting beads (DEB-TACE) or conventional TACE (cTACE) containing Lipiodol, follow-up contrast-enhanced CT or MRI should be performed to assess areas lacking enhancement [43]. Areas that increase in size over time or develop arterial enhancement after contrast injection (image 3) should raise suspicion for recurrent disease. cTACE using Lipiodol can lead to interference with CT; however, this is not an issue with DEB-TACE. In general, MRI is more useful in comparison with CT in cTACE setting because of the lack of interference of Lipiodol and the use of subtraction images to evaluate for residual/recurrent disease [11]. Because of this, many centers now utilize DEB- rather than Lipiodol-TACE, and in these instances MRI with contrast, or multiphasic CT would be most useful for response evaluation. (See 'MRI' above.)
●Radiation-based treatments
•TARE – In general, patients treated with transarterial radioembolization (TARE) have a delayed tumor response; the median time to develop evidence of necrosis (decreased enhancement) and tumor shrinkage is approximately 30 and 120 days, respectively [54]. Because of this, we perform our first radiographic assessment for disease response at three months.
As with other ablative locoregional therapies, rim contrast enhancement can persist over time and is not indicative of viable tumor; in one radiologic-pathologic correlation study, complete pathologic necrosis was seen in 93 percent of the cases with thin rim enhancement on post-treatment imaging, while complete pathologic necrosis was seen in only 38 percent of the lesions that showed peripheral nodular enhancement [55,56]. Enhancement patterns after TARE are highly variable. Heterogeneous enhancement of the liver in a perivascular distribution in a treated segment or lobe reflects radiation injury and should not be mistaken for tumor progression (image 5) [55,57]. This is particularly a problem with the Liver Reporting and Data System (LI-RADS) treatment response algorithm. Key features suggesting residual tumor after TARE includes a new or enlarging nodular or mass-like area of arterial phase hyperenhancement within or around the treated tumor, and growth over time [58]. (See 'LI-RADS algorithm' below.)
•EBRT and SBRT – Patients treated with external beam radiotherapy (EBRT), including hypofractionated stereotactic body radiation therapy (SBRT) also have a delayed tumor response, with persisting arterial phase hyperenhancement with or without washout for a year or more, followed by gradual decrease over time [58,59]. In one report, radiographic "complete response" rates at 3, 6, and 12 months post-treatment were 24, 67, and 71 percent, respectively [60]. Furthermore, irradiated nontumorous hepatic parenchyma may also show arterial hypervascularity after treatment, and this might interfere with the evaluation of treatment response on dynamic cross-sectional imaging [61]. These issues create diagnostic dilemma during image interpretation, particularly using the LI-RADS treatment response algorithm, because persistent arterial enhancement, while expected post-treatment, denotes viable disease in the setting of other locoregional therapies. (See 'LI-RADS algorithm' below.)
In our experience, these findings are usually equivocal and a decision for further treatment is usually deferred until the six-month scan. At our institution, we prefer MRI over CT unless there is a contraindication for MRI. MRI is generally more helpful than CT in this setting, although difficulties remain with differentiating residual arterial enhancement from radiation changes. In general, these distinctions become clearer with time, usually greater than three months.
Response to systemic therapy — Early assessment of response to molecularly targeted therapies including tyrosine kinase inhibitors (TKIs) can be challenging since tumor necrosis, extension, and radiologic appearance can be inhomogeneous, and radiologic response rates are very low [62-65]. As with locoregional therapies, residual or progressive thick and nodular foci of arterial enhancement within HCC suggest persistent or progressing viable tumor. In clinical practice, efficacy for molecularly targeted agents like TKIs tends to be measured in lack of progression rather than true radiologic "response." (See "Systemic treatment for advanced hepatocellular carcinoma", section on 'Response assessment' and "Systemic treatment for advanced hepatocellular carcinoma", section on 'First-line therapy'.)
Notably, an increase in tumor size due to necrosis has been reported in patients treated with multikinase inhibitors, such as sorafenib, a phenomenon referred to as "pseudoprogression" [21]. Therefore, quantification of necrosis/tumor viability rather than size measurement appears to be a better method of treatment response assessment in HCC treated with these biologic agents [21,62]. (See 'Measuring tumor dimensions versus tumor viability' below.)
For patients treated with checkpoint inhibitor immunotherapy, patterns of response to treatment differ from those with molecularly targeted agents. Patients may appear to have a transient worsening of disease, manifested either by progression of known lesions or the appearance of new lesions, before disease stabilizes or regresses, and responses may be delayed. (See "Principles of cancer immunotherapy", section on 'Patterns of response'.)
Immune-related response criteria have been proposed to properly recognize the nontraditional patterns of response occasionally seen with checkpoint inhibitors and some other immunotherapies. Pseudoprogression has also been reported with immune checkpoint inhibitors. Issues specific to assessing response to immunotherapy are addressed below. (See 'Response criteria for immunotherapy' below.)
CRITERIA FOR RESPONSE ASSESSMENT — In patients with advanced HCC who have completed therapy (either nonsurgical locoregional or systemic therapy), various classification systems are used to assess treatment response.
For patients treated with systemic therapy, treatment response is typically classified using either the modified Response Evaluation Criteria In Solid Tumor for HCC (mRECIST) (table 1), or RECIST (table 2). Patients treated with immunotherapy are assessed using consensus based criteria for response to immunotherapy (iRECIST) (table 3). For patients treated with locoregional therapy, treatment response is classified using LI-RADS (algorithm 2).
Measuring tumor dimensions versus tumor viability — Anatomic measurements of tumor dimension before and after treatment is one radiographic method to assess treatment response. A decrease in tumor size is often used as a surrogate endpoint to predict survival outcome in patients undergoing systemic treatment for a variety of solid tumors [66-68]. Tumor size is still an important parameter in measuring treatment response to HCC, and there is evidence that the more rapidly a tumor shrinks, the more durable the response. For advanced HCC, response criteria based on tumor dimension include the World Health Organization (WHO) [69] and RECIST 1.1. (See 'Response Evaluation Criteria In Solid Tumor (RECIST)' below.)
However, the antitumor effect of many nonsurgical locoregional and systemic therapies, (such as antiangiogenic multikinase inhibitors and immune checkpoint inhibitors) is not always accurately reflected by changes in tumor dimension [70-73]. Such measurements do not incorporate tumor viability and the extent of necrosis resulting from locoregional ablative therapies, such as transarterial chemoembolization (TACE) [72] or systemic therapy with antiangiogenic therapy. Necrosis may be associated with no change in the lesion size, or it may precede a change in measurable dimensions [73]. (See 'Response criteria for immunotherapy' below.)
This led to the development of criteria that classify tumor response for HCC based on tumor viability, including tumor necrosis, the ablative margin, and lesion vascularity (as reflected by contrast enhancement). Criteria have evolved from those that utilize only bidimensional measurements to those that evaluate the post-treatment changes on contrast-enhanced cross-sectional imaging scans (such as mRECIST for HCC and LI-RADS). (See 'Modified RECIST for HCC' below and 'LI-RADS algorithm' below.)
Response criteria that assess tumor viability are typically preferred over those based on tumor dimension. Using contrast-enhanced cross-sectional imaging, the absence of contrast uptake within the tumor is thought to reflect tumor necrosis, while the persistence of contrast uptake indicates persistent disease. Recurrence of tumor in the treated area (or elsewhere) is signaled by the reappearance of vascular enhancement. Most (but not all [30,74,75]) studies support the view that classifications systems of tumor response to both systemic and nonsurgical locoregional treatments for HCC according to tumor viability are a better predictor of outcomes compared with those that use tumor size alone [76-83].
Following systemic therapy — Response assessment criteria are available that are applicable to systemic therapies, although some may also be used to assess treatment response with nonsurgical liver-directed therapies.
Modified RECIST for HCC — In 2000, a panel of experts convened by the EASL recommended a modification of the Response Evaluation Criteria In Solid Tumor (RECIST) criteria to take into account tumor necrosis induced by HCC treatment, whether locoregional or systemic [84]. The panel adopted the concept of viable tumor (defined as uptake of contrast agent in the arterial phase of dynamic CT or MRI) as endorsed by the EASL [85] and the American Association for the Study of Liver Disease [84], and proposed amendments to RECIST criteria for the determination of tumor response for target lesions in HCC, termed the mRECIST for HCC (table 1) [86]. (See 'Response Evaluation Criteria In Solid Tumor (RECIST)' below.)
The expert panel emphasized the importance of standardization of response assessment, definition of treatment response and tumor progression, and overall response assessment [84,86]:
●Standardized response assessment – The key to proper application of mRECIST for HCC is optimization of image acquisition protocol and consistent use of the same protocol for response assessment throughout follow-up evaluation. Modality-wise, either CT or MRI can be used. Intravenous contrast administration is recommended unless medically contraindicated. Dual phase imaging is mandatory, including arterial and portal venous phase. Delayed phase imaging is not mandatory. Multidetector CT acquiring thin, contiguous slices with no reconstruction gap is mandatory. For image interpretation, the panel recommended a "centralized radiologic review" to ensure comparability across studies. Uniform image acquisition parameters, rigorous quality control, and independent blinded multireader assessment are mandatory. Independent radiologists will assess the baseline imaging and follow-up imaging according to mRECIST.
At baseline, tumors are characterized as measurable (lesion accurately measured in at least one dimension equal to or greater than 1 cm) or nonmeasurable (all other lesions, lesions smaller than 1 cm). Importantly, mRECIST for HCC criteria can only be used for typical hypervascular HCC and not for atypical HCC, including poorly differentiated HCC and diffuse-type HCC. For these nonenhancing atypical lesions, standard RECIST criteria are used (table 2). HCC lesions that were treated in the past with locoregional or systemic treatments may or may not be considered suitable targets for mRECIST for HCC; if the lesion shows a well-defined area of arterial enhancement, which correlates with viable tumor, and is at least 1 cm in the longest dimension, it can be considered a target lesion. Lesions showing atypical enhancement with poor delineation due to prior treatment cannot be selected for mRECIST for HCC evaluation.
Lesions that can be selected as a target lesion as per mRECIST for HCC should meet the following criteria:
•They can be classified as a RECIST measurable lesion
•They are suitable for repeat measurement
•They show intratumoral arterial enhancement on contrast-enhanced CT or MRI
●Definitions of treatment response and tumor progression – As outlined in the table, these are classified as target lesion response, nontarget lesion response, and new lesions (table 1).
Target lesion response is classified as complete response, partial response, progressive disease, and stable disease.
Following treatment, likely due to necrosis, viable tumor measurement can be a challenge. Arterial phase measurement of CT or MRI was recommended since the contrast between the viable and nonenhancing necrotic area is highest at this time. Measure the longest diameter of viable tumor without any intervening major area of necrosis. It should be mentioned that the longest diameter of enhancement on post-treatment scans may not be in the same plane as the baseline measurement. The panel mentioned that volumetric assessment should be a priority for future research, since there is a 138 percent difference in tumor volume between partial response (at least a 30 percent decrease in the sum of the diameters of the target lesions, amounting to a 65 percent change in volume) and progressive disease (increase of at least 20 percent in the diameter of the viable tumor, amounting to a 73 percent change in volume).
For nontarget lesions, persistent enhancement in even one lesion should be considered an incomplete response or stable disease, and a lack of contrast enhancement is consistent with a complete response. Appearance of even one new lesion or unequivocal progression of the older nontarget lesion should be called progressive disease. Some nontarget lesions (eg, malignant portal vein thrombus) are considered nonmeasurable. For a porta hepatis lymph node to be considered malignant, it should measure at least 20 mm in the short axis. In patients showing worsening of ascites or pleural effusion who have otherwise met criteria of response or stable disease, the nature of ascites/pleural effusion should be confirmed by cytopathology.
A newly detected hepatic nodule is classified as HCC (and declared evidence of progression) when its longest diameter is at least 1 cm and the nodule has a typical vascular pattern on dynamic imaging (hypervascular in arterial phase with washout in the portal venous or late venous phase). Liver lesions >1 cm that do not show a typical vascular pattern can be diagnosed as HCC if there is at least 1 cm of interval growth on subsequent scans. An individual radiologic event will be called disease progression in retrospect at the time the lesion was first detected, even if the lesion formally fulfilled the criteria for disease progression only on subsequent imaging scans.
●Overall response assessment – Similar to conventional RECIST, mRECIST defines overall patient response as a result of the combined assessment of the target lesions, nontarget lesions, and new lesions (table 1). Appearance of any new lesion equates to disease progression, irrespective of the target and nontarget lesions' response to therapy. Thus, strict criteria should be enforced, and any lesion not meeting strict criteria should be considered equivocal and not conclusive for disease progression.
Although data are limited, a preliminary report from the IMbrave150 trial of atezolizumab plus bevacizumab plus a report from the REFLECT trial of sorafenib or lenvatinib for systemic treatment of HCC both support the view that decreases in tumor size as assessed by mRECIST represent a surrogate endpoint to predict survival outcome in patients undergoing systemic treatment for HCC using these agents [87-89]. (See "Systemic treatment for advanced hepatocellular carcinoma", section on 'Sorafenib' and "Systemic treatment for advanced hepatocellular carcinoma", section on 'Lenvatinib'.)
Response Evaluation Criteria In Solid Tumor (RECIST) — RECIST 1.1 represents the most widely used criteria for response assessment in solid tumors (table 4).
Anatomic criteria, such as RECIST, are still used in virtually every clinical trial of HCC therapeutics as bidimensional measurements of response on imaging are essentially always measured. Furthermore, in many countries, regulatory agencies use RECIST as the standard response assessment criteria in clinical trials of HCC; alternative methods of response assessment (eg, mRECIST for HCC) have not yet achieved widespread adoption.
Standard RECIST 1.1 guidelines are also used for nonenhancing atypical lesions, including poorly differentiated HCC and diffuse-type HCC. Radiographic response criteria that assess tumor viability cannot be used for atypical HCC.
Response criteria for immunotherapy — Immunotherapy for HCC is a rapidly evolving area of treatment in advanced HCC. The combinations of atezolizumab plus bevacizumab and durvalumab plus tremelimumab are now preferred first-line regimens over treatment with a molecularly targeted agent such as sorafenib or lenvatinib. (See "Systemic treatment for advanced hepatocellular carcinoma", section on 'Atezolizumab plus bevacizumab'.)
The patterns of response to treatment with immunotherapy differ from those with molecularly targeted agents or cytotoxic chemotherapy. As an example, in early trials of immunotherapy in melanoma, a unique response pattern was seen, termed "pseudoprogression," in which patients appear to have a transient worsening of disease, manifested either by progression of known lesions or the appearance of new lesions, before disease stabilizes or regresses. In addition, responses can take appreciably longer to become apparent compared with cytotoxic therapy. These issues are all addressed elsewhere. (See "Principles of cancer immunotherapy", section on 'Patterns of response'.)
In clinical practice, patients receiving any immune-based therapy for advanced HCC whose tumors show initial growth should be assessed carefully for signs and symptoms of clinical benefit or progression; the majority of patients will have true progressive disease. In the absence of symptomatic progression, however, a short-term repeat scan is reasonable prior to considering immune-based therapy a failure.
Accumulating data suggest that on-treatment AFP changes over time can complement imaging findings and provide prognostic information for AFP-producing HCCs treated with immune checkpoint inhibitor immunotherapy [90-93].
Other criteria — The European Association for the Study of the Liver (EASL) criteria are not routinely used to measure treatment response in HCC [85]. These criteria incorporated estimated bidimensional reductions in viable tumor burden using dynamic imaging into response definitions. The most important limitation is the lack of guidelines regarding the choice of target lesions (number and size of measurable lesions). These criteria have fallen out of favor since the publication of the mRECIST in HCC criteria, which is endorsed by most of the leading groups in the field of HCC management. (See 'Modified RECIST for HCC' above.)
Following nonsurgical locoregional therapies — The antitumor effect of many nonsurgical locoregional treatment modalities for HCC is not accurately reflected by conventional bidimensional radiographic tumor measurements. Accurate radiographic assessment of response following nonsurgical locoregional treatment requires evaluation of residual tumor enhancement after therapy. Furthermore, the treatment response classification systems described above are based on tumor response assessment at the patient level rather than at the lesion level. For individual therapies that target specific lesions (eg, radiofrequency or microwave ablation), this may not provide a suitable assessment of treatment outcomes. (See "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates who are eligible for local ablation" and "Localized hepatocellular carcinoma: Liver-directed therapies for nonsurgical candidates not eligible for local thermal ablation".)
Assessment of treatment response after nonsurgical local therapies should be performed with multiphase, contrast-enhanced, cross-sectional imaging (CT with iodinated contrast or MRI with gadolinium-based agents). MRI is the preferred imaging modality with a sensitivity and specificity for lesion detection of 88 and 94 percent, respectively [94]. Furthermore, in patients undergoing locoregional therapy, contrast-enhanced MRI is more sensitive than CT for evaluating tumor viability using the LI-RADS treatment response algorithm [95]. (See 'LI-RADS algorithm' below.)
As tumor enhancement characteristics post-therapy are essential in assessment for tumor viability [96], contrast should be administered unless there are contraindications. Patients with iodinated contrast allergies should be routed to MRI and those with gadolinium allergies to CT. Preexisting kidney impairment is not considered a contraindication for contrast-enhanced MRI [97].
LI-RADS algorithm — The American College of Radiology has developed the Liver Reporting and Data System (LI-RADS) treatment response algorithm after locoregional therapies for HCC (algorithm 2) [96,98]. The most important feature of this treatment response algorithm is persistent lesion contrast enhancement and washout. These same characteristics are used for early detection of residual/recurrent tumor and new areas of tumor involvement. Using the LI-RADS treatment response algorithm, tumors are categorized as viable, nonviable, or equivocal [58,96]. The absence of contrast uptake within the tumor reflects nonviable tumor, while the persistence of contrast uptake indicates persistent disease, referred to as viable. If a tumor has enhancement atypical for treatment, it is deemed equivocal. Recurrence of tumor in the treated area (or elsewhere) is signaled by the reappearance of vascular enhancement or nodular, mass-like, or thick irregular tissue in or along the treated lesion.
An important point is that LI-RADS provides guidelines only to identify viable tissue, and not to classify changes over time. An integration of LI-RADS with mRECIST for HCC could be necessary for best response evaluation over time.
LI-RADS is best suited to evaluate response to nonradiation arterial and ablation procedures, in which tumor necrosis is expected immediately after locoregional therapy. It should be applied only cautiously to patients undergoing stereotactic body radiation therapy and selective internal radiation therapy in which tumor necrosis develops over time, and early post-treatment persistent arterial phase hyperenhancement is common, expected, and can confound the assessment of treatment response. (See 'Response after locoregional therapy' above.)
Future trends
Tumor volume measurement — Estimates of tumor volume may prove more useful than planar methods of tumor measurement in evaluating tumor response. In a retrospective study of 45 HCCs, the largest tumor diameter based upon three-dimensional measurements was significantly different than the diameter based upon conventional two-dimensional measurements [99]. Where available, volumetric evaluation of HCC and its necrotic component offers the most comprehensive anatomic evaluation for determining treatment response [100], and it is more accurate and reproducible than two-dimensional analysis [99,101]. Volumetric quantification is particularly helpful in cases in which areas of necrosis are heterogeneously distributed within the tumor, making it difficult to assess using mRECIST for HCC guidelines.
Although semiautomatic volumetric evaluation is feasible in clinical settings of HCC treated with locoregional therapies [102] and sorafenib [103], volumetry has not yet been incorporated into any of the commonly used viability or size-based response criteria in HCC.
Functional imaging — In contrast to anatomic imaging, functional imaging provides information on tumor viability, cellularity, vascularity, and metabolism [104]. In general, changes such as these can be detected earlier than anatomic changes and are more accurate in assessing treatment response after nonsurgical locoregional ablative therapy or systemic therapy of HCC [105]. Contrast enhancement on CT or MRI is a good marker for function, as is diffusion-weighted MRI. (See 'MRI' above.)
Other than these methods, functional imaging of response to HCC therapy can also be achieved by quantifying liver perfusion using fluorodeoxyglucose-positron emission tomography imaging, perfusion CT, and multiparametric MRI using gadolinium contrast with quantitative perfusion and diffusion-weighted imaging [106-118]. While functional imaging is promising, none of these modalities have been validated for clinical use. Their role as novel markers for response to nonsurgical locoregional or systemic therapy in HCC awaits evaluation in prospective studies.
GUIDELINES FOR POST-TREATMENT IMAGING — For most patients, we perform dynamic (contrast-enhanced) cross-sectional imaging with either computed tomography or magnetic resonance imaging approximately eight weeks after the start of systemic therapy and four weeks after nonsurgical locoregional therapy for HCC. However, if radioembolization or stereotactic body radiation therapy was undertaken, we delay the first assessment until three months. In patients whose disease is treated with radiation and imaged earlier than three months, it may be confusing to interpret radiographic studies because they often demonstrate arterial hyperenhancement both in the tumor and other non-target liver regions. It may take three to six months to observe the optimal response [119,120].
For most patients, we then continue monitoring for recurrent or new disease every three months for at least the first year. For patients treated with transarterial chemoembolization (TACE), imaging to assess for treatment response is offered at one month post treatment completion and then at three-month intervals subsequently for one year. For patients treated with transarterial radioembolization (TARE) and stereotactic body radiation therapy (SBRT), imaging is generally performed three months after treatment and at three-month intervals for one year.
If there has been no recurrence of disease after a year, we then perform the same imaging every six months. After two years with no evidence of disease recurrence some institutions, including ours, revert back to standard HCC surveillance with cross-sectional imaging and alpha-fetoprotein (AFP) assay every six months. At many institutions, cross-sectional imaging every six months is continued beyond the two-year cutoff, given the continued elevated risk for HCC in these patients. The use of ultrasound (US) in this situation is controversial, and many believe that the risk remains sufficiently high that periodic cross-sectional imaging should be continued beyond two years. (See "Surveillance for hepatocellular carcinoma in adults", section on 'Summary and recommendations'.)
These recommendations are consistent with guidelines for post-treatment imaging after systemic therapy or nonsurgical locoregional therapy from the European Association for the Study of the Liver (EASL) [121]. Specific recommendations for post-treatment imaging are also provided in the consensus-based guidelines of the National Comprehensive Cancer Network [122].
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hepatocellular carcinoma".)
SUMMARY AND RECOMMENDATIONS
●Methods for response assessment – Adequate response assessment of hepatocellular cancer (HCC) is a requirement for clinical evaluation of the effect of nonsurgical locoregional and systemic therapies, and for use as endpoints in clinical trials as a surrogate for survival. (See 'Techniques for assessment of treatment response' above.)
●Radiographic assessment
•For most patients, the efficacy of nonsurgical local ablation therapy or systemic therapy is assessed by dynamic contrast-enhanced CT or contrast-enhanced MRI. (See 'Imaging studies' above.)
Following nonsurgical local therapy, MRI is generally preferred over CT due to its inherent superior tissue contrast resolution and sensitivity for detection of both the lesion and associated post-treatment changes. We use the American College of Radiology Liver Reporting and Data System (LI-RADS) treatment response algorithm after locoregional therapies for HCC (algorithm 2). The most important feature of this treatment response algorithm is persistent lesion contrast enhancement and washout. (See 'LI-RADS algorithm' above.)
●Frequency of imaging
•We perform the first cross-sectional imaging study four to eight weeks after nonsurgical locoregional therapy. For most patients, we continue imaging monitoring every three months for at least the first year. Thereafter, either cross-sectional imaging could be performed every six months. (See 'Guidelines for post-treatment imaging' above.)
Radiographic responses may be delayed in patients treated with radioembolization or stereotactic body radiation therapy, and we delay the first post-treatment radiographic assessment for disease response to three months. (See 'Guidelines for post-treatment imaging' above.)
•For patients receiving systemic therapy, radiographic assessment for tumor response is typically carried out after the initial six to eight weeks of therapy, and then every two to three months thereafter.
●Monitoring serum AFP – For the subset of patients with an elevated alpha-fetoprotein (AFP), this marker can be useful to follow response; elevations may precede radiographic imaging changes of progression by several weeks or months. (See 'Tumor markers' above.)
●Response criteria for hypervascular HCC – For patients with typical hypervascular tumors who are treated with nonsurgical locoregional or systemic therapy, response criteria that assess tumor viability by cross-sectional imaging are typically preferred over those that measure tumor dimension. Response criteria based on tumor dimension do not incorporate tumor viability and the extent of necrosis caused by locoregional or systemic therapies. (See 'Measuring tumor dimensions versus tumor viability' above.)
•Following systemic therapy – For patients treated with systemic therapy, treatment response is typically classified using either the modified Response Evaluation Criteria In Solid Tumor for HCC (mRECIST) (table 1), or unidimensional Response Evaluation Criteria In Solid Tumors 1.1 (RECIST 1.1) (table 2). (See 'Modified RECIST for HCC' above.)
•Following immunotherapy – Patients treated with immunotherapy are assessed using consensus based criteria for response to immunotherapy (iRECIST) (table 3). (See 'Response criteria for immunotherapy' above.)
•Following locoregional therapy – For patients treated with locoregional therapy, treatment response is classified using LI-RADS (algorithm 2). (See 'LI-RADS algorithm' above.)
●Response criteria for nonenhancing atypical HCCs – Standard RECIST 1.1 guidelines are used for nonenhancing atypical lesions, including poorly differentiated HCC and diffuse-type HCC. Radiographic response criteria that assess tumor viability cannot be used for atypical HCC. (See 'Response Evaluation Criteria In Solid Tumor (RECIST)' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
