The researchers review intra-arterial therapies and the latest literature supporting their use, categorized by different tumor types.
Alexander Y. Kim, MD
Intra-arterial therapies (IATs) are taking on an increasingly prominent role in the treatment of patients with primary and secondary liver cancers. Although the mechanism and rationale behind these treatments are similarcatheter-based delivery from the hepatic artery to augment local treatment response and reduce systemic adverse events—IATs are heterogeneous. The underlying principle of an IAT may be to occlude the arterial blood supply to the tumor, to deliver a high local concentration of chemotherapy, and/or to deliver tumor-selective radiation. All of these treatments can be and have been used to treat various liver cancer types. The heterogeneous nature of IATs and the overlapping diseases they treat makes it challenging to understand which type of IAT to incorporate for specific cancer types. Here, we review IATs and the latest literature supporting their use, categorized by different tumor types.
Intra-arterial therapies (IATs) are taking on an increasingly prominent role in the treatment of patients with primary and secondary liver cancers. Although the mechanism and rationale behind these treatments are similarcatheter-based delivery from the hepatic artery to augment local treatment response and reduce systemic adverse events—IATs are heterogeneous. The underlying principle of an IAT may be to occlude the arterial blood supply to the tumor, to deliver a high local concentration of chemotherapy, and/or to deliver tumor-selective radiation. All of these treatments can be and have been used to treat various liver cancer types. The heterogeneous nature of IATs and the overlapping diseases they treat makes it challenging to understand which type of IAT to incorporate for specific cancer types. Here, we review IATs and the latest literature supporting their use, categorized by different tumor types. Introduction Intra-arterial therapies (IATs) are regional therapies designed to augment local treatment effects while minimizing systemic adverse effects (AEs). Although IATs have been delivered for treatment of disease in organs beyond the liver1, the predominant targets of IAT are primary and secondary liver cancers.
The physiologic basis for liver-directed IAT relies on the dual hepatic blood supply: The normal liver parenchyma is supplied primarily by the portal circulation (75% to 80%), while the blood supply to highly vascular malignant hepatic cells is almost exclusively derived from the hepatic artery. By taking advantage of this differential blood flow between tumor and normal tissue, the goal of IAT is to maximize delivery of cytotoxic agents to the tumor while minimizing toxicity to the background liver.2
Characterized by catheter-based arterial delivery, IAT encompasses a heterogeneous set of treatments including transarterial chemoembolization (TACE), transarterial embolization (TAE), yttrium-90 radioembolization (Y-90 RE), and hepatic artery infusion (HAI).
The purpose of this article is to review the literature for use of IAT in treatment of hepatocellular carcinoma (HCC), metastatic colorectal cancer (mCRC), and other secondary liver metastases.
Types of IAT
Transarterial Embolization (TAE)
TAE, colloquially known as “bland” embolization, is a procedure wherein embolic particles, such as polyvinyl alcohol (PVA) or tris-acryl gelatin microspheres (TAGMs) are infused through a catheter into the hepatic artery. The goal of TAE is to occlude the arterial tumor supply and cause ischemic tissue death. TAE is a treatment strategy used not only for cancer care but also in symptomatic benign disease such as uterine fibroids3or lower urinary symptoms due to benign prostate hypertrophy.4
Transarterial Chemoembolization (TACE)
Similar to the goal of TAE, the goal of TACE is to inhibit the arterial blood flow to tumors. However, in TACE, chemotherapy is also delivered through the hepatic artery to infuse a high concentration of chemotherapy locally into tumor tissue. The goal is to maximize the chemotherapeutic effects while minimizing potential systemic AEs.
Traditionally, chemotherapeutic agents are mixed and delivered with an ethiodized oil, a radio-opaque medium thought to enhance the tumor uptake of the chemotherapeutic agent.5This is followed by delivery of an embolic material such as PVA or TAGM. Recent developments have allowed for coinfusion of chemotherapy along with embolic particles, which are bound through a chemical bond. To differentiate the 2 methods, the sequential delivery of chemotherapyethiodized oil followed by an embolic is referred to as conventional TACE (cTACE), whereas the delivery of the bound chemotherapy bead is known as drugeluting bead TACE (DEB-TACE). To date, there is no clear evidence demonstrating superior outcomes of one method over the other.6
Serious AEs after TACE are uncommon.7The rate of liver failure in well-selected patients after TACE is <5%.8 Potential chemotherapy-related AEs include pancytopenia and alopecia. The most common treatment-related AE is termed postembolization syndrome, which is a triad of abdominal pain, nausea, and/or fevers. This may last for a few days following treatment and tends to be self-limiting.9
In RE, a radioactive isotope is delivered intra-arterially bound to a particle. Yttrium-90 (Y-90) is the most commonly used isotope for this treatment, although other isotopes such as iodine-131 and holmium-166 are also under investigation.10,11Unlike TAE or TACE, the goal of RE is to use the hepatic arterial tumor supply to deliver radiation rather than to occlude tumor-feeding vessels. Transarterial radioembolization and selective internal radiation therapy are other commonly used terms for this treatment.
The 2 dominant Y-90 RE products are marketed in the United States under the trade names Thera- Spheres and SIR-Spheres (see Table). SIR-Spheres has an FDA indication for treatment of mCRC; Thera-Spheres has been approved for use for HCC under a humanitarian device exemption. As with TACE, serious AEs after Y-90 RE administration are uncommon. The most feared complications include radiation-induced gastrointestinal ulceration, RE-induced liver dysfunction, and radiationinduced cholecystitis of pneumonitis.12,13The results of most recent studies indicate that the incidence of these potential AEs is about <5%. In combination with systemic therapy, Y-90 RE can worsen pancytopenia. More commonly, patients can develop fatigue, abdominal pain, and/or nausea, all of which tend to be self-limiting.
Hepatic Artery Infusion (HAI)
In HAI, a catheter is placed in the hepatic artery and attached to a port to facilitate direct chemotherapy infusion in the liver. The goal of HAI is to augment the intra-tumoral delivery of chemotherapy, and it has been demonstrated to increase the chemotherapy uptake in liver tumors by a factor of 15.14Potential serious AEs associated with HAI include gastrointestinal ulceration and hepatic artery thrombosis/occlusion.15,16Because HAI can be technically challenging to perform,17it is offered in seelcted centers only. Comprehensive review of HAI is available elsewhere18and will not be further discussed in this review.
Patient Selection for IAT
Similar criteria are used in patient selection for TAE, TACE, and Y-90 RE. Eligible patients have primary liver cancer or a liver-dominant secondary liver cancer with adequate hepatic reserve and functional status.
As IAT may lead to worsening of hepatic function, assessment of adequate hepatic reserve may be the most important component in determining patient eligibility for IAT. This is most commonly accomplished by assessment of liver function tests, including levels of total bilirubin, aspartate aminotransferase (AST), alanine aminotransferase (ALT), and albumin. Total bilirubin of >3.0 mg/dL, albumin level <3.0 g/dL and/or AST/ALT >5 times the upper normal limit may preclude patients from undergoing IAT.19,20Selected patients may still be eligible for IAT, despite suboptimal liver function, if the liver disease burden can be successfully treated in a subselective fashion. Other factors to consider when assessing adequacy of hepatic reserve include overall liver disease burden, prior liver-directed radiotherapy, and amount of prior systemic chemotherapy received by the patient.
Adequate functional reserve is also important to consider when determining a patient’s eligibility to undergo IAT. Because IAT can lead to significant fatigue post procedure, patients with Eastern Cooperative Oncology Group (ECOG) scores >2 or Karnofsky scores <60 are generally considered to be ineligible for IAT.
IAT for HCC
TACE for HCC
TACE is the standard of care for patients with Barcelona Clinic Liver Cancer (BCLC) intermediate-stage HCC. The basis for this recommendation is largely supported by the results of 2 randomized controlled trials from 2002.
In a 3-arm trial, Llovet et al compared the outcomes of TACE versus TAE versus best supportive care (BSC) for treatment of HCC.21A total of 903 patients were assessed; 112 patients (12%) with unresectable HCC were randomized. Patients with unresectable HCC not suitable for curative treatment, Child-Pugh class A or B, and Okuda stage I or II, were eligible for the study. The results demonstrated that TACE led to a survival benefit over BSC (P = .025). Survival probabilities at 1 and 2 years were, respectively, 82% and 63% for TACE, 75% and 50% for TAE, and 63% and 27% for BSC.
Lo et al performed a similar trial in 80 patients with unresectable HCC in an Asian population.22 All patients had newly diagnosed, unresectable hepatocellular carcinoma and were assigned to treatment with TACE using a variable dose of an emulsion of cisplatin in ethiodized oil and gelatin-sponge particles injected through the hepatic artery (chemoembolization group; 40 patients) or symptomatic treatment (control group; 40 patients). TACE was repeated every 2 to 3 months unless there was evidence of contraindications or progressive disease. OS was the primary endpoint. Survival was significantly improved in the TACE group (1 year, 57%; 2 years, 31%; 3 years, 26%) versus the control group (1 year, 32%; 2 years, 11%; 3 years, 3%) (P = .002). Although death from liver failure was more frequent in patients who received chemoembolization, the liver functions of the survivors were not significantly different compared with the control group.
In clinical practice, TACE is often used to treat patients in BCLC stages A and B. For patients with early-stage HCC, TACE is a common “bridging” tool treatments used to prevent disease burden from progressing beyond the United Network for Organ Sharing transplant criteria. TACE can also be used to downstage some patients with intermediate-stage HCC into transplant criteria.23
Y-90 RE for HCC
Increasing evidence supports the role of Y-90 RE in the treatment of patients with advanced-stage HCC.24The results of the SorAfenib Versus RE in Advanced Hepatocellular Carcinoma (SARAH) study were recently released at the 2017 European Association for the Study of the Liver International Liver Congress. In this multicenter, randomized controlled trial in France, more than 400 patients with advanced HCC were recruited and randomized to receive sorafenib or Y-90 RE with SIR-Spheres. Patients eligible for this trial included adults with unresectable HCC who had a life expectancy of >3 months, ECOG performance status ≤1, and advanced HCC (BCLC stage C) or recurrent HCC after surgical or thermoablative treatment. Patients were randomized 1:1 to receive either Y-90 RE or sorafenib 400 mg twice daily, with the primary endpoint being OS. Patients were not allowed to cross over to the other treatment arm. Results showed a nonsignificant difference in OS and median progression-free survival (PFS) in the Y-90 and sorafenib groups in both the intentionto- treat (P = .18) and per-protocol (P = .92) analyses. There was an advantage in favor of Y-90 RE in treatment tolerance, with fewer treatment-related AEs (P <.001), fewer grade 3 or higher AEs (P <.001), and significantly improved quality of life.
Two other single-arm phase II trials also demonstrated promising outcomes for patients with advanced HCC treated with TheraSpheres, with both studies showing a median OS of 13 months in this population.25,26 Mazzaferro et al studied 52 patients with intermediate (n = 17) to advanced (n = 35) HCC, assessing the efficacy of Y-90 RE on time-toprogression (TTP). All patients were ECOG score ≤1 and Child-Pugh class A/B. The median TTP was 11 months with no significant difference between portal vein thrombosis (PVT) versus no PVT (7 vs 13 months). On multivariate analysis, tumor response was the sole variable affecting TTP (P <.001).
Kokabi et al studied the safety and efficacy of Y-90 therapy for unresectable infiltrative HCC with PVT. In a prospective single-center study, patients with unresectable (BCLC stage C) infiltrative HCC with PVT were recruited. OS and TTP were measured from the first Y-90 therapy. Forty-five patients were recruited, and 30 patients who met the study’s inclusion criteria underwent glass-based Y-90 therapy. Four patients (13%) had transient hepatobiliary toxicity (grade ≥2). Ten patients (33%) had related emergency department visits, with 5 patients (17%) requiring shortterm hospitalization. No radiation pneumonitis, gastrointestinal ulceration, or procedure-related mortality occurred. The median OS was 13 months (95% CI, 4.4-22 months) with a TTP of 9 months (95% CI, 6.2-13.1 months).
The aim of 2 ongoing prospective evaluations (SORAMIC and STOP-HCC) is to assess the potential benefit of Y-90 RE when incorporated with sorafenib for treatment in patients with advancedstage HCC. The results of these studies should clarify where Y-90 RE belongs in the treatment of patients with advanced HCC.
The potential role of Y-90 RE in treatment of early- or intermediate-stage HCC remains ill defined. In their single-center, randomized phase II trial, Salem et al compared the outcomes of patients with BCLC A- or Bstage HCC (78% BCLC A) treated with cTACE or Y-90 RE.27 Inclusion criteria were image-/biopsy-proven HCC, unablatable/unresectable disease, no vascular invasion, Child-Pugh A/B, bilirubin level of 2.0 mg/ dL or less, and AST/ALT 5 times the upper limit of normal or less. Twenty-four patients were randomized to receive Y-90 RE, and 21 patients to receive cTACE. They met their primary outcome of TTP, >26 months versus 6.8 months in favor of Y-90 RE. However, this did not translate into a significant difference in OS between the 2 groups (18.6 months for Y-90 RE vs 17.7 months for cTACE; P = .99). A significantly greater proportion of patients in the cTACE group developed diarrhea (21%) than in the Y-90 group (0%; P = .031); the situation was similar for hypoalbuminemia (58% in the cTACE group vs 4% in the Y-90 group; P <.001).
IAT for Metastatic Colorectal Cancer
TACE for mCRC
TACE is a common strategy for the treatment of secondary liver cancers. Among metastatic disease, TACE has been best studied in the patient population with a colorectal primary.
In a multicenter Italian study, Fiorentini and colleagues compared the outcomes of patients randomized to receive irinotecan, folinic acid, and leucovorin (in combination, FOLFIRI) versus DEB-TACE loaded with irinotecan (DEBIRI-TACE).28All patients had received at least 2 prior lines of chemotherapy and a majority had received 3 lines of chemotherapy. Thirty-eight patients were randomized to receive DEBIRI-TACE and 36 received FOLFIRI. The primary endpoints were tumor shrinkage, safety, feasibility, compliance, and OS. They found a statistically significant survival advantage in the DEBIRI-TACE group: 22 months versus 15 months for the FOLFIRI group (P = .031). Observed treatment-related AEs included fever (80%), increased transaminases (70%), right upper quadrant pain (40%), and nausea (27%).
Martin et al studied the safety, response, and AE rates of DEBIRI-TACE delivered in a transarterial approach as first-line treatment for unresectable colorectal liver metastases.29Seventy patients were selected. All had histologically proven colorectal cancer to the liver; chemotherapy-naïve, liver-dominant disease (defined as ≥80% of the tumor body burden being confined to the liver) but less than 60% liver replacement by the tumor; and had an ECOG performance status score ≤2. Patients were randomly assigned to receive folinic acid, leucovorin and oxaliplatin (mFOLFOX) and bevacizumab or mFOLFOX and bevacizumab plus DEBIRI-TACE (FOLFOX-DEBIRI). Based on modified Response Evaluation Criteria in Solid Tumors, they found that the overall response rate (ORR) was significantly greater in the FOLFOX-DEBIRI arm versus the control arm at 2 months (78% vs 54%; P = .02), 4 months (95% vs 70%; P = .03), and 6 months (76% vs 60%; P = .05). There was significantly more downsizing to resection in the FOLFOX-DEBIRI arm versus the FOLFOX-bevacizumab arm (35% vs 16%; P = .05), and there was improved median PFS (15.3 vs 7.6 months).
Y-90 Radioembolization for Metastatic Colorectal Cancer
Several prospective evaluations have demonstrated the efficacy of Y-90 RE in treatment of mCRC with liver-dominant disease.
Hendlisz et al conducted a prospective, multicenter, randomized trial in 46 patients with histologically proven adenocarcinoma of the colon or rectum metastasized to the liver only, not amenable to curative surgery or local ablation and resistant or intolerant to standard chemotherapy (FU, oxaliplatin, and irinotecan).30Forty-six patients (23 in each arm) were randomly assigned to 1 of 2 arms: Patients assigned to arm A received protracted intravenous infusion of FU (300 mg/m2) until progression. Patients assigned to arm B received Y-90 RE plus intravenous FU (225 mg/m2) for 14 days followed by 1 week of rest. The study met its primary endpoint of timeto- liver progression (TTLP) with median TTLPs of 2.1 and 5.5 months in favor of the Y-90 cohort (P = .003). They found no statistically significant median OS benefit (7.3 and 10 months for groups A and B, respectively) (P = .80); however, this result was confounded by 10 of 23 (43%) patients in the control arm crossing over to receive Y-90 RE after demonstrating progression. Grade 3 or 4 toxicities were recorded in 6 patients after FU monotherapy and in 1 patient after RE plus FU treatment (P = .10).
In their single-arm phase II trial, Cosimelli also studied the efficacy of incorporating resin Y-90 RE in the treatment of chemorefractive colorectal liver metastases.31All 50 eligible patients had demonstrated progression of disease after receiving at least 3 lines of systemic chemotherapy and 38 of 50 (76%) had received 4 or more lines of systemic treatment. Most patients presented with synchronous disease (72%), >4 hepatic metastases (58%), 25% to 50% replacement of total liver volume (60%), and bilateral spread (70%). In this heavily treated population, 12 patients (24%) demonstrated response to treatment per RECIST criteriaincluding 1 patient with complete imaging response—and disease control was demonstrated in 24 of 50 (48%) patients. The median OS was 12.6 months from treatment. One patient died 40 days after treatment from acute renal failure and another responding patient died 60 days after treatment due to liver failure. Liver and kidney function tests in both patients were normal before treatment. Both deaths were classified as possibly related to treatment.
The SIRFLOX trial assessed the efficacy of incorporating Y-90 RE to standard first-line treatment in patients with liver-dominant mCRC. The goal of this 530-patient trial was to study the safety and efficacy of incorporating Y-90 RE into standard FOLFOX-based chemotherapy in patients with previously untreated mCRC, having World Health Organization performance status of 0 to 1, and with a life expectancy of >3 months. PFS was the primary endpoint.32RE was delivered prior to initiation of chemotherapy in the treatment arm with the dose of oxaliplatin reduced from 85 mg/m2 to 60 mg/m2 for the first 3 cycles. Approximately half of the patients in each cohort also received bevacizumab. The addition of Y-90 to FOLFOX-based first-line chemotherapy did not improve PFS at any site (10.2 vs 10.7 months; P = .43), but significantly delayed disease progression in the liver (12.6 vs 20.5 months, P = .002). Treatmentemergent grade ≥3 AEs were reported in 73.3% of control and 85.4% of RE patients. Hematologic toxicities were reported at a higher rate in the RE group compared with control (P <.05). Known RE-associated AEs were reported in RE patients only. Serious AEs were reported less frequently in control patients (41.6%) than in RE patients (54.1%; P = .005).
The results of the FOXFIRE trial were recently released.33 This was an 1100-patient, multicenter trial designed to assess for the survival benefit of incorporating Y-90 RE to standard first-line therapy in patients with liver-only or liver-dominant mCRC not amenable to curative surgical resection. Eligible patients were randomized 1:1 to receive either systemic chemotherapy with oxaliplatin, 5-FU, and folic acid (OxMdG), or single-session whole-liver Y-90 RE + OxMdG. The study failed to meet its primary objective of demonstrating an OS advantage for the treatment group. Subgroup analysis did demonstrate a 4.9-month median OS advantage in patients with right-sided primary colon cancer (P = .007). The full manuscript has yet to be published at the time of this writing.
IAT for Other Secondary Liver Cancers
The literature supporting the use of IAT for secondary liver cancers from a noncolorectal origin are largely limited to small, nonrandomized prospective or retrospective case series. The physiologic basis and rationale for incorporating IAT into the treatment of this patient population remains the same.
IAT for Neuroendocrine Liver Metastases
After HCC and colorectal cancers, IAT is most frequently used to treat liver metastases from neuroendocrine tumors. TAE, TACE, and Y-90 RE are the most frequently used strategies. Various case series and meta-analyses demonstrate a wide range of disease response rates (15% to 90%), disease control rates (50% to 100%), and symptom control rates (>80%) after these treatments. One large retrospective review of 148 patients treated with Y-90 RE for neuroendocrine hepatic metastases demonstrated complete and partial response rates of 2.7% and 60.5% and a disease control rate of 95.1% in a pretreated population. No radiation liver failure was observed.34 There is no clear evidence supporting the use of one type of IAT versus another, nor the order in which these treatments should be utilized if multiple treatments are to be delivered. With the expected FDA approval of peptide receptor radionucleotide therapy (PRRT) in the United States, more centers are moving toward treatment of neuroendocrine liver metastases with either TAE or TACE rather than Y-90 RE.
IAT for Breast Cancer Liver Metastases
Various studies have demonstrated a potential role of IAT in patients with liver metastases from a primary breast cancer. A multicenter registry evaluating DEB-TACE studied 40 patients with metastatic breast cancer to the liver and administered a total of 75 image-guided procedures with hepatic arterial drug-eluting beads loaded with doxorubicin. Results demonstrated an ORR of 58% at 3 months and 50% at 12 months following treatment. The median OS was 47 months in this cohort, with overall and liver-specific PFS of 17 and 26 months, respectively. Treatment was well tolerated with a total of 8 patients sustaining 13 AEs within the 30 days of each treatment session. All AEs were either grade 1 or 2 in toxicity.35A number of retrospective evaluations of Y-90 RE for breast metastases have also been performed. These studies show disease response rates ranging from 25% to 78% and disease control rates of 71% to 98%.36
IAT for Intrahepatic Cholangiocarcinoma
Many investigations have assessed the potential role of IAT in the treatment of intrahepatic cholangiocarcinoma (ICC). The largest prospective study investigated cTACE for treatment of 115 patients with unresectable ICC.37In this single-center study, patients were repeatedly treated with TACE. Various combinations of mitomycin C, gemcitabine, and cisplatin were delivered. In total, 819 chemoembolization sessions were performed in 4-week intervals with a mean of 7.1 (range, 3-30) sessions per patient. The median OS from start of treatment was 13 months. No major treatment complications were reported. Similar median OS rates have been reported after Y-90 for ICC.38
IAT for Other Liver Metastases
The potential roles of IAT from metastatic disease from other primaries including melanoma,39 renal cell,40 and pancreatic adenocarcinoma41 have been reported. The role of IAT in treatment of these disease processes is still evolving, but IAT may play a role in selected patients with chemorefractory disease.
IATs are a heterogeneous group of treatments increasingly being utilized to treat patients with primary and secondary liver malignancies. There are well-established data supporting their use in the treatment of HCC and colorectal liver metastases. There are accumulating data to support their use in a number of other secondary liver cancers.