DOI: 10.1148/rg.25si055515
RadioGraphics 2005;25:S41-S55
© RSNA, 2005
Yttrium-90 Microsphere Therapy for Hepatic Malignancy: Devices, Indications, Technical Considerations, and Potential Complications1
Ravi Murthy, MD,
Rodolfo Nunez, MD,
Janio Szklaruk, MD,
William Erwin, PhD,
David C. Madoff, MD,
Sanjay Gupta, MD,
Kamran Ahrar, MD,
Michael J. Wallace, MD,
Alan Cohen, MD,
Douglas M. Coldwell, PhD, MD2,
Andrew S. Kennedy, MD and
Marshall E. Hicks, MD
1 From the Interventional Radiology Section (R.M., D.C.M., S.G., K.A., M.J.W., M.E.H.) and the Departments of Nuclear Medicine (R.N.), Body Imaging (J.S.), and Imaging Physics (W.E.), Division of Diagnostic Imaging, University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Unit 325, Houston, TX 77030; Department of Radiology and Nuclear Medicine, University of Texas Hermann Hospital, Houston, Tex (A.C.); Department of Radiology, University of Mississippi Medical Center, Jackson, Miss (D.M.C.); and Wake Radiation Oncology, Cary, NC (A.S.K.). Presented as an education exhibit at the 2004 RSNA Annual Meeting. Received March 23, 2005; revision requested April 25 and received June 7; accepted June 15. The article discusses an investigational or unlabeled use of a commercial device or pharmaceutical that has not been approved for such purpose by the FDA. The glass microsphere device (TheraSphere; MDS Nordion, Ottawa, Ontario, Canada) has received humanitarian device exemption approval from the FDA for treatment of unresectable hepatocellular carcinoma and can be used only with investigational review board oversight. The resin microsphere device (SIR-Spheres; Sirtex Medical, Lake Forest, Ill) has received premarket approval from the FDA for use in combination with hepatic arterial floxuridine therapy to treat colorectal metastasis to the liver; its use in any other manner for treatment of hepatic neoplastic disease is an off-label application. R.M., A.S.K., and D.M.C. have received honoraria from Sirtex Medical; all remaining authors have no financial relationships to disclose.
Address correspondence to R.M. (e-mail: rmurthy{at}di.mdacc.tmc.edu).
 |
Abstract
|
|---|
Management of hepatic malignancies is a ubiquitous medical problem. Surgical resection of primary or metastatic liver cancer, with or without adjuvant chemotherapy, is the most effective method for enhancing survival; however, hepatic malignancies in the vast majority of patients are unresectable both at initial manifestation and at recurrence. In these patients, palliative cytoreductive therapies may help to retard tumor progression and therefore favorably alter the course of the disease. Since hepatic neoplasms are principally supplied by the hepatic artery, various arterially delivered cytotoxic agents have been developed to achieve these objectives. Recently, the Food and Drug Administration approved the transarterial administration of yttrium-90 microspheres for liver-directed therapy. Effective use of these devices requires knowledge of the accumulated clinical experience and a dedicated multidisciplinary effort to ensure optimal outcomes and avoid therapy-specific life-threatening complications.
© RSNA, 2005
|
LEARNING OBJECTIVES FOR TEST 2
|
|---|
After reading this article and taking the test, the reader will be able to:
- Identify the principles underlying yttrium-90 microsphere therapy for hepatic malignancy and patient selection criteria.
- Describe similarities and differences between types of microspheres, the process of therapy planning, and published outcomes.
- Discuss methods for avoiding complications and mitigating toxic effects of therapy.
|
Introduction
|
|---|
The liver is a predominant focus of metastatic disease from a wide variety of neoplasms. Autopsy studies have demonstrated that the liver is involved in 50%–70% of metastases from melanoma, lymphoma, and common malignancies that originate in the breast, lung, and gastrointestinal tract (1), including colorectal cancer. There were estimated to be 145,290 new cases of colorectal cancer in 2005 (2), and 5-year survival rates are dismal for patients with unresectable disease (3).
The liver also may be the locus of various primary neoplasms. Hepatocellular carcinoma is the visceral malignancy with the highest annual incidence worldwide. The incidence of hepatocellular carcinoma, predominantly secondary to hepatitis C, is increasing in the United States, where more than 18,000 largely unresectable malignancies of this type are diagnosed per year (2). Since the median survival is less than 1 year, the incidence is approximately equivalent to the prevalence of this disease (4). Most of those affected by hepatocellular carcinoma die either from cirrhosis or from local tumor progression (5). Surgical resection and transplantation produce reasonable 5-year survival rates; however, fewer than 10% of those in whom hepatocellular carcinoma is diagnosed are candidates for either of these procedures. There is a need for more effective treatments of this disease in these patients.
Because the liver may be both a transit point for metastasis and the location of substantial tumor burden, effective hepatic control of cancer may favorably alter the overall progression of disease. Since hepatic neoplasms are supplied principally by the hepatic artery (6), various cytotoxic agents have been developed for transarterial delivery to the neoplasms. Recently, the Food and Drug Administration approved the transarterial administration of yttrium-90 (90Y) microspheres for liver-directed cancer therapy. This article is focused on two important current applications of this technology: the treatment of hepatic meta-static colorectal cancer and of hepatocellular carcinoma.
The use of 90Y to treat these diseases began with its topical application and direct intratumoral injection. Transarterial administration of 90Y-bearing microspheres in humans was described in the early literature as feasible (7–9), and, somewhat later, as capable of inducing tumor regression (10). The results of numerous subsequent studies have supported the use and confirmed the effectiveness of 90Y-bearing microspheres in the treatment of hepatic primary neoplasms (11–15) and metastatic neoplasms (16–18).
Therapeutic Advantages of 90Y Microspheres
Radiation, if delivered in sufficient doses, is lethal to neoplastic tissue. Normal hepatocytes have an even lower tolerance of the effects of radiation than does neoplastic tissue. When the whole liver is exposed to external-beam radiation at a mean radiation dose of more than 43 Gy, more than 50% of patients develop liver dysfunction (19). Conformal and stereotactic radiation therapy techniques can be used to deliver much higher radiation doses for focal treatment of disease; however, since hepatic metastases and primary neoplasms are most often multifocal and irregular in shape and may replace large parts of the liver volume, only a small minority of patients are optimal candidates for such therapies (20). 90Y-bearing microspheres, unlike external-beam radiation sources, are point sources of radiation that preferentially localize in the peritumoral and intratumoral arterial vasculature (21). This characteristic makes them suitable vehicles for selective delivery of very high radiation doses to tumors while radiation exposure to the normal hepatic parenchyma remains within tolerable limits.
90Y is produced by bombardment of yttrium 89 with neutrons in a nuclear reactor. It has a physical half-life of 64.2 hours (2.67 days), and it decays to stable zirconium 90. 90Y emits pure high-energy beta rays (energy maximum, 2.27 MeV; mean, 0.9367 MeV) with an average penetration range of 2.5 mm and a maximum range of 11 mm in tissue. One gigabecquerel (27 mCi) of 90Y delivers a total absorbed radiation dose of 50 Gy/kg. In therapeutic use in which the isotope decays to infinity, 94% of the radiation is delivered in 11 days.
There are currently two commercially available microsphere devices in which 90Y is incorporated: one with microspheres made of glass (Thera-Sphere; MDS Nordion, Ottawa, Ontario, Canada) and the other with microspheres made of resin (SIR-Spheres; Sirtex Medical, Sydney, Australia). The characteristics of the two devices are compared in Table 1.
Glass Microsphere Device.—
90Y is an integral constituent of the insoluble and nonbiodegradable glass microspheres. The mean diameter of each microsphere is 20–30 µm, and an aggregate of 22,000–73,000 spheres weighs just 1 mg (Fig 1). Use of the glass microsphere device initially was approved in Canada, in 1991, for the intraarterial treatment of hepatic neoplasia. Later, the Food and Drug Administration (FDA) Office of Orphan Products Development granted a humanitarian device exemption for designated use in the presence of a specific indication: for radiation therapy or as a neoadjuvant to surgery or transplantation in patients with unresectable hepatocellular carcinoma in whom hepatic arterial catheters can be appropriately positioned. The FDA-conferred humanitarian device exemption is intended to benefit patients in the United States by enabling use of a device for treatment or diagnosis of a disease or condition that affects fewer than 4000 individuals in the United States per year. This use may take place only with oversight by the institutional review board. The glass microsphere device was approved by the FDA in 2000 for humanitarian use in the United States.
Resin Microsphere Device.—
The resin microsphere device consists of biocompatible 90Y-bearing microspheres with diameters of 20–40 µm (Fig 2). The spheres are neither metabolized nor excreted but remain in the liver as a permanent implant. The device is approved for use in treating unresectable metastatic liver tumors from primary colorectal cancer and is used with adjuvant chemotherapy with floxuridine administered via the hepatic artery. The device was granted pre-market approval by the FDA in 2002 for use in the United States.
Contraindications
There are two absolute contraindications to microsphere-based 90Y treatment: exaggerated hepatopulmonary shunting, and reflux into the arteries that supply the gastroduodenal region. The additional effect of internal irradiation from 90Y after therapeutic external-beam irradiation of the liver has not been studied, and therefore such use of 90Y-bearing microspheres is not advocated. Prior, concurrent, or subsequent treatment with capecitabine may increase the risk of toxic effects in the liver. The presence of ascites or an elevated bilirubin level usually indicates decreased functional hepatocellular reserve; after treatment, arterial ischemia, combined with the effects of radiation on nontumoral tissue, usually results in further deterioration of hepatic function. Portal venous thrombosis is listed in the package insert as a contraindication to use of the resin microsphere device, but treatment with the glass microsphere device has been successful in patients with this condition (22).
Federal Regulations of Use
A detailed explanation of the regulatory issues can be found in the Code of Federal Regulations from the U.S. Nuclear Regulatory Commission. 90Y-bearing microspheres are brachytherapy devices that are permanently implanted by using manual techniques. An authorized user, typically a nuclear medicine physician or radiation oncologist, must fulfill the requirements with regard to training and experience in the handling of radioactive materials as well as vendor-specific training in the use of the microspheres and the microsphere delivery system.
For 90Y-bearing microspheres, the prescribed dose is the total dose documented in the prescription, which, in regulatory parlance, is referred to as the "written directive." The written directive should include the total radiation dose to be administered to the liver, the physical form of the 90Y microspheres, and the maximum acceptable dose at extrahepatic sites to which the microspheres might be shunted (specifically, the lungs and gastrointestinal tract). A description of the mechanism used is mandatory to confirm that the radiation dose administered accords with the written directive. The patient may be released from the hospital only when the total effective radiation dose equivalent from bremsstrahlung emission from the patient to any other individual is not likely to exceed 5 mSv (0.5 rem).
Radiation Dosimetry
Compared with external-beam radiation therapy, therapy with 90Y microspheres results in a significantly higher dose of radiation, as well as greater variations in dose because of the nonuniform distribution of microspheres in the liver, with some tumors receiving more than 3000 Gy. The dose variation is a subject of intense investigation (23). Investigators in some studies have used a partition model to calculate the dose difference between nontumoral and tumoral tissues (24,25) on the basis of information extrapolated from technetium 99m (99mTc)-labeled macroaggregated albumin (MAA) scans coregistered with computed tomographic (CT) scans. A three-dimensional dose calculation technique has been developed for calculating the radiation dose to the liver and adjacent organs (26). Until subsequent developments in therapy planning are standardized and validated, use of the dose calculation method described in the manufacturer’s package insert is recommended.
|
Therapy Planning
|
|---|
Treatment with 90Y microspheres must be based on specific findings on cross-sectional images and arteriograms in the individual patient. The initial work-up should include CT or magnetic resonance (MR) imaging of the liver for assessment of tumoral and nontumoral volume, portal vein patency, and presence of extrahepatic disease. Serum chemical analyses also are performed to evaluate hepatic and renal function and to determine the presence of tumor markers. In general, a serum bilirubin level of more than 2 mg/dL is considered a contraindication to treatment with 90Y microspheres. In the presence of renal insufficiency, care must be taken to avoid or minimize the use of iodinated contrast material.
Arteriography for Hepatopulmonary Shunt Calculation
The aim of pretherapeutic assessment of the hepatic arterial vasculature is to ensure delivery of the microspheres to the target. The superior mesenteric, celiac, and hepatic arterial branches are evaluated. Evaluation should include a determination of the arterial location and any consequent necessity for embolization of the gastroduodenal artery, right gastric artery, and any other accessory arteries to prevent gastrointestinal deposition of the microspheres (Fig 3). The presence of variant hepatic arterial anatomy may alter the treatment plan (Fig 4). Treatment with 90Y microspheres is precluded by stenosis or slow antegrade flow within the hepatic arteries that results in embolic occlusion of the vessel and, therefore, reflux in extrahepatic territories. Complete portal vein thrombosis with absence of hepatopedal flow may indicate that the patient has a high risk for ischemic complications.
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a. Images obtained for therapy planning show normal anatomy in a 69-year-old man. (a) Celiac arteriogram obtained before treatment shows gastroduodenal artery (arrowhead), right gastric artery (white arrow), and hepatic artery branches (black arrows). Note that the right hepatic artery arises early in the branching of the common hepatic artery, before the origin of the gastroduodenal artery. (b, c) Arteriograms show selective catheterization and coil embolization of the right gastric artery (arrow in b) and gastroduodenal artery (arrow in c) to avert reflux and thereby decrease the risk of gastric ulcer. (d) Indirect portal venogram depicts patency of the main (arrow), left, and right portal venous branches.
|
|
View larger version (127K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b. Images obtained for therapy planning show normal anatomy in a 69-year-old man. (a) Celiac arteriogram obtained before treatment shows gastroduodenal artery (arrowhead), right gastric artery (white arrow), and hepatic artery branches (black arrows). Note that the right hepatic artery arises early in the branching of the common hepatic artery, before the origin of the gastroduodenal artery. (b, c) Arteriograms show selective catheterization and coil embolization of the right gastric artery (arrow in b) and gastroduodenal artery (arrow in c) to avert reflux and thereby decrease the risk of gastric ulcer. (d) Indirect portal venogram depicts patency of the main (arrow), left, and right portal venous branches.
|
|
View larger version (119K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3c. Images obtained for therapy planning show normal anatomy in a 69-year-old man. (a) Celiac arteriogram obtained before treatment shows gastroduodenal artery (arrowhead), right gastric artery (white arrow), and hepatic artery branches (black arrows). Note that the right hepatic artery arises early in the branching of the common hepatic artery, before the origin of the gastroduodenal artery. (b, c) Arteriograms show selective catheterization and coil embolization of the right gastric artery (arrow in b) and gastroduodenal artery (arrow in c) to avert reflux and thereby decrease the risk of gastric ulcer. (d) Indirect portal venogram depicts patency of the main (arrow), left, and right portal venous branches.
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3d. Images obtained for therapy planning show normal anatomy in a 69-year-old man. (a) Celiac arteriogram obtained before treatment shows gastroduodenal artery (arrowhead), right gastric artery (white arrow), and hepatic artery branches (black arrows). Note that the right hepatic artery arises early in the branching of the common hepatic artery, before the origin of the gastroduodenal artery. (b, c) Arteriograms show selective catheterization and coil embolization of the right gastric artery (arrow in b) and gastroduodenal artery (arrow in c) to avert reflux and thereby decrease the risk of gastric ulcer. (d) Indirect portal venogram depicts patency of the main (arrow), left, and right portal venous branches.
|
|
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Images obtained for therapy planning show common anatomic variations in a 42-year-old man with bilo-bar disease, findings that resulted in alteration of the treatment plan. (a) Celiac arteriogram depicts an accessory left hepatic artery (curved arrow), which originates from the left gastric artery (straight arrow), and the splenic and common hepatic arteries (arrowheads). (b) Superior mesenteric arteriogram shows a replacement of the right hepatic artery (arrowhead) and a normal left hepatic artery (arrow). Embolization of the accessory left hepatic artery was necessary to augment intrahepatic collateral flow before treatment. A single transarterial infusion of 90Y microspheres was then administered via the proper hepatic artery to treat the entire liver.
|
|
View larger version (135K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Images obtained for therapy planning show common anatomic variations in a 42-year-old man with bilo-bar disease, findings that resulted in alteration of the treatment plan. (a) Celiac arteriogram depicts an accessory left hepatic artery (curved arrow), which originates from the left gastric artery (straight arrow), and the splenic and common hepatic arteries (arrowheads). (b) Superior mesenteric arteriogram shows a replacement of the right hepatic artery (arrowhead) and a normal left hepatic artery (arrow). Embolization of the accessory left hepatic artery was necessary to augment intrahepatic collateral flow before treatment. A single transarterial infusion of 90Y microspheres was then administered via the proper hepatic artery to treat the entire liver.
|
|
Hepatopulmonary shunting secondary to tumor-related pathologic arteriovenous channels, as well as reflux toward the gastrointestinal region, may be detected at scintigraphy with the injection of 5–6 mCi (185–222 MBq) of 99mTc-labeled MAA as a microsphere surrogate into the hepatic arterial territory. The hepatopulmonary shunt fraction then can be calculated as the ratio of the gamma emission count in the lung to that in the liver in regions of interest on planar scintigrams (Fig 5). The ratio is calculated as a percentage that is rounded to the nearest whole percentage point. Patients in whom the hepatopulmonary shunt fraction is greater than 20% of the injected dose or in whom the shunt fraction indicates potential exposure of the lung to an absorbed radiation dose of more than 30 Gy should not be considered for treatment with 90Y microspheres (27). Additional exclusion criteria include blood flow toward the gastrointestinal region that is not correctable by means of embolization or catheter positioning.
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Scintigraphy performed for treatment planning in a 54-year-old man with metastasis of neuroendocrine cancer to the liver. Planar scintigram obtained after hepatic arterial injection of 5 mCi (185 MBq) of 99mTc-labeled MAA shows radionuclide activity in regions of interest in the liver and lungs but not in the gastrointestinal region. The hepatopulmonary shunt fraction, calculated on the basis of scintigraphic findings, was 10.2%, and the patient therefore underwent treatment with 90Y resin microspheres at a reduced dose. Treatment was successful.
|
|
Radiation Dose Calculation
To enable calculation of the appropriate radiation dose, all patients should undergo cross-sectional imaging with CT to determine the volume of the whole liver and of the right and left lobes. Calculations of dose with the glass microsphere device are based on the assumptions of a nominal target dose of 150 Gy/kg and a uniform distribution of microspheres throughout the liver volume. The actual dose delivered may be calculated by using the following equation: Aglass = (D • M)/(50 Gy • kg • GBq–1), where Aglass is the activity of the 90Y content in the glass microspheres (in gigabecquerels), D is the nominal target dose (in grays), and M is the liver mass (in kilograms).
The calculation of the dose to be delivered with the resin microsphere device is based on the assumption of a nonuniform distribution of the microspheres, with the degree of nonuniformity assumed to be dependent on the extent of tumor replacement of the liver volume. Two methods are currently used to calculate the actual dose with resin microspheres: the body surface area method and the empirical method. The empirical method, which is the one that was used in the pivotal clinical trial, is presented in Table 2a. (Depending on the magnitude of hepatopulmonary shunting, the dose may be decreased as indicated in Table 2b.) Alternatively, the dose may be based on the body surface area and the percentage of tumor involvement of the liver. The body surface area (BSA) is calculated in square meters as BSA = 0.20247 • h0.725 • w0.425, where h is height in meters and w is weight in kilograms. The percentage of tumor involvement of the liver (TI) is calculated as TI = (TV • 100)/(TV + LV), where TV is the volume of the tumor and LV is the volume of the liver. The dose, then, may be calculated by using the following equation: Aresin = (BSA – 0.2) + (TI/100), where Aresin is the activity of the 90Y content of the resin microspheres (in gigabecquerels).
|
Delivery of Microspheres
|
|---|
Before the microsphere devices were approved for use in the United States, the published data regarding the technical details of delivery were derived mostly from clinical trials in which hepatic arterial ports were used. In the United States, the delivery method of choice is currently via catheter inserted percutaneously with fluoroscopic guidance (Figs 6, 7). Infusion is performed in an angiography suite, primarily by an interventional radiologist who has experience with chemoembolization (the procedures are similar, although there are also important dissimilarities, which are outlined in Table 3).
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. Photograph shows the infusion set used with the glass microsphere device. The radioactive microspheres are suspended in normal saline in a vial that is shielded by a lead pig (curved black arrow) adjacent to a separate venting vial inside another lead pig. The suspension is administered with high-pressure infusion by using an inflation device (straight black arrow) to agitate and disperse the microspheres via the efferent tubing into the catheter (white arrow).
|
|
View larger version (87K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7. Photograph shows the infusion set used with the resin microsphere device. The radioactive microspheres are suspended in sterile water inside a vial (curved black arrow) that is housed in a shielded container (straight black arrow). A three-way stopcock (white arrow) allows sequential infusion of the 90Y microspheres and contrast material injection for monitoring the progress of infusion.
|
|
The catheter is usually positioned in essentially the same location as that used at arteriography for therapy planning. The experience gained in the delivery of the glass microsphere device after its initial approval was later applied in the delivery of the resin microsphere device. A critical aspect of this experience was a perceived increase in gastrointestinal toxic effects and a resultant need for empirical embolization of extrahepatic vessels before delivery of a single whole-liver infusion of microspheres, usually via the common or proper hepatic artery. This development led in turn to the adoption of a more conservative approach to treatment of bilobar disease, whereby each lobe is treated separately, with infusion via the right hepatic artery separated by a period of 4 weeks from infusion via the left hepatic artery (28). The use of a single whole-liver infusion is becoming more common, however, as familiarity with microsphere-based therapy increases. Because the resin microspheres have an embolic tendency, there is a risk of reflux during the last stage of the infusion. Therefore, embolization of the extrahepatic arteries leading to the gastroduodenal region (ie, the gastroduodenal artery and right gastric artery) is strongly advised, regardless of findings on 99mTc-labeled MAA images.
|
Follow-up Imaging of Bremsstrahlung
|
|---|
Images of bremsstrahlung—broad-spectrum secondary gamma-ray emissions produced as a result of the interaction of the high-energy beta emissions with tissue—may be obtained to enable detection of suspected extrahepatic deposition or to elucidate the distribution of microspheres in the liver (29). For this purpose, planar scintigraphy and/or single photon emission computed tomography (SPECT) is performed within 30 hours after infusion therapy, by using a dual-head gamma camera equipped with medium-energy collimators (Fig 8).
View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8. Bremsstrahlung imaging helps confirm successful targeting of hepatic metastases from pancreatic islet cell tumor in a 54-year-old man. Representative axial, coronal, and sagittal CT images (first column), SPECT images (second column), and SPECT/CT fusion images (third column) obtained with a dual-modality imaging system (Hawkeye; GE Medical Systems, Milwaukee, Wis) show selective activity in the right hepatic lobe approximately 24 hours after intraarterial infusion of 90Y-bearing resin microspheres at a dose of 55 mCi (2035 MBq).
|
|
|
Clinical Outcomes
|
|---|
Glass Microsphere Device
The initial development of the glass microsphere device and its application for treatment of hepatocellular carcinoma largely took place in Canada. The results of safety trials (11,29) demonstrated that treatment with doses up to 100 Gy was well tolerated. On the basis of these encouraging results, a larger phase II trial was conducted with a planned dose of 100 Gy (12). Outcomes in the phase II trial indicated a difference in survival that favored those who received larger doses (635 days with a dose of >104 Gy vs 323 days with a dose of <104 Gy), with a 20% response rate for the entire group. The results of subsequent studies performed in the United States have shown that this treatment is safe even in patients with portal vein thrombosis (22). In a recent large pooled analysis of 80 patients who received absorbed radiation doses of 47–270 Gy, a favorable median duration of survival was found for patients with Okuda I disease (628 days) and for those with Okuda II disease (384 days) (30). These survival times are similar to those reported for chemoembolization.
The 90Y-bearing glass microsphere device also has been used in the treatment of metastatic colorectal cancer. In a phase I dose escalation trial, five of 12 patients who received a dose of as much as 100 Gy experienced no progression of disease for a period of 7.5 months (31). In two separate studies, after the administration of a dose of as much as 150 Gy, more than 50% of patients experienced either a partial response or no progression of disease (17). These included patients in whom disease did not respond to chemotherapy (32).
Resin Microsphere Device
Most clinical trials of the resin microsphere device and its use in treatment of hepatocellular carcinoma were performed in Hong Kong. In a phase II trial that included 18 patients with inoperable hepatocellular carcinoma (33), 
-fetoprotein levels returned to baseline in 10 patients, and a partial response was seen in seven others who had received a dose of more than 120 Gy. The median duration of survival in this group was 35 weeks, but the patients who had received more than 120 Gy had a median survival of 55 weeks. It is noteworthy that a similar relationship between dose response and survival was observed in a separate study of the glass microsphere device (12). In an efficacy trial that included 71 patients who were predominantly hepatitis B carriers with cirrhosis, 19 (26.7%) partial responses were observed, and the tumors in four patients became resectable (14).
Clinical trials of the resin microsphere device for treatment of colorectal cancer took place mainly in Australia and New Zealand. In these trials, the device was found efficacious when used alone or in combination with prevailing chemo-therapeutic agents as both a first- and a second-line therapy (34–39).
The device was approved by the FDA in March 2002 on the basis of outcomes in a phase III trial in 71 patients, in which treatment with intraarterial floxuridine alone was compared with combined treatment with intraarterial floxuridine and the 90Y-bearing resin microspheres. There was a significant difference in both the response rate and the median time to disease progression in the group that received the combined therapy (9.7 vs 15.9 months, P = .001) (36). Since then, the results of additional clinical trials have been published, including those of a randomized phase II trial in which outcomes after chemotherapy alone (with fluorouracil and leucovorin) were compared with those after chemotherapy combined with brachytherapy. The results showed significant differences, in favor of the combined therapy arm, in three clinically important indicators: response rate, time to disease progression (18.6 vs 3.6 months), and survival benefit (29.4 vs 12.8 months) (35) (Fig 9). Favorable outcomes with use of the resin microsphere device also have been observed in phase I dose escalation studies with irinotecan and with oxaliplatin in combination with fluorouracil and leucovorin. It is noteworthy that a partial response was demonstrated in 55%–90% of patients and that the incidence of toxic effects was within reasonable limits (40,41).
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Complete response to brachytherapy with use of the resin microsphere device as an adjunct to systemic chemotherapy. Axial contrast-enhanced CT scans show numerous low-attenuation hepatic metastases before treatment (a) and absence of metastases 15 months after treatment (b). (Courtesy of Bruce Gray, MD, Sirtex Medical, Australia.)
|
|
View larger version (131K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Complete response to brachytherapy with use of the resin microsphere device as an adjunct to systemic chemotherapy. Axial contrast-enhanced CT scans show numerous low-attenuation hepatic metastases before treatment (a) and absence of metastases 15 months after treatment (b). (Courtesy of Bruce Gray, MD, Sirtex Medical, Australia.)
|
|
|
Imaging Findings
|
|---|
The most common change in the CT appearance of the liver after irradiation with 90Y is decreased attenuation in the affected hepatic areas. This and other physiologic changes are thought to represent liver edema, congestion, and microinfarction. Changes are more noticeable on the scans obtained soon after therapy than on those obtained later, a fact that suggests that such changes are reversible.
The decrease in attenuation on CT scans obtained after 90Y therapy is similar to that observed on scans obtained after external-beam irradiation. On scans of 90Y-treated livers that received an absorbed radiation dose of 100 Gy or less, the low-attenuation areas were heterogeneous. However, on scans of livers that received a dose of 125 Gy or more, the changes were diffuse. These changes were seen at 8 weeks after radiation therapy and had diminished at 16-week follow-up (42). It is most important that these changes not be mistaken for evidence of recurrent disease (Fig 10). The possibility of making that mistake should be kept in mind, particularly during the interpretation of follow-up images in patients who have undergone therapy at low doses, in whom decreased liver attenuation may be heterogeneous.
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Radiation-induced change in appearance of liver parenchyma on axial contrast-enhanced CT images in a 49-year-old woman who underwent treatment with the glass microsphere device for metastatic neuroendocrine cancer. (a) Image obtained before treatment shows low-attenuation regions consistent with metastases to the liver. (b) Image obtained 3 months after treatment shows areas with low attenuation (arrows) in the liver parenchyma adjacent to the metastases. Low attenuation in this case was treatment related but could be erroneously interpreted as evidence of disease progression. (Courtesy of Barry Daly, MD, University of Maryland, Baltimore, Md.)
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Radiation-induced change in appearance of liver parenchyma on axial contrast-enhanced CT images in a 49-year-old woman who underwent treatment with the glass microsphere device for metastatic neuroendocrine cancer. (a) Image obtained before treatment shows low-attenuation regions consistent with metastases to the liver. (b) Image obtained 3 months after treatment shows areas with low attenuation (arrows) in the liver parenchyma adjacent to the metastases. Low attenuation in this case was treatment related but could be erroneously interpreted as evidence of disease progression. (Courtesy of Barry Daly, MD, University of Maryland, Baltimore, Md.)
|
|
Effects on other organs may include thickening of the duodenum, stomach, and gallbladder wall, all of which have been observed in patients who underwent treatment with 90Y microspheres. The effects of 90Y therapy on liver metastases have been compared at CT, MR, and positron emission tomography (PET) in small study populations. PET images can show decreased metabolic activity, a finding that indicates treatment response (Fig 11) but that often conflicts with findings on CT and MR images.
View larger version (105K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Representative coronal images from two fluorine 18 fluorodeoxyglucose whole-body PET examinations performed 6 months apart in a 55-year-old man who was undergoing systemic chemotherapy for metastatic colorectal cancer. (a) Initial scan shows extensive hepatic metastases. (b) Follow-up scan obtained after liver-directed therapy with the resin microsphere device shows a marked decrease in metabolic activity in the metastatic hepatic lesions but increased metabolic activity at sites in the sternum, lungs, and left side of the groin. Local tumor control was achieved, but systemic therapy failed.
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Representative coronal images from two fluorine 18 fluorodeoxyglucose whole-body PET examinations performed 6 months apart in a 55-year-old man who was undergoing systemic chemotherapy for metastatic colorectal cancer. (a) Initial scan shows extensive hepatic metastases. (b) Follow-up scan obtained after liver-directed therapy with the resin microsphere device shows a marked decrease in metabolic activity in the metastatic hepatic lesions but increased metabolic activity at sites in the sternum, lungs, and left side of the groin. Local tumor control was achieved, but systemic therapy failed.
|
|
|
Complications and Toxic Effects
|
|---|
Overall, the incidence of complications after 90Y microsphere therapy, if patients are selected appropriately and target delivery is performed meticulously, is low. Gastroduodenal complications occur in less than 5% of those treated and are largely preventable. Pancytopenia as a result of bone marrow suppression from leaching of 90Y was reported after use of the earliest microsphere device (43). The device subsequently underwent multiple iterations, and this complication has not been reported since that time. A life-threatening complication, progressive pulmonary insufficiency secondary to radiation pneumonitis, can be avoided by avoiding the use of 90Y therapy in any patient with marked hepatopulmonary shunting (27). More than 1500 doses of 90Y-bearing microspheres of both types have been administered in the United States to date, and there have been no reported occurrences of this complication. The gallbladder also may receive radioactive microspheres through a patent cystic artery; a characteristic thick-walled appearance of the gall-bladder is observed on cross-sectional images in such cases. For avoidance of this complication, infusion distal to the cystic artery is preferred; however, the risk of cholecystitis requiring cholecystectomy is low (44) (Fig 12).
View larger version (156K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12. Cholecystitis and perforated duodenal ulcer in a 43-year-old man who underwent four treatments with the glass microsphere device for hepatic metastasis from colorectal cancer. Photomicrograph of a gallbladder specimen (hematoxylineosin stain; original magnification, x40) obtained at cholecystectomy performed during exploratory laparotomy shows glass microspheres (arrows) in the Rokitansky-Aschoff sinuses, with associated inflammation in the gallbladder wall and ulcerated duodenum.
|
|
Gastric and duodenal ulceration have been reported to occur after the use of 90Y microspheres and are related to the inadvertent intestinal deposition of microspheres via extrahepatic visceral arterial branches. Even in the absence of extrahepatic activity on 99mTc-labeled MAA and bremsstrahlung emission images, gastrointestinal symptoms have been reported to develop. The risk of ulceration can be minimized with empirical embolic occlusion of the visceral arteries before infusion of the microsphere device (45) (Fig 13 ). Similar gastrointestinal complications have been observed after treatment with transcatheter arterial chemoembolization (46) and hepatic artery pump infusion (47).
View larger version (143K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. Gastric ulcer in a 67-year-old man who underwent treatment with the resin microsphere device for metastatic colorectal cancer. Ulceration occurred despite prior coil embolization of the gastroduodenal artery. (a, b) Images obtained at common hepatic arteriography show a prominent gastroduodenal artery (black arrow) and a right gastric artery (white arrow) that originates from the proximal portion of the left hepatic artery (black arrowhead), a common anatomic variant. Because of the variant anatomy, infusion via the proper hepatic artery resulted in resin microsphere deposition and consequent ulceration in the stomach. (c) Photomicrograph (hematoxylineosin stain; original magnification, x200) of a gastric biopsy specimen shows staining of resin microspheres (arrows) in the submucosa, with associated eosinophil infiltration.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. Gastric ulcer in a 67-year-old man who underwent treatment with the resin microsphere device for metastatic colorectal cancer. Ulceration occurred despite prior coil embolization of the gastroduodenal artery. (a, b) Images obtained at common hepatic arteriography show a prominent gastroduodenal artery (black arrow) and a right gastric artery (white arrow) that originates from the proximal portion of the left hepatic artery (black arrowhead), a common anatomic variant. Because of the variant anatomy, infusion via the proper hepatic artery resulted in resin microsphere deposition and consequent ulceration in the stomach. (c) Photomicrograph (hematoxylineosin stain; original magnification, x200) of a gastric biopsy specimen shows staining of resin microspheres (arrows) in the submucosa, with associated eosinophil infiltration.
|
|
View larger version (176K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13c. Gastric ulcer in a 67-year-old man who underwent treatment with the resin microsphere device for metastatic colorectal cancer. Ulceration occurred despite prior coil embolization of the gastroduodenal artery. (a, b) Images obtained at common hepatic arteriography show a prominent gastroduodenal artery (black arrow) and a right gastric artery (white arrow) that originates from the proximal portion of the left hepatic artery (black arrowhead), a common anatomic variant. Because of the variant anatomy, infusion via the proper hepatic artery resulted in resin microsphere deposition and consequent ulceration in the stomach. (c) Photomicrograph (hematoxylineosin stain; original magnification, x200) of a gastric biopsy specimen shows staining of resin microspheres (arrows) in the submucosa, with associated eosinophil infiltration.
|
|
When 90Y microspheres are administered via a hepatic arterial port, patients undergo an exploratory laparotomy that includes a cholecystectomy, with ligation of all vessels that pass from the hepatic artery to the pancreas, stomach, and duodenum and that are depicted on images. Isolation of the liver may be confirmed intraoperatively with an injection of 3–5 mL of methylene blue through the port. A lesser-curvature omentopexy assists in shielding the stomach from the effects of hepatic radiation. When percutaneous transfemoral catheter administration is performed, embolization of extrahepatic vessels that originate from the hepatic arteries, or balloon occlusion (48), may help achieve the same objective.
Radiation-induced liver disease is a rare complication of 90Y microsphere treatment. It results in various degrees of hepatic decompensation and is indistinguishable from hepatic veno-occlusive disease. Radiation-induced liver disease is manifested clinically by the development of anicteric ascites. High doses of corticosteroids traditionally are administered in an attempt to decrease intra-hepatic inflammation. Treatment results are variable and mostly nongratifying, as the condition progresses in some patients to hepatic insufficiency of various degrees. When the resin microsphere device is used, according to the manufacturer, concomitant chemotherapy with capecitabine should be avoided because of the increased risk of liver disease. Findings in liver biopsy specimens from patients with radiation-induced liver injury are typically few and nonspecific; the clinical status of the patient is typically disproportionately worse than the pathologic findings (Fig 14).
View larger version (123K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14a. Radiation-induced liver disease in a 44-year-old woman who underwent treatment for bilobar cholangiocarcinoma with 90Y-bearing resin microspheres at a total radiation dose of 50.1 mCi (1853.7 MBq). Although the blood level of CA 19–9 antigen decreased from 980.2 to 408.2 U/mL after treatment, the patient developed worsening symptoms of hepatic decompensation. (a) Axial contrast-enhanced CT scan of the liver, obtained 4 weeks after treatment, shows ascites (arrow). (b) Photomicrograph (hematoxylineosin stain; original magnification, x200) of a liver biopsy specimen obtained for histopathologic analysis depicts hepatocyte swelling, mild microvesicular steatosis (black arrows), and a resin microsphere (white arrow) without any associated inflammatory response.
|
|
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14b. Radiation-induced liver disease in a 44-year-old woman who underwent treatment for bilobar cholangiocarcinoma with 90Y-bearing resin microspheres at a total radiation dose of 50.1 mCi (1853.7 MBq). Although the blood level of CA 19–9 antigen decreased from 980.2 to 408.2 U/mL after treatment, the patient developed worsening symptoms of hepatic decompensation. (a) Axial contrast-enhanced CT scan of the liver, obtained 4 weeks after treatment, shows ascites (arrow). (b) Photomicrograph (hematoxylineosin stain; original magnification, x200) of a liver biopsy specimen obtained for histopathologic analysis depicts hepatocyte swelling, mild microvesicular steatosis (black arrows), and a resin microsphere (white arrow) without any associated inflammatory response.
|
|
|
Conclusions
|
|---|
The initial results of liver-directed therapy with 90Y microspheres appear very promising and include significant improvements in response rate, in the interval to disease progression, and in survival. Accumulated experience with these devices has allowed treating physicians to develop methods for avoiding or mitigating complications and toxic effects and for delivering therapy on an out-patient basis. Further experience will help to determine whether the use of 90Y microspheres provides therapeutic benefits beyond those achievable with other currently available therapies.
|
Footnotes
|
|---|
Abbreviations: FDA = Food and Drug Administration, MAA = macroaggregated albumin
2 Current address: Department of Radiology, Division of Vascular and Interventional Radiology, University of Texas Southwestern Medical Center, Dallas, Tex. 
|
References
|
|---|
- Gilbert HA. Metastasis: incidence, detection and evaluation without histological confirmation. In: Weiss L, ed. Fundamental aspects of metastasis. New York, NY: American Elsevier, 1976; 385–405.
- Jemal A, Murray T, Ward E, et al. Cancer statistics, 2005. CA Cancer J Clin
2005;55:10–30.[Abstract/Free Full Text]
- Chorost MI, Datta R, Santiago RC, et al. Colon cancer screening: where have we come from and where do we go? J Surg Oncol
2004;85:7–13.[CrossRef][Medline]
- El-Serag HB, Mason AC, Key C. Trends in survival of patients with hepatocellular carcinoma between 1977 and 1996 in the United States. Hepatology
2001;33:62–65.[CrossRef][Medline]
- Carr BI. Hepatic artery chemoembolization for advanced stage HCC: experience of 650 patients. Hepatogastroenterology
2002;49:79–86.[Medline]
- Breedis C, Young G. The blood supply of neoplasms in the liver. Am J Pathol
1954;30:969–977.
- Ariel IM. Treatment of inoperable primary and metastatic cancer of the lung by internal irradiation. Surg Clin North Am
1967;47:1483–1505.[Medline]
- Ariel IM, Pack GT. Treatment of inoperable cancer of the liver by intra-arterial radioactive isotopes and chemotherapy. Cancer
1967;20:793–804.[CrossRef][Medline]
- Blanchard RJ, Grotenhuis I, Lafave JW, Frye CW, Perry JF Jr. Treatment of experimental tumors: utilization of radioactive microspheres. Arch Surg
1964;89:406–410.
- Grady ED. Intrahepatic arterial 90-yttrium resin spheres to treat liver cancer. Int J Nucl Med Biol
1978;5:253–254.
Top
Abstract
LEARNING OBJECTIVES FOR TEST...
