Although treatments for hepatocellular carcinoma (HCC) at the early stage are well established1, 2 treatment allocation remains unsatisfactory for a number of patients in Barcelona Cancer Liver Clinic (BCLC) intermediate and advanced stages.3
For intermediate HCCs, transarterial chemoembolization (TACE) is the only therapy with proven survival benefit, thus becoming the only recommended treatment in several guidelines.1, 2 Median overall survival (OS) increased from 16 months in untreated patients to 19-20 months with TACE1 and may reach 36-45 months in best responders. However, results of TACE are variable, because patients with intermediate HCC represent a heterogeneous population, including very different tumor burdens, liver function, and possible comorbidities. There is no universally accepted subgroup stratification of these patients allowing tailored treatments. However, recent literature has outlined clinical situations where TACE is poorly effective or contraindicated and where treatment repetition should be stopped.4, 5 For the advanced stage, where the expected median OS of untreated patients is 6-7 months, the recommended therapy is sorafenib, given the evidence of efficacy on survival. Again, the results of this therapy are variable, depending on individual sensitivity and severity of adverse events.
Choice of proper treatment is sometimes complicated by the difficult diagnosis of neoplastic peripheral portal vein thrombosis (PVT), which may confound stage classification. TACE is still recommended for these patients by some guidelines,6, 7 whereas American Association for the Study of Liver Diseases (AASLD) and European Association for the Study of the Liver/European Organization for Research and Treatment of Cancer (EASL-EORTC) guidelines include this situation in the advanced stage and suggest only sorafenib.
This multifaced clinical setting of intermediate and advanced HCCs leaves a large space for tailoring treatments and for experimenting with new strategies. Transarterial radioembolization (TARE) with yttrium-90 (Y90RE) has recently been increasingly performed in many clinical centers. It has been investigated in several cohorts of patients with intermediate and advanced HCC,8, 9 with promising antitumoral effect—median OS of 17.2 months for intermediate stage and 12 months for advanced stage—and acceptable safety profile. The procedure was approved for treatment of HCC with or without PVT both in Europe and America, but definite recommendation for its use outside clinical trials is not given by either AASLD or EASL-EORTC guidelines.
Indeed, all randomized, controlled trials (RCTs) are still ongoing, including the phase III PREMIERE study investigating TARE versus TACE (NCT00956930), STOP-HCC investigating sorafenib with or without TARE (NCT01556490), the SORAMIC trial comparing sorafenib with or without TARE (NCT01126645), and the SIRveNIB (NCT 01135056) and SARAH (01482442) trials, comparing sorafenib versus TARE. Their final results are not yet available.
Thus, its application in clinical practice is still based on the availability of the procedure in highly specialized centers, local clinical expertise, and variable allowance for such expensive treatment by healthcare systems worldwide.
A study giving further contribution to understanding the possible role of Y90RE is published in the current issue of HEPATOLOGY.10 This is the first prospective phase II trial that has sought to determine the efficacy and safety of Y90RE in intermediate and advanced HCC. Of note is that the time to progression (TTP) of BCLC-B patients (13 months) was nearly identical to what was previously described in retrospective series. The limited number of grade 3 or 4 toxicities in properly selected patients (mostly, but not all, Child-Pugh A) was also confirmed prospectively. However, the significant rate of liver decompensation (treatment and nontreatment related) deserves a comment. It corresponded to approximately 35% within 6 months from treatment, with a mortality at 90 days of 3.8% (no case was treatment related). Decompensation figures apparently differ from those reported for conventional TACE (30.6%) and TACE with drug-eluting beads (20.4%), whereas mortality was <2%, ranging from 0% to 10% in selected candidates. However, populations submitted to the latter therapies were expected to have better survival than in this study, which included mostly advanced HCCs.
The prospective design of this study is also relevant for outlining patient disposition. A rate of 15%-20% of potential candidates are not finally treated because of excessive intrahepatic arteriovenous shunting. This finding cannot be ignored when considering patient outcomes. The analysis of survival of only patients who were finally treated might, in fact, lead to a survival overestimation: A definite answer to this issue will come only from the intention-to-treat analysis of RCTs.
This study has also dissected the group of patients with PVT into four different subgroups according to location and extension of thrombosis. This is relevant, because no distinction has been made by the BCLC in patients with PVT. Results of this study show that all endpoints, including TTP, objective response, disease control rate and OS, are different according to type of PVT, with survival of treated patients with PVT 1-2 similar to that of patients without PVT. Overall, the investigators state that TTP and survival achieved in patients with PVT compare favorably with those of sorafenib in similar patients.
How can the results of this study affect clinical practice? The critical issue still remains regarding the selection of patients who would achieve greater benefit from Y90RE, in comparison to conventional treatments. Again, RCTs are lacking and at present this question cannot be answered. As stated by the investigators, this study captured patients accepting Y90RE as an experimental treatment, being suboptimal candidates for both TACE and sorafenib, although “suboptimal” was not defined and the inclusion criteria only specified that these patients were originally referred for transplantation and excluded for tumor extension. The majority of these patients had 2-3 (48.1%) and >3 nodules (21.1%), exceeding both the Milan criteria and most likely the up-to-seven criterion,11 with largest tumor diameter up to 150 mm in the whole population and 120 mm in intermediate-stage HCC. Because it has been demonstrated and it is common clinical experience that tumor burden greatly affects the efficacy of TACE for intermediate HCCs, it might be reasonable to consider Y90RE as a valid alternative option for these patients. For selecting patients based on tumor burden, the up-to seven criterion, which has been previously proposed by the same investigators of this study, could be reasonably adopted. This criterion combines the number of nodules and the size of the largest tumor, with the sum being no more than 7. Then, patients in Child-Pugh A status, exceeding the up-to-seven limit, without ascites or jaundice and Eastern Cooperative Oncology Group PS0 could be considered for Y90RE. The possibility of using Y90RE as a second-line option for selected patients not responding to two to three TACE cycles, still maintaining a preserved liver function, should also be investigated in the future. In advanced HCCs, the subgroup of patients with peripheral limited PVT in Child-Pugh class A, which may be considered an overlap between intermediate and advanced stages, represents the clinical setting where Y90RE has provided the most favorable results, in comparison to systemic treatment, and should strongly be considered.
In conclusion, the study by Mazzaferro et al.10 adds further data for proposing the application of this expensive procedure into the appropriate clinical setting, but further relevant data for defining the eventual introduction in the therapeutic algorithm of the guidelines are expected from ongoing clinical trials.