Trial Design and Endpoints in Hepatocellular Carcinoma:... : Hepatology (original) (raw)

Potential conflict of interest: Dr. Llovet consults for and received grants from Bayer, Eisai, Ipsen, and Bristol‐Myers Squibb. He consults for Merck, Celsion, Eli Lilly, Roche, Genentech, Glycotest, Nucleix, Can‐Fite, Sirtex, and AstraZeneca. He received grants from Boehringer Ingelheim. Dr. Villanueva consults for NGM, Gilead, and Nucleix. He advises Fuji Wako and Exact Sciences. Dr. Marrero consults for Glycotest. Dr. Meyer consults for Ipsen, Bristol‐Myers Squibb, MSD, Roche, Eisai, and Beigene. He received grants from Bayer and BTG. Dr. Galle consults for, advises, is on the speakers’ bureau for, and received grants from Bayer. He consults for, advises, and is on the speakers’ bureau for AstraZeneca, Bristol‐Myers Squibb, Eisai, Ipsen, Eli Lilly, MSD, Roche, and Sirtex. Dr. Kudo consults for, is on the speakers’ bureau for, and received grants from Eisai, Ono, MSD, Bristol‐Myers Squibb, Roche, Bayer, EA Pharma, Gilead, Otsuka, Sumitomo, Dainippon, Taiho, Takeda, AbbVie, and Eli Lilly. Dr. Zhu advises Eli Lilly, Bayer, Eisai, Merck, Sanofi, Exelixis, and Roche. Dr. Finn consults for AstraZeneca, Bayer, CStone, Bristol‐Myers Squibb, Eisai, Eli Lilly, Pfizer, Merck, Novartis, Genentech, and Roche. Dr. Roberts advises and received grants from Bayer, Exact Sciences, and Gilead. He advises GRAIL, QED, and TAVEC. He received grants from BTG, Glycotest, RedHill, TARGET, and Wako.

Hepatocellular carcinoma (HCC) is the fourth leading cause of cancer‐related mortality worldwide.(1,2) This neoplasm has some unique characteristics. It occurs in most cases complicating underlying cirrhosis, has specific noninvasive criteria for diagnosis, follows a unique staging system, and historically has been resistant to conventional chemotherapy. Several treatments have been adopted as standard of care according to clinical practice guidelines, including potentially curative therapies (i.e., resection, liver transplantation, and local ablation) for early tumors, transarterial chemoembolization (TACE) for intermediate‐stage tumors, and systemic drugs for advanced tumors in the front line (sorafenib and lenvatinib) and second line (regorafenib, cabozantinib, and ramucirumab).(3‐5) Life expectancy has improved progressively in all stages of the disease. Effective implementation of surveillance for patients at risk of developing HCC and access to current proven therapies have been milestones. Overall, median survival times beyond 5 years are expected for early stages, about 20‐30 months for intermediate stages, and 10‐16 months for advanced‐stage HCC(3,4,6) (Fig. 1). Drugs and combinations continue to enter the research arena to address unmet medical needs. All these research activities require precise endpoints and tools for measuring clinical benefits.

FIG. 1:

Modified BCLC staging system considering effective therapies in advanced stages (modified and updated from EASL guidelines( 3 )). Management of patients with HCC is guided by the BCLC staging system, which takes into account both tumor extent and the severity of the underlying liver disease and defines five prognostic subgroups with respective treatments. Treatment for early‐stage tumors is with curative intent, and options include RFA, hepatic resection, and liver transplantation. Patients with intermediate or advanced HCC are candidates for chemoembolization or systemic therapies, respectively. *Patients with end‐stage liver disease if Child‐Pugh class C should first be considered for liver transplantation. **Patients with preserved hepatic function Child‐Pugh class A with normal bilirubin and no portal hypertension are optimal candidates for hepatic resection. ‡Atezolizumab plus bevacizumab has been approved as new first‐line treatment for advanced HCC. Nonetheless, sorafenib and lenvatinib are still considered first line options when there is a contraindication for the combination treatment.( 134 ) Abbreviations: DDLT, deceased donor liver transplantation; LDLT, living donor liver transplantation; M1, distant metastasis; N1, lymph node metastasis.

Thus, clinical trial design has become a major focus of attention in HCC research. Because randomized controlled clinical trials (RCTs) are the main source of evidence for drug approvals in oncology, it is of paramount relevance to understand the critical endpoints and tools for measuring them, as well as optimal strategies for identifying and enrolling target populations and for patient stratification. It has become evident that a deep understanding of factors determining HCC outcomes and trial design is required to achieve optimal results. There are trials with a noninferiority design that have been positive and led to drug approval, while others demonstrate superior outcomes in surrogate endpoints, such as progression‐free survival (PFS) or patient‐reported outcomes (PROs), but not in the primary endpoint of overall survival (OS). Some recent trials have also been reported to be statistically negative but “clinically positive.” It is important to understand the reasons for the failure or success of a trial in order to move the field forward. In addition, while several positive phase 3 trials for advanced HCC have recently changed clinical practice,(7‐11) no major advances have occurred in the area of surveillance and early detection, adjuvant therapies after resection/ablation, or management of intermediate‐stage HCC over the last 15 years. The lack of effective drugs/devices may be the cause of negative studies in these areas, but suboptimal trial design may also have jeopardized the likelihood of a positive result. With this challenge in mind, the Hepatobiliary Neoplasia Special Interest Group of the American Association for the Study of Liver Diseases (AASLD) organized a single topic conference in Atlanta in 2019 to address these issues. This position paper summarizes the major concepts discussed in the conference with the aim of updating the proposals previously reported by a similar AASLD panel in 2008.(12)

Overview of Trial Design and Endpoints

Clinical trials are essential to establish the clinical efficacy of therapeutic interventions. They are instrumental in developing clinical practice guidelines and form the basis for evidence‐based medicine.(13) An adequate clinical trial design is crucial as an effective drug can be discarded due to a poor trial design and vice versa. The main considerations when designing a clinical trial are to (1) select a well‐defined target patient population (i.e., inclusion and exclusion criteria), (2) prespecify clear endpoints (primary and secondary) and a data analysis plan, (3) specify randomization and allocation methods, and (4) secure efficacy of randomization (stratification at enrollment for prognostic variables). Based on these and other variables, the quality of clinical trials can be quantified using different scores such as the Jadad score,(14) the Delphi List,(15) the CONSORT statement,(16,17) and the Cochrane Back Review Group criteria.(18) Until the Sorafenib HCC Assessment Randomized Protocol (SHARP) trial,(11) which established the benefit of sorafenib in advanced‐stage patients with HCC, the quality of the trials conducted in HCC was commonly modest. A systematic review found that only 50% of the clinical trials reported between 2002 and 2005 in HCC were deemed high quality per the modified Jadad score.(19) The 2008 position paper resulting from the AASLD conference provided a useful framework for academic centers, industry partners, and regulators on the design of trials in HCC.(12) Subsequently, the quality of clinical trials assessing systemic therapies has significantly improved. There has been less activity in terms of high‐end clinical trials in other treatment areas. This position paper will extensively discuss the singularities of trial design in every clinical aspect of HCC management.

In clinical trials, the benefit of an intervention is quantified using endpoints, which are predefined events that, once reached, exclude the patient from further evaluation within the trial. There are three main types of endpoints: hard, surrogate, and patient‐reported, all extensively described elsewhere.(20) Hard endpoints are well defined and easy to measure objectively. The archetypes of a hard endpoint are OS or cancer‐related survival. Surrogate endpoints, such as PFS or time‐to‐progression (TTP) partially rely on the quantification of tumor response, generally using imaging techniques and prespecified criteria.(21) Surrogate endpoints are more vulnerable than hard endpoints, but they have several advantages including their convenience in terms of event accumulation and trial feasibility. Patient‐reported endpoints, sometimes referred to as soft endpoints, are subjective measures such as quality of life (QoL), in most instances obtained from questionnaires. Overall recommendations of trial design and endpoints in HCC are detailed in Table 1, whereas expected outcomes for standard of care therapeutic interventions within these trials are summarized in Table 2.

TABLE 1 - Recommendations for Trial Design and Endpoints in Patients With HCC by AASLD Panel of Experts

*Not always recommended.

†Asian guidelines recommend additional treatments for intermediate HCC: Japan, hepatic arterial infusion chemotherapy; China, resection; Taiwan, resection/Y90. For advanced HCC: Japan, hepatic arterial infusion chemotherapy/resection/TACE; China, FOLFOX4, resection/TACE; Korea, TACE.

Abbreviations: EHS, extrahepatic spread; HCV, hepatitis C virus; MELD, Model for End‐Stage Liver Disease; MVI, macrovascular invasion; RFS, recurrence‐free survival; TTR, time to recurrence.

TABLE 2 - Expected Outcomes Reported in Phase 3 Trials in HCC Research

*ORR per mRECIST.

†Atezolizumab+bevacizumab is expected to be first‐line, while sorafenib and lenvatinib will be second‐line therapies, see Fig. 1.

Abbreviation: RFS, recurrence‐free survival.

Endpoints

OS is defined by the time between patient randomization and death from any cause. OS is usually recommended as the primary endpoint for randomized phase 3 clinical trials.(12) OS is the endpoint most frequently used by regulatory agencies to approve drugs as it is objective and clinically relevant. In HCC, as most patients suffer from concomitant cirrhosis, death can result from competing risks, mainly liver toxicity and failure. This fact underscores the need for detailed assessments of safety with any intervention in this population. It is important to capture adverse events in early‐phase clinical studies as well as in larger randomized studies. Failures in phase 3 studies have been seen from agents that are more toxic in an HCC population than in other tumor types.(22) The competing risk of cirrhosis can introduce bias when evaluating the antitumoral activity of a therapeutic intervention, but it can be easily controlled by imposing stringent inclusion criteria in terms of liver function (i.e., Child‐Pugh score A without hepatic decompensation). OS has some limitations such as the long follow‐up time required to capture the number of events needed to verify significantly improved survival in the experimental arm.(23) This can be a critical limitation when exploring interventions at early or intermediate stages. Also, OS can be confounded by sequential therapies received by patients after tumor progression, which, for instance, affected 30% of patients enrolled in the lenvatinib trial(7) and up to 50% of patients in the Checkmate 459 comparing nivolumab versus sorafenib.(24) Hence, there is a need to develop surrogate endpoints, which are defined as outcomes not inherently meaningful from the clinical standpoint but thought to accurately predict hard outcomes such as OS.(25)

The main surrogate endpoints in oncology are PFS, TTP, and objective response rate (ORR) (Table 1). PFS is the time between patient randomization and death or radiological tumor progression, whichever occurs first. There are different tools to assess tumor response with imaging. The most established tool for measuring tumor response in oncology is the set of Response Evaluation Criteria in Solid Tumors (RECIST) criteria,(26) initially developed to evaluate response to cytotoxic drugs. These criteria were adapted to account for HCC singularities in the modified RECIST (mRECIST) version, which incorporates viable tumor detected with arterial enhancement as a key component to evaluate response.(21,27) Using mRECIST criteria increases the percentage of subjects who achieve an objective response compared to standard RECIST, as shown in different studies of systemic therapies.(28‐32) A recent meta‐analysis evaluated the power of PFS to predict OS in phase 3 trials testing systemic therapies in advanced HCC.(20) The study found a moderate correlation between PFS and OS in 21 RCTs. The authors proposed a conservative surrogate threshold of ≤0.6 for a hazard ratio (HR) of PFS to predict clinically relevant improvements in OS.(33) TTP is defined as the time elapsed between patient randomization and radiological tumor progression. Scheduling repeated radiological assessment of response every 6‐8 weeks is mandatory for patients included in trials. Data from SHARP and subsequent studies challenge the implied correlation between TTP and OS. The type of progression may also have clinical implications.(34) Survival is worse if patients develop an extrahepatic lesion and/or vascular invasion as opposed to tumor progression resulting from growth of an existing lesion or a new intrahepatic lesion. Lastly, ORR is the percentage of patients with an objective tumor response, and its correlation with OS is worse than for PFS or TTP.(20) This is partially inherent to the use of odds ratios instead of HRs for ORR and to the fact that only a small proportion of patients achieve an objective response (≤27%, for approved drugs in advanced HCC), which is the event that correlates with OS(31) (Table 2). Nonetheless, ORR has been reported as an independent predictor of survival in early HCC treated with radiofrequency ablation (RFA), intermediate HCC treated with TACE, and advanced HCC treated with a tyrosine kinase inhibitor (TKI).(27) The impact of the duration of response, which has been reported to be around 12 months for checkpoint inhibitors versus less than 6 months for TKIs, has not yet been properly incorporated into response assessment. The same is true for small reductions in tumor size not reaching standard thresholds for objective response. In some cases, duration of disease control may be more clinically relevant than the extent of reduction in tumor size. Also, in the case of immune checkpoint inhibitors, tumor response can have a longer lag time compared to other molecular therapies and can even mimic progression shortly after treatment initiation (i.e., pseudoprogression(35)). This has led to the development of immune‐related response criteria,(36) which require confirmation of progression at least 4 weeks after progressive disease is first documented.

Surrogate endpoints are frequently used by the Food and Drug Administration (FDA) to approve drugs under the accelerated program, which was initially developed to facilitate early access to antivirals during the worst years of the human immunodeficiency virus epidemic.(37) In HCC, the FDA has used ORR and duration of response to grant accelerated approval of the immune checkpoint inhibitors nivolumab,(38) pembrolizumab,(39) and recently the combination of ipilimumab and nivolumab.(40) Accelerated approval is not universal and includes some subjectivity from regulators in regard to the strength of the evidence to support approval without a randomized phase 3 study.(24,41) In addition, while the ORR and other surrogate endpoints may be used to support regulatory approval, they do not necessarily support inclusion in guidelines, which often adhere to a higher level of evidence. Despite their common use, surrogate endpoints are vulnerable to interpretation bias. Besides the strength of the endpoint, it is key to determine when the benefit provided by a therapy is really clinically meaningful. This can be controversial, depending on factors such as the perceptions of patients, providers, health insurers, and regulators. In HCC, there is no set threshold that defines a clinically meaningful benefit, but some authors have suggested an HR cutoff of OS ≤0.8 as a starting point for clinical trial design.(42) In fact, all positive trials in HCC have led to significant differences in survival with an HR below this threshold.

Surveillance for HCC: Design and Endpoints

Surveillance for HCC is one of the milestones advancing the management of HCC, despite the fact that there are no unquestionable data directly supporting a decrease in cancer‐related death in persons on surveillance.(43) Ultrasound (US), with or without alpha‐fetoprotein (AFP), performed every 6 months is the current standard and is recommended for surveillance of patients with cirrhosis of any cause or chronic hepatitis B without cirrhosis above a regional and gender‐appropriate age cutoff determined by expert liver societies.(3,4) Overall, the implementation of those programs to all targeted populations is modest, and current data report detection of HCC in the setting of surveillance in between 30% and 50% of cases.(44) In meta‐analysis, the pooled sensitivity and specificity of US alone has been shown to be 53% (95% confidence interval [CI], 35‐70) and 91% (95% CI, 86‐94), respectively, while the combination of US and AFP has a sensitivity of 63% (95% CI, 48‐75) and a specificity of 85% (95% CI, 77‐89).(45) Due to the relatively low sensitivity and specificity of this approach for detecting early‐stage HCC, particularly in North America, where high rates of central obesity decrease the performance of US, a recent study showed that this strategy leads to 27% of patients with cirrhosis experiencing harms such as follow‐up testing (computed tomography [CT], magnetic resonance imaging [MRI], liver biopsy) performed for false‐positive or indeterminate results.(46) Further, due to low implementation of comprehensive strategies for HCC surveillance, more than 60% of HCCs in North America, Europe, Africa, and large parts of Asia, excepting Taiwan and Japan, are diagnosed with intermediate‐stage or advanced‐stage HCC.(47) There is, therefore, an urgent need for better‐performing, low‐cost surveillance strategies in HCC, and accounting for both the benefits and harms of surveillance strategies is important.

Within this overriding context, there is excitement because advances in genetic, epigenetic, proteomic, glycoproteomic, and metabolomic analyses have enabled large‐scale multi‐omic analyses of HCC tissues, circulating tumor DNA (ctDNA), plasma, and serum, resulting in the accelerated identification of biomarkers.(48‐51) Models using standard biostatistical and machine learning and artificial intelligence approaches are using biomarkers combined with clinical parameters to identify persons at highest risk for HCC. Models and biomarkers under active exploration include the GALAD (gender, age, AFP‐L3, AFP, and des‐carboxy‐prothrombin) score,(52) glycoproteins (fucosylated kininogen),(53) liquid biopsy analyses of ctDNA for differentially methylated regions,(54,55) and imaging with abbreviated MRI.(56) Creating the framework for validation of future surveillance is critically important.

To guide the development and evaluation of surveillance strategies for clinical use, a five‐phase program has been developed by the National Cancer Institute’s Early Detection Research Network for biomarkers that uses human samples (blood or human tissue) as well as imaging tests.(57) Table 3 shows recommended phases of surveillance test validation, including trial design for studies for HCC surveillance. Phase 1 includes biomarker discovery or exploratory studies. Phase 2 studies estimate the ability of a test to distinguish early‐stage HCC from those with cirrhosis without HCC. It is important to test for confounders such as age, etiology of liver disease, and liver function and to have adequate sample size and power. Phase 3 studies enroll at‐risk individuals and follow them for clinical diagnosis of HCC using a prospective‐specimen collection, retrospective‐blinded evaluation (PRoBE) design.(58) The aim is to evaluate, as a function of time before clinical diagnosis, the capacity of the test to detect preclinical HCC;and to define the criteria for a positive surveillance test in preparation for phase 4 and 5 studies. Thus, phases 1‐3 rely on retrospective analysis of stored data and specimens. Phase 4 studies require the test be applied to patients with cirrhosis in the clinical setting to assess test performance in HCC detection and false positive and negative rates. Depending on the test under study, it may be possible to skip phase 4 if the test is already used for patient care, for example, evaluating an MRI for surveillance of HCC. Phase 5 studies are randomized trials comparing the surveillance tests against the standard of care; in the case of HCC, the standard should be US with or without AFP, with the aim of determining whether the test can reduce mortality at the population level.

TABLE 3 - Phases of Surveillance Test Validation in HCC

Adapted from Pepe et al.(57)

When performing surveillance studies in patients with cirrhosis it is important to enrich the at‐risk population in order to achieve a sufficient number of incident HCCs in a reasonable time period. Enriching cohorts with patients of older age, viral hepatitis, male sex, Hispanic ethnicity, history of diabetes, and family history of HCC should be considered.(59,60) Alternatively, known independent factors associated with HCC development are abnormal bilirubin and platelet count <100,000/mm3. There is also a need to study currently important populations such as those with nonalcoholic fatty liver disease–related cirrhosis, those with hepatitis C–related cirrhosis who have achieved a sustained virological response after antiviral treatment, and those with suppressed hepatitis B infection on antiviral treatment. These three specific populations will be the most important etiologically in the next decade, and their HCC incidence rates (around 1%/year) appear lower than in previous at‐risk populations.(61) Methods for risk stratification within these populations will therefore become increasingly important for improving the effectiveness of surveillance strategies and programs. Models such as the REAL‐B and PAGE‐B scores, incorporating male sex, age, alcohol use, baseline cirrhosis, diabetes, platelet count, and AFP, allow improved risk stratification of patients on oral antiviral therapy for chronic hepatitis B and could potentially be incorporated into surveillance programs.(62)

An important potential confounder in studies that compare the performance of biomarkers to current surveillance strategies is the incorporation of imaging by ultrasound or other radiologic modalities into the standard of care. This may confound the results if US is also used as part of the control arm for the study as ultrasonography is itself typically part of the gold standard process for determining whether a patient has HCC. Thus, patients with HCCs that are not visible by US may be falsely determined to be negative for cancer, and a positive biomarker test may erroneously be labeled as a false positive. It is therefore important to use a different high‐accuracy imaging modality such as multiphasic MRI as a gold standard in studies for which US is part of the surveillance strategy. However, use of MRI may add substantial cost to the study and may result in visualization of a number of small indeterminate false positive lesions that are seen on MRI and require follow‐up investigation, a component of the harms associated with surveillance. While studies of the performance of US with or without AFP in the clinical care setting have shown suboptimal performance in detection of HCC in at‐risk individuals, it is not clear what the performance characteristics are for phase 2, 3, or 4 biomarker studies that would meet the threshold for FDA approval as a surveillance test. In general, the FDA guidelines for supporting biomarker qualification recommend that analyses intended to support biomarker qualification should be specified in an analysis plan with a prospective‐retrospective design before analyzing the data. The FDA provides no set quantitative criteria for determining the relationship between the biomarker and clinical outcome, such as diagnosis of HCC, within a particular context of use. Overall, the goals for in vitro diagnostic biomarker studies are that they should produce valid scientific evidence demonstrating reasonable assurance of the safety and effectiveness of the product and protect the rights and welfare of study subjects.(63,64)

Key unmet needs in the field of chemoprevention include an improved understanding of the potential for HCC risk reduction by chemoprevention using commonly used medications such as aspirin and other antiplatelet agents, statins, metformin, and similar agents.(65‐68) In order to build a robust evidence base through chemoprevention trials, a number of key hurdles need to be crossed, including better definition of target populations; trial enrichment or stratification prior to randomization using clinical, genetic, or other molecular risk stratifying strategies; and careful delineation of appropriate and clinically meaningful endpoints for both biomarker‐based and chemoprevention trials. Enrichment of populations included in chemopreventive trials should aim to a reasonable time‐to‐event (occurrence of HCC) endpoint, certainly within the threshold of 5 years. Stratification factors for at‐risk populations have been outlined below and are mandatory to prevent imbalances. Finally, one of the bottlenecks of these trials is that the accepted adverse events for maximum tolerated doses (grade 3 toxicities are unacceptable) are completely different compared to those accepted for primary treatments of advanced tumors, where grade 3‐4 adverse events at the level of 30%‐50% are common for currently accepted drug treatments.

Early HCC Stages: Design of Trials for Resection, Transplantation, and Local Ablation

Hepatic resection is the treatment of choice for patients with preserved liver function (Child class A, bilirubin <1.0 mg/dL, no evidence of portal hypertension) who have a solitary HCC >2 cm without macrovascular invasion(3,4,69) (Fig. 1, Table 4). Outcomes of ideal candidates treated following these criteria are significantly better compared with outcomes not following the guidelines.(70) Recent guidelines accepted expanding criteria to include patients with HCC within the Milan criteria.(3) While 5‐year survival rates are in the range of 70% after resection, recurrence of HCC is also around 70% at 5 years.(71) Early (within 2 years) recurrence is most commonly due to the appearance of preexisting undetected metastatic disease, with the most common site in the remaining liver; late recurrence is predominantly the result of de novo development of HCC in the remaining liver. There is, thus, a critical unmet need for therapy that can reduce the incidence of HCC recurrence after resection. A study demonstrating benefit of retinoid administration(72) was not confirmed in a subsequent multicenter trial(73) and small studies suggesting benefit from adoptive immunotherapy(74) and I‐131 lipiodol embolization of the liver remnant,(75) the results of which have not been duplicated. To this point, all phase 3 high‐quality adjuvant trials conducted in this area have been negative, A large RCT of sorafenib after resection or thermal ablation demonstrated no benefit.(76) Current attention is largely focused on immunotherapy. Treatment of advanced HCC with anti–programmed death 1 (PD‐1) or PD‐1 ligand (PD‐L1) antibodies has consistently yielded responses in the range of 15%‐20%(38,39,41) that are often quite durable. In non‐small cell lung cancer similar response rates are seen in advanced disease, and a neoadjuvant trial for resectable tumors resulted in a roughly doubled response rate.(77)

TABLE 4 - Guideline Recommendations for Treatment According to Levels of Evidence* and Strength of Recommendation†

Treatments accepted in guidelines (EASL(3) and AASLD(4)) and level of evidence (modified from Llovet et al.(119)).

*National Cancer Institute classification: Strength of evidence: level 1, RCT or meta‐analysis; level 2, nonrandomized controlled studies; level 3, case series. Strength of endpoint: A, survival; B, cancer‐specific survival; C, QoL; D, others.

†Table modified from EASL‐EORTC guidelines.(5)

Abbreviations: 3D, three‐dimensional; MWA, microwave ablation; SBRT, stereotactic body radiotherapy.

Phase 3 trials are currently under way with single‐agent immunotherapy or combination therapies. In advanced disease combination therapy, an anti‐PD‐1/PD‐L1 plus either a TKI (e.g., sorafenib, lenvatinib), an anti–vascular endothelial growth factor (VEGF) antibody (e.g., bevacizumab), or a second checkpoint inhibitor (e.g., anti–cytotoxic T lymphocyte antigen 4 [CTLA4] antibody) appears to significantly raise response rates; and if established in the advanced setting, combination therapy will no doubt be studied in adjuvant/neoadjuvant trials. The ultimate hope is that effective adjuvant/neoadjuvant therapy will be able to substantially improve recurrence‐free survival. It is the consensus of the panel that entry criteria for adjuvant/neoadjuvant studies in HCC resection should conform to the criteria for resectability currently espoused in AASLD guidelines(4,69) and prevent a broadening of the tumor eligibility for resection (e.g., multiple tumors, presence of vascular invasion) observed in some currently running adjuvant trials. While all patients undergoing resection for HCC have significant risk of recurrence, studies should stratify for known risk factors including tumor size (>3 cm), microvascular invasion, differentiation degree, and serum AFP > 400 ng/mL (Table 1). Neoadjuvant studies provide a unique opportunity to better understand what factors are associated with response to immunotherapy or lack thereof. Pretreatment biopsy should be mandatory, and thorough characterization of the tumor immune microenvironment should be built into these trials.

Liver transplantation is the treatment of choice for HCC within the Milan criteria in patients who are not candidates for resection(78) (Fig. 1, Table 4). These criteria led to median OS of 10 years and a recurrence rate of <20%. In the United States it has been accepted that patients with more extensive disease (one nodule between 5 and 8 cm, two or three nodules ≤5 cm, or four or five nodules <3 cm with sum of diameters <8 cm) downstaged to Milan criteria are acceptable for transplantation.(79) Downstaging is not accepted by European guidelines, although it is performed in some countries such as Italy. A significant number of patients who enter the waiting list or a downstaging protocol drop out and do not ultimately undergo transplantation. Locoregional therapies using thermal ablation or TACE have been the modalities traditionally applied to maintain HCC within the Milan criteria while awaiting transplant or to downstage patients to eligibility. With the advent of effective systemic therapies, their role in the pretransplant setting vis‐à‐vis locoregional treatment warrants exploration in clinical trials. Locoregional treatment should be the control arm, compared to systemic therapy either alone or in combination with locoregional, with the primary endpoint of dropout/transplantability. Stratification should be according to whether patients were initially within or beyond the Milan criteria or downstaged to Milan and baseline AFP levels >400 ng/mL.

Treatment of HCC recurrence following transplantation is largely unstudied. The rate of recurrence in properly selected patients is low (10%‐20%), and these patients have been routinely excluded from studies of systemic therapies. TKIs have been shown to be safe and are commonly used in an uncontrolled manner.(80) There is considerable reluctance to use immunotherapy with anti‐PD‐1/PD‐L1 antibodies due to reports of treatment‐related organ rejection, though there are reports of successful treatment.(81) As HCC now accounts for nearly 25% of liver transplants in the United States, it is time for trials to be implemented studying treatment of posttransplant HCC recurrence.

Local ablation is the mainstay treatment for nonsurgical candidates with early‐stage HCC(3,4) (Fig. 1, Table 4). Tumor size (up to 4‐5 cm), number (up to three tumors), and location (accessiblility with US, CT, or MRI guidance) limit the applicability of percutaneous ablation. Several randomized studies have demonstrated a significant benefit of RFA over percutaneous ethanol injection in terms of complete response rate and time to recurrence.(82,83) Consequently, RFA is the standard ablative therapy at early stages (Table 1). Median OS with RFA is 60 months, with a recurrence rate ranging 50%‐70%.(3,4,82,83) AASLD and European Association for the Study of the Liver (EASL) guidelines have adopted RFA as a front‐line therapy for single tumors <2 cm, but in tumors beyond this threshold resection it remains as a first treatment option.(3,4) Randomized phase 3 trials are scarce in this arena and are mostly currently focused on adjuvant therapies to prevent recurrence than in challenging the ablative treatment. Microwave ablation has largely supplanted RFA in the United States,(84) whereas ethanol injection is restricted to HCC <2 cm in difficult locations. Cryoablation and irreversible electroporation are still under investigation.(3,4,85) Clinical benefit associated with the use of thermally sensitive carriers loaded with liposomal doxorubicin in conjunction with RFA is currently tested in phase 3.

Overall, the main criteria for trial design in the neoadjuvant/adjuvant after resection/local ablation or liver transplantation setting are as follows (Table 1):

1. Target populations for neoadjuvant and adjuvant trials: For resection, trials should include patients meeting current AASLD guidelines and should not include patients with more advanced HCC, e.g., macrovascular invasion. For transplantation, trials should include patients meeting criteria for listing (i.e., Milan criteria) or meeting established criteria for entry into downstaging protocols. For local ablation the target population should follow AASLD guidelines.
2. Endpoints: The appropriate endpoint for adjuvant trials in the setting of either resection or transplant is recurrence‐free survival or time to recurrence. For neoadjuvant trials, pathological response or 1‐year recurrence can also be considered. For treatments challenging locoregional therapies, OS remains the primary endpoint, but PFS is also recommended as a coprimary endpoint. Secondary endpoints should at least include objective response rates.
3. Stratification prior to randomization: Appropriate stratification parameters for neoadjuvant/adjuvant studies in the setting of early‐stage HCC should include geographical region, tumor size and number, AFP > 400 ng/mL, type of curative treatment, and pathological features of high risk (size >3 cm, microvascular invasion, differentiation degree, and tumor satellites).
4. Control arms: For neoadjuvant/adjuvant studies in the setting of resection, a placebo control arm is appropriate. Adjuvant studies in transplantation should also include placebo controls. Defining the control arm for neoadjuvant studies in transplantation remains problematic as there is no evidence‐based standard, but there is a general acceptance of the need to include locoregional therapies to limit tumor progression in patients awaiting transplant that precludes including placebo or untreated patients. Control arms for devices or drugs challenging local ablation should be radiofrequency. Of note, because RFA has been considered effective in nodules up to 4 cm, trials exploring treatments for single nodules beyond this size should consider chemoembolization as the best standard control.
5. Unmet needs: HCC recurrence rates after resection or local ablation are unacceptably high. Key needs include biomarkers to improve case selection and effective neoadjuvant/adjuvant therapies. With regard to transplantation for HCC, key needs include definition of optimal neoadjuvant (waiting list) strategies and identification of useful biomarkers to refine candidate selection beyond algorithms based on tumor size and number.

Trial Design and Endpoints in Intermediate‐Stage HCC

TACE was established as the standard of care for intermediate‐stage HCC in 2002 following the publication of two small RCTs for which OS was the primary endpoint (Table 4). The first trial, conducted in Barcelona, demonstrated an HR of 0.47 (95% CI, 0.25‐0.91; P = 0.025) in favor of TACE and a 2‐year survival of 63% compared with 23% for supportive care.(86) In the second, TACE was associated with an improvement in 2‐year survival from 11% with supportive care to 31% with TACE and a reduction in relative risk of death: 0.49 (95% CI, 0.29‐0.81; P = 0.006).(87) Response using World Health Organization criteria was evaluated as a secondary endpoint and was shown to be associated with a better survival.(86) On the basis of these trials and a subsequent meta‐analysis,(88) the Barcelona Clinic Liver Cancer (BCLC) algorithm recommends TACE for those with intermediate‐stage disease HCC defined by liver confinement, multinodular disease, performance status of 0, Child‐Pugh A or B cirrhosis, and absence of portal vein invasion(3,5) (Fig. 1). Chemoembolization was subsequently adopted by AASLD and EASL guidelines of management of HCC, and no other therapy has so far replaced this standard of care. However, since 2003(86,88) there have been further innovations, guidelines, and therapeutic advances which need to be considered in the design of current and future trials. Finally, radioembolization with Y90 for intermediate HCC has produced positive efficacy signals coming from phase 2 investigations,(89) but they have not been adopted by guidelines awaiting phase 3 positive data for this specific population.

Eligibility Criteria and Stratification Factors

It is increasingly recognized that the BCLC B stage is heterogeneous, and this likely accounts for the wide spectrum of reported survival outcomes, which range from 12 to 48 months. Consequently, there have been several proposals to subdivide the BCLC group; but to date, none have been widely adopted.(90,91) Additionally, patients who have a performance status of 1 but otherwise conform to the BCLCB criteria are routinely treated with TACE, and many clinicians regard Child‐Pugh B disease as a relative contraindication. Applicability of TACE in BCLC B is 50%, with the excluded patients having relative contraindications for the procedure due to advanced liver dysfunction or technical issues.(92) Recent large RCTs have included patients with performance status 0‐1, Child‐Pugh A, and absence of portal vein thrombosis (Tables 1 and 2).(93‐96) Stratification factors have been less consistent with the exception of AFP, for which a threshold of 400 ng/mL has been commonly applied. Composite and fully objective prognostic systems may provide a more feasible and consistent method by which to stratify patients. The albumin–bilirubin (ALBI) score allocates a grade based on bilirubin and albumin and provides a more objective measure of liver function compared with Child‐Pugh class.(97) A direct comparison between ALBI and Child‐Pugh has shown that ALBI grade 1 is 92% Child‐Pugh A5, ALBI 2 spans a wide range from A5 to B9, and ALBI 3 is B7 and above.(98) However, tumor characteristics such as size and AFP are also prognostic, and this has been addressed by the hepatoma arterial‐embolization prognostic (HAP) score, which provides a four‐class prognostic system using bilirubin, albumin, tumor size, and AFP as categorical variables.(99) The HAP score has been validated in the TACE‐treated population, most recently within a cohort of 3,000 patients.(100) Applying the HAP score resulted in four distinct groups, with survival ranging from 33 months for HAP A to 12 months for HAP D. HAP appears to be a simple and robust stratification factor that might be incorporated into TACE trials

TACE Procedure

The TACE technique provides another source of heterogeneity and potential bias.(101) There remains no consensus regarding the optimal embolic particle, the role of lipiodol, or the type of chemotherapy used. Indeed, there are no trials demonstrating the superiority of TACE over bland particle embolization (transarterial embolization), and a meta‐analysis of five trials including 582 patients showed no difference in survival, although TACE was the only procedure superior to placebo.(102) It is unlikely that further technical innovation to the TACE procedure will result in significantly improved outcomes, and the future generation of TACE trials will continue to evaluate the combination of TACE and systemic therapy or to compare TACE with systemic therapy. In both cases, TACE will be the control arm, and it is important that this is standardized. To achieve this, some of the recent randomized trials have mandated use of drug‐eluting beads (DEB TACE).(93,94) Trials comparing DEB TACE with conventional TACE have failed to show a survival benefit, but systemic toxicity from chemotherapy is reduced with DEB TACE.(103,104) If the technique is not standardized, stratification according to center is an alternative way to reduce bias. Another area of contention is the schedule of TACE administration. In clinical practice, TACE is usually performed on demand according to radiological response rather than according to a fixed interval, and it is reasonable to recapitulate this in clinical trials. However, an effective systemic therapy may reduce the requirement for TACE. In the TACE‐2 trial, there were 18% fewer TACE procedures performed in 12 months in the sorafenib arm compared with the placebo arm(93); and in the Orantinib versus Placebo Combined with TACE in Patients with Unresectable HCC (ORIENTAL) trial, the median number of procedures was 3.2 versus 3.7 in the orantinib and placebo arms, respectively.(96) Recording the number of procedures over the first 12 months or the mean number of procedures should be considered mandatory for randomized trials of TACE versus TACE plus systemic therapy. In this sense, the reduction in frequency and number of TACE procedures may have implications for health economics and preservation of liver function.

Response Assessment

Radiological response is an important indicator of therapeutic activity and can be a surrogate marker of long‐term outcomes. Response assessment is addressed in the next section, but a few concepts regarding locoregional therapies are summarized here. In the TACE‐related population, mRECIST demonstrated a higher response rate compared with RECIST 1.1.(105) Moreover, there was a significant association between survival and overall response according to mRECIST but not with RECIST 1.1. The association between mRECIST response and survival has subsequently been confirmed in multiple other studies, and a recent meta‐analysis of seven studies including 1,357 patients reported an HR for survival of 0.39 (95% CI, 0.26‐0.61) for those with mRECIST response.(106) Unfortunately, not all the recently reported phase 3 studies reported response, and only TACE‐2 ascertained response by both RECIST 1.1 and mRECIST. Best response by RECIST 1.1 was higher than first response but still less than response by mRECIST. Guidelines recommend capturing response per mRECIST in clinical practice and both RECIST 1.1 and mRECIST as secondary endpoints trials targeting intermediate‐stage tumors.(3)

Primary Endpoints

In recent trials, OS for intermediate‐stage patients receiving TACE was of 21‐33 months(93‐96) (Table 2). Over the past 10 years, there have been major advances in systemic therapy, and many patients now transition from TACE to first‐line and increasingly second‐line systemic therapy. In TACE‐2, patients were unblinded on progression, and 36% of those on placebo subsequently received sorafenib.(93) Similarly, in the Brivanib versus Placebo as Adjuvant Therapy to TACE in Patients with Unresectable HCC (BRISK TA) trial, 21% of placebo‐treated patients had a postprogression systemic therapy(95); and in the ORIENTAL trial, 66% of patients in the placebo arm received poststudy therapy.(96) Use of postprogression therapy may confound OS as an endpoint and increases the duration of follow‐up required to meet the survival endpoint. To address this, a variety of surrogate endpoints have been proposed including PFS, TTP, time‐to‐disease progression, time‐to‐extrahepatic spread and vascular invasion (TTES/VI), and time‐to‐unTACEable progression (TTUP). Recent trials reporting these potential surrogates in addition to survival have allowed evaluation of their performance. The BRISK TA trial reported a promising HR of 0.61 for TTP, but the trial missed its primary endpoint for survival (HR, 0.9).(95) Overall, the correlation coefficient of TTP and OS is 0.77. A major limitation of TTP is that it fails to capture death, which is an important indication of toxicity as well as lack of efficacy. By contrast, PFS, which is the most commonly applied surrogate endpoint used in oncology, captures disease progression and death and has been reported to correlate with OS in the TACE 2 trial. Composite endpoints have also been explored. TTES/VI (or macrovascular invasion and extrahepatic spread) showed promising HRs of 0.64 and 0.62 in the BRISK TA and Sorafenib or placebo plus TACE (SPACE) trials that did not correlate with OS benefit.(94,95) Particularly, TTUP, a composite endpoint defined as failure of response after treatments, or emerging contraindications for TACE was tested in the SPACE trial but failed to identify benefits for the combination of TACE plus sorafenib versus TACE (HR, 1.586). Recently, other endpoints were incorporated into the TACE Plus Sorafenib as Compared with TACE Alone in Patients with HCC (TACTICS) trial comparing TACE plus sorafenib versus TACE alone.(107,108) In this study, PFS and OS were coprimary end points; but progression was defined as unTACEable progression, and response evaluation criteria in cancer of the liver(109) were used to define progression rather than RECSIT 1.1 or mRECIST. Applying these criteria, PFS was superior in the combination arm (HR, 0.59; 95% CI, 0.41‐0.87; P = 0.006), but further follow‐up is required to establish wether this translates into a survival benefit. In the meantime, for RCTs testing devices alone or in combination with systemic therapies, it is recommended that PFS should be the coprimary endpoint along with OS, while ORR should be included as a secondary endpoint (Tables 1 and 2). Additional composite endpoints can be included as exploratory endpoints until they are properly validated.

A challenging question for the future is how TACE compares to systemic therapy. TACE was developed at a time when systemic therapy was virtually nonexistent. With the advent of first‐line, second‐line, and even third‐line systemic therapies and achievement of OS beyond 2 years in selected patients receiving two lines of therapy(23) systemic therapy can be discussed not only following TACE but also as an alternative to TACE. This is particularly relevant as transarterial therapies impair liver function and may render many patients no longer eligible for systemic therapy. For patients with limited tumor burden and nodules accessible superselectively by TACE, locoregional TACE may still be the best approach. In contrast, patients exceeding the up‐to‐seven criteria may be better suited for clinical trials exploring up front systemic therapy.(110) To answer this question a head‐to‐head comparison of TACE versus systemic therapy (or versus TACE plus systemic therapy) in defined patient subgroups will be needed, making the endpoint discussion even more complex.

Radiologic Assessment of Response

The RECIST criteria are the standard imaging approach for assessing tumor response in oncology. The original RECIST panel acknowledged that amendments could be needed for tumors with unique complexities and for evaluating noncytotoxic drugs.(111) Both issues are highly relevant for HCC: (1) the association of HCC with an underlying chronic liver disease complicates image assessment because pathologic and hemodynamic changes in cirrhosis and extrahepatic manifestations of chronic liver disease may mimic tumor progression and (2) nonsurgical treatments for HCC, including locoregional and systemic therapies, achieve improvements in survival without inducing sizeable tumor shrinkage, frustrating attempts to capture tumor response using standard RECIST metrics.(12)

In 2010, mRECIST criteria for HCC were proposed,(21) addressing confounding factors related to cirrhosis using specific amendments for the assessment of lymph nodes, ascites, portal vein thrombosis, and newly detected hepatic nodules (Table 5). These recommendations were made primarily to prevent “overcalls” of progressive disease. In addition, the absence of substantial tumor shrinkage was addressed by introducing the concept of “viable tumor” in the measurement of intrahepatic HCC lesions, enabling the classification of treatment‐induced intratumoral necrosis in the absence of significant changes in overall tumor diameter as objective responses.(21)

TABLE 5 - Basic Concepts and Key Points for Standard RECIST 1.1 and mRECIST Assessment in HCC(27)

Abbreviations: CR, complete response; PD, progressive disease; PR, partial response; SD, stable disease.

During the past decade, mRECIST for HCC has been used extensively in HCC clinical research, and its performance has been reviewed elsewhere.(27) The proposed mRECIST refinements for assessment of lymph nodes, ascites, portal vein thrombosis, and newly detected hepatic nodules were progressively incorporated into radiology charters of HCC clinical trials, even when the criteria were named RECIST or RECIST 1.1.(112) This process homogenized radiologic interpretation of these findings, improving consistency and reliability in the assessment of tumor progression. Consequently, recent studies have reported similar results for standard RECIST 1.1 and mRECIST in assessment of progression‐driven endpoints, such as PFS and TTP.(7,8) Currently, the main difference between standard RECIST and mRECIST is the approach to measurement of intrahepatic lesions, which primarily affects the ability to capture an objective response. Use of the mRECIST viable tumor concept results in identification of 2‐3 times more responders than standard RECIST, not only in patients receiving locoregional treatments but also in those receiving systemic therapies.(7,32)

With the advent of immune checkpoint inhibitors, changes to the RECIST model have been proposed.(35,36,113‐115) Response to immunotherapy can manifest after imaging features that meet current RECIST criteria for progression. Pseudoprogression has been defined as increase in tumor size of existing lesions or the appearance of new lesions, followed by a response.(35) Differentiating pseudoprogression from true progression is challenging but important: while early discontinuation of an effective drug is not desirable, continued long‐term treatment with a noneffective drug past true progression might delay the initiation of potentially effective therapies. Pseudoprogression has been described as a marginal event in phase 3 investigations with anti PD‐L1/PD‐1 checkpoint inhibitors in HCC. The incidence of this phenomenon with anti‐CTLA4 and other inhibitors is unknown.

Limited information is available on use of immune‐related criteria in HCC. In a phase 2 study of 104 patients who received pembrolizumab in second line after sorafenib, the use of immune‐related RECIST (irRECIST) did not affect response rate or time to response compared to mRECIST; however, median PFS was 7.0 months (95% CI, 4.9‐8.0) when assessed by irRECIST versus 3.2 months (95% CI, 2.2‐4.1) when assessed by mRECIST.(116) In a phase 2b study(117) investigating a vaccinia virus–based oncolytic immunotherapy, pexastimogene devacirepvec, in advanced HCC, changes to mRECIST were implemented because the treatment induces a flare with swelling and edema.(118) These changes included the confirmation of progression at 4 weeks, either by further increase in size or by additional signs of progression such as emergence of new lesions.(117) Overall, to assess response to checkpoint inhibitors or immunotherapies in HCC, evaluation by CT/MRI at 8‐12 weeks after treatment can be recommended, as opposed to the usual interval of 6‐8 weeks for TKIs. This window was used in phase 2 studies testing nivolumab (12 weeks)(38) and pembrolizumab (9 weeks),(116) where the phenomenon of pseudoprogression was reported as a marginal event.

Design and Endpoints for Systemic Therapies in HCC

Standard of Care with Systemic Therapies in HCC

Current estimates suggest that around 50% of patients with HCC will receive effective systemic therapies during their life span.(3,119,120) Several trials have tried to show survival benefits of systemic agents in advanced disease (Tables 2 and 4), a traditionally challenging setting due to the limited efficacy and high toxicity of conventional systemic chemotherapy.(121‐124) Randomized studies for antiestrogen therapies also failed to prove any clinical efficacy.(125) In 2008, the landmark SHARP trial assessing the multi‐TKI sorafenib was the first to significantly improve survival with manageable adverse events.(11) Afterward, five treatments have succeeded, while several other drugs failed.(7,22,122,126‐133) In first line, atezolizumab (anti‐PD‐L1 inhibitor) plus bevacizumab (VEGFA inhibitor) have been shown to be superior to sorafenib in a recently reported RCT.(134) The study was stopped at the first interim analysis by showing an HR of 0.58 for OS (median not reached for combination versus 13.2 months for sorafenib) and an HR of 0.59 for PFS (median 6.8 months for combination versus 4.3 for sorafenib). These results will pose this combination as standard of care first‐line therapy for advanced HCC. Also, lenvatinib (multikinase inhibitor: VEGF receptors, fibroblast growth factor [FGF] receptors, RET, KIT and platelet‐derived growth factor receptor A [PDGFRA]) has become an option equal to sorafenib, after the positive result of the noninferiority Randomized Phase 3 Trial to compare the Efficacy and Safety of Lenvatinib versus Sorafenib in Patients with Unresectable HCC (REFLECT) study (HR, 0.92; 95% CI 0.79‐1.06)(7) (Table 2, Fig. 2A). Because this trial excludes patients with main portal vein invasion, tumor involvement >50% of the liver, and clear bile duct invasion, the relative benefit of lenvatinib versus sorafenib in these patients remains uncertain.

FIG. 2:

OS outcomes (HR, 95% CI) of phase 3 clinical trials testing molecularly targeted therapies, checkpoint inhibitors, and radioembolization in patients with advanced‐stage HCC. Randomized controlled trials in first line (A) and second line (B). Green, positive trials for superiority; orange, positive trials for noninferiority; black, negative trials for the primary endpoint; red, tested drug was significantly worse than the standard of care. Vertical red line at HR = 1.08 defines the upper boundary of 95% CI accepted by the FDA for a positive noninferior study.

In second line, the phase 3 trial testing regorafenib (VEGF receptors, PDGFRs, KIT, and TEK receptor tyrosine kinase) improved OS compared to placebo from 7.8 to 10.6 months (HR, 0.63) in patients who progressed and were tolerant to sorafenib.(8) The sequential treatment sorafenib–regorafenib led to a median OS of 26 months compared to 19 months for sorafenib–placebo.(23) These results need to be taken with caution because they will not apply to all patients receiving sorafenib but only to those able to receive the sequential treatment. The Cabozantinib versus Placebo in Subjects With HCC Who Have Received Prior Sorafenib (CELESTIAL) study showed median OS of 10.2 months with cabozantinib (VEGFreceptors, MET, and AXL) versus 8 months with placebo (HR, 0.76)(9); and the Ramucirumab and Best Supportive Care (BSC) versus Placebo and BSC as Second‐Line Treatment in Patients With HCC and Elevated Baseline AFP Following First‐Line Therapy With Sorafenib (REACH‐2) study, where ramucirumab (VEGFreceptor 2 monoclonal antibody) provided a median OS of 8.5 months in patients with AFP ≥ 400 ng/mL versus 7.3 months with placebo (HR, 0.71).(10,135) AFP is well known for its independent prognostic capacity in HCC.(136) As such, REACH‐2 was the first and only positive phase 3 trial in a biomarker‐driven population of patients with HCC (Fig. 2B). In contrast, three phase 3 trials testing internal radiation with Y‐90 for advanced HCC, either as single treatment (Efficacy and Safety of Selective Internal Radiotherapy with Yttrium‐90 Resin Microspheres Compared with Sorafenib in Locally Advanced and Inoperable HCC [SARAH’(137) and Selective Internal Radiation Therapy Versus Sorafenib in Asia‐Pacific Patients with HCC [SIRveNIB](138)) or in combination Y‐90 with sorafenib,(139) did not meet the primary endpoint of improved OS compared to sorafenib (Fig. 2A). As a result, Y‐90 was discouraged for the management of advanced HCC in the EASL guidelines (Fig. 1).(3) Despite an appealing ORR of 15% with durable response for nivolumab and 18% for pembrolizumab, phase 3 trials comparing the former with sorafenib(24) in front‐line and the latter with placebo in second‐line resulted negative. Particularly, the latter trial showed an HR of 0.78 with an upper boundary of the 95 CI below 1, but the prespecified P value (P < 0.0178) was not hit.(41)

Trial Design in Advanced HCC

OS remains as the primary endpoint for advanced HCC research(1,3) (Table 1). It has driven clinical research in HCC for more than 40 years and has been the gold standard for measuring benefits at all stages of the disease. Nonetheless, the emergence of several effective drugs in advanced HCC has exposed the need for alternative endpoints that can capture the benefits of a treatment before they can be diluted by postprogression therapy.(3) PFS, TTP, and ORR are now emerging as tools to (1) identify early strong signals of efficacy that lead to accelerated regulatory approval (particularly ORR and PFS)(6,88) and (2) test interventions in which benefit can be assessed prior to additional sequential drugs being received beyond progression that might mask the actual benefit of the tested drug. In this sense, a recent investigation analyzing 21 reported phase 3 studies(7‐11,121‐123,126‐131,133,135,137,138,140) in advanced HCC proposed PFS (with a restrictive HR criteria ≤0.6) as a surrogate endpoint for survival when testing kinase inhibitors or monoclonal antibodies, and thus as a potential primary endpoint in advanced HCC trials(3) (Table 4). Subsequently, six phase 3 studies have been released that confirm the hypothesis: two positive studies, one testing atezolizumab plus bevacizumab versus sorafenib(134) and the second sorafenib plus hepatic arterial infusion of FOLFOX (folinic acid, fluorouracil, and oxaliplatin) versus sorafenib(141) (both show an HR for PFS ≤0.6 and significant survival benefits), and four negative trials for survival testing nivolumab,(24) sorafenib plus pravastatin,(142) sorafenib plus doxorubicin,(143) and pembrolizumab,(41) in which the HR for PFS in all cases was >0.6 (Fig. 3). Considering the special circumstances of the two negative trials testing anti‐PD‐1 inhibitors, we should be cautious when applying this rule for testing immune therapies as single agents of immune regimes.

FIG. 3:

Correlation between PFS and OS in 27 phase 3 trials of advanced HCC (modified from Llovet et al.( 20 )). Trial‐level correlation between endpoints using criteria from the Institute for Quality and Efficiency in Health Care. Each dot represents phase 3 clinical trials conducted on advanced HCC. Size of the dot is proportional to the total number of patients enrolled in the trial. First 21 phase 3 trials defined a cutoff of 0.6 for PFS to correlate with a significant OS (in gray).( 20 ) Afterward, six additional phase 3 trials were reported: two positive for survival show an HR for PFS <0.6 (green) and four negative for OS show a PFS HR >0.6 (red). _x_‐axis and _y_‐axis depict the value of the HR for the surrogate and the hard endpoint, respectively. Gray‐shaded areas represent the upper and lower limits of the 95% CIs for the regression.

Trial design in HCC has been evolving, and challenges emerge as different therapies become standard of care. Although there might be distinct approaches to trial design in HCC, there has been a consensus on the basic principles that have been recently reported in guidelines and critical appraisals.(3,139,144,145) The key points are summarized below (Table 1):

1. Phase 2 and phase 3 trials: The panel recommends assessing drugs in the setting of randomized phase 2 studies before moving to phase 3 trials. Nonetheless, for some therapies, a large single‐arm phase 2 with a strong signal of efficacy might suffice to justify a phase 3 study. Thresholds for defining signals of efficacy are not clearly established, but for molecular therapies the ORR should likely be above 20%‐30%.(146)
2. Selection of the target population: Clinical trials should consider BCLC staging system, Child‐Pugh class, and Eastern Cooperative Oncology Group (ECOG) performance status for selection of the target population. In principle, for advanced HCC almost all RCTs include patients with well‐preserved liver function (Child‐Pugh A) and good performance status (ECOG 0 and 1).
3. Control arm: The control arm of randomized phase 2 and 3 studies should be the standard of care established according to guidelines. Although sorafenib and lenvatinib in front line(7,11) and regorafenib,(8) cabozantinib,(9) and ramucirumab (in patients with AFP ≥ 400 ng/mL)(10) are accepted as standard of care, this has changed with atezolizumab plus bevacizumab being approved by regulatory agencies. Thus, this combination has become the standard of care for comparison in front line, and subsequent lines of therapy will move downward. Double‐blind trials (as opposed to open label trials) are recommended to prevent selection and allocation biases.
4. Stratification for prognostic factors prior to randomization: Stratification is critical in randomized studies to warrant balanced comparisons. For advanced HCC the recommendation is as follows: region, macrovascular invasion, extrahepatic spread, AFP > 400 ng/mL, and ECOG 0 versus 1‐2. Etiology should also be considered as studies with sorafenib and atezolizumab and bevacizumab suggest an influence of this factor in response.
5. Endpoints: OS: For systemic therapies the primary endpoint should be OS, and PFS is proposed as a coprimary endpoint. To date, all regular FDA and European Medicines Agency drug conventional approvals in advanced HCC were based upon improvements in OS. Surrogate endpoints: OS has limitations as a sole endpoint in cancer research: it might require a long follow‐up to capture adequate numbers and can be affected by sequential therapies. Thus, surrogate endpoints that are more practical for trial execution are needed. There are no optimal surrogate endpoints that can recapitulate OS in HCC, and thus clinical practice guidelines do not recommend ORR, TTP, and PFS as primary endpoints in phase 3 investigations.(144,145) ORR is an independent predictor of OS in three phase 2 and 3 trials(7,123,127) but is still considered a suboptimal primary endpoint for phase 3 investigations. Nonetheless, an ORR of 16%‐18% resulted in accelerated FDA approval of nivolumab and pembrolizumab in second‐line treatment of advanced HCC.(38,39) PFS was formerly discarded as a primary endpoint of phase 3 investigations due to the concept of competing risk of survival (competing between death due to tumor progression and due to the natural history of cirrhosis).(12) However, this competing risk drawback has been reduced by the universal selection of Child‐Pugh A patients for these investigations, thus reducing the 1‐year risk of death due to decompensation to <5%. Stringent criteria for accepting PFS as a primary endpoint have been proposed (HR ≤ 0.6), and it is adopted in the current guidelines (Table 1); but this point is still controversial. Regarding ORR, use of both RECIST1.1. and mRECIST is proposed for the assessment of response in HCC treated with systemic therapy, whereas changes in serum biomarker levels (i.e., AFP levels) are not supported.(3)
6. Magnitude of benefit: In HCC, there is no consensus on what absolute survival benefit (or magnitude of benefit in OS according to HR) is clinically relevant. Reported thresholds of OS with HR <0.8 are sound for capturing the benefit for patients in advanced HCC trials.(20) This figure needs to be taken with caution because other variables can impact the overall benefit of a given drug, such as QoL, safety profile, and availability of alternative therapies in distinct countries.
7. Checkpoint inhibitors and other immunotherapies have unique features and generally produce higher ORR and longer duration of response, as measured by RECIST1.1. The values of mRECIST and irRECIST in assessing checkpoint inhibitor–mediated responses remain investigational.

Immune Treatments: Overview of Results and Specific Endpoints

The initial clinical experience with checkpoint inhibitors in HCC was with a phase 2 study testing tremelimumab, a CTLA4 antibody leading to an objective response of 18% of patients and TTP of 6.5 months.(147) Immunotherapy has drawn significant attention in HCC with the approval of nivolumab and pembrolizumab by the FDA based on promising results obtained in different phase 2 studies.(38,39) A phase 1/2 open‐label, noncomparative trial (CheckMate 040) assessing the efficacy of nivolumab in advanced HCC reported ORRs of 20% in the dose‐expansion phase (n = 214) and 15% in the dose‐escalation phase (n = 48). Duration of response was 9.9 months and median PFS as 4.0 months in the dose‐expansion cohort. Nivolumab treatment was well tolerated.(38) Pembrolizumab, another PD‐1‐specific antibody, was tested in phase 2 in patients with HCC progressing or intolerant to sorafenib (Keynote 224). Pembrolizumab was effective and tolerable, with one complete response and 17 partial responses out of 104 patients. The median PFS was 4.9 months, and median OS was 12.9 months.(39) Camrelizumab, another fully humanized anti‐PD‐1 antibody, was evaluated in a randomized phase 2 trial in Chinese patients with advanced HCC after failure of at least one line of therapy.(148) The ORR was 13.8%, and the 6‐month OS was 74.7%.

Nivolumab and pembrolizumab failed in phase 3 trials (Fig. 2A,B). Pembrolizumab was tested in a randomized, double‐blind phase 3 trial against placebo in 443 patients with HCC who progressed on or were intolerant to sorafenib (Keynote 240). The coprimary endpoints of PFS and OS failed to reach the prespecified level of statistical significance, although median OS was prolonged from 10.6 to 13.9 months (HR, 0.781; 95% CI, 0.611‐0.998; P = 0.0238).(149) Nivolumab was tested against sorafenib in a phase 3 trial (CheckMate 459) but did not reach survival differences for superiority.(24) In this RCT including around 750 patients, median OS for nivolumab was 16.4 versus 14.7 months for the sorafenib arm (HR, 0.85; 95% CI, 0.72‐1.02). ORRs were 15% and 7%, respectively.

Anti‐CTLA4 antibodies have been tested as single agents(147) or in combination with locoregional therapies(150) and are under investigation in combination with anti‐PD‐1 drugs.(151) In this regard, very recently the combination ipilimumab and nivolumab received FDA approval based on an ORR of 31%.(40) Currently, phase 3 trials are ongoing which either test the combination of two immune checkpoint inhibitors, immune checkpoint inhibitors plus TACE, immune checkpoint blockade in the adjuvant setting, or immune checkpoint inhibitors plus vascular targeting agents.(152) While the overall response to immune checkpoint inhibition (15%‐20%) may not be as dramatic as initially hoped, complete responses are seen in a small number of cases in almost every trial. This observation together with the recent results from two phase 3 trials testing anti‐PD1 antibodies in the first‐line and second‐line settings raise the important question of what endpoint to use in future trials. While OS remains the “gold standard,” it should be noted that HCC is not the only cancer for which this question is being asked. Due to the unique mechanism of action of immune checkpoint inhibitors,(153) new endpoints such as ORR and surrogate biomarkers have been tested and new immune‐related RECIST criteria devised to capture distinctive patterns and timing of response to immunotherapy.(35,115) Finally, while immunological endpoints may be helpful as surrogates,(154) they cannot be recommended at this time.

A systematic review and meta‐analysis of 87 phase 2 trials with the goal of defining the most appropriate primary endpoint in phase 2 trials of immune checkpoint inhibitors for advanced solid cancers has been reported. Correlations between ORR odds ratios and HRs for PFS and OS were examined for randomized comparisons. Within checkpoint inhibitor treatment arms, correlations of ORR with 6‐month PFS and 12‐month OS rates were examined. All analyses were weighted by trial size. Multivariable models to predict 6‐month PFS and 12‐month OS rates from ORR were developed, and their performance was validated in an independent sample of trials. The authors demonstrated that ORR correlated poorly with OS and recommended a 6‐month PFS rate as an endpoint for future phase 2 studies.(155) Thus, the PFS endpoint can also be recommended for studies evaluating immune checkpoint inhibitors in HCC (Table 1).

Biomarker‐Driven Trials in HCC: Results and Endpoints

Recent clinical trials in advanced HCC are demonstrating that the sequential use of systemic agents is changing the natural history of the disease. Still, these results are incremental, and the incorporation of biomarker driven strategies has generally been unsuccessful. Unlike other solid tumors such as breast, lung, colon, and others, where therapeutic decisions are driven by an understanding of a patient given molecular features, in HCC “one‐size‐fits‐all” is still the usual approach to patients. This applies to all therapies so far accepted in guidelines, except for ramucirumab.

Biomarkers provide the distinct possibility of supplementing existing anatomic and/or pathologic information to provide a more accurate assessment of prognosis (to be used for patient stratification) and/or to identify individuals who are more likely to respond to specific therapy (predictive of response).(156‐158) There is a plethora of literature on the different predictive biomarker validation designs.(159) The National Cancer Institute defines a biomarker as a biological molecule found in blood, other body fluids, or tissues that is a sign of a normal or abnormal process, or of a condition or disease.(160) A biomarker may be used to determine how well the body responds to a treatment for a disease or condition.(160)

In HCC, numerous studies have defined the molecular heterogeneity of the disease and specific genetic alterations and subtypes. These data are fertile ground for testing biomarker hypotheses as prognostic and/or predictive markers in prospective studies, but so far these data have largely been ignored in clinical development in HCC.(6) To date, two phase 3 studies have tested biomarker‐driven approaches. Firstly, tivantinib, a small molecule inhibitor of the hepatocyte growth factor/c‐MET type was evaluated in patients who had progressed on sorafenib and had elevated expression of c‐MET in their tissue. This was a placebo‐controlled study that yielded negative results.(131) The possible reasons for failure highlight the challenges with this approach including (1) validity of the target, (2) robustness of the assay for patient selection, and (3) ability of the agent to inhibit the target successfully in tumor tissue. The latter may be a plausible reason for failure of the trial, considering that the anti‐MET activity of this drug has been challenged in experimental studies.(161) Secondly, ramucirumab, which initially failed in an “all‐comers” study,(162) demonstrated an improvement in OS for selected patients with AFP ≥ 400 ng/mL. Proof‐of‐concept studies testing small molecule inhibitors of FGF receptor 4 using biomarker‐enriched populations based on FGF‐19 expression have been reported with an ORR of 16%.(163)

Recently, immunotherapeutic approaches have garnered high interest in the management of HCC, and the PD‐1‐directed antibodies nivolumab and pembrolizumab have received accelerated approval by the US FDA.(38,39) However, unlike in other cancers, PD‐1 and/or PD‐L1 expression has not correlated with outcome. This has likely contributed to the recent negative results from phase 3 studies with these agents.(149) Ongoing work is focused on further refining biomarker development and evaluating other inflammatory markers including incorporation of more broad‐based assessment tools such as an immune‐enriched signature identified through molecular profiling of HCC.(164)

Several studies have incorporated biomarker assessments into the trial design. While tissue collection is often optional and therefore limited, serum assays have served to generate hypotheses for further study. In the pivotal SHARP study, baseline levels of angiopoietin‐2 and VEGF were prognostic but not predictive of benefit from sorafenib.(136) Relevant biomarkers in the FGF and VEGF pathways were analyzed in the REFLECT study and identified differences in the modulation of these pathways by lenvatinib and sorafenib, but no biomarker could define a group receiving differential benefit from either compound.(165) In the REACH‐2 study, decreases in AFP correlated with better outcome from ramucirumab.(10) Study designs evaluating biomarker assessments pretreatment and posttreatment are being performed. These so‐called presurgical studies are designed to acquire tissue at baseline from patients with resectable tumors, expose the patient to an agent for a short period, and then collect tissue at the time of resection. Molecular studies comparing the pretreatment and posttreatment tissue provide an opportunity to understand the effects of therapeutics on relevant pathways in the tumor. These studies can provide critical information that could guide a patient selection strategy in conventional efficacy studies. One such study with nivolumab is producing interesting insights into tumor characteristics that may correlate to response to this drug.(166)

Despite the recent successes in clinical trials of agents for HCC, the improvements in survival are modest. Throughout cancer medicine, the largest impacts in outcomes have been by biomarker‐driven drug development. Examples include activin receptor‐like kinase(167) and epidermal growth factor receptor (EGFR)(168) testing in lung cancer, HER‐2(169) and estrogen receptor(170) testing in breast cancer, c‐KIT testing in gastrointestinal tumors,(171) and BCR‐ABL testing in chronic myelogenous leukemia.(172) By enriching for the population most likely to benefit, studies can be conducted with smaller numbers of patients and minimize risk for failure. While historically predictive marker testing is done on tumor tissue obtained by biopsy, newer technologies are now allowing biomarker detection in peripheral blood. The practice of not obtaining biopsies for diagnosis of HCC, the fact that most common driver mutations in HCC are nonactionable, and the observation that only 25% of HCCs harbor at least one actionable mutation,(172) in contrast to the majority of solid tumors,(173) have hindered development of biomarker‐driven precision treatment to date. Nonetheless, there is now renewed interest in incorporating tissue acquisition into clinical trials, not only in the early part of drug development but in later studies as well. Clinical trial designs for predictive marker validation are inherently complex and require data from an RCT.(153) There is a plethora of literature on the different predictive biomarker validation designs, including articles that specifically focus on the statistical and clinical properties and assumptions of these different trial designs.(156)

Trial design in the precision medicine era require a platform for biomarker profiling.(173,174) The ultimate clinical utility of a biomarker will depend on (1) its added value in every patient in the context of the marker’s prevalence, (2) its incremental benefit for treatment selection when considering the added costs and complexity induced by the use of the marker, and (3) the added effectiveness of the treatment option in all patients versus biomarker‐defined subgroups.

Liquid Biopsy in Early HCC Detection, Prediction of Response and Tumor Relapse

Liquid biopsy entails the analysis of tumor components released by cancer cells to biological fluids such as blood, saliva, or cerebrospinal fluid.(175) The concept includes the analysis of actual cancer cells (i.e., circulating tumor cells [CTCs]), fragments of DNA from necrotic or apoptotic cancer cells (i.e., ctDNA), and extracellular vesicles.(176) Compared to conventional tissue biopsies, the main advantages of liquid biopsy using samples from peripheral blood are (1) it is minimally invasive, which eliminates the complications associated with invasive tissue biopsies; (2) it facilitates sequential sampling, which is crucial to better select therapies in patients receiving multiple lines of treatment; (3) it enables tracking tumor clonal composition in heterogeneous tumors, a feature that allows earlier detection mechanisms of treatment resistance; and (4) it can be implemented as a point‐of‐care diagnostic. Potential clinical applications include cancer surveillance, early detection of minimal residual disease after curative therapies, prognostic prediction, and molecular monitoring of therapeutic response.(177) In an early sign of impact on patient care, the FDA recently approved the use of a ctDNA‐based test to detect mutations of EGFR in lung cancer patients who are candidates to receive EGFR‐based TKIs.(178)

In HCC, liquid biopsy has been evaluated for three clinical applications: early HCC detection in the context of surveillance, as a prognostic biomarker after surgical resection, and to predict response to systemic therapies. Mutation profiling of ctDNA is feasible and confidently detects tissue mutations in early‐stage HCC.(179) A recent report combining data from ctDNA and protein markers had a sensitivity and specificity of 85% and 93%, respectively, for the detection of HCC.(180) Also, analysis of DNA methylation alterations in ctDNA has high accuracy for HCC diagnosis.(55,181) A study that included a gene signature derived from CTCs was able to accurately discriminate between patients with HCC and controls.(182) Higher CTC count correlates with a greater risk of tumor recurrence after surgical resection.(175) There are few studies using liquid biopsy to predict response to systemic therapies in HCC. A retrospective study suggested that patients with HCC with ctDNA detectable VEGFA DNA amplifications have better outcomes when treated with sorafenib.(183) Also, RAS mutation analysis of ctDNA was used to determine eligibility to receive refametinib in a phase 2 clinical trial.(184) Thus, there is increasing interest in applying this technology to predict response to systemic therapies.

QoL and PROs

Systematic capture of the patient perspective can inform the development of cancer therapies. The US FDA’s Office of Hematology and Oncology Products has identified symptomatic adverse events as a central PRO using the National Cancer Institute’s Patient‐Reported Outcomes version of the Common Terminology Criteria for Adverse Events to provide a standard yet flexible method to assess symptomatic adverse events from the patient perspective. The FDA’s patient‐focused drug development program has made ongoing efforts to improve methods around the collection, analysis, and interpretation of PRO data, as well as initiatives to identify patient‐friendly language and leverage digital health tools. In 2016, the 21st Century Cures Act tasked the FDA to consider the patient experience in the risk–benefit determination. The FDA draft guidance outlines the use of PRO to assess symptomatic side effects and the core set of clinical outcomes to measure in cancer trials, including design considerations and assessment frequency.(11)

The purpose of measuring QoL should be to compare outcomes between treatment arms, even if one is a placebo. There are two methods of measuring QoL specific to HCC: the European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire(185) and the Functional Assessment of Cancer Therapy‐Hepatobiliary(186) questionnaire. Few studies have adequately assessed PRO using these tools in HCC research, a recommendation that is endorsed by the panel.

Most phase 3 trials for HCC were designed primarily to compare two different treatments in patients with similar‐stage disease. For example, one study compared QoL after resection with QoL following RFA.(187) As expected, QoL was much better after RFA than after resection and remained superior up to 36 months posttreatment. In addition, QoL following radioembolization has been compared with TACE.(188) In this study, there was no overall difference in QoL between the two groups, but the sample size was small. Despite the lack of statistically significant differences, in the TACE group QoL was decreased at 2 and 4 weeks, whereas in the radioembolization group some aspects of QoL actually improved. Similarly, QoL measures favoring Y‐90 versus sorafenib have been claimed to support the former treatment in three negative RCTs, the SARAH trial,(134) the SIRveNIB trial,(135) and the SORAMIC trial. However, because indication of a drug/device should be based upon the primary endpoint (survival), no actual indication can be claimed if the result is negative for the primary endpoint. Finally, the SHARP trial, demonstrating a survival benefit of sorafenib, also tested time‐to‐symptomatic progression—as measured by the Functional Assessment of Cancer Therapy–Hepatobiliary Symptom Index 8—as a coprimary endpoint. The negative results of this endpoint contrasted with the survival benefit obtained by sorafenib, thus challenging the accuracy of the tool used.(11) More recently, PROs have been tested in the setting of phase 3 investigations showing significant results in positive RCTs in advanced HCC. This is the case of lenvatinib compared to sorafenib and of atezolizumab plus bevacizumab compared to sorafenib, where the tested arms showed better QoL parameters compared with the standard of care.(7,134) The panel encourage the integration of these endpoints in all investigations in HCC (Table 6).

TABLE 6 - Unmet Needs in Trial Design in HCC

Abbreviations: PRO, patient‐reported outcomes; SBRT, stereotactic body radiotherapy.

Implications of Trial Design in Asia

Differences Between AASLD, EASL, and Asian Guidelines

Recommendations in Western guidelines (AASLD and EASL) are based upon evidence from clinical trials (Table 4), while Asian guidelines integrate evidence with expert consensus and clinical practice. Applicability of those guidelines varies according to region and treatment stage.(47) Asian guidelines(189) in general recommend ablation or resection for very early‐stage (stage 0) disease but differ from Western guidelines in the recommendations at other stages of disease. For instance, TACE or yttrium‐90 (Y90‐ selective internal radiation therapy) is recommended for single large tumors, and systemic therapies—i.e., FOLFOX(141) or hepatic arterial infusion chemotherapy(189)—are recommended for advanced stages, along with liver transplantation, mostly living donor transplantation. Similarly, in Asia, patients with portal vein invasion and well‐preserved liver function might be considered for TACE, resection, or radiotherapy.(190‐194)

Specificities of Trial Design in Asia

Considering all these guidelines, trial design in Asia has some specificities. For instance, resection in very high‐risk patients (multinodular tumors, macrovascular invasion) is common in Asia, and thus adjuvant trials might consider this indication with a recurrence‐free survival endpoint. Similarly, studies exploring the role of systemic therapies plus TACE in patients with advanced stages might also be considered in Asia with a primary endpoint of PFS. Whether these approaches should be tested in specific trials or as part of global trials needs further consideration.

Future Prospects

The Dawn of a New Era: Combination Therapies

When the report of the first AASLD conference, Design and Endpoints of Clinical Trials in HCC, was published,(12) the field was still heady with excitement from the first positive trial of a systemic agent for advanced HCC, which established sorafenib as the first FDA‐approved systemic therapy for HCC.(11) However, enthusiasm was also tempered by the subsequent negative results of trials of sorafenib as adjuvant therapy after resection or ablation(76) or in combination with TACE.(94) It was recognized then that a unique challenge is posed by the combination of underlying chronic liver disease with a very heterogeneous and variably aggressive primary HCC. It is therefore important that treatment strategies account for both the liver disease and malignancy, and thus variables capturing both diseases should be considered in the publication of clinical trials for HCC (Table 7). Discussions at the previous AASLD endpoints conference set the framework for subsequent attempts to bring additional agents to approval, which were met with uniformly disappointing results over the next several years, with trial failures due to unacceptable toxicity or inadequate efficacy.(12) While disappointing, these failures led to robust examination of the optimal approach to trial design and catalyzed a more rigorous approach that contributed to the successes in phase 3 HCC clinical trials. With the positive results and approvals of lenvatinib, regorafenib, cabozantinib, and ramucirumab based on phase 3 studies and of the checkpoint inhibitors nivolumab and pembrolizumab based on convincing phase 2 data, we appear to be poised for success in the next most logical treatment paradigms using combination therapies. Indeed, the recent positive phase 3 study demonstrating superior OS for atezolizumab plus bevacizumab versus sorafenib(134) represents the dawn of a new era of combination therapies in all stages of HCC research (Fig. 1). This combination is certainly first in‐class of this approach. Whether other combinations may become best‐in‐class will depend upon the ability of specific TKIs and/or monoclonal antibodies to transform “cold tumors,” which are primarily resistant to immunotherapy, into “hot‐immune‐active tumors,” allowing checkpoint inhibitors to optimally unleash immune attack against cancer cells.(195‐197)

TABLE 7 - Variables to be Included in Clinical Trials Assessing Treatments for Patients with HCC

*Modified from Llovet et al.(12)

Abbreviations: HBV, hepatitis B virus; HCV, hepatitis C virus; INR, international normalized ratio; MELD, Model of End‐Stage Liver Disease; NAFLD, nonalcoholic fatty liver disease; NASH, nonalcoholic steatohepatitis; pTNM, pathological tumor–node–metastasis.

The advent of combination therapies achieving response rates of 30% and survival rates above 20 months in advanced HCC provides the rationale for testing these combinations in earlier HCC stages. Currently, phase 3 trials testing combination therapies are ongoing for early stages (neoadjuvant or adjuvant approaches), intermediate HCC (in combination with TACE or in comparison to TACE), and front‐line trials for advanced HCC (Table 7). It is conceivable that systemic therapies may be incorporated in all areas of HCC management in the near future. Thus, the updated target population and endpoints described here should be valuable in this endeavor.

Understanding Tumor Biology Remains Critical: Tissue and Blood Samples Are Needed

It is likely that the next key advances in HCC therapy will emanate from an improved understanding of HCC biology and the ability to predict the response of specific HCC subgroups to particular therapies. Until now, most HCC therapy has been applied in a biologically indiscriminate fashion. The biological heterogeneity of HCC has been evident for many years, demonstrated by differences in phenotypes, tumor growth rates, numbers of tumor nodules, discrete versus infiltrative appearance, propensity for microvascular or macrovascular invasion, propensity for distant metastasis, and association with elevation of AFP, AFP‐L3, des‐γ‐carboxyprothrombin, and other biomarkers. Apart from the limitations that multifocal, invasive, or metastatic disease have placed on the application of potentially curative treatments such as surgical resection, liver transplantation, and local ablation, we have only recently begun to incorporate markers of tumor biology into therapeutic decision‐making. Applications of tumor biologic characteristics into therapeutic approaches have been scarce in HCC and mostly focused on using AFP levels for selection policy for transplantation, as a stratification factor in most of trials, and finally for selecting candidates to ramucirumab in the management of advanced HCCs in second line.

With the advent of next‐generation DNA, RNA and noncoding RNA sequencing, and similar genome‐wide methodologies for copy number variation, methylation, and proteomic characterization, we now stand ready to translate information from these technologies to the care of patients with HCC, transforming the selection of systemic therapy and the selection of optimal candidates for locoregional therapies. Results suggesting that catenin beta 1 gene–mutated HCCs are immune‐excluded and potentially resistant to immune checkpoint inhibitors(198‐200) but potentially susceptible to mammalian target of rapamycin inhibitors are an early indication of the potential value of genomics in personalizing HCC therapy. These studies may also provide us with tools for better understanding the recent borderline negative results of phase 3 trials with single‐agent immune checkpoint inhibitors. Personalization of therapy using molecular and genomic signatures will require integration of molecular subclasses into clinical staging systems, to better guide treatment selection. Optimal treatment selection will depend on the ability to target oncogenic signaling pathways that drive tumorigenesis, tumor progression, and metastasis. The development of preclinical tumor models, including organoids, patient‐derived xenografts, and syngeneic models that preserve aspects of the immune response will be critical for the testing of agents and combinations. Ideally, integration of molecular profiling into the HCC treatment paradigm will require genomic data derived in real time from patients, either by tissue biopsy or through liquid biopsy–based access to ctDNA or other analytes. This will require a cultural change in the care of patients with HCC, shifting from a state in which the diagnosis and evaluation of patients are performed noninvasively to regular use of tissue biopsy and highly sensitive liquid biopsy assays. Development of robust, reproducible predictive biomarkers of high reliability is a key priority to facilitate this transition (Table 6). The first implementations of the biomarker‐based approaches should be within RCTs, which should now routinely require tissue biopsy and liquid biopsy collection as a condition of trial enrollment. Tumor biopsy at screening for trial entry and liquid biopsy at different time points should be mandatory in clinical trials for advanced HCC, to allow identification of prognostic and predictive biomarkers, guide clinical decision‐making, and improve patient outcomes.

Endpoints That Might Be Adopted

The revolution in drug development in HCC has created the need to revisit established conventions in trial design. OS is regarded as a core endpoint. Nonetheless, the realization that more than 60% of patients progressing after TACE and 50% of patients progressing after first‐line systemic therapies receive effective next‐line therapies may compel the adoption of PFS as an acceptable primary endpoint for major trials (Table 1). In this position paper, we already are recommending PFS as a coprimary endpoint for intermediate HCC trials and for phase 2/3 trials assessing systemic therapies, with restrictive cut‐points. Similarly, PROs should be pursued as a relevant endpoint in HCC trials, particularly as we enter an era of potent but seemingly toxic dual or triple combination therapies possibly associated with serious adverse events. It is important to recognize that currently many patients with HCC reach a point in their therapeutic journey when they elect to forgo potentially life‐extending therapy in favor of approaches that optimize their QoL. It is therefore critical to extend decisions about HCC trial design and endpoints to incorporate elements that reflect the importance of patient well‐being.

Author Contributions

All authors contributed substantially to drafting of the article and its final approval.

References

1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394‐424.

2. Villanueva A. Hepatocellular carcinoma. N Engl J Med 2019;380:1450‐1462.

3. Galle PR, Forner A, Llovet JM, Mazzaferro V, Piscaglia F, Raoul J‐L, et al. EASL clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 2018;56:908‐943.

4. Marrero JA, Kulik LM, Sirlin CB, Zhu AX, Finn RS, Abecassis MM, et al. Diagnosis, staging, and management of hepatocellular carcinoma: 2018 practice guidance by the American Association for the Study of Liver Diseases. Hepatology 2018;68:723‐750.

5. Llovet JM, Ducreux M, Lencioni R, Di Bisceglie AM, Galle PR, Dufour JF, et al. EASL‐EORTC clinical practice guidelines: management of hepatocellular carcinoma. J Hepatol 2012;56:908‐943.

6. Llovet JM, Montal R, Sia D, Finn RS. Molecular therapies and precision medicine for hepatocellular carcinoma. Nat Rev Clin Oncol 2018;15:599‐616.

7. Kudo M, Finn RS, Qin S, Han K‐H, Ikeda K, Piscaglia F, et al. Lenvatinib versus sorafenib in first‐line treatment of patients with unresectable hepatocellular carcinoma: a randomised phase 3 non‐inferiority trial. Lancet 2018;391:1163‐1173.

8. Bruix J, Qin S, Merle P, Granito A, Huang Y‐H, Bodoky G, et al. Regorafenib for patients with hepatocellular carcinoma who progressed on sorafenib treatment (RESORCE): a randomised, double‐blind, placebo‐controlled, phase 3 trial. Lancet 2017;389:56‐66.

9. Abou‐Alfa GK, Meyer T, Cheng A‐L, El‐Khoueiry AB, Rimassa L, Ryoo B‐Y, et al. Cabozantinib in patients with advanced and progressing hepatocellular carcinoma. N Engl J Med 2018;379:54‐63.

10. Zhu AX, Kang Y‐K, Yen C‐J, Finn RS, Galle PR, Llovet JM, et al. Ramucirumab after sorafenib in patients with advanced hepatocellular carcinoma and increased α‐fetoprotein concentrations (REACH‐2): a randomised, double‐blind, placebo‐controlled, phase 3 trial. Lancet Oncol 2019;20:282‐296.

11. Llovet JM, Ricci S, Mazzaferro V, Hilgard P, Gane E, Blanc J‐F, et al. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med 2008;359:378‐390.

12. Llovet JM, Di Bisceglie AM, Bruix J, Kramer BS, Lencioni R, Zhu AX, et al. Design and endpoints of clinical trials in hepatocellular carcinoma. J Natl Cancer Inst 2008;100:698‐711.

13. Lee DH, Vielemeyer O. Analysis of overall level of evidence behind Infectious Diseases Society of America practice guidelines. Arch Intern Med 2011;171:18‐22.

14. Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996;17:1‐12.

15. Verhagen AP, de Vet HC, de Bie RA, Kessels AG, Boers M, Bouter LM, et al. The Delphi list: a criteria list for quality assessment of randomized clinical trials for conducting systematic reviews developed by Delphi consensus. J Clin Epidemiol 1998;51:1235‐1241.

16. Altman DG, Schulz KF, Moher D, Egger M, Davidoff F, Elbourne D, et al. The revised CONSORT statement for reporting randomized trials: explanation and elaboration. Ann Intern Med 2001;134:663‐694.

17. Piaggio G, Elbourne DR, Altman DG, Pocock SJ, Evans SJW; CONSORT Group . Reporting of noninferiority and equivalence randomized trials: an extension of the CONSORT statement. JAMA 2006;295:1152‐1160.

18. van Tulder M, Furlan A, Bombardier C, Bouter L; Editorial Board of the Cochrane Collaboration Back Review Group . Updated method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group. Spine 2003;28:1290‐1299.

19. Lopez PM, Villanueva A, Llovet JM. Systematic review: evidence‐based management of hepatocellular carcinoma—an updated analysis of randomized controlled trials. Aliment Pharmacol Ther 2006;23:1535‐1547.

20. Llovet JM, Montal R, Villanueva A. Randomized trials and endpoints in advanced HCC: role of PFS as a surrogate of survival. J Hepatol 2019;70:1262‐1277. Accessed June 18th 2020.

21. Lencioni R, Llovet JM. Modified RECIST (mRECIST) assessment for hepatocellular carcinoma. Semin Liver Dis 2010;30:52‐60.

22. Cheng AL, Kang YK, Lin DY, Park JW, Kudo M, Qin S, et al. Sunitinib versus sorafenib in advanced hepatocellular cancer: results of a randomized phase III trial. J Clin Oncol 2013;31:4067‐4075.

23. Finn RS, Merle P, Granito A, Huang Y‐H, Bodoky G, Pracht M, et al. Outcomes of sequential treatment with sorafenib followed by regorafenib for HCC: additional analyses from the phase 3 RESORCE trial. J Hepatol 2018;69:353‐358.

24. Bristol‐Myers Squibb . Bristol‐Myers Squibb announces results from Checkmate‐459 study evaluating Opdivo (nivolumab) as a first‐line treatment for patients with unresectable hepatocellular carcinoma. Published June 24, 2019. Available from: https://news.bms.com/press‐release/bmy/bristol‐myers‐squibb‐announces‐results‐checkmate‐459‐study‐evaluating‐opdivo‐nivol. Accessed June 2020.

25. Prasad V, Kim C, Burotto M, Vandross A. The strength of association between surrogate end points and survival in oncology: a systematic review of trial‐level meta‐analyses. JAMA Intern Med 2015;175:1389‐1398.

26. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228‐247.

27. Llovet JM, Lencioni R. mRECIST for HCC: performance and novel refinements. J Hepatol 2020;72:288‐306.

28. Edeline J, Boucher E, Rolland Y, Vauléon E, Pracht M, Perrin C, et al. Comparison of tumor response by Response Evaluation Criteria in Solid Tumors (RECIST) and modified RECIST in patients treated with sorafenib for hepatocellular carcinoma. Cancer 2012;118:147‐156.

29. Ronot M, Bouattour M, Wassermann J, Bruno O, Dreyer C, Larroque B, et al. Alternative response criteria (Choi, European Association for the Study of the Liver, and Modified Response Evaluation Criteria in Solid Tumors [RECIST]) versus RECIST 1.1 in patients with advanced hepatocellular carcinoma treated with sorafenib. Oncologist 2014;19:394‐402.

30. Takada J, Hidaka H, Nakazawa T, Kondo M, Numata K, Tanaka K, et al. Modified response evaluation criteria in solid tumors is superior to response evaluation criteria in solid tumors for assessment of responses to sorafenib in patients with advanced hepatocellular carcinoma. BMC Res Notes 2015;8:609.

31. Lencioni R, Montal R, Torres F, Park J‐W, Decaens T, Raoul J‐L, et al. Objective response by mRECIST as a predictor and potential surrogate end‐point of overall survival in advanced HCC. J Hepatol 2017;66:1166‐1172.

32. Meyer T, Palmer DH, Cheng A‐L, Hocke J, Loembé A‐B, Yen C‐J. mRECIST to predict survival in advanced hepatocellular carcinoma: analysis of two randomised phase II trials comparing nintedanib versus sorafenib. Liver Int 2017;37:1047‐1055.

33. Burzykowski T, Buyse M. Surrogate threshold effect: an alternative measure for meta‐analytic surrogate endpoint validation. Pharm Stat 2006;5:173‐186.

34. Reig M, Rimola J, Torres F, Darnell A, Rodriguez‐Lope C, Forner A, et al. Postprogression survival of patients with advanced hepatocellular carcinoma: rationale for second‐line trial design. Hepatology 2013;58:2023‐2031.

35. Wolchok JD, Hoos A, O’Day S, Weber JS, Hamid O, Lebbé C, et al. Guidelines for the evaluation of immune therapy activity in solid tumors: immune‐related response criteria. Clin Cancer Res 2009;15:7412‐7420.

36. Seymour L, Bogaerts J, Perrone A, Ford R, Schwartz LH, Mandrekar S, et al. iRECIST: guidelines for response criteria for use in trials testing immunotherapeutics. Lancet Oncol 2017;18:e143‐e152.

37. Gyawali B, Kesselheim AS. Reinforcing the social compromise of accelerated approval. Nat Rev Clin Oncol 2018;15:596‐597.

38. El‐Khoueiry AB, Sangro B, Yau T, Crocenzi TS, Kudo M, Hsu C, et al. Nivolumab in patients with advanced hepatocellular carcinoma (CheckMate 040): an open‐label, non‐comparative, phase 1/2 dose escalation and expansion trial. Lancet 2017;389:2492‐2502.

39. Zhu AX, Finn RS, Edeline J, Cattan S, Ogasawara S, Palmer D, et al. Pembrolizumab in patients with advanced hepatocellular carcinoma previously treated with sorafenib (KEYNOTE‐224): a non‐randomised, open‐label phase 2 trial. Lancet Oncol 2018;19:940‐952.

40. US Food and Drug Administration . FDA grants accelerated approval to nivolumab and ipilimumab combinatio. Published March 2020. Accessed March 26, 2020. http://www.fda.gov/drugs/resources‐information‐approved‐drugs/fda‐grants‐accelerated‐approval‐nivolumab‐and‐ipilimumab‐combination‐hepatocellular‐carcinoma.

41. Finn RS, Ryoo B‐Y, Merle P, Kudo M, Bouattour M, Lim HY, et al. Pembrolizumab as second‐line therapy in patients with advanced hepatocellular carcinoma in KEYNOTE‐240: a randomized, double‐blind, phase III trial. J Clin Oncol 2020;38:193‐202.

42. Ellis LM, Bernstein DS, Voest EE, Berlin JD, Sargent D, Cortazar P, et al. American Society of Clinical Oncology perspective: raising the bar for clinical trials by defining clinically meaningful outcomes. J Clin Oncol 2014;32:1277‐1280.

43. Nault J‐C, Cheng A‐L, Sangro B, Llovet JM. Milestones in the pathogenesis and management of primary liver cancer. J Hepatol 2020;72:209‐214.

44. Singal AG, Lampertico P, Nahon P. Epidemiology and surveillance for hepatocellular carcinoma: new trends. J Hepatol 2020;72:250‐261.

45. Tzartzeva K, Obi J, Rich NE, Parikh ND, Marrero JA, Yopp A, et al. Surveillance imaging and alpha fetoprotein for early detection of hepatocellular carcinoma in patients with cirrhosis: a meta‐analysis. Gastroenterology 2018;154:1706‐1718.e1.

46. Atiq O, Tiro J, Yopp AC, Muffler A, Marrero JA, Parikh ND, et al. An assessment of benefits and harms of hepatocellular carcinoma surveillance in patients with cirrhosis. Hepatology 2017;65:1196‐1205.

47. Park J‐W, Chen M, Colombo M, Roberts LR, Schwartz M, Chen P‐J, et al. Global patterns of hepatocellular carcinoma management from diagnosis to death: the BRIDGE Study. Liver Int 2015;35:2155‐2166.

48. Totoki Y, Tatsuno K, Covington KR, Ueda H, Creighton CJ, Kato M, et al. Trans‐ancestry mutational landscape of hepatocellular carcinoma genomes. Nat Genet 2014;46:1267‐1273.

49. Schulze K, Imbeaud S, Letouzé E, Alexandrov LB, Calderaro J, Rebouissou S, et al. Exome sequencing of hepatocellular carcinomas identifies new mutational signatures and potential therapeutic targets. Nat Genet 2015;47:505‐511.

50. Ally A, Balasundaram M, Carlsen R, Chuah E, Clarke A, Dhalla N, et al. Comprehensive and integrative genomic characterization of hepatocellular carcinoma. Cell 2017;169:1327‐1341.e23.

51. Gao Q, Zhu H, Dong L, Shi W, Chen R, Song Z, et al. Integrated proteogenomic characterization of HBV‐related hepatocellular carcinoma. Cell 2019;179:561‐577.e22.

52. Johnson PJ, Pirrie SJ, Cox TF, Berhane S, Teng M, Palmer D, et al. The detection of hepatocellular carcinoma using a prospectively developed and validated model based on serological biomarkers. Cancer Epidemiol Biomarkers Prev 2014;23:144‐153.

53. Wang M, Sanda M, Comunale MA, Herrera H, Swindell C, Kono Y, et al. Changes in the glycosylation of kininogen and the development of a kininogen‐based algorithm for the early detection of HCC. Cancer Epidemiol Biomarkers Prev 2017;26:795‐803.

54. von Felden J, Craig AJ, Villanueva A. Role of circulating tumor DNA to help decision‐making in hepatocellular carcinoma. Oncoscience 2018;5:209‐211.

55. Kisiel JB, Dukek BA, Kanipakam RVSR, Ghoz HM, Yab TC, Berger CK, et al. Hepatocellular carcinoma detection by plasma methylated DNA: discovery, phase I pilot, and phase II clinical validation. Hepatology 2019;69:1180‐1192.

56. Khatri G, Pedrosa I, Ananthakrishnan L, de Leon AD, Fetzer DT, Leyendecker J, et al. Abbreviated‐protocol screening MRI versus complete‐protocol diagnostic MRI for detection of hepatocellular carcinoma in patients with cirrhosis: an equivalence study using LI‐RADS v2018. J Magn Reson Imaging 2020;51:415‐425.

57. Pepe MS, Etzioni R, Feng Z, Potter JD, Thompson ML, Thornquist M, et al. Phases of biomarker development for early detection of cancer. J Natl Cancer Inst 2001;93:1054‐1061.

58. Pepe MS, Feng Z, Janes H, Bossuyt PM, Potter JD. Pivotal evaluation of the accuracy of a biomarker used for classification or prediction: standards for study design. J Natl Cancer Inst 2008;100:1432‐1438.

59. Ioannou GN, Green P, Lowy E, Mun EJ, Berry K. Differences in hepatocellular carcinoma risk, predictors and trends over time according to etiology of cirrhosis. PLoS One 2018;13:e0204412.

60. Kanwal F, Kramer JR, Mapakshi S, Natarajan Y, Chayanupatkul M, Richardson PA, et al. Risk of hepatocellular cancer in patients with non‐alcoholic fatty liver disease. Gastroenterology 2018;155:1828‐1837.e2.

61. Kanwal F, Singal AG. Surveillance for hepatocellular carcinoma: current best practice and future direction. Gastroenterology 2019;157:54‐56.

62. Yang H‐I, Yeh M‐L, Wong GL, Peng C‐Y, Chen C‐H, Trinh HN, et al. Real‐world effectiveness from the Asia Pacific Rim Liver Consortium for HBV risk score for the prediction of hepatocellular carcinoma in chronic hepatitis B patients treated with oral antiviral. Therapy J Infect Dis 2020;221:389‐399.

63. US Food and Drug Administration . Biomarkers at FDA. http://www.fda.gov/science‐research/about‐science‐research‐fda/biomarkers‐fda. Published March 29, 2018. Accessed November 15, 2019.

64. Goodsaid FM. The labyrinth of product development and regulatory approvals in liquid biopsy diagnostics. Clin Transl Sci 2019;12:431‐439.

65. Simon TG, Ma Y, Ludvigsson JF, Chong DQ, Giovannucci EL, Fuchs CS, et al. Association between aspirin use and risk of hepatocellular carcinoma. JAMA Oncol 2018;4:1683‐1690.

66. Malehmir M, Pfister D, Gallage S, Szydlowska M, Inverso D, Kotsiliti E, et al. Platelet GPIbα is a mediator and potential interventional target for NASH and subsequent liver cancer. Nat Med 2019;25:641‐655.

67. Simon TG, Duberg A‐S, Aleman S, Hagstrom H, Nguyen LH, Khalili H, et al. Lipophilic statins and risk for hepatocellular carcinoma and death in patients with chronic viral hepatitis: results from a nationwide Swedish population. Ann Intern Med 2019;171:318‐327.

68. Tseng C‐H. Metformin and risk of hepatocellular carcinoma in patients with type 2 diabetes. Liver Int 2018;38:2018‐2027.

69. Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, et al. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology 2018;67:358‐380.

70. Roayaie S, Jibara G, Tabrizian P, Park J‐W, Yang J, Yan L, et al. The role of hepatic resection in the treatment of hepatocellular cancer. Hepatology 2015;62:440‐451.

71. Roayaie S, Obeidat K, Sposito C, Mariani L, Bhoori S, Pellegrinelli A, et al. Resection of hepatocellular cancer ≤2 cm: results from two Western centers. Hepatology 2013;57:1426‐1435.

72. Muto Y, Moriwaki H, Ninomiya M, Adachi S, Saito A, Takasaki KT, et al. Prevention of second primary tumors by an acyclic retinoid, polyprenoic acid, in patients with hepatocellular carcinoma. Hepatoma Prevention Study Group. N Engl J Med 1996;334:1561‐1567.

73. Okita K, Izumi N, Matsui O, Tanaka K, Kaneko S, Moriwaki H, et al. Peretinoin after curative therapy of hepatitis C‐related hepatocellular carcinoma: a randomized double‐blind placebo‐controlled study. J Gastroenterol 2015;50:191‐202.

74. Takayama T, Sekine T, Makuuchi M, Yamasaki S, Kosuge T, Yamamoto J, et al. Adoptive immunotherapy to lower postsurgical recurrence rates of hepatocellular carcinoma: a randomised trial. Lancet 2000;356:802‐807.

75. Lau WY, Lai ECH, Leung TWT, Yu SCH. Adjuvant intra‐arterial iodine‐131‐labeled lipiodol for resectable hepatocellular carcinoma: a prospective randomized trial‐update on 5‐year and 10‐year survival. Ann Surg 2008;247:43‐48.

76. Bruix J, Takayama T, Mazzaferro V, Chau GY, Yang J, Kudo M, et al. Adjuvant sorafenib for hepatocellular carcinoma after resection or ablation (STORM): a phase 3, randomised, double‐blind, placebo‐controlled trial. Lancet Oncol 2015;16:1344‐1354.

77. Bott MJ, Yang SC, Park BJ, Adusumilli PS, Rusch VW, Isbell JM, et al. Initial results of pulmonary resection after neoadjuvant nivolumab in patients with resectable non‐small cell lung cancer. J Thorac Cardiovasc Surg 2019;158:269‐276.

78. Mazzaferro V, Regalia E, Doci R, Andreola S, Pulvirenti A, Bozzetti F, et al. Liver transplantation for the treatment of small hepatocellular carcinomas in patients with cirrhosis. N Engl J Med 1996;334:693‐699.

79. Yao FY, Mehta N, Flemming J, Dodge J, Hameed B, Fix O, et al. Downstaging of hepatocellular cancer before liver transplant: long‐term outcome compared to tumors within Milan criteria. Hepatology 2015;61:1968‐1977.

80. Sposito C, Mariani L, Germini A, Flores Reyes M, Bongini M, Grossi G, et al. Comparative efficacy of sorafenib versus best supportive care in recurrent hepatocellular carcinoma after liver transplantation: a case‐control study. J Hepatol 2013;59:59‐66.

81. Morales RE, Shoushtari AN, Walsh MM, Grewal P, Lipson EJ, Carvajal RD. Safety and efficacy of ipilimumab to treat advanced melanoma in the setting of liver transplantation. J Immunother Cancer 2015;3:22.

82. Breen DJ, Lencioni R. Image‐guided ablation of primary liver and renal tumours. Nat Rev Clin Oncol 2015;12:175‐186.

83. Cho YK, Kim JK, Kim MY, Rhim H, Han JK. Systematic review of randomized trials for hepatocellular carcinoma treated with percutaneous ablation therapies. Hepatology 2009;49:453‐459.

84. Glassberg MB, Ghosh S, Clymer JW, Qadeer RA, Ferko NC, Sadeghirad B, et al. Microwave ablation compared with radiofrequency ablation for treatment of hepatocellular carcinoma and liver metastases: a systematic review and meta‐analysis. Onco Targets Ther 2019;12:6407‐6438.

85. Nault J‐C, Sutter O, Nahon P, Ganne‐Carrié N, Séror O. Percutaneous treatment of hepatocellular carcinoma: state of the art and innovations. J Hepatol 2018;68:783‐797.

86. Llovet JM, Real MI, Montaña X, Planas R, Coll S, Aponte J, et al. Arterial embolisation or chemoembolisation versus symptomatic treatment in patients with unresectable hepatocellular carcinoma: a randomised controlled trial. Lancet 2002;359:1734‐1739.

87. Lo C‐M, Ngan H, Tso W‐K, Liu C‐L, Lam C‐M, Poon RT‐P, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002;35:1164‐1171.

88. Llovet JM, Bruix J. Systematic review of randomized trials for unresectable hepatocellular carcinoma: chemoembolization improves survival. Hepatology 2003;37:429‐442.

89. Salem R, Mazzaferro V, Sangro B. Yttrium 90 radioembolization for the treatment of hepatocellular carcinoma: biological lessons, current challenges, and clinical perspectives. Hepatology 2013;58:2188‐2197.

90. Bolondi L, Burroughs A, Dufour J‐F, Galle PR, Mazzaferro V, Piscaglia F, et al. Heterogeneity of patients with intermediate (BCLC B) hepatocellular carcinoma: proposal for a subclassification to facilitate treatment decisions. Semin Liver Dis 2012;32:348‐359.

91. Kudo M, Arizumi T, Ueshima K, Sakurai T, Kitano M, Nishida N. Subclassification of BCLC B stage hepatocellular carcinoma and treatment strategies: proposal of modified Bolondi’s subclassification (Kinki criteria). Dig Dis 2015;33:751‐758.

92. Forner A, Gilabert M, Bruix J, Raoul JL. Treatment of intermediate‐stage hepatocellular carcinoma. Nat Rev Clin Oncol 2014;11:525‐535.

93. Meyer T, Fox R, Ma YT, Ross PJ, James MW, Sturgess R, et al. Sorafenib in combination with transarterial chemoembolisation in patients with unresectable hepatocellular carcinoma (TACE 2): a randomised placebo‐controlled, double‐blind, phase 3 trial. Lancet Gastroenterol Hepatol 2017;2:565‐575.

94. Lencioni R, Llovet JM, Han G, Tak WY, Yang J, Guglielmi A, et al. Sorafenib or placebo plus TACE with doxorubicin‐eluting beads for intermediate stage HCC: the SPACE trial. J Hepatol 2016;64:1090‐1098.

95. Kudo M, Han G, Finn RS, Poon RT, Blanc JF, Yan L, et al. Brivanib as adjuvant therapy to transarterial chemoembolization in patients with hepatocellular carcinoma: a randomized phase III trial. Hepatology 2014;60:1697‐1707.

96. Kudo M, Cheng A‐L, Park J‐W, Park JH, Liang P‐C, Hidaka H, et al. Orantinib versus placebo combined with transcatheter arterial chemoembolisation in patients with unresectable hepatocellular carcinoma (ORIENTAL): a randomised, double‐blind, placebo‐controlled, multicentre, phase 3 study. Lancet Gastroenterol Hepatol 2018;3:37‐46.

97. Johnson PJ, Berhane S, Kagebayashi C, Satomura S, Teng M, Reeves HL, et al. Assessment of liver function in patients with hepatocellular carcinoma: a new evidence‐based approach‐the ALBI grade. J Clin Oncol 2015;33:550‐558.

98. Edeline J, Blanc J‐F, Johnson P, Campillo‐Gimenez B, Ross P, Ma YT, et al. A multicentre comparison between Child Pugh and albumin‐bilirubin scores in patients treated with sorafenib for hepatocellular carcinoma. Liver Int 2016;36:1821‐1828.

99. Kadalayil L, Benini R, Pallan L, O’Beirne J, Marelli L, Yu D, et al. A simple prognostic scoring system for patients receiving transarterial embolisation for hepatocellular cancer. Ann Oncol 2013;24:2565‐2570.

100. Waked I, Berhane S, Toyoda H, Chan SL, Stern N, Palmer D, et al. Transarterial chemo‐embolisation of hepatocellular carcinoma: impact of liver function and vascular invasion. Br J Cancer 2017;116:448‐454.

101. Lencioni R, de Baere T, Soulen MC, Rilling WS, Geschwind J‐FH. Lipiodol transarterial chemoembolization for hepatocellular carcinoma: a systematic review of efficacy and safety data. Hepatology 2016;64:106‐116.

102. Meyer T, Kirkwood A, Roughton M, Beare S, Tsochatzis E, Yu D, et al. A randomised phase II/III trial of 3‐weekly cisplatin‐based sequential transarterial chemoembolisation versus embolisation alone for hepatocellular carcinoma. Br J Cancer 2013;108:1252‐1259.

103. Lammer J, Malagari K, Vogl T, Pilleul F, Denys A, Watkinson A, et al. Prospective randomized study of doxorubicin‐eluting‐bead embolization in the treatment of hepatocellular carcinoma: results of the PRECISION V study. Cardiovasc Intervent Radiol 2010;33:41‐52.

104. de Baere T, Arai Y, Lencioni R, Geschwind J‐F, Rilling W, Salem R, et al. Treatment of liver tumors with lipiodol TACE: technical recommendations from experts opinion. Cardiovasc Intervent Radiol 2016;39:334‐343.

105. Gillmore R, Stuart S, Kirkwood A, Hameeduddin A, Woodward N, Burroughs AK, et al. EASL and mRECIST responses are independent prognostic factors for survival in hepatocellular cancer patients treated with transarterial embolization. J Hepatol 2011;55:1309‐1316.

106. Vincenzi B, Di Maio M, Silletta M, D’Onofrio L, Spoto C, Piccirillo MC, et al. Prognostic relevance of objective response according to EASL criteria and mRECIST criteria in hepatocellular carcinoma patients treated with loco‐regional therapies: a literature‐based meta‐analysis. PLoS One 2015;10:e0133488.

107. Kudo M. Proposal of primary endpoints for TACE combination trials with systemic therapy: lessons learned from 5 negative trials and the positive TACTICS trial. Liver Cancer 2018;7:225‐234.

108. Kudo M, Ueshima K, Ikeda M, Torimura T, Tanabe N, Aikata H, et al. Randomised, multicentre prospective trial of transarterial chemoembolisation (TACE) plus sorafenib as compared with TACE alone in patients with hepatocellular carcinoma: TACTICS trial. Gut 2019 Dec 4. https://doi.org/10.1136/gutjnl‐2019‐318934. [Epub ahead of print]

109. Kudo M, Kubo S, Takayasu K, Sakamoto M, Tanaka M, Ikai I, et al. Response Evaluation Criteria in Cancer of the Liver (RECICL) proposed by the Liver Cancer Study Group of Japan (2009 revised version). Hepatol Res 2010;40:686‐692.

110. Kudo M. A new treatment option for intermediate‐stage hepatocellular carcinoma with high tumor burden: initial lenvatinib therapy with subsequent selective TACE. Liver Cancer 2019;8:299‐311.

111. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205‐216.

112. Lencioni R. New data supporting modified RECIST (mRECIST) for hepatocellular carcinoma. Clin Cancer Res 2013;19:1312‐1314.

113. Nishino M, Giobbie‐Hurder A, Gargano M, Suda M, Ramaiya NH, Hodi FS. Developing a common language for tumor response to immunotherapy: immune‐related response criteria using unidimensional measurements. Clin Cancer Res 2013;19:3936‐3943.

114. Bohnsack O, Hoos A, Ludajic K. Adaptation and modification of the immune related response criteria (IRRC): irRECIST. J Clin Orthod 2014;32:e22121.

115. Hodi FS, Ballinger M, Lyons B, Soria J‐C, Nishino M, Tabernero J, et al. Immune‐modified Response Evaluation Criteria in Solid Tumors (imRECIST): refining guidelines to assess the clinical benefit of cancer immunotherapy. J Clin Oncol 2018;36:850‐858.

116. Ikeda M, Sung MW, Kudo M, Kobayashi M, Baron AD, Finn RS, et al. A phase 1b trial of lenvatinib (LEN) plus pembrolizumab (PEM) in patients (pts) with unresectable hepatocellular carcinoma (uHCC). J Clin Oncol 2018;36:4076.

117. Moehler M, Heo J, Lee HC, Tak WY, Chao Y, Paik SW, et al. Vaccinia‐based oncolytic immunotherapy pexastimogene devacirepvec in patients with advanced hepatocellular carcinoma after sorafenib failure: a randomized multicenter phase IIb trial (TRAVERSE). Oncoimmunology 2019;8:1615817.

118. Heo J, Reid T, Ruo L, Breitbach CJ, Rose S, Bloomston M, et al. Randomized dose‐finding clinical trial of oncolytic immunotherapeutic vaccinia JX‐594 in liver cancer. Nat Med 2013;19:329‐336.

119. Llovet JM, Zucman‐Rossi J, Pikarsky E, Sangro B, Schwartz M, Sherman M, et al. Hepatocellular carcinoma. Nat Rev Dis Prim 2016;2:16018‐16023.

120. Bruix J, Sherman M; American Association for the Study of Liver Diseases . Management of hepatocellular carcinoma: an update. Hepatology 2011;53:1020‐1022.

121. Qin S, Bai Y, Lim HY, Thongprasert S, Chao Y, Fan J, et al. Randomized, multicenter, open‐label study of oxaliplatin plus fluorouracil/leucovorin versus doxorubicin as palliative chemotherapy in patients with advanced hepatocellular carcinoma from Asia. J Clin Oncol 2013;31:3501‐3508.

122. Abou‐Alfa GK, Niedzwieski D, Knox JJ, Kaubisch A, Posey J, Tan BR, et al. Phase III randomized study of sorafenib plus doxorubicin versus sorafenib in patients with advanced hepatocellular carcinoma (HCC): CALGB 80802 (Alliance). J Clin Orthod 2016;34:192.

123. Kudo M, Ueshima K, Yokosuka O, Ogasawara S, Obi S, Izumi N, et al. Sorafenib plus low‐dose cisplatin and fluorouracil hepatic arterial infusion chemotherapy versus sorafenib alone in patients with advanced hepatocellular carcinoma (SILIUS): a randomised, open label, phase 3 trial. Lancet Gastroenterol Hepatol 2018;3:424‐432.

124. Yeo W, Mok TS, Zee B, Leung TWT, Lai PBS, Lau WY, et al. A randomized phase III study of doxorubicin versus cisplatin/interferon alpha‐2b/doxorubicin/fluorouracil (PIAF) combination chemotherapy for unresectable hepatocellular carcinoma. J Natl Cancer Inst 2005;97:1532‐1538.

125. Chow PKH, Tai B‐C, Tan C‐K, Machin D, Win KM, Johnson PJ, et al. High‐dose tamoxifen in the treatment of inoperable hepatocellular carcinoma: a multicenter randomized controlled trial. Hepatology 2002;36:1221‐1226.

126. Johnson PJ, Qin S, Park J‐W, Poon RTP, Raoul J‐L, Philip PA, et al. Brivanib versus sorafenib as first‐line therapy in patients with unresectable, advanced hepatocellular carcinoma: results from the randomized phase III BRISK‐FL study. J Clin Oncol 2013;31:3517‐3524.

127. Llovet JM, Decaens T, Raoul JL, Boucher E, Kudo M, Chang C, et al. Brivanib in patients with advanced hepatocellular carcinoma who were intolerant to sorafenib or for whom sorafenib failed: results from the randomized phase III BRISK‐PS study. J Clin Oncol 2013;31:3509‐3516.

128. Cainap C, Qin S, Huang W‐T, Chung IJ, Pan H, Cheng Y, et al. Linifanib versus sorafenib in patients with advanced hepatocellular carcinoma: results of a randomized phase III trial. J Clin Oncol 2015;33:172‐179.

129. Zhu AX, Rosmorduc O, Evans TRJ, Ross PJ, Santoro A, Carrilho FJ, et al. SEARCH: a phase III, randomized, double‐blind, placebo‐controlled trial of sorafenib plus erlotinib in patients with advanced hepatocellular carcinoma. J Clin Oncol 2015;33:559‐566.

130. Zhu AX, Kudo M, Assenat E, Cattan S, Kang Y‐K, Lim HY, et al. Effect of everolimus on survival in advanced hepatocellular carcinoma after failure of sorafenib: the EVOLVE‐1 randomized clinical trial. JAMA 2014;312:57‐67.

131. Rimassa L, Assenat E, Peck‐Radosavljevic M, Pracht M, Zagonel V, Mathurin P, et al. Tivantinib for second‐line treatment of MET‐high, advanced hepatocellular carcinoma (METIV‐HCC): a final analysis of a phase 3, randomised, placebo‐controlled study. Lancet Oncol 2018;19:682‐693.

132. Merle P, Blanc J‐F, Phelip J‐M, Pelletier G, Bronowicki J‐P, Touchefeu Y, et al. Doxorubicin‐loaded nanoparticles for patients with advanced hepatocellular carcinoma after sorafenib treatment failure (RELIVE): a phase 3 randomised controlled trial. Lancet Gastroenterol Hepatol 2019;4:454‐465.

133. Abou‐Alfa GK, Qin S, Ryoo B‐Y, Lu S‐N, Yen C‐J, Feng Y‐H, et al. Phase III randomized study of second line ADI‐PEG 20 plus best supportive care versus placebo plus best supportive care in patients with advanced hepatocellular carcinoma. Ann Oncol 2018;29:1402‐1408.

134. Finn RS, Qin S, Ikeda M, Galle PR, Ducreux M, Kim TY et al. Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma. N Engl J Med 2020;382:1894‐1905.

135. Zhu AX, Park JO, Ryoo B‐Y, Yen C‐J, Poon R, Pastorelli D, et al. Ramucirumab versus placebo as second‐line treatment in patients with advanced hepatocellular carcinoma following first‐line therapy with sorafenib (REACH): a randomised, double‐blind, multicentre, phase 3 trial. Lancet Oncol 2015;16:859‐870.

136. Llovet JM, Pena CEA, Lathia CD, Shan M, Meinhardt G, Bruix J, et al. Plasma biomarkers as predictors of outcome in patients with advanced hepatocellular carcinoma. Clin Cancer Res 2012;18:2290‐2300.

137. Vilgrain V, Pereira H, Assenat E, Guiu B, Ilonca AD, Pageaux G‐P, et al. Efficacy and safety of selective internal radiotherapy with yttrium‐90 resin microspheres compared with sorafenib in locally advanced and inoperable hepatocellular carcinoma (SARAH): an open‐label randomised controlled phase 3 trial. Lancet Oncol 2017;18:1624‐1636.

138. Chow PKH, Gandhi M, Tan S‐B, Khin MW, Khasbazar A, Ong J, et al. SIRveNIB: selective internal radiation therapy versus sorafenib in asia‐pacific patients with hepatocellular carcinoma. J Clin Oncol 2018;36:1913‐1921.

139. Ricke J, Sangro B, Amthauer H, Bargellini I, Bartenstein P, De Toni E, et al. The impact of combining selective internal radiation therapy (SIRT) with sorafenib on overall survival in patients with advanced hepatocellular carcinoma: the Soramic trial palliative cohort. J Hepatol 2018;68(Suppl. 1):S102.

140. Cheng A‐L, Kang Y‐K, Chen Z, Tsao C‐J, Qin S, Kim JS, et al. Efficacy and safety of sorafenib in patients in the Asia‐Pacific region with advanced hepatocellular carcinoma: a phase III randomised, double‐blind, placebo‐controlled trial. Lancet Oncol 2009;10:25‐34.

141. He M, Li Q, Zou R, Shen J, Fang W, Tan G, et al. Sorafenib plus hepatic arterial infusion of oxaliplatin, fluorouracil, and leucovorin versus sorafenib alone for hepatocellular carcinoma with portal vein invasion. A randomized clinical trial. JAMA Oncol 2019;5:953.

142. Jouve J‐L, Lecomte T, Bouché O, Barbier E, Khemissa Akouz F, Riachi G, et al. Pravastatin combination with sorafenib does not improve survival in advanced hepatocellular carcinoma. J Hepatol 2019;71:516‐522.

143. Abou‐Alfa GK, Shi Q, Knox JJ, Kaubisch A, Niedzwiecki D, Posey J, et al. Assessment of treatment with sorafenib plus doxorubicin versus sorafenib alone in patients with advanced hepatocellular carcinoma: phase 3 CALGB 80802 randomized, clinical trial. JAMA Oncol 2019;5:1582.

144. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet‐Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin 2015;65:87‐108.

145. Gan HK, You B, Pond GR, Chen EX. Assumptions of expected benefits in randomized phase III trials evaluating systemic treatments for cancer. J Natl Cancer Inst 2012;104:590‐598.

146. Oxnard GR, Wilcox KH, Gonen M, Polotsky M, Hirsch BR, Schwartz LH. Response rate as a regulatory end point in single‐arm studies of advanced solid tumors. JAMA Oncol 2016;2:772‐779.

147. Sangro B, Gomez‐Martin C, de la Mata M, Iñarrairaegui M, Garralda E, Barrera P, et al. A clinical trial of CTLA4 blockade with tremelimumab in patients with hepatocellular carcinoma and chronic hepatitis C. J Hepatol 2013;59:81‐88.

148. Qin SK, Ren ZG, Meng ZQ, Chen ZD, Chai XL, Xiong JP, et al. A randomized multicentered phase II study to evaluate SHR‐1210 (PD‐1 antibody) in subjects with advanced hepatocellular carcinoma (HCC) who failed or intolerable to prior systemic treatment [Abstract]. Ann Oncol 2018. Abstract LBA27. https://academic.oup.com/annonc/article/29/suppl_8/mdy424.029/5141688. Accessed September 23, 2019.

149. Finn RS, Ryoo B‐Y, Merle P, Kudo M, Bouattour M, Lim H‐Y, et al. Results of KEYNOTE‐240: phase 3 study of pembrolizumab (Pembro) versus best supportive care (BSC) for second line therapy in advanced hepatocellular carcinoma (HCC). J Clin Orthod 2019;37:4004.

150. Duffy AG, Ulahannan SV, Makorova‐Rusher O, Rahma O, Wedemeyer H, Pratt D, et al. Tremelimumab in combination with ablation in patients with advanced hepatocellular carcinoma. J Hepatol 2017;66:545‐551.

151. Yau T, Kang Y‐K, Kim T‐Y, El‐Khoueiry AB, Santoro A, Sangro B, et al. Nivolumab (NIVO) + ipilimumab (IPI) combination therapy in patients (pts) with advanced hepatocellular carcinoma (aHCC): results from CheckMate 040. J Clin Orthod 2019;37:4012.

152. Greten TF, Lai CW, Li G, Staveley‐O’Carroll KF. Targeted and immune‐based therapies for hepatocellular carcinoma. Gastroenterology 2019;156:510‐524.

153. Greten TF, Sangro B. Targets for immunotherapy of liver cancer. J Hepatol 2018;68:157‐166.

154. Anagnostou V, Yarchoan M, Hansen AR, Wang H, Verde F, Sharon E, et al. Immuno‐oncology trial endpoints: capturing clinically meaningful activity. Clin Cancer Res 2017;23:4959‐4969.

155. Ritchie G, Gasper H, Man J, Lord S, Marschner I, Friedlander M, et al. Defining the most appropriate primary end point in phase 2 trials of immune checkpoint inhibitors for advanced solid cancers: a systematic review and meta‐analysis. JAMA Oncol 2018;4:522‐528.

156. Simon RM, Paik S, Hayes DF. Use of archived specimens in evaluation of prognostic and predictive biomarkers. J Natl Cancer Inst 2009;101:1446‐1452.

157. Mandrekar SJ, Sargent DJ. Clinical trial designs for predictive biomarker validation: theoretical considerations and practical challenges. J Clin Oncol 2009;27:4027‐4034.

158. Mandrekar SJ, Sargent DJ. Genomic advances and their impact on clinical trial design. Genome Med 2009;1:69.

159. Renfro LA, An M‐W, Mandrekar SJ. Precision oncology: a new era of cancer clinical trials. Cancer Lett 2017;387:121‐126.

160. National Cancer Institute . NCI dictionary of cancer terms. Biomarker. n.d. https://www.cancer.gov/publications/dictionaries/cancer‐terms/def/biomarker. Accessed June 2020.

• Cited Here

161. Rebouissou S, La Bella T, Rekik S, Imbeaud S, Calatayud A‐L, Rohr‐Udilova N, et al. Proliferation markers are associated with MET expression in hepatocellular carcinoma and predict tivantinib sensitivity in vitro. Clin Cancer Res 2017;23:4364‐4375.

162. Zhu AX, Baron AD, Malfertheiner P, Kudo M, Kawazoe S, Pezet D, et al. Ramucirumab as second‐line treatment in patients with advanced hepatocellular carcinoma: analysis of REACH trial results by Child‐Pugh score. JAMA Oncol 2017;3:235‐243.

163. Kim RD, Sarker D, Meyer T, Yau T, Macarulla T, Park J‐W, et al. First‐in‐human phase I study of fisogatinib (BLU‐554) validates aberrant fibroblast growth factor 19 signaling as a driver event in hepatocellular carcinoma. Cancer Discov 2019;9:1696‐1707.

164. Sia D, Jiao Y, Martinez‐Quetglas I, Kuchuk O, Villacorta‐Martin C, Castro de Moura M, et al. Identification of an immune‐specific class of hepatocellular carcinoma, based on molecular features. Gastroenterology 2017;153:812‐826.

165. Finn RS, Kudo M, Cheng A‐L, Wyrwicz L, Ngan R, Blanc J‐F, et al. Analysis of serum biomarkers (BM) in patients (pts) from a phase 3 study of lenvatinib (LEN) versus sorafenib (SOR) as first‐line treatment for unresectable hepatocellular carcinoma (uHCC) [Abstract]. Ann Oncol 2017. Abstract LBA30. https://www.annalsofoncology.org/article/S0923‐7534(20)39122‐5/fulltext. Accessed 2019 July 26, 2019.

166. Kaseb AO, Carmagnani Pestana R, Vence LM, Blando JM, Singh S, Ikoma N, et al. Randomized, open‐label, perioperative phase II study evaluating nivolumab alone versus nivolumab plus ipilimumab in patients with resectable HCC. J Clin Orthod 2019;37:185.

167. Kwak EL, Bang Y‐J, Camidge DR, Shaw AT, Solomon B, Maki RG, et al. Anaplastic lymphoma kinase inhibition in non‐small‐cell lung cancer. N Engl J Med 2010;363:1693‐1703.

168. Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non‐small‐cell lung cancer to gefitinib. N Engl J Med 2004;350:2129‐2139.

169. Slamon DJ, Leyland‐Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001;344:783‐792.

170. Finn RS, Martin M, Rugo HS, Jones S, Im S‐A, Gelmon K, et al. Palbociclib and letrozole in advanced breast cancer. N Engl J Med 2016;375:1925‐1936.

171. Demetri GD, von Mehren M, Blanke CD, Van den Abbeele AD, Eisenberg B, Roberts PJ, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med 2002;347:472‐480.

172. Druker BJ, Guilhot F, O’Brien SG, Gathmann I, Kantarjian H, Gattermann N, et al. Five‐year follow‐up of patients receiving imatinib for chronic myeloid leukemia. N Engl J Med 2006;355:2408‐2417.

173. Pantel K. Blood‐based analysis of circulating cell‐free DNA and tumor cells for early cancer detection. PLoS Med 2016;13:e1002205.

174. Mandrekar SJ, An M‐W, Sargent DJ. A review of phase II trial designs for initial marker validation. Contemp Clin Trials 2013;36:597‐604.

175. Labgaa I, Villanueva A. Liquid biopsy in liver cancer. Discov Med 2015;19:263‐273.

176. Corcoran RB, Chabner BA. Application of cell‐free DNA analysis to cancer treatment. N Engl J Med 2018;379:1754‐1765.

177. Wan JCM, Massie C, Garcia‐Corbacho J, Mouliere F, Brenton JD, Caldas C, et al. Liquid biopsies come of age: towards implementation of circulating tumour DNA. Nat Rev Cancer 2017;17:223‐238.

178. US Food and Drug Administration . FDA approves first blood test to detect gene mutation associated with non‐small cell lung cancer. https://www.fda.gov/news‐events/press‐announcements/fda‐approves‐first‐blood‐test‐detect‐gene‐mutation‐associated‐non‐small‐cell‐lung‐cancer. Published March 1, 2019. Accessed May 16, 2019.

179. Labgaa I, Villacorta‐Martin C, D’Avola D, Craig AJ, von Felden J, Martins‐Filho SN, et al. A pilot study of ultra‐deep targeted sequencing of plasma DNA identifies driver mutations in hepatocellular carcinoma. Oncogene 2018;37:3740‐3752.

180. Qu C, Wang Y, Wang P, Chen K, Wang M, Zeng H, et al. Detection of early‐stage hepatocellular carcinoma in asymptomatic HBsAg‐seropositive individuals by liquid biopsy. Proc Natl Acad Sci USA 2019;116:6308‐6312.

181. Xu R‐H, Wei W, Krawczyk M, Wang W, Luo H, Flagg K, et al. Circulating tumour DNA methylation markers for diagnosis and prognosis of hepatocellular carcinoma. Nat Mater 2017;16:1155‐1161.

182. Bhan I, Mosesso K, Goyal L, Philipp J, Kalinich M, Franses JW, et al. Detection and analysis of circulating epithelial cells in liquid biopsies from patients with liver disease. Gastroenterology 2018;155:2016‐2018.e11.

183. Oh CR, Kong S‐Y, Im H‐S, Kim HJ, Kim MK, Yoon K‐A, et al. Genome‐wide copy number alteration and VEGFA amplification of circulating cell‐free DNA as a biomarker in advanced hepatocellular carcinoma patients treated with Sorafenib. BMC Cancer 2019;19:292.

184. Lim HY, Merle P, Weiss KH, Yau TC, Ross P, Mazzaferro V, et al. Phase II studies with refametinib or refametinib plus sorafenib in patients with RAS‐mutated hepatocellular carcinoma. Clin Cancer Res 2018;24:4650‐4661.

185. Chie W‐C, Blazeby JM, Hsiao C‐F, Chiu H‐C, Poon RT, Mikoshiba N, et al. International cross‐cultural field validation of an European Organization for Research and Treatment of Cancer questionnaire module for patients with primary liver cancer, the European Organization for Research and Treatment of Cancer quality‐of‐life questionnaire HCC18. Hepatology 2012;55:1122‐1129.

186. Blazeby JM, Currie E, Zee BCY, Chie W‐C, Poon RT, Garden OJ, et al. Development of a questionnaire module to supplement the EORTC QLQ‐C30 to assess quality of life in patients with hepatocellular carcinoma, the EORTC QLQ‐HCC18. Eur J Cancer 2004;40:2439‐2444.

187. Huang G, Chen X, Lau WY, Shen F, Wang RY, Yuan SX, et al. Quality of life after surgical resection compared with radiofrequency ablation for small hepatocellular carcinomas. Br J Surg 2014;101:1006‐1015.

188. Salem R, Gilbertsen M, Butt Z, Memon K, Vouche M, Hickey R, et al. Increased quality of life among hepatocellular carcinoma patients treated with radioembolization, compared with chemoembolization. Clin Gastroenterol Hepatol 2013;11:1358‐1365.e1.

189. Kudo M, Trevisani F, Abou‐Alfa GK, Rimassa L. Hepatocellular carcinoma: therapeutic guidelines and medical treatment. Liver Cancer 2016;6:16‐26.

190. Omata M, Cheng A‐L, Kokudo N, Kudo M, Lee JM, Jia J, et al. Asia‐Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int 2017;11:317‐370.

191. Kokudo N, Takemura N, Hasegawa K, Takayama T, Kubo S, Shimada M, et al. Clinical practice guidelines for hepatocellular carcinoma: the Japan Society of Hepatology 2017 (4th JSH‐HCC guidelines) 2019 update. Hepatol Res 2019;49:1109‐1113.

192. Surveillance Group, Diagnosis Group, Staging Group, Surgery Group, Local Ablation Group, TACE/TARE/HAI Group , et al. Management consensus guideline for hepatocellular carcinoma: 2016 updated by the Taiwan Liver Cancer Association and the Gastroenterological Society of Taiwan. J Formos Med Assoc 2016;2018:381‐403.

193. Zhou J, Sun H‐C, Wang Z, Cong W‐M, Wang J‐H, Zeng M‐S, et al. Guidelines for diagnosis and treatment of primary liver cancer in China (2017 edition). Liver Cancer 2018;7:235‐260.

194. Korean Liver Cancer Association, National Cancer Center . 2018 Korean Liver Cancer Association‐National Cancer Center Korea practice guidelines for the management of hepatocellular carcinoma. Gut Liver 2019;13:227‐299.

195. Sharma P, Allison JP. The future of immune checkpoint therapy. Science 2015;348:56‐61.

196. Fukumura D, Kloepper J, Amoozgar Z, Duda DG, Jain RK. Enhancing cancer immunotherapy using antiangiogenics: opportunities and challenges. Nat Rev Clin Oncol 2018;15:325‐340.

197. Kalbasi A, Ribas A. Tumour‐intrinsic resistance to immune checkpoint blockade. Nat Rev Immunol 2020;20:25‐39.

198. Ruiz de Galarreta M, Bresnahan E, Molina‐Sanchez P, Lindblad KE, Maier B, Sia D, et al. β‐catenin activation promotes immune escape and resistance to anti‐PD‐1 therapy in hepatocellular carcinoma. Cancer Discov 2019;9:1124‐1141.

199. Pinyol R, Sia D, Llovet JM. Immune exclusion‐Wnt/CTNNB1 class predicts resistance to immunotherapies in HCC. Clin Cancer Res 2019;25:2021‐2023.

200. Harding JJ, Nandakumar S, Armenia J, Khalil DN, Albano M, Ly M, et al. Prospective genotyping of hepatocellular carcinoma: clinical implications of next generation sequencing for matching patients to targeted and immune therapies. Clin Cancer Res 2019;25:2116‐2126.

Author names in bold designate shared co‐first authorship.

© 2020 by the American Association for the Study of Liver Diseases.