Practical Considerations When Choosing Chemoembolization versus Radioembolization for Hepatocellular Carcinoma

Abstract

Transarterial chemoembolization (TACE) and transarterial radioembolization (TARE) are common liver-directed therapies (LDTs) for unresectable HCC. While both deliver intra-arterial treatment directly to the site of the tumor, they differ in mechanisms of action and side effects. Several studies have compared their side effect profile, time to progression, and overall survival data, but often these lack practical considerations when choosing which treatment modality to use. Many factors can impact operator's choice for treatment, and the choice depends on treatment availability, cost, insurance coverage, operator's comfort level, patient-specific factors, tumor location, tumor biology, and disease stage. This review discusses survival data, time to progression data, as well as more practical patient and tumor characteristics for personalized LDT with TACE or TARE.

Hepatocellular carcinoma (HCC) constitutes 90% of primary hepatic cancers, with 906,000 new cases and 830,000 deaths in 2020. ¹ Many HCC patients are diagnosed at more advanced stages, necessitating downstaging for definitive treatment or liver transplant. ² Locoregional therapy aids in tumor reduction and improvement of overall survival. The Barcelona Clinic Liver Cancer (BCLC) guidelines are the most widely accepted guideline for the management of HCC. ³ Transarterial chemoembolization (TACE) is currently BCLC recommended for intermediate-stage HCC. ⁴ Recent updates in 2022 to the BCLC guidelines include recommending TACE or transarterial radioembolization (TARE) for very early (stage 0) and early (stage A) disease when other treatments are not feasible or have failed. ⁴ When both TACE and TARE are an acceptable option, nuances of patient selection should guide which technique to use.

Mechanisms of Action

Chemoembolization

With TACE, increased arterial supply to HCC compared to surrounding parenchyma leads to preferential deposition of the chemoembolic mixture in the tumor, leading to tumor response from cytotoxicity and ischemic necrosis ⁵ ( Fig. 1 ). In TACE, chemotherapy is either mixed with lipiodol or bound to microspheres; however, these approaches lack standardized protocols and vary across centers. No conclusive evidence supports lipiodol TACE's (conventional, cTACE) superiority over drug-eluting bead TACE (DEB-TACE).

Fig. 1.

A 63-year-old man with a 6-cm hepatocellular carcinoma near the dome of his liver had TACE. ( a ) Axial post-contrast T1-weighted image shows a 6.0-cm HCC in the hepatic dome (segment 8) (arrow). ( b ) A hepatic angiogram shows a hypervascular tumor supplied by the right hepatic arteries (arrows). ( c ) Same date axial CT post-TACE shows 6.0-cm mass of the right lobe of the liver with lipiodol staining within the mass. ( d ) 36-month posttreatment axial MRI images showed necrotic lesion post-TACE therapy in the segment 7/8 (arrow).

Various factors, such as extrahepatic tumor blood supply and insufficient treatment uptake, can limit efficacy in TACE, and lead to an incomplete response, especially for larger tumors. ⁶ TACE theoretically causes the release of tumor antigens, triggering local and systemic immune responses for enhanced chemotherapy efficacy. However, due to the ischemia induced, TACE may cause vascular endothelial growth factor (VEGF) production, promoting angiogenesis and leading to incomplete response. ^{7 8} There is a suggestion that elevated VEGF levels post-TACE correlate with poorer overall survival.

Radioembolization

TARE uses yttrium-90 (Y90), a pure beta-emitter with a short tissue penetration depth (2.5 mm mean, 11 mm maximum) and with a half-life of 64 hours. Y90 is administered via resin or glass microspheres ⁹ ( Fig. 2 ). By injecting radioactive microspheres into the hepatic arteries, localized radiation causes direct DNA damage and tumor cell death. Most cases require a single session, but, if needed, repeat sessions may be performed in patients with preserved liver, hematologic, and renal function. ¹⁰

Fig. 2.

An 81-year-old man with unresectable hepatocellular carcinoma (ECOG 1, Child-Pugh A/5, MELD 8) was maintained on palliative Y90 therapy. ( a ) Axial CT image showed 12.0-cm biopsy-proven HCC in the posterior inferior aspect of the right liver lobe (arrow). ( b ) Celiac angiogram shows a hypervascular tumor supplied by the right hepatic arteries. ( c ) Liver and lung perfusion with SPECT/CT shows successful delivery of 80 mCi Y90 Sir-sphere to the right hepatic lobe. ( d ) Axial CT 12 months post-radioembolization shows posttreatment changes with a 9.0-cm necrotic lesion in the right hepatic lobe HCC bed (arrow). ( e ) Axial CT 24 months post-radioembolization shows stable posttreatment changes in the right hepatic lobe with the treatment zone measuring 8.5 cm.

Theoretical mechanisms for TARE treatment extend beyond cellular damage, including signal transmission from irradiated to non-irradiated cells, p53 production, and immune modulation. Irradiated tumor cells affect the tumor microenvironment and may act as an immunogenic hub, inducing an abscopal effect. Further research is required to confirm systemic effects. TARE can be combined with systemic treatments, such as immunotherapy, with promising results. ^{9 11 12}

Other than discussing outcomes, for simplicity this article will refer to cTACE and DEB-TACE as TACE, and glass and resin microsphere TARE as TARE. While differences exist, specific considerations for the intricacies within these techniques for various disease states fall beyond the scope of this review article.

Outcomes and as a Bridge to Transplant

Evidence from multiple recent randomized controlled trials (RCT) and meta-analyses supports TARE over TACE in lengthening progression-free survival and improving treatment response for similarly staged tumors. ^{13 14 15} In the TRACE 2 RCT, improved local tumor control and overall survival were seen when comparing TARE to DEB-TACE for early/intermediate HCC. TARE showed a median time to progression (TTP) of 17.1 months, whereas DEB-TACE achieved 9.5 months ( p  < 0.001). TARE also had increased success in downstaging for transplantation when compared to DEB-TACE. ¹³ Additionally, TARE exceeded the expected target for intermediate-stage HCC overall survival at 30.2 months, while DEB-TACE underperformed at 15.6 months.

Another phase 2 RCT by Salem et al in 2016 demonstrated that TARE significantly extended TTP and improved local tumor control compared to cTACE for early-/intermediate-stage HCC, despite similar overall survival outcomes between the two techniques. ¹⁴

A propensity score-matched study of 138 HCC patients in 2021 by Kim et al showed superior overall survival with initial TARE (mean OS endpoint not reached) compared to initial TACE treatment (20.8 months) with a total median follow-up time of the study being 27.6 months. ¹⁵ This was coupled with favorable intrahepatic tumor control (HR 0.60) and reduced toxicity of TARE compared to TACE. TARE was identified as an independent prognostic factor for better overall, progression-free, and intrahepatic progression-free survival.

Multiple meta-analyses have also provided evidence supporting superior survival outcomes with TARE. A meta-analysis in 2020 reported significantly higher 3-year overall survival rates with TARE in comparison to TACE. ¹⁶ Pooled results from a 2015 meta-analysis conducted by Zhang et al (encompassing 1,499 patients) indicated that TARE was associated with improved overall survival, 3-year survival, TTP, shorter hospitalization, and fewer complications compared to TACE. ¹⁷ Additionally, a 2018 meta-analysis by Yang and Si (involving 1,652 patients) demonstrated that TARE resulted in a 2-year increase in overall survival and achieved a better objective response according to mRECIST criteria, while carrying a reduced risk of adverse events compared to cTACE. ¹⁸

It is worth noting that some studies have reported comparable survival rates between TARE and TACE. For instance, the SIRTACE 2015 trial by Kolligs et al did not show a significant difference in progression-free survival between TARE and cTACE (3.6 vs. 3.7 months), although TARE required fewer repeat procedures. ¹⁹ Similarly, a 2015 study by EI Fouly et al in intermediate-stage HCC patients revealed similar survival outcomes between TACE and TARE despite the more advanced disease state in the TARE group, which was also better tolerated. ²⁰ Other meta-analyses have reported similar survival outcomes between TACE and TARE but acknowledged benefits favoring TARE in terms of tumor response and reduced adverse events. ^{16 17 18}

Several studies favor TARE for more effective downstaging and transplant eligibility. A prospective observational study from Lewandowski involving 86 patients demonstrated significantly greater event-free survival for TARE compared to TACE (17.7 vs. 7.1 months) and overall survival also favoring TARE (censored at 35.7/18.7 months; uncensored at 41.6/19.2 months). ² Additionally, another analysis of 172 posttransplant patients found that despite a longer time to transplant for TARE (6.5 vs. 4.8 months for TACE), posttransplant outcomes were similar for patients bridged or down-staged by TACE or TARE, supporting the use of TARE as a viable bridging or downstaging option for HCC. ²¹

These collective findings suggest that while data comparing overall survival rates are conflicting, TARE is often used in more advanced disease cases. These data strongly indicate that TARE offers improved progression-free survival, enhanced tumor control, better downstaging to transplant, and reduced toxicity in the treatment of unresectable hepatocellular carcinoma when compared to TACE. However, despite the apparent superiority in TTP and overall survival for TARE, many patients still can benefit from TACE over TARE, and this will subsequently be discussed.

Risks and Side Effect Profile

Postembolization syndrome (PES) occurs in >60% of patients who undergo TACE, causing fever, pain, nausea, and elevated liver enzymes. ²² Patients may require hospital admission for overnight observation and pain management. Later complications can include hepatic decompensation and hepatic abscess. ²³ Multiple sessions can lead to liver functional decline, TACE-resistant tumors, and reduced quality of life, ^{24 25} After more than two ineffective TACE procedures, alternatives such as TARE should be considered.

TARE has an overall well-tolerated side-effect profile, with many patients experiencing fatigue. The minimal embolic effect of TARE is associated with a lower incidence of PES (20–55%). ²⁶ Postradiation syndrome usually presents with fatigue as the main side effect, but can also include nausea, anorexia, and fever and can last several weeks.

TARE has a favorable mechanism in those with portal vein thrombosis as well and is the preferred liver-directed therapy (LDT) in these patients, although it has yet to receive BCLC guideline recommendations for use in intermediate- or advanced-staged patients. ³ TARE should be used with caution in patients who are receiving hepatotoxic chemotherapy, have undergone multiple prior LDT sessions, or have received radiation to the liver previously, as this can potentiate liver injury and lead to radioembolization-induced liver disease (REILD). ²⁷ Care should also be taken with whole bilobar treatment for the same reason. REILD can lead to significant hepatic decompensation, portal hypertension, and even death in the most severe form. ²⁶

Non-target radiation delivery to extrahepatic organs, such as the skin, lungs, or gastrointestinal tract, is a risk with TARE. The location of the microcatheter, embolization extent, and the presence of aberrant vessel branches on angiograms are factors associated with such complications. The calculated total lung dose should be kept to less than 30 Gy for a single treatment, and less than 50 Gy for cumulative treatments, to avoid radiation pneumonitis. ²⁸

Contraindications

For TACE, absolute contraindications include decompensated cirrhosis (Child-Pugh B ≥ 8), the presence of multinodular tumors that extensively replace both hepatic lobes, severely reduced portal vein flow, and technical impediments to safe hepatic intra-arterial treatment. ²⁵ Relative contraindications for TACE include factors such as tumor size ≥10 cm, the occurrence of bile-duct occlusion or other biliary drainage issues, and untreated varices at high risk of bleeding. ²⁶

Absolute contraindications for TARE include inadequate functional liver reserve as indicated by a significantly elevated total bilirubin level and severely reduced albumin levels. TARE should be avoided when non-target embolization cannot be prevented, or pathological lung shunting could result in a lung radiation dose of ≥ 30 Gy in a single treatment or 50 Gy over cumulative treatment sessions. ²⁹ An early trial involving resin microspheres showed that treatment with capecitabine within 3 months prior to TARE is an absolute contraindication for the use of resin microspheres. ³⁰ However, a more recent study in patients with neuroendocrine tumors showed that the combination is safe in the non-HCC, non-cirrhotic population. ³¹

Factors Influencing the Decision of TACE versus TARE

The selection of optimal LDT for individual patients can be complex. When deciding between TACE and TARE, several critical factors must be considered, including tumor-related factors (tumor number, size, vascular invasion, lung shunt fraction), patient-specific factors (age, sex, liver functional status, comorbidities, geographic location, economic status, insurance coverage, regional time to transplant), and institutional factors (availability, operator experience). By considering these factors, a more personalized therapy selection can optimize outcomes.

Patient-Specific Factors

Age and Sex

Age alone should not be a determining factor in HCC patient management with LDT. However, it should be noted that in the management of HCC, advanced age is considered a relative contraindication for TACE, and poorer outcomes are often seen in age >60. ^{25 32} A retrospective study by the European Network on Radioembolization with Yttrium-90 resin microspheres (ENRY) across eight European centers indicated that radioembolization is well-tolerated and effective for both the elderly (≥70 years) and younger patients (<70 years) with HCC. ³³

In the United States, sex disparities exist in liver transplantation data, with women being two times less likely to receive a transplant than men. ³⁴ Some studies have reported higher mortality rates in women with similar Model for End-Stage Liver Disease (MELD) scores, possibly due to worse overall clinical status with the same MELD scores from lower creatinine levels in patients with less relative muscle mass leading to falsely lowered MELD scores. ^{34 35 36} However, sex disparities for LDT are less definitive. For TACE and TARE, the data on sex-based outcome differences are inconclusive with some showing similar outcomes, and others worse overall survival outcomes for women, but are not statistically significant, likely due to small sample sizes. ^{37 38}

Race

Racial disparities in receiving a liver transplant are well described. ³⁹ Black and Hispanic patients are more likely to present with more advanced HCC than White and Asian patients and are therefore less likely to undergo definitive therapy. Patients of a lower socioeconomic status also often present to hospitals with fewer resources, and can be denied transfer for nonemergent conditions, limiting the treatment options available to them. Further research is needed to determine if this trend is consistent within options for LDT. In most publications comparing Y90 to TACE, and in those evaluating the outcomes of Y90, the treatment population is skewed toward the White population. No studies have specifically evaluated race-based outcomes of LDT.

Within the context of practical considerations, as we will discuss further, stage of disease may limit treatment options based on size, lobar involvement, anatomic location of the tumor, lung shunt fraction, and liver function. Therefore, poorer outcomes may be seen in the Black and Hispanic patient population due to their more advanced stage at presentation. Additionally, socioeconomic factors may limit treatment options available for some patients.

Liver Function

Evaluation of liver function prior to LDT is critical. Studies comparing TACE and TARE show no difference in rates of hepatotoxicity. ^{13 15 32 38 40} In practical terms, TACE carries risks of hepatic decompensation with Child-Pugh score ≥B8. ^{25 41} Recently, the albumin–bilirubin (ALBI) score has been shown to be beneficial in assessing liver function prior to TARE. Patients with lower ALBI scores (ALBI 1 vs. 2/3) have a lower incidence of REILD (3.4 vs. 16%) and better overall survival (26.4 months vs. 17.3 and 8.1 months) after treatment. ⁴²

Of note, in a retrospective review of 102 TARE patients by Kim et al, a low MELD score independently correlated with greater likelihood of a complete imaging response, in addition to smaller tumor size, post-TARE in patients with HCC. ⁴³

Selective treatment should be performed wherever possible to minimize hepatic decompensation with TACE and TARE. If nonselective bilobar treatment is required in cirrhotic patients, caution should be exercised with TARE, as mentioned previously, and in this setting at our institution we typically treat the lobe with the largest tumor burden with TARE and treat the other lobe with TACE with or without ablation where possible.

Tumor-Specific Factors

Stage of Disease

TACE remains the standard care for patients within the intermediate-stage HCC in the BCLC staging system. These are patients presenting with multinodular and/or large HCC, no cancer-related symptoms (ECOG 0), and no evidence of vascular invasion or extrahepatic dissemination. ⁴ It is also recommended for very early and early-stage (BCLC 0-A) patients for which definitive treatment has failed or is not feasible. ⁴

As of the 2022 update, TARE is now recommended within the BCLC guidelines for BCLC 0-A solitary tumors <8 cm in size in patients with preserved liver function and no cancer-related symptoms (ECOG 0). ⁴ In this setting, TARE has evidence as a potential curative strategy but is not currently guideline recommended as first-line therapy. ³⁸ As previously discussed, in BCLC stage B, TARE may offer improved tumor control and opportunity for liver transplantation through downstaging compared to TACE. For BCLC stage B or C patients, TARE is linked with fewer adverse outcomes and reduced hospital admissions than its TACE counterpart. ⁴

Tumor Size and Number

Patients with larger tumors (≥5 cm) or more lesions (four or more) have lower 5-year survival rates compared to those with smaller, solitary tumors with TACE. Recurrence-free survival following initial complete response is also reduced in cases of larger, multifocal HCCs when compared to smaller, fewer tumors. ^{30 39}

In patients with microvascular invasion, adjuvant TACE has been linked to improved outcomes compared to more conservative systemic management. ⁴⁴ However, a study by Liu et al suggests no such association for tumors exceeding 5 cm in size. ⁴⁴ PES was also found to be more severe following TACE for tumors larger than 5 cm. As such, a cutoff of >5 cm tumors and more than five lesions when advocating for TARE over TACE seems reasonable if the operator is concerned about poor PES tolerance after embolization.

A prospective, nonrandomized trial examined 86 intermediate-stage patients treated either with TARE ( n  = 44) or TACE ( n  = 42). Although tumor burden, as indicated by median tumor size, multiplicity, and bilobar distribution, appeared more favorable in the TACE group, the median overall survival rates remained comparable (16.4 months for TARE vs. 18 months for TACE). Additionally, the median TTP was longer for TARE (13.3 months) compared to TACE (6.8 months) but lacked statistical significance. The number of treatment sessions, total hospitalization time, and adverse event rates were significantly elevated in the TACE cohort. ³⁰

Lung Shunting

Historically, TARE was contraindicated if the shunt fraction on the 99mTc-labeled macroaggregated albumin mapping angiogram exceeded 20% due to the risk of radiation-induced pneumonitis ⁸ ( Fig. 3 ). However, with recent advances in dosimetry, the focus has shifted to the total lung dose. Treatment with TARE can proceed if the calculated projected lung dose remains below 30 Gy for a one-time dose, or 50 Gy over cumulative treatment sessions. ²⁹ When exceeding this threshold, physicians may consider pre-TARE bland or chemoembolization of the tumor, to decrease the shunt, or simply transition to TACE for treatment.

Fig. 3.

A 79-year-old woman who presents with the diagnosis of unresectable hepatocellular carcinoma on a background of autoimmune hepatitis and cirrhosis and a 7-cm segment 6/7 right hepatic lobe mass. ( a ) NM Y90 SPECT mapping shows intense radiotracer uptake in the right hepatic lobe with shunt fraction of 12%. ( b ) SPECT/CT shows precise arterial localization to the right lobe of the liver with Y90 SIR-Spheres (Sirtex Medical, Woburn, MA). The lesion was treated with an intended dose of 200 Gy, leading to a total lung dose of 5.65 Gy.

Portal Vein Tumor Invasion

One-third of the normal vascular supply to the liver is through the hepatic artery, while two-thirds is through the portal veins. While HCC has preferential hepatic arterial flow, the HCC microenvironment may induce portal vein thrombosis and portal vein invasion in more advanced stages. In these patients, systemic therapy is the standard of care in this setting in an effort to preserve liver function. HCC treatment guidelines recommend against TACE out of fear of worsening ischemia and further impairment of liver function. However, the results of several retrospective studies emphasize that outcomes vary based on factors such as tumor size, the selectivity of embolization, thrombus location, and the degree of cavernous transformation. ^{45 46 47 48}

Given that TARE is embolic at the microvascular level rather than the macrovascular level, tumor necrosis is thought to be secondary to radiation effect from beta emission and direct cell damage rather than ischemia. Because of this, TARE is felt to be more appropriate in this patient population due to a theoretically reduced risk of liver injury and further hepatic functional impairment. The National Comprehensive Cancer Network (NCCN) guidelines endorse TARE in select patients with advanced HCC and segmental or lobar portal vein thrombosis. ⁴⁹

Economic Factors

A systematic review demonstrated that TARE's relative cost-effectiveness increases with more advanced HCC stages. ⁵⁰ While difficult to determine exact costs, in the United States, TARE's average cost is around $32,000 per treatment, compared with TACE at $20,000. TACE is also readily covered by Medicare and Medicaid. ⁵¹

Unfortunately, the high one-time cost and inconsistent insurance coverage make TARE cost-prohibitive for some patients. However, considering the increased frequency of TACE sessions, associated overnight hospital stays, and costs associated with adverse events relative to a single TACE procedure, the cumulative costs on a per-lesion basis are likely to be comparable. Data show that 60% of TACE patients need multiple treatments, whereas 70% of TARE patients receive a single treatment for similar lesions. A detailed cost–benefit analysis is essential, and although the U.S. healthcare payment complexities cause a challenge to clear cost determination, initial findings indicate TARE's fewer treatment sessions and infrequent overnight admissions likely balance its higher initial costs, potentially alleviating healthcare system strain and improve the patient experience. ⁵²

Future Directions

The potential of merging genetic and clinical parameters to develop personalized prognostic models for HCC LDT treatment is an exciting future frontier. Such models may determine patients poised to benefit from hypoxia induced by TACE versus DNA damage through TARE, helping to tailor the most appropriate therapy for individual patients to improve outcomes and provide better prognostic information.

Gene expression signatures related to hypoxia and angiogenesis pathways may differentiate TACE responders from non-responders. Elevated expression of genes, including VEGF, HIF-1α, EPO, HMOX1, and SERPINE1, prior to treatment correlates with unfavorable outcomes. ^{53 54} TACE non-responders show increased macrophage M0 cells, reduced mast cells, and a heightened stemness index. ⁵⁵ In TARE, factors such as VEGF and PDGF-BB, along with Tsp-1, and factors such as IL-8 and follistatin may rise post-radioembolization, correlating with reduced survival. ⁵⁶

Research is needed to validate these biomarkers, especially the hypoxia-induced gene panels, and refine models combining genetic and clinical indicators. This could improve patient selection and predict therapeutic response. Identifying optimal biomarkers might also guide combination therapies, such as radioembolization and specific immunotherapy treatments, to improve outcomes.

Discussion

Hepatocellular carcinoma is characterized by significant heterogeneity in tumor characteristics and patients' underlying liver function, leading to challenges with selecting appropriate LDT in these patients. ⁵⁷ Studies comparing TACE and TARE underscore a progression-free survival benefit of TARE over TACE, as well as improved local tumor control and downstaging to transplant, and suggestion of a safer adverse event profile for TARE.

Common criticisms against TARE include the need for two angiographic sessions for one treatment, its increased per-treatment costs compared to TACE, and the absence of definitive data showing improved overall survival compared to TACE for similarly staged disease. TARE advocates argue its overall treatment cost to be more economical than TACE due to fewer overall treatments yielding comparable or superior outcomes, fewer hospitalizations for pain and PES, and improved response rates with more personalized dosimetry.

Choosing between TACE and TARE also requires a comprehensive assessment of multiple factors such as patient comorbidities, underlying liver function, the liver volume being treated, performance status, socioeconomic status, tumor stage, the presence of vascular invasion, degree of lung shunting, and total lung radiation dose. Other important considerations are operator expertise and comfort, as well as economic factors specific to institutions, regions, or even countries. The ultimate decision of TACE or TARE should be tailored to each patient rather than a one-size-fits-all approach.

Future directions of research should focus on a better understanding of HCC at a genetic level in order to recognize biomarkers, radiosensitivity, and ischemic sensitivity, and other factors that may predict which patients are likely to respond to TACE or TARE, as well as effective combinations of transarterial therapy with systemic therapies, allowing for a more personalized approach to LDT.