Embolization Therapy: Principles and Clinical Applications, 1 Ed.

Radioembolization

Ryan M. Hickey • Robert J. Lewandowski • Riad Salem

Radioembolization refers to the transcatheter, intra-arterial injection of micrometer-sized embolic particles loaded with the high-dose radioisotope yttrium 90. Radioembolization combines the selective angiographic techniques essential to all embolization procedures with an understanding of radiation administration and safety, including radiation dosimetry and radiation dose modification based on tumor characteristics and the patient’s clinical profile.

Radioembolization delivers an internal radiation source and is therefore considered brachytherapy. Whereas the radiosensitive nature of normal liver tissue has limited the role of external beam radiation in the treatment of primary and metastatic hepatic malignancies, radioembolization allows for the safe administration of high and therapeutic doses of radiation. More specifically, the likelihood of developing severe radiation-induced liver disease may exceed 50% for external beam radiation doses greater than 35 to 40 Gy, whereas radiation doses greater than 150 Gy have been safely administered with radioembolization.14

The high tumoral radiation doses achieved with radioembolization rely on the selective intra-arterial injection of the radioactive particles and the preferential deposition of these particles within tumor as opposed to normal liver tissue due to differences in normal and tumor tissue perfusion. Whereas normal liver parenchyma receives most (75%) of its blood supply from the portal vein, hepatic malignancies, particularly hypervascular tumors such as hepatocellular carcinoma (HCC), derive most of their blood supply from the hepatic arteries.5,6

The radioisotope yttrium 90 is irreversibly incorporated into glass or resin microspheres that range in size from 20 to 30 µm (glass) or 20 to 60 µm (resin). Yttrium 90 is a pure beta emitter with a half-life of 64.2 hours and tissue penetration ranging from 2.5 to 11 mm.711 Glass microspheres (TheraSphere; Nordion, Kanata, Ontario, Canada) were approved in 1999 by the U.S. Food and Drug Administration (FDA) under a humanitarian device exemption (HDE) for the treatment of unresectable HCC.12 Resin microspheres (SIR-Spheres; SIRTeX Medical Limited, New South Wales, Australia) were granted full premarketing approval in 2002 by the FDA for the treatment of unresectable colorectal metastases in conjunction with intrahepatic floxuridine.13

PATIENT SELECTION

The patient selection process for yttrium 90 (Y90) radioembolization is multifactorial and involves an assessment of the patient’s burden of disease, biochemical profile, and performance status. Ideally, patients should have liver-only disease with a tumor burden comprising less than 70% of the liver volume. There should be adequate hepatic reserve, as indicated by the patient’s liver function tests with a bilirubin less than or equal to 2 mg/dL and albumin greater than 3 g/dL. The prothrombin time is a sensitive indicator of hepatic synthetic function and should correspond to a normal international normalized ratio (INR). Cancer-related symptoms should be minimal, corresponding to an Eastern Cooperative Oncology Group (ECOG) score of 0 to 2. Although the presence of portal vein thrombosis (PVT) has been traditionally considered a contraindication to hepatic arterial embolization procedures, radioembolization has been shown to be safe and effective in the setting of partial and branch PVT.14,15Radioembolization should be performed with caution in patients who have had intervention or surgery involving the ampulla of Vater due to the risk of developing hepatic abscesses. These patients should receive aggressive antibiotic coverage at the time of the procedure that is continued into the postprocedural period.

TECHNIQUE

Y90 radioembolization is performed on an outpatient basis and involves two separate angiography procedures. An initial mapping angiography is performed, at which time a radioisotope lung shunt fraction is also determined. Y90 microspheres are then administered during a separate treatment angiography.

Due to the frequency of hepatic arterial variants and the propensity of hepatic tumors to exhibit arteriovenous shunting, the mapping angiography and lung shunt fraction calculation are integral to radioembolization. Meticulous visceral and hepatic angiography is required to establish the planned treatment volumes that will be used for radiation dosimetry as well as to identify extrahepatic arterial pathways that could result in devastating toxicities and complications from nontarget radioembolization.

HCC is associated with a relatively high incidence of direct arteriovenous shunts that bypass the tumor capillary bed.16 The administration of microspheres smaller than these shunts could therefore result in direct shunting of the radioactive microspheres to the lungs, which can cause radiation pneumonitis at sufficient doses.17

99mTc-macroaggregated albumin (MAA) is of similar size to the Y90 microspheres and expected to mirror their distribution, including in pulmonary shunting. 99mTc-MAA 2 to 4 mCi is administered via the proper, right, or left hepatic artery, depending on the planned treatment site, at the conclusion of the initial angiography. The patient is then transferred to nuclear medicine for acquisition of planar and/or single-photon emission computed tomography (SPECT) gamma camera images that are used to calculate the fraction of administered 99mTc-MAA activity to arrive in the lungs. It is worth mentioning that the time required to transfer patients from the angiography suite to nuclear medicine is an important factor, as the normal, time-dependent degradation of 99mTc-MAA over a prolonged period can result in a falsely increased calculation of lung shunting.18 For the most accurate calculation of lung shunting, it is recommended that the 99mTc-MAA be administered at the end of the mapping angiography and patients be transferred to nuclear medicine as soon as hemostasis has been achieved at the femoral access site. When lung shunting is identified, the cumulative pulmonary dose must be calculated along with the hepatic treatment doses. Pulmonary doses greater than 30 Gy per treatment or greater than 50 Gy cumulatively have been associated with the development of radiation pneumonitis.19

Hepatic arterial anatomy and corresponding treatment volumes are determined with the mapping angiography. Because the administered radiation dose depends on the activity of Y90 administered to a volume of tissue, common variants of hepatic vasculature can significantly alter the treatment plan. For example, in a patient with multinodular HCC limited to the right hepatic lobe, standard anatomy may allow for Y90 administration via the right hepatic artery. However, in the presence of an accessory hepatic artery, which typically perfuses the posterior segments 6 and 7, one injection must be performed via the accessory right hepatic artery according to the volume of segments 6 and 7 while a separate injection must be delivered via the right hepatic artery according to the volume of segments 5 and 8. The target lobar and/or segmental volumes corresponding to the angiographic anatomy are measured from the patient’s preceding CT or magnetic resonance imaging (MRI) on a three-dimensional (3-D) workstation.

In addition to treatment planning, the mapping angiography provides a critical opportunity to identify extrahepatic arterial flow, particularly to the gastrointestinal tract, that could result in nontarget radioembolization. These vessels can either be eliminated with coil embolization during the mapping angiography or avoided by alterations in the treatment plan. A left gastric artery may have to be embolized to avoid reflux of Y90 to the stomach during treatment via a replaced or accessory left hepatic artery. The gastroduodenal and right gastric arteries are often prophylactically embolized, particularly when using the more embolic resin microspheres. The relative location of the cystic, supraduodenal, retroduodenal, falciform, and phrenic arteries should also be considered in determining the site of Y90 injection. An in-depth review of angiographic considerations and relevant variant anatomy has been previously published.20

A simplified matrix has been described for the treatment planning of patients with unresectable HCC based on the extent of disease and the patient’s total bilirubin level.18 For patients with uninodular disease and a normal bilirubin, Y90 can be administered via a lobar or segmental injection. For patients with uninodular disease but elevated bilirubin, treatment should only proceed if a segmental perfusing vessel can be isolated, resulting in high-dose radiation to the tumor and minimal distribution to the normal liver parenchyma. For patients with multinodular and/or bilobar disease and a normal bilirubin level, staged lobar treatments may be performed. In the setting of abnormal bilirubin and multinodular/bilobar disease, the risk of Y90 treatment may be unacceptably high.

DOSIMETRY

Y90 radiation dosimetry uses volumetric calculations of liver tissue (glass microspheres) or tumor burden and body surface area (resin microspheres). Several different software packages are available to calculate 3-D tissue volumes from triphasic CT or multiphasic contrast-enhanced MRI. Volumetric calculation and treatment planning require a sound understanding of the Couinaud hepatic segments and their anatomic landmarks on cross-sectional imaging.

For glass microsphere (TheraSphere) dosimetry, the volume that is measured and used in the dosimetry calculation is the volume of liver tissue that is perfused by the vessel infused. In other words, it is the volume of the liver segments being perfused by the vessel of interest.

Glass microspheres (TheraSphere) are available in vials of several different activities that are dispensed weekly on Wednesday by the manufacturer (Nordion) and calibrated at 12:00 noon (Eastern Standard Time) of the following Sunday. The microspheres have an approximate activity of 2,500 Bq per sphere.1

The recommended activity that should be administered to a tumor-containing hepatic lobe should correspond to a dose between 80 and 150 Gy. Patients with significant cirrhosis should be treated conservatively with doses between 80 and 100 Gy, whereas patients without cirrhosis may be treated at higher doses between 100 and 150 Gy. The most commonly used dose range at our institution is between 100 and 120 Gy. Activity required to deliver the desired dose can be calculated according to the equation:

Activity is measured in GBq infused to the target liver, D is the absorbed dose (Gy) to the target liver mass, M (kg). Liver volume (mL) is calculated with 3-D software and converted to mass using a conversion factor of 1.03 mg/mL. Lung shunt fraction (%LSF) is calculated from nuclear medicine images acquired following the initial mapping angiography. Residual activity within the vial (%R) is measured after Y90 administration and approximated to be 2% for pretreatment dosing calculations.18

Resin microspheres (SIR-Spheres) are available in 3 GBq vials dispensed three times per week and calibrated at 6:00 pm EST on the date of treatment. Each vial contains 40 to 80 million microspheres with an activity per sphere of 50 Bq.1

Resin microsphere dosimetry can be calculated using body surface area and estimates of tumor burden according to the equation:

Activity is GBq infused and BSA is the body surface area in square meters.

For resin microsphere dosimetry calculation, activity is decreased depending on the degree of LSF: less than 10% LSF, no reduction; 10% to 15% LSF, 20% reduction; 15% to 20% LSF, 40% reduction; greater than 20% LSF, no treatment.18

CLINICAL OUTCOMES

Several large studies have recently been published describing long-term outcomes following radioembolization in patients with intermediate and advanced stage HCC. Table 37.1 provides a summary of these findings. For context of the natural disease of HCC, meta-analysis of the reported survival rates of untreated HCC patients included in randomized controlled trials revealed 1-year survival rates for intermediate (Barcelona Clinic Liver Cancer [BCLC B]) and advanced (BCLC C) stage HCC of approximately 50% and 25%, respectively21 (Table 37.1).

The long-term outcomes of a comprehensive, 291-patient cohort of intermediate and advanced stage HCC patients were recently reported and is the first study to describe toxicity, imaging, and survival outcomes stratified according to BCLC staging, United Network for Organ Sharing (UNOS) tumor stage, and Child-Pugh classification.22 Response rates to radioembolization were 42% and 57% according to World Health Organization (WHO) and European Association for the Study of Liver (EASL) criteria, respectively. Time to tumor progression (TTP) was 7.9 months. Child-Pugh A patients benefitted most from radioembolization, whereas Child-Pugh B patients with PVT had poor outcomes. Child-Pugh A patients had better response rates than Child-Pugh B patients, with median survival of 17.2 months and 7.7 months, respectively (P = .002). Child-Pugh B patients with PVT had a median survival of 5.6 months.

These outcomes were validated by a subsequent study of 108 patients with advanced HCC, confirming both the reproducibility of Y90 treatment in advanced HCC as well as outcomes equivalent to conventional transarterial chemoembolization (cTACE) and TACE with drug-eluting beads (DEB-TACE).23 The response rate according to EASL criteria was 40%. TTP was 10.0 months, with median overall survival of 16.4 months.

A 325-patient study has further confirmed long-term survival outcomes stratified by BCLC stages.24 Median overall survival was 12.8 months, which varied across BCLC stages (BCLC A, 24.4 months; BCLC B, 16.9 months; BCLC C, 10.0 months). The strongest independent prognostic factors for survival on multivariate analysis were ECOG status, tumor burden (>5 nodules), INR greater than 1.2, and extrahepatic disease.

A phase 2 study of 52 patients prospectively evaluated the efficacy of Y90 radioembolization using TTP as the primary end point for patients with intermediate and advanced stage HCC, the findings of which have led to a randomized phase 3 trial comparing radioembolization to sorafenib.25 Overall response rate was 40.4%. TTP was 11 months, with no significant difference between patients with and without PVT. Median overall survival was 15 months.

Although there has been no randomized study comparing radioembolization to chemoembolization, a comparative effectiveness report described outcomes following radioembolization and cTACE in a 245-patient cohort. The authors determined that adverse events, clinical toxicities, response rate, and TTP were improved with radioembolization compared to cTACE. Overall survival was no different between radioembolization and cTACE, although likely as a result of the competing risks of death of HCC and cirrhosis. Post hoc analyses concluded that a sample size larger than 1,000 patients would be required to establish survival equivalence between cTACE and radioembolization.26

Radioembolization has shown favorable rates of downstaging HCC patients to transplantation, resection, or ablation and superior rates of downstaging to transplantation when compared to cTACE. In one study, 66% of patients who were initially not candidates for transplantation, resection, or ablation were successfully downstaged to transplantation, resection, or ablation following radioembolization.27 In another study, 58% of patients with UNOS T3 disease (outside transplant criteria) treated with radioembolization were downstaged to T2 disease, compared to 31% of patients treated with cTACE.28

Diversion of portal venous flow away from the liver parenchyma with a transjugular intrahepatic portosystemic shunt (TIPS) has raised concern about the use of embolic transarterial therapies in patients with TIPS due to further reduction in liver perfusion resulting in hepatic ischemia. A retrospective comparison of patients with and without TIPS undergoing chemoembolization reported significantly higher rates of severe hepatotoxicity in patients with TIPS.29 Radioembolization, however, is minimally embolic and has previously been shown to be safe and effective in the setting of partial and branch PVT.14,15Additionally, a recent report of Y90 radioembolization in the presence of TIPS indicated rates of hepatotoxicity comparable to those previously reported for Y90 in patients without TIPS. The authors concluded that radioembolization may be safely performed in patients with unresectable HCC and TIPS, particularly as a bridge to liver transplantation.30

SIDE EFFECTS AND COMPLICATIONS

The most common side effect of Y90 radioembolization is fatigue, which occurs in 50% to 60% of patients during 1 to 2 weeks following treatment. Approximately 20% of patients experience low-grade abdominal pain or nausea/vomiting, both of which are typically well controlled with oral medications.22

Meticulous angiographic technique and careful identification of potential routes of extrahepatic flow are absolutely essential to avoid gastrointestinal (GI) ulcers from nontarget radioembolization. Radiation-induced GI ulcers generally do not respond to proton pump inhibitors and therefore may require surgical resection in patients who are poor surgical candidates.

Biliary complications of Y90 include biliary dyskinesia, radiation cholecystitis, biliary stricture, and biliary necrosis. Biliary dyskinesia presents with postprandial abdominal pain without fever or leukocytosis and is usually self-limited. In the largest reported series evaluating biliary sequelae after Y90 radioembolization for the treatment of HCC as well as metastatic disease to the liver, 10% of patients demonstrated imaging findings related to the biliary tree, of which 1.8% required an unplanned interventional or surgical procedure. Findings included biliary necrosis (3.9%), biloma (1%), biliary stricture (2.4%), gallbladder wall enhancement (1.8%), and gallbladder wall disruption (0.9%).31

Radiation-induced liver disease (RILD), previously referred to as radiation hepatitis, was first described in patients undergoing external beam radiation and is the most severe potential hepatotoxicity to result from radioembolization. Sinusoidal congestion, venous occlusion, and hepatic fibrosis are the pathologic findings of RILD. The clinical manifestations include nausea, vomiting, abdominal pain, jaundice, and ascites, typically presenting 4 to 8 weeks after radiation exposure. Greater than twofold elevation of alkaline phosphatase is the most specific of liver chemistry abnormalities. Outcomes are variable, with a minority of patients dying of fulminant hepatic failure during the acute phase and most patients surviving with chronic liver failure. Rates of RILD following radioembolization have been reported between 4% and nearly 7%, which include patients with metastatic disease to the liver previously treated with chemotherapy.32,33 There is no predictive model for the development of RILD following radioembolization. The use of an empiric model in the calculation of resin microspheres dosimetry, which is no longer recommended, has been associated with RILD.32 Pathologically confirmed RILD has not been reported in patients receiving glass microsphere radioembolization for unresectable HCC.34 Nonetheless, careful patient selection and appropriate radiation dosimetry are critical to minimize the risk of radiation injury to the liver.

POSTTREATMENT IMAGING

Imaging evaluation following Y90 radioembolization includes assessing tumor response as well as distinguishing benign from more worrisome imaging findings. Anatomic response assessment is based on changes in tumor size (WHO, Response Evaluation Criteria In Solid Tumors [RECIST]) and degree of necrosis as indicated by tumor tissue enhancement (EASL, modified RECIST). Because changes in tumor size alone often underestimate the actual response rate and outcomes following locoregional liver therapies, many clinical trials use the EASL criteria and modified RECIST (mRECIST) to assess residual viable tumor. Indeed, EASL response has been shown to be more consistent than WHO in predicting survival outcomes for HCC following locoregional therapy.35 Functional imaging with diffusion-weighted MRI (DWI) can provide evidence of metabolic tumor responses on early posttreatment imaging that correlate with anatomic responses on subsequent imaging.36

Several benign incidental findings may be seen on imaging after Y90 radioembolization as a result of local radiation effects, including peritumoral edema, ring enhancement, perivascular edema, perihepatic fluid, and pleural effusion. Over time, atrophy of the treated lobe with hypertrophy of the untreated lobe, liver capsular retraction, fibrosis, and evidence of portal hypertension may be seen. More worrisome findings such as hepatic abscess, biloma, radiation cholecystitis, or radiation hepatitis warrant appropriate clinical investigation.37

REFERENCES

 1. Kennedy AS, Nutting C, Coldwell D, et al. Pathologic response and microdosimetry of (90)Y microspheres in man: review of four explanted whole livers. Int J Radiat Oncol Biol Phys. 2004;60(5):1552–1563.

 2. Emami B, Lyman J, Brown A, et al. Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys. 1991;21(1):109–122.

 3. Ingold JA, Reed GB, Kaplan HS, et al. Radiation hepatitis. Am J Roentgenol Radium Ther Nucl Med. 1965;93:200–208.

 4. Lawrence TS, Robertson JM, Anscher MS, et al. Hepatic toxicity resulting from cancer treatment. Int J Radiat Oncol Biol Phys. 1995;31(5):1237–1248.

 5. Gyves JW, Ziessman HA, Ensminger WD, et al. Definition of hepatic tumor microcirculation by single photon emission computerized tomography (SPECT). J Nucl Med. 1984;25(9):972–977.

 6. Bierman HR, Byron RL Jr, Kelley KH, et al. Studies on the blood supply of tumors in man. III. Vascular patterns of the liver by hepatic arteriography in vivo. J Natl Cancer Inst. 1951;12(1):107–131.

 7. Geschwind JF, Salem R, Carr BI, et al. Yttrium-90 microspheres for the treatment of hepatocellular carcinoma. Gastroenterology. 2004;127(5)(suppl 1):S194–S205.

 8. Andrews JC, Walker SC, Ackermann RJ, et al. Hepatic radioembolization with yttrium-90 containing glass microspheres: preliminary results and clinical follow-up. J Nucl Med. 1994;35(10):1637–1644.

 9. Sarfaraz M, Kennedy AS, Cao ZJ, et al. Physical aspects of yttrium-90 microsphere therapy for nonresectable hepatic tumors. Med Phys. 2003;30(2):199–203.

10. Dancey JE, Shepherd FA, Paul K, et al. Treatment of nonresectable hepatocellular carcinoma with intrahepatic 90Y-microspheres. J Nucl Med. 2000;41(10):1673–1681.

11. Salem R, Thurston KG, Carr BI, et al. Yttrium-90 microspheres: radiation therapy for unresectable liver cancer. J Vasc Interv Radiol. 2002;13(9, pt 2):S223–S229.

12. TheraSphere Yttrium-90 microspheres [package insert]. Kanata, Canada: Nordion; 2004.

13. SIR-Spheres Yttrium-90 microspheres [package insert]. New South Wales, Australia: SIRTeX Medical; 2004.

14. Kulik LM, Carr BI, Mulcahy MF, et al. Safety and efficacy of 90Y radiotherapy for hepatocellular carcinoma with and without portal vein thrombosis. Hepatology. 2008;47(1):71–81.

15. Iñarrairaegui M, Thurston KG, Bilbao JI, et al. Radioembolization with use of yttrium-90 resin microspheres in patients with hepatocellular carcinoma and portal vein thrombosis. J Vasc Interv Radiol. 2010;21(8):1205–1212.

16. Chen JH, Chai JW, Huang CL, et al. Proximal arterioportal shunting associated with hepatocellular carcinoma: features revealed by dynamic helical CT. AJR Am J Roentgenol. 1999;172(2):403–407.

17. Ho S, Lau, WY, Leung TW, et al. Partition model for estimating radiation doses from yttrium-90 microspheres in treating hepatic tumours. Eur J Nucl Med. 1996;23(8):947–952.

18. Salem R, Thurston KG. Radioembolization with 90Yttrium microspheres: a state-of-the-art brachytherapy treatment for primary and secondary liver malignancies. Part 1: technical and methodologic considerations. J Vasc Interv Radiol. 2006;17(8):1251–1278.

19. Ho S, Lau WY, Leung TW, et al. Clinical evaluation of the partition model for estimating radiation doses from yttrium-90 microspheres in the treatment of hepatic cancer. Eur J Nucl Med. 1997;24(3):293–298.

20. Lewandowski RJ, Sato KT, Atassi B, et al. Radioembolization with 90Y microspheres: angiographic and technical considerations. Cardiovasc Intervent Radiol. 2007;30(4):571–592.

21. Cabibbo G, Enea M, Attanasio M, et al. A meta-analysis of survival rates of untreated patients in randomized clinical trials of hepatocellular carcinoma. Hepatology. 2010;51(4):1274–1283.

22. Salem R, Lewandrowski RJ, Mulcahy MF, et al. Radioembolization for hepatocellular carcinoma using yttrium-90 microspheres: a comprehensive report of long-term outcomes. Gastroenterology. 2010;138(1):52–64.

23. Hilgard P, Hamami M, Fouly AE, et al. Radioembolization with yttrium-90 glass microspheres in hepatocellular carcinoma: European experience on safety and long-term survival. Hepatology. 2010;52(5):1741–1749.

24. Sangro B, Carpanese L, Cianni R, et al. Survival after yttrium-90 resin microsphere radioembolization of hepatocellular carcinoma across Barcelona clinic liver cancer stages: a European evaluation. Hepatology. 2011;54(3):868–878.

25. Mazzaferro V, Sposito C, Bhoori S, et al. Yttrium-90 radioembolization for intermediate-advanced hepatocellular carcinoma: a phase 2 study. Hepatology. 2013;57(5):1826–1837.

26. Salem R, Lewandowski RJ, Kulik L, et al. Radioembolization results in longer time-to-progression and reduced toxicity compared with chemoembolization in patients with hepatocellular carcinoma. Gastroenterology. 2011;140(2):497–507.e2.

27. Kulik LM, Atassi B, van Holsbeeck L, et al. Yttrium-90 microspheres (TheraSphere) treatment of unresectable hepatocellular carcinoma: downstaging to resection, RFA and bridge to transplantation. J Surg Oncol. 2006;94(7):572–586.

28. Lewandowski RJ, Kulik LM, Riaz A, et al. A comparative analysis of transarterial downstaging for hepatocellular carcinoma: chemoembolization versus radioembolization. Am J Transplant. 2009;9(8):1920–1928.

29. Kohi MP, Fidelman N, Naeger DM, et al. Hepatotoxicity after transarterial chemoembolization and transjugular intrahepatic portosystemic shunt: do two rights make a wrong? J Vasc Interv Radiol. 2013;24(1):68–73.

30. Donahue LA, Kulik L, Baker T, et al. Yttrium-90 radioembolization for the treatment of unresectable hepatocellular carcinoma in patients with transjugular intrahepatic portosystemic shunts. J Vasc Interv Radiol. 2013;24(1):74–80.

31. Atassi B, Bangash AK, Lewandowski RJ, et al. Biliary sequelae following radioembolization with yttrium-90 microspheres. J Vasc Interv Radiol. 2008;19(5):691–697.

32. Kennedy AS, McNellie P, Dezarn WA, et al. Treatment parameters and outcome in 680 treatments of internal radiation with resin 90Y-microspheres for unresectable hepatic tumors. Int J Radiat Oncol Biol Phys. 2009;74(5):1494–1500.

33. Sangro B, Gil-Alzugaray B, Rodriguez J, et al. Liver disease induced by radioembolization of liver tumors: description and possible risk factors. Cancer. 2008;112(7):1538–1546.

34. Goin JE, Salem R, Carr BI, et al. Treatment of unresectable hepatocellular carcinoma with intrahepatic yttrium 90 microspheres: factors associated with liver toxicities. J Vasc Interv Radiol. 2005;16(2, pt 1):205–213.

35. Memon K, Kulik L, Lewandowski RJ, et al. Radiographic response to locoregional therapy in hepatocellular carcinoma predicts patient survival times. Gastroenterology. 2011;141(2):526–535, 535.e1–535.e2.

36. Rhee TK, Naik NK, Deng J, et al. Tumor response after yttrium-90 radioembolization for hepatocellular carcinoma: comparison of diffusion-weighted functional MR imaging with anatomic MR imaging. J Vasc Interv Radiol. 2008;19(8):1180–1186.

37. Atassi B, Bangash AK, Bahrani A, et al. Multimodality imaging following 90Y radioembolization: a comprehensive review and pictorial essay. Radiographics. 2008;28(1):81–99.