Hepatocellular carcinoma (HCC) is the most common primary liver cancer and is the second most common cause of death from cancer worldwide, with an estimated 745000 deaths in 2012 (1). It is the fifth most common cancer among men (554000 cases) and the ninth most common cancer in women (228000) (1). The age-adjusted worldwide incidence rates per 100000 for men and women are 15.3 and 5.3, respectively (1). In men, the incidence rates are highest in Eastern and South-Eastern Asia (31.9 and 22.22, respectively), intermediate in Southern Europe (9.5) and Northern America (9.3), and the lowest in Northern Europe (4.6) and South-Central Asia (3.7) (1). In women, the rates are generally lower than those in men with the highest rates in Eastern Asia and Western Africa (10.2 and 8.1, respectively) and the lowest in Northern Europe and Micronesia/Polynesia (1.9 and 1.4, respectively) (1). In the United States, an estimated 23000 deaths from liver cancer occurred in 2014 (2). In this same year, an estimated 33190 new cases of liver cancer were expected (2). From 2006 to 2010, the rates of liver cancer increased by 3.7% per year in men and by 2.9% per year in women (2).

Hepatocellular carcinoma has a poor prognosis with an overall 5-year relative survival rate of 16% and survival decreases as patients are diagnosed with regional and distant stages of the disease (2). Only 41% of liver cancer patients are diagnosed at an early stage, while the majority of patients present with advanced disease. Curative treatments indicated for early HCC include resection, liver transplantation, and percutaneous ablation (34). Patients with early HCC can achieve a survival rate of 50–70% at 5 years after undergoing resection, liver transplantation, or percutaneous ablation (5). Non-curative treatments to improve survival in patients with unresectable HCC include locoregional therapies such as transarterial chemoembolization (TACE) and selective internal radiation therapy (SIRT) for intermediate HCC and sorafenib for advanced HCC (34).

A systematic literature search for yttrium-90 (Y90)-SIRT with glass microsphere studies (published, unpublished and ongoing) was carried out from August 19, 2014 to July 19, 2015, and it was updated on August 30, 2015. Two investigators developed and conducted the literature search. Studies were identified using database searches and citation searches of selected articles. The electronic database searched was PubMed (1965-present). Only articles in English were selected for the review and no date restrictions were applied to the search. The search was conducted using free-text terms and standardized subject terms appropriate for the specific database. Combinations of the following search terms were used: HCC, hepatocellular carcinoma, radioembolization, TARE, SIRT, Yttrium-90, Y90-SIRT, TheraSphere®, glass microspheres, therapy/treatment, safety, efficacy, and Sorafenib (Table 1). Ongoing and completed clinical trials were identified through searches of clinical trial databases (6). Study eligibility was assessed by a single investigator in an un-blinded manner. Studies were retained if they met the following criteria: 1) TheraSphere® or glass microspheres use in the treatment for HCC, 2) effect of Y90-SIRT with glass microspheres on early, intermediate or advanced HCC, or 3) safety and/or efficacy of Y90-SIRT with glass microspheres or Sorafenib for HCC. All study designs were accepted. Studies were excluded if their aim was to evaluate: 1) SIR-Spheres® or 2) resin-based microspheres as our main focus in this literature review was the treatment of HCC with glass microspheres which is currently approved by the Food and Drug Administration (FDA) under an humanitarian device exemption (HDE).

The technique uses Y90 impregnated glass or resin microspheres, which are delivered through a catheter directly into the hepatic arteries (78). It takes advantage of the fact that primary and secondary hepatic tumors are vascularized mostly by arterial blood flow in comparison to healthy liver parenchyma which obtains its blood supply mainly from portal venous blood (7). The diameters of the glass and resin microspheres (20–30 microns versus 20–60 microns, respectively) allow them to become permanently embolized in the terminal arterioles of tumor (9). Y90 is a pure beta-emitter that disintegrates into stable zirconium-90 and has a half-life of 64.24 ± 0.30 hours (7810). The average β-emission is 0.9367 MeV, with a mean tissue penetration of 2.5 mm and a maximum tissue penetration of 10 mm (78). This allows delivery of high radiation doses to hepatic tumors while minimally affecting the healthy surrounding liver parenchyma unlike external radiation (3711).

There are two commercially available microspheres: TheraSphere® (MDS Nordion, Ottawa, Canada) and SIR-Spheres® (Sirtex Medical Ltd., Lane Cove, Australia) (8). In 1999, the United States FDA approved the glass microspheres (TheraSphere®), under an HDE, for radiation treatment or as a neo-adjuvant to surgery or transplantation in patients with unresectable HCC who could have placement of appropriately positioned hepatic arterial catheters, and later it approved the use of TheraSphere® in patients with portal vein thrombosis (12). The FDA approved the use of resin microspheres (SIR-Spheres®) in 2002 via a premarket approval for its use in the treatment of unresectable metastatic colorectal cancer to the liver with adjuvant chemotherapy with floxuridine (13).

Prior to undergoing Y90-SIRT, patients must undergo pre-treatment evaluation to ensure that the therapy is successful (1415). This pre-treatment evaluation is performed 1–2 weeks prior to the procedure and it consists of the following studies: clinical studies, laboratory tests, and imaging (triphasic liver CT or MRI, angiography, and a technetium-99m labeled macroaggregated albumin [^99mTc-MAA] scan) (1416).

To determine the liver volume for which the Y90 glass microspheres are delivered (i.e., volume of distribution) a 3-dimensional reconstruction of the target site is performed using CT or MR imaging. To calculate the mass of infused liver tissue (in kg), a conversion factor of 1.03 g/cm³ is used. The required activity for injection and the dose delivered to the target are calculated by the following formula:

Where A is the net TheraSphere® activity delivered to the liver, D is the dose administered to the target liver mass, and M is the target liver mass.

When the LSF and residual activity (R) in the vial after treatment are taken into account, the actual dose delivered to the target mass (Gy) is calculated by the following formula:

After pre-treatment evaluation, the patient will undergo Y90-SIRT which is done on an outpatient basis (21). Under fluoroscopic guidance, glass microspheres (about 1–8 million glass spheres) are delivered to the liver via a catheter placed into the femoral artery, which is then guided into the hepatic artery and eventually positioned into the diseased lobe/segment of the liver by following the branch of the hepatic artery (21). Once the catheter is properly positioned, the microspheres are infused at a rate similar to that of the hepatic arterial flow over a period of 3–5 minutes (9). Delivery of the glass microspheres is dependent on the hepatic arterial flow distal to the catheter tip. It is necessary to ensure that the catheter does not occlude the vessel in which it is positioned in order to prevent reflux secondary to vessel spasm (9).

After patients receive their treatment, they undergo post-treatment imaging and laboratory evaluation to assess the response to treatment or lack thereof. Y90-SIRT is associated with low toxicity and patients who receive an administered activity of less than 3 GBq can be released without contact restrictions according to the Nuclear Regulatory Commission contact scenario (25). The time interval for obtaining post-treatment imaging and laboratory evaluation varies among authors and clinical institutions (20). The study by Riaz, Awais, and Salem obtained triphasic CT or MRI at 1 month following treatment and at 3 month intervals following the first post-treatment as it may take 3–6 months for the optimal response (size reduction) to occur (14). Laboratory studies obtained 1 month following Y90-SIRT include liver function studies, complete blood count and tumor markers (alpha-fetoprotein for HCC) (14).

Tumor response assessment following Y90-SIRT can be performed with conventional cross-sectional imaging such as CT and MRI (917). There is a lack of standardization of functional imaging in HCC and response assessment based on anatomical methods is considered as the standard (2627).

Four commonly used guidelines to assess tumor response following therapy are: World Health Organization (WHO) (28) (bidimensional), Response Evaluation Criteria in Solid Tumors (RECIST) (29) (unidimensional), Modified Response Evaluation Criteria in Solid Tumors (mRECIST) (30) (unidimensional), and the European Association for the Study of the Liver (EASL) (31) (necrosis). The RECIST guideline was amended in 2010 to become the mRECIST (30). The mRECIST adopted the concept of viable tumor, which shows arterial enhancement on contrast-enhanced radiologic imaging techniques, thus enabling an evaluation of the tumor response after therapeutic strategy (30). The EASL guidelines address the limitations of the guidelines of the WHO and RECIST, which based their guidelines on systemic therapies resulting in limitations when applied to locoregional therapies such as Y90-SIRT (31).

The study by Rhee et al. (32) showed that diffusion-weighted imaging (DWI) functional MR may assist in early determination of the response or failure of Y90-SIRT in HCC. Unlike MR anatomic imaging studies which are able to assess tumor response until 3 months after treatment, DWI showed that imaging changes at 1 month preceded anatomic size changes seen at 3 months following Y90-SIRT (32). Earlier detection would allow for repeat treatment or alternative therapy such as TACE to be given sooner to patients rather than waiting 3 months to determine the treatment response (32).

In an effort to better identify the tumor response and therapy-related changes in serum markers after Y90-SIRT, a retrospective single-center study assessed the tumor response by using the RECIST 1.1, mRECIST, and Choi criteria (33). Using these three guidelines in addition to the model for end stage liver disease and the serum C-reactive protein, the study was able to create a prognostic model which could predict survival probability for individual patients as early as one month after treatment (33). However, this prognostic model needs to be validated further by performing larger prospective treatment studies.

Although studies have shown that Y90-SIRT can be considered a safe and efficacious treatment for HCC, complications associated with the therapy can occur (Table 2) (34). Some patients may experience a post-radioembolization syndrome (PRS) which consists of nausea, vomiting, fatigue, abdominal pain/discomfort, and/or cachexia (14). Hospitalization is rarely required and the incidence of PRS ranges from 10% to 70% (1435). Extrahepatic microsphere delivery or radioactivity affecting surrounding structures can result in hepatic dysfunction, biliary system complications, radiation pneumonitis, gastrointestinal (GI) complications, acute pancreatitis, radiation dermatitis, and lymphopenia (14). Other complications include: thrombocytopenia, vascular injury, contrast-induced nephrotoxicity, and allergic reaction to iodinated contrast media (14).

A serious hepatic complication after Y90-SIRT is radiation-induced liver disease (RILD) (14). The incidence of RILD ranges from 0% to 4% and it occurs due to radiation exposure of healthy hepatic parenchyma (143637). Clinically, patients develop hepatomegaly, ascites, jaundice and elevated serum transaminases, especially alkaline phosphatase (37). Treatment involves aggressive therapy to control symptoms of abdominal ascites (37). Sangro et al. (38) further described radioembolization-induced liver disease (REILD) in 20% of patients who had undergone chemotherapy either before or after radioembolization. Patients with REILD presented clinically with jaundice and ascites as late as 1 to 2 months post-radioembolization. Post-radioembolization biliary complications include radiation cholecystitis (may be prevented by identifying and coiling the cystic artery (1819)), radiation-induced cholangitis, and bilomas/abscess (14). The incidence of post-radioembolization biliary complications is less than 10% and patients with prior surgeries involving the ampulla of Vater have a higher risk (1439). Less than 1% of patients will present with radiation pneumonitis and it is recommended that delivery of Y90 to the lungs should be less than 30 Gy in one treatment or the cumulative dose should be less than 50 Gy in multiple treatments (40). Patients with GI complications may present with diarrhea or ulcers. GI ulcers occur in less than 5% of patients and prophylactic coil embolization of the GDA and the RGA may help prevent sequelae (71441). Acute pancreatitis may present with severe epigastric pain and with elevated serum lipase and amylase levels (14). Although diffusion of microspheres in the extrahepatic arteries such as the falciform artery is rare, it may result in radiation dermatitis with periumbilical pain (714). Patients may have a greater than 25% decrease in their lymphocyte count following Y90-SIRT without an increase in the incidence of opportunistic infections (111434).

The majority of published studies on the treatment of HCC patients with Y90-SIRT are retrospective or small, non-controlled prospective studies, and therefore, they are only supported by level II-2 and II-3 evidence (42). There is a lack of large, randomized, controlled trials (Table 3). Studies have reported improved median survival in patients with intermediate- to advanced-stage HCC following Y90-SIRT (median 7–41.6 months) (24384344). Objective response rates vary between 35–70% among studies depending on the guideline criteria used (24384445). Approximately 20% of patients present with liver-associated toxicity and treatment-related deaths are estimated to occur in about 3% of patients (46).

There are currently 16 active clinical trials investigating the use of Y90-SIRT with glass microspheres for treatment of HCC (Table 4) (6). In the United States, there are studies investigating the use of Y90-SIRT for the treatment of unresectable HCC (NCT01349075, NCT02072356, NCT00906984, NCT01176604, and NCT00877136) and assessing the safety of Y90-SIRT compared to radiofrequency prior to surgical resection in patients with HCC (NCT01686880) (6). The PREMIERE trial in the United States is comparing Y90-SIRT with TACE in patients who are not candidates for radiofrequency ablation or in patients with unresectable HCC (NCT00956930) (6). The European TRACE trial is a multicenter randomized control study comparing TARE with TACE for the treatment of HCC (NCT01381211) (6).

The good toxicity profile of Y90-SIRT makes it an alternative or adjunct to treatment with sorafenib (70). In the United States, a prospective randomized trial is studying the role of Y90-SIRT alone or in combination with sorafenib in the treatment of HCC patients who are awaiting liver transplantation (NCT00846131) and a phase II study is investigating the role of Y90-SIRT with sorafenib in advanced HCC (NCT01900002) (6). The STOP-HCC study, which is being conducted in the United States and Europe, is comparing sorafenib with and without Y90-SIRT (NCT01556490) (6). The outcomes of these trials may result in the incorporation of Y90-SIRT with glass microspheres into the treatment guidelines as a standard therapy option for patients with intermediate and advanced stage HCC.

Although the use of Y90-SIRT in treating liver malignancies dates back to the 1960s (717273), it has recently begun to gain clinical acceptance as a promising treatment option for patients with intermediate and advanced HCC (8). Multiple studies have provided compelling data which suggest that Y90-SIRT in comparison to TACE, has a higher tumor response, less post-embolization symptoms, better downstaging outcomes, and provides a higher QoL. However, the EASL, the European Organization for Research and Treatment of Cancer, and the American Association for the Study of Liver Diseases do not currently recommend Y90-SIRT as a standard therapy for intermediate or advanced HCC outside clinical trials (70). This clinical lack of enthusiastic approval stems from the fact that there is a lack of research and evidence from large-scale randomized controlled trials. Nonetheless, with the growing body of level 2 and level 3 evidence, Y90-SIRT has found a place in the guidelines adopted by the European Society for Medical Oncology, the European Society of Digestive Oncology, and the National Comprehensive Cancer Network (21). Y90-SIRT may become widely accepted and may be incorporated into the treatment guidelines if further research shows reproducibility, multicenter implementation, and economic feasibility.