Radioembolization with Use of Yttrium-90 Resin Microspheres in Patients with Hepatocellular Carcinoma and Portal Vein Thrombosis

Transcription

1 Radioembolization with Use of Yttrium-90 Resin Microspheres in Patients with Hepatocellular Carcinoma and Portal Vein Thrombosis Mercedes Iñarrairaegui, MD, PhD, Kenneth G. Thurston, MA, Jose I. Bilbao, MD, PhD, Delia D Avola, MD, Macarena Rodriguez, MD, Javier Arbizu, MD, PhD, Antonio Martinez-Cuesta, MD, MSc, FRCR, and Bruno Sangro, MD, PhD PURPOSE: Intraarterial delivery of yttrium-90 ( 90 Y) bound microspheres (ie, radioembolization) is a promising treatment for hepatocellular carcinoma (HCC). An early concern was the embolic nature of the microspheres, and their potential to reduce hepatic arterial blood flow in patients with compromised portal blood flow secondary to portal vein thrombosis/occlusion (PVT). In this situation, the risk of liver failure could be enhanced, particularly in patients with cirrhosis who have increased hepatic arterial blood flow. This retrospective analysis was undertaken to assess the safety and clinical benefits of radioembolization with 90 Y resin microspheres in HCC with branch or main PVT. MATERIALS AND METHODS: A total of 25 patients presenting with unresectable HCC and compromised portal flow received segmental, lobar, or whole-liver infusion of 90 Y resin microspheres. For the analysis of tumor response, changes in target lesions, appearance of new lesions, and changes in portal vein thrombus were studied. Controlled disease was defined by absence of progression in all these components. RESULTS: Globally, controlled disease was achieved in 66.7% of patients at 2 months and 50% of patients at 6 months. No significant changes were observed in liver-related toxicities according to Common Toxicity Criteria (version 3.0) at 1 and 2 months after treatment. Median survival time was 10 months (95% CI, months). CONCLUSIONS: Radioembolization of unresectable HCC and branch or main PVT with 90 Y resin microspheres was associated with minimal toxicity and a favorable median survival time. Further prospective studies are warranted to validate the findings in this clinically challenging patient population. J Vasc Interv Radiol 2010; 21: Abbreviations: HCC hepatocellular carcinoma, MAA macroaggregated albumin, PVT portal vein thrombosis/occlusion From the Liver Unit (M.I., D.D., B.S.) and Departments of Interventional Radiology (J.I.B., A.M.C.) and Nuclear Medicine (M.R., J.A.), Clinica Universitaria de Navarra, Pamplona;Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD) (M.I., D.D., B.S.), Barcelona, Spain; and Sirtex Medical United States Holdings (K.G.T.), Wilmington, Massachusetts. Received September 22, 2009; final revision received February 22, 2010; accepted April 5, Address correspondence to B.S, Liver Unit, Clinica Universitaria de Navarra, Avda. Pio XII, Pamplona, Spain; bsangro@unav.es RADIOEMBOLIZATION with yttrium- 90 ( 90 Y) resin microspheres is an emerging treatment for unresectable liver disease in patients who are not amenable to Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD) (M.I., D.D., B.S.) is funded by Instituto de Salud Carlos III. K.G.T. is a paid employee of Sirtex Medical United States Holdings (Wilmington, Massachusetts). None of the other authors have identified a conflict of interest. SIR, 2010 DOI: /j.jvir liver transplantation, resection, or selective ablative techniques (eg, radiofrequency ablation or percutaneous ethanol injection) (1,2). The microspheres loaded with 90 Y are delivered to liver tumors by means of endovascular catheters selectively placed within the hepatic arterial vasculature. The microspheres lodge preferentially within the neovessels of the tumor(s) and deliver high-energy -radiation over a limited range (mean penetration of radiation into tissues is 2.4 mm), thereby minimizing radiation exposure to normal liver parenchyma. Patients suitable for radioembolization are those with good per- 1205

2 1206 Radioembolization with 90 Y Microspheres in HCC and PV Thrombosis August 2010 JVIR formance status (Eastern Cooperative Oncology Group score 0 2), adequate hepatic functional reserve (total bilirubin 2.0 mg/dl, albumin 3.0 mg/ dl, platelet count ) and acceptable renal function (creatinine 2.0 mg/dl) (3). With meticulous angiographic technique and conservative models for dose calculation, patients experience minimal side effects (mild abdominal pain, nausea, and fatigue are most common) and are typically discharged from the hospital within 24 hours of treatment. Liver damage is infrequent, and in the absence of cirrhosis, it seems to be associated with bilobar treatment of patients heavily pretreated with antineoplastic chemotherapy (4). Median survival of patients with hepatocellular carcinoma (HCC) who are treated with 90 Y microspheres varies widely (between 5 and 27 months) in phase II and retrospective studies, depending on performance status, degree of hepatic functional reserve, extent of disease involvement, and presence or absence of cirrhosis (3 11). When radioembolization with 90 Y microspheres was first introduced into the clinic for the treatment of unresectable HCC, there was some concern regarding the potential for liver failure in patients presenting with portal vein thrombosis/occlusion (PVT). Portal vein thrombus arises as a result of the vascular invasion of primary tumor or as a consequence of sluggish blood flow. A variable but significant proportion of patients with HCC (range, 15% 40%) present with PVT (12,13), and the actuarial probabilities of developing vascular invasion at 1, 2, and 3 years are 21%, 29%, and 46%, respectively, in patients with cirrhosis and unresectable HCC (12). These patients have a poor prognosis (14), and locoregional therapies including transarterial chemoembolization or bland embolization are usually contraindicated (15). In radioembolization treatment, although most microspheres are ultimately embedded in the small vessels of the periphery of tumor nodules, aggregation and occlusion of medium-sized vessels can also occur, resulting in reduced arterial blood flow (16). Therefore, although the principal mode of action for radioactive microspheres relies on irradiation, a combined radioembolic effect cannot be ruled out. Medium-sized vessel occlusion can also occur in the nontumoral liver parenchyma regardless of the presence of cirrhosis (4). Hence, in patients presenting with PVT, there is a concern of an increased risk of liver failure as a result of the preexistent reduction of portal flow and the incremental compromise of arterial hepatic perfusion caused by the embolization of microspheres in the hepatic micro- and macrovessels. As a result, 90 Y microsphere therapy has historically been contraindicated in patients presenting with PVT (17,18). Recent studies with 90 Y glass microspheres have suggested that the original safety concerns regarding risk of hepatic compromise in patients presenting with PVT may have been unfounded (19). An intense embolizing effect of glass microspheres has been recently ruled out by analyzing the pattern of arterial blood flow after contrast agent power injection immediately after treatment (20). In addition, a phase II study in 37 patients with HCC presenting with PVT (21) showed a favorable toxicity profile after radioembolization with 90 Y glass microspheres. Based on the lower radiation activity per microsphere, a greater number of resin microspheres are needed for a standard treatment, and this may translate into a more intense embolizing effect, although this issue has not been previously analyzed to our knowledge. In animal models, 90 Y-free resin microspheres do not cause significant acute ischemia and the microspheres are usually incorporated into the vessel wall or excluded from the circulation after 2 months (22). However, evidence from the clinical setting is lacking. The current retrospective analysis was conducted to assess the safety regarding a hypothetical embolizing effect and potential clinical benefit of 90 Y resin microspheres (SIR-Spheres; Sirtex Medical, Bonn, Germany) in the treatment of patients with unresectable HCC who present with PVT. MATERIALS AND METHODS Patient Cohort Twenty-five patients with unresectable HCC presenting with branch or main PVT were treated between September 2003 and September 2009 at a single institution. Beginning in September 2003, all consecutive patients with HCC who were ineligible for radical treatments (surgery, liver transplantation, or percutaneous ablation) or chemoembolization as a result of the presence of PVT or extensive tumor burden were evaluated for radioembolization. Eligibility criteria for this study included (i) age of at least 18 years; (ii) preserved liver function with a Child- Pugh score no greater than 7 and no ascites; (iii) platelet count greater than ; (iv) no contraindication to angiography; and (v) signed informed consent. Thirty-two patients presented with branch or main PVT, seven of whom were not treated for the following reasons: intense mismatch between intrahepatic technetium Tc 99m macroaggregated albumin (MAA) distribution and liver lesions (local exclusion criterion; n 3), high lung shunting (n 2), massive intratumoral arterioportal shunt (n 1), and clinical deterioration before treatment (n 1). Treatment Protocol The protocol for 90 Y resin microsphere therapy is provided elsewhere (23). In summary, patients underwent a planning angiographic study to (i) identify vascular anatomy, HCC feeding vessels, aberrant vessels and extrahepatic collateral vessels feeding extrahepatic organs (especially the gastrointestinal tract), and the presence of intrahepatic or intratumoral arterioportal shunting; and (ii) evaluate the direction of portal blood flow (hepatopetal or hepatofugal) and, in case of PVT, the presence of bypassing blood flow through collateral vessels). Aberrant hepatic vessels and extrahepatic collaterals were coil-embolized to prevent the inadvertent misplacement of 90 Y resin microspheres into the gastrointestinal tract or pancreas. Technetium Tc 99m labeled MAA particles were then injected with the delivery catheter in the intended position for 90 Y resin microsphere infusion. All the patients underwent 99m Tc-MAA planar imaging and single photon emission computed tomography (CT) to assess pulmonary shunt and any unintended flow to other extrahepatic organs. Any patient who exhibited an intense mismatch between intrahepatic 99m Tc-MAA distribution and liver lesions as viewed on fusion imaging with single photon emission CT and CT or magnetic resonance (MR) imaging was ineligible for treatment. Dosimetry was calculated to maximize the therapeutic activity to tu-

3 Volume 21 Number 8 Iñarrairaegui et al 1207 morous tissue and minimize exposure of nontumoral parenchyma and lung tissue. Within 2 weeks of the planning angiography study, the prescribed activity of 90 Y resin microspheres was administered by placing the tip of the delivery catheter in the same anatomic position as that used for the 99m Tc-MAA injection. Treatment protocol was approved institutionally. Data Collection and Follow-up Ethics committee approval was obtained for data analysis. Medical records were reviewed, and the following data were collected at baseline: age, sex, presence of cirrhosis per histologic or clinical/imaging criteria, extent of portal vein occlusion (segmental, branch, or main) and liver disease involvement (left lobe, right lobe, both lobes), extent of hepatic involvement (number and size of nodules) via MR imaging or CT, laboratory parameters including complete blood cell count, liver function tests, serum creatinine, anti hepatitis C virus antibodies and hepatitis B surface antigen, Eastern Cooperative Oncology Group performance status, and Cancer of the Liver Italian Program score (24). To assess changes in liver function secondary to the embolizing effect of radioembolization, serum levels of bilirubin and prothrombin activity and the occurrence of ascites and hepatic encephalopathy were recorded at baseline and 1 and 2 months after treatment. The same parameters were also recorded at month 6 to evaluate long-term effects. Adverse events were classified and coded for severity according to Common Terminology Criteria for Adverse Events, version 3.0 (25). Tumor response following radioembolization was assessed at 2 and 6 months by MR imaging or CT scan. In the evaluation of tumor response, the parenchymal and intravascular components of disease were assessed. For the former, changes in size of target lesions were assessed according to Response Evaluation Criteria In Solid Tumors (26) in those patients with measurable disease, and the development of new lesions was assessed in all patients. For the latter, changes in the extent of PVT were evaluated and classified as partial response (ie, clearance or regression of tumor thrombus into a more distal portal vein segment), stable disease (ie, no changes), or progressive disease (ie, progression of tumor thrombus into a more proximal portal vein segment). Finally, a global response was established as progressive disease whenever a patient had tumor progression in any of the three individual parameters (measurable target lesions, new lesions, or tumor thrombus) and controlled disease in any other case. Survival was determined from the time of treatment until death. Patients lost to follow-up were censored at the date of their last known visit. Statistical Analysis Descriptive statistics for nominal, ordinal, and continuous variables, including frequency, median, and average, were used as appropriate. Proportions of patients with liver toxicities before and after treatment were compared by using the McNemar test. Survival from date of treatment was calculated with the Kaplan-Meier method. A P value lower than.05 was considered statistically significant. A commercial statistical software package (SPSS version ; SPSS, Chicago, Illinois) was used for data analysis. RESULTS Patient Characteristics and Treatment Table 1 presents the baseline characteristics of the 25 patients (median age, 66 years) with PVT who received 90 Y resin microsphere treatment. Median follow-up was 8 months (interquartile range, months), and only one patient was lost to follow-up 7 months after treatment. Most patients were men (88%) and presented with cirrhosis (92%) and bilobar disease (56%). Vessel occlusion was identified in the main portal vein in 24% of patients and in the right or left branches in 64% of patients. Two patients presented with segmental portal vein occlusion. In five patients, collateral vessels were embolized during the arteriographic evaluation. Adequate liver reserve was reflected by median serum bilirubin level of 1.15 mg/dl (interquartile range, mg/dl) and absence of ascites. All patients had Barcelona Clinic Liver Cancer stage C disease. Patients presented with a high tumor burden (median tumor volume, 241 ml; interquartile range, ml; median tumor involvement, 13.6%; interquartile range, 7.9% 40.4%), and 44% of patients had more than five nodules. The treatment parameters for the patient cohort are outlined in Table 2. Yttrium-90 resin microspheres were injected into the right or the left hepatic artery in 16 patients (64%), the tumorfeeding vessel in three patients (12%), and the right hepatic and segment IV arteries in three patients (12%). Wholeliver treatment was performed in three patients (12%): one patient received 90 Y resin microsphere infusion from the common hepatic artery and the other two received sequential infusions into the right and left hepatic branches with a 3-week and 8-week interval, respectively. The median activity administered was 1.8 GBq (range, GBq). All patients successfully received the whole dose prescribed, without stasis during infusion. This wide range in administered activity relies on the different tumor burden treated in each patient. The average pulmonary shunt fraction was 9% and no patient received an excessive (ie, 30 Gy) radiation dose to the lung. Tumor Response The clinical outcomes for the patients treated are presented in Table 3. One patient died from adhesive large bowel obstruction before initial assessment. Seven of the remaining 24 patients had nonmeasurable parenchymal target disease (ie, diffuse or massive tumors) that could not be evaluated for response. Globally, progressive disease was observed at 2 months in eight patients (33.3%). Sites of progression included target lesions (in one patient who had an internal hemorrhage in one of the nodules that was responsible for the increased size), new lesions (n 6), and tumor thrombus (n 1). Controlled disease was observed in 16 patients (66.7%), including a partial response of the target lesion in one patient. Among the 16 patients who reached 6-month follow-up and were evaluable for response, eight (50%) had progressive disease. Sites of progression included new lesions (n 7) and tumor thrombus (n 1). Controlled disease was observed in eight patients (50%), including a partial response of the target lesions in two patients. Among the 19 patients with branch and segmental PVT, one patient showed a complete resolution of portal vein in-

4 1208 Radioembolization with 90 Y Microspheres in HCC and PV Thrombosis August 2010 JVIR Table 1 Baseline Characteristics of Study Patients (N 25) Characteristic Value Median (range) age (y) 66 (46 75) Male sex 22 (88) Portal vein occlusion Right PVT 10 Left PVT 6 Segmental PVT 2 Right and left PVT 1 Main PVT 2 Main and right PVT 2 Main, right, and left PVT 2 CLIP tumor stage 1 3 (12) 2 10 (40) 3 10 (40) 4 2 (8) Cirrhosis 23 (92) No. of nodules (56) 5 11 (44) Median (range) tumor volume (ml) 241 (18 2,000) Median (range) tumor involvement* 13.6 ( ) Median (range) total bilirubin (mg/dl) 1.15 ( ) Note. Values in parentheses are percentages unless specified otherwise. CLIP Cancer of the Liver Italian Program. * Calculated as tumor volume divided by total liver volume. Table 2 Characteristics of Treatment Characteristic Value Extent of treatment* Selective 3 (12) Lobar 16 (64) Extended lobar 3 (12) Whole liver 3 (12) Median (range) activity administered (GBq) 1.8 ( ) Median (range) pulmonary shunt fraction (%) 9 (2 32.4) Note. Values in parentheses are percentages unless specified otherwise. * Selective treatment involved infusion of microspheres via the feeding vessel to the tumor; lobar treatment included infusion to a single lobe via the right or left hepatic arteries; extended lobar included infusion from the right hepatic and segment IV arteries; whole-liver treatment was accomplished via (i) infusion of microspheres through the proper hepatic artery or (ii) double infusion via the right and left branches beyond the proper hepatic artery. vasion 4 months after treatment, 13 had not shown a progression to complete PVT after a median of 8 months (range, 1 26 months), two patients showed progression to complete PVT 8 months and 3 years after treatment, respectively, and one patient with left PVT showed progression to segmental right PVT 2 months after treatment. Additionally, two patients with right PVT developed new lesions with vascular invasion in the left lobe 2 and 17 months after treatment, respectively. Side Effects and Long-term Outcome Although some patients experienced mild pain during the infusion of 90 Y resin microspheres, there was no evidence of postembolization syndrome. There was also no evidence of altered liver function as a result of sustained reduction in vascular supply to nontumorous liver or acute ischemic hepatitis caused by severe, immediate reduction in vascular supply. As shown in Table 4, relevant changes in liver function were not observed in the first 2 months after radioembolization. One patient developed hepatic encephalopathy because of an adhesive large bowel obstruction that will be explained later in detail, two patients developed ascites 2 months after treatment, and no differences were observed in prothrombin activity. Regarding total bilirubin, although no differences were observed in the proportion of patients with abnormal levels according to Common Toxicity Criteria, a statistically significant but clinically irrelevant increase was observed after radioembolization. Median values were 1.15 mg/dl at baseline, 1.21 mg/dl at 1 month (P.01 vs baseline), and 1.4 mg/dl at 2 months (P.007 vs. baseline). Six months after treatment, when tumors had already progressed in some patients, median bilirubin level was 1.32 mg/dl, four patients developed grade 2 and 3 ascites, and two patients developed grade 3 encephalopathy. As illustrated in the Figure, median overall survival time was 10 months (95% CI, months). Eighteen patients died 2 44 months after radioembolization and, as of the time of manuscript preparation, seven patients are alive between 2 and 24 months after treatment. In 11 of the 18 patients who died, tumor progression was the suspected cause of death. In one of them, despite no major change in size, the tumor progressed toward the inferior cava vein and the patient developed ascites and renal failure. Among four patients with diffuse single-lobe tumors that were treated from the corresponding lobar artery, two died 8 months after treatment from tumor progression that was predominant in the contralateral lobe and two died 16 and 45 months after treatment, respectively, from hepatic and extrahepatic tumor progression after having received sorafenib, without response. PVT progressed in two of these four patients. Three patients with bilobar disease treated by whole-liver or extended lobar infusion died from tumor progression 14, 24, and 26 months after treatment, respectively. Three additional patients with bilobar involve-

5 Volume 21 Number 8 Iñarrairaegui et al 1209 Table 3 Treatment Outcomes Outcome 2 Months (n 24) 6 Months (n 16) Parenchymal component Target lesions* Controlled disease 16 (94) 9 (100) Progressive disease 1 (6) 0 New lesions 6 (24) 7 (43.7) Intravascular component Controlled disease 22 (91.6) 15 (93.7) Progressive disease 1 (4.1) 1 (6.3) Global response Controlled disease 16 (66.7) 8 (50) Progressive disease 8 (33.3) 8 (50) Note. Values in parentheses are percentages. RECIST Response Evaluation Criteria in Solid Tumors. Controlled disease means (i) partial response or stable disease according to RECIST for target measurable lesions in parenchymal component, (ii) absence of progression of tumor thrombus into a more proximal portal vein segment for intravascular component, and (iii) controlled disease in the parenchymal and intravascular components for global response. Progressive disease means absence of controlled disease. * Target lesions were evaluated according to RECIST only in patients with measurable disease (n 17 at month 2 and n 9 at month 6). Table 4 Liver-related Side Effects* Toxicity/CTC Grade Basal (N 25) Month 1 (N 25) Month 2 (n 24) Month 6 (n 16) Bilirubin Total 12 (48) 11 (51) 15 (64) 10 (61) 1 9 (36) 6 (28) 8 (34) 5 (31) 2 3 (12) 5 (23) 6 (26) 3 (18) (4) 2 (12) INR Total 1 (5) 1 (4) 1 (4) (5) 1 (4) 1 (4) Ascites Total (8) 4 (24) (8) 2 (12) (12) Encephalopathy Total 0 1 (4) 0 2 (12) (4) (12) Note. Values in parentheses are percentages. CTC Common Toxicity Criteria (version 3); INR International Normalized Ratio. * Differences between time points were not significant for any of the toxicities. Twenty-one patients were available for assessment of blood parameters at 1 month. Twenty-three patients were available for assessment of blood parameters at 2 months. ment that included massive lesions in one lobe and few lesions in the contralateral lobe were treated from the corresponding lobar artery. In all cases, the targeted disease remained stable but tumors progressed in the form of new lesions in the contralateral lobe (two patients also developed extrahepatic metastases) and patients died 6, 11, and 14 months after radioembolization. All six of these patients received sorafenib but showed no tumor response. In seven patients, the cause of death was not likely to be tumor progression. One patient developed an adhesive large bowel obstruction 10 days after treatment, underwent a subtotal colectomy that had a problematic course complicated with infection, cardiac arrhythmia, and hepatic encephalopathy, and finally died 2 months after treatment. A second patient developed obstructive jaundice after 3 months and died from variceal bleeding nearly 4 months after radioembolization. In a third patient who underwent selective treatment, the treated lesion remained stable but the patient developed biliary tract compression by hilar lymph node metastases and died of cholangitis in another hospital. A fourth patient treated from the right hepatic artery whose disease had been controlled despite a slightly increased bilirubin level of 3.2 mg/dl at 3 months died from an unknown cause at a different center 4 months after radioembolization. A fifth patient received treatment from the left artery and had stable disease as the better response in the treated area but showed disease progression in the right lobe and received a second radioembolization procedure 5 months later. By this second treatment, the patient s basal bilirubin level was 3.35 mg/dl, and 3 months later, worsening liver function and ascites were seen. The patient died 11 months after the first treatment (5 months after the second treatment). A sixth patient treated from a right segmental artery showed bilirubin levels increased to 3.17 mg/dl 2 months after radioembolization, and 1 month later the patient developed refractory ascites and died 8 months after treatment. The final patient in this group was treated from the right hepatic artery and the targeted lesion remained stable without progression, but 6 months after treatment, the patient developed ascites with worsening liver function and died 8 months after radioembolization. In the latter four patients, it was difficult to identify the likely cause of liver failure, but tumor progression was not considered a major contributor. Although cirrhosis itself is a progressive disease and decompensation is not uncommon,

6 1210 Radioembolization with 90 Y Microspheres in HCC and PV Thrombosis August 2010 JVIR Figure. Overall survival from the date of treatment with 90 Y radioembolization among 25 patients with HCC who presented with PVT. treatment-induced liver disease cannot be ruled out. At the time of evaluation, seven patients were alive and had good liver function. One patient with a large single tumor that replaced virtually the entire right lobe showed an intense response and received two additional radioembolization procedures 7 and 18 months after the first treatment. The patient continues to have a stable partial response 24 months after the first treatment. A second patient with a right-sided lesion treated from the right artery showed stable disease during 17 months, but then new lesions developed in the left lobe with thrombosis of the left hepatic artery, and the patient started sorafenib therapy. He was alive 19 months after treatment with preserved liver function. Another patient was lost to follow-up 7 months after radioembolization, and the last scan performed at 2 months showed progression of target lesions as a result of internal hemorrhage of one nodule. Another patient showed lung progression with stable disease in the liver 6 months after treatment. Three additional patients were alive and with good liver function 2 months after treatment (two with stable disease and one with a new portal invading lesion in the contralateral lobe). DISCUSSION The introduction of radioembolization for the treatment of patients with HCC and PVT has been tempered by concerns regarding the potential increased risk of liver failure, complications, and death as a result of impairment of hepatic vascular supply in the presence of compromised portal blood flow (18,19). These concerns have been driven by the previous experience with intense embolic therapies such as chemoembolization. Randomized studies involving chemoembolization have typically excluded patients with vascular invasion, including segmental portal obstruction, in view of the increased risk of compromised hepatic arterial flow (27). A randomized study of chemoembolization versus untreated controls found the relative risk of death for patients with unilobar (ie, branch) portal vein obstruction treated with chemoembolization was 2.71 (P.004) compared with patients without PVT (28). By contrast, the current study suggests that 90 Y resin microspheres may be used with a decreased risk of complication and may provide some clinical benefit in this challenging patient population. If there were an incremental effect of significant impairment of hepatic arterial blood flow in concert with preexisting portal flow compromise, one would expect three possible scenarios after radioembolization: (i) postembolization syndrome resulting from tumor ischemia; (ii) altered liver function resulting from minor but sustained reduction in vascular supply to nontumorous tissue; or (iii) acute ischemic hepatitis resulting from severe, immediate reduction in vascular supply. None of these clinical phenomena have been observed in these patients with PVT who were treated with 90 Y resin microspheres, although the slight increase in serum bilirubin may reflect a subclinical embolic effect. In 2004, Salem and colleagues (19) reported clinical outcomes after treatment with glass 90 Y microspheres in 15 patients with HCC presenting with branch PVT. Other than increased posttreatment bilirubin level in five patients (four of whom had evidence of intrahepatic disease progression), there were no significant clinical sequelae from treatment. A second phase II study (20) showed that patients with branch PVT (n 25) exhibited a similar incidence of increased bilirubin compared with those without PVT (n 71; 42% vs 35%, respectively). For patients presenting with main PVT, the incidence of increased bilirubin level was 64% higher after radioembolization; although this was not statistically significant (P.20) (20). Increased bilirubin level has been shown to be a good indicator of overall hepatic function. Nevertheless, our study showed only a mild increase at 1 and 2 months after treatment, which would suggest an absence of clinically relevant ischemia-induced effect. Indeed, this same phenomenon is observed in patients with PVT who are treated with 90 Y microspheres and could reflect a subtle effect of radiation. A study of treatment with 90 Y glass microspheres in 37 patients with PVT (21) indicated a posttreatment increase in bilirubin level (Southwest Oncology Group grade 3 5) in 32% of branch PVT cases (n 25) and 58% of main PVT cases (n 12). Clinically detectable posttreatment ascites (Southwest Oncology Group grade 2 4) was evident in 8% of branch PVT cases and 50% of main PVT cases. The authors attributed the higher incidence of liver-related events in the main PVT group to their more advanced disease stage and worsened prognosis (21). In the current study, controlled disease was observed in the majority of patients (66.7%) at 2 months and in 50% at 6 months. Notably, although main PVT usually leads to an increased risk of liver decompensation and ascites, only

7 Volume 21 Number 8 Iñarrairaegui et al 1211 two cases of branch PVT evolved to main PVT. The median survival time in the current study cohort was 10.0 months, which compares favorably with the median survival of 10.1 months for patients with branch PVT who are treated with glass microspheres, and exceeds the 4.5- month median survival time of patients with main PVT (21). The results of this study compare favorably with the median survival time of 2.7 months in untreated patients with HCC and PVT (14) and the survival times of 8.9 and 6.7 months in patients with PVT or extrahepatic disease and Child-Pugh class A disease treated with sorafenib or placebo, respectively, in a recent controlled study (29). One of the major drawbacks of the present study is its retrospective design. After 90 Y radioembolization, all patients did not have imaging procedures done at the same intervals, so time to progression could not be calculated. Systemic therapy with sorafenib was not started at the same time after 90 Y treatment in all patients who showed progressive disease, and this may have influenced overall survival. Finally, the small number of patients, especially in the main PVT group (n 6), did not allow the calculation of differences in survival between branch and main PVT cases. Based on the results of the present study, it appears that radioembolization with 90 Y resin microspheres offers a favorable risk/benefit profile for patients presenting with unresectable HCC and branch or main PVT. Unlike other embolic therapies such as chemoembolization, radioembolization appears to be an effective treatment for patients who otherwise have limited treatment options and present with a poor prognosis. Further prospective clinical studies are required to validate the findings from our small patient sample and to evaluate radioembolization with and without sorafenib compared with sorafenib alone. Acknowledgments: The authors acknowledge the contribution of Carmen Fuertes to the conduct of this study. References 1. 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 methodological considerations. J Vasc Interv Radiol 2006; 17: Salem R, Thurston KG. Radioembolization with 90yttrium microspheres: a state-of-the-art brachytherapy treatment for primary and secondary liver malignancies: part 3: comprehensive literature review and future direction. J Vasc Interv Radiol 2006; 17: Kennedy A, Nag S, Salem R, et al. Recommendations for radioembolization of hepatic malignancies using yttrium-90 microsphere brachytherapy: a consensus panel report from the Radioembolization Brachytherapy Oncology Consortium (REBOC). Int J Radiat Oncol Biol Phys 2007; 68: 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: Dancey JE, Shepherd FA, Paul K, et al. Treatment of nonresectable hepatocellular carcinoma with intrahepatic 90Y microspheres. J Nucl Med 2000; 41: Geschwind JF, Salem R, Carr BI, et al. Yttrium-90 microspheres for the treatment of hepatocellular carcinoma. Gastroenterology 2004; 127 (Suppl):S194 S Goin JE, Salem R, Carr BI, et al. Treatment of unresectable hepatocellular carcinoma with intrahepatic yttrium 90 microspheres: a risk-stratification analysis. J Vasc Interv Radiol 2005; 16: Salem R, Lewandowski RJ, Atassi B, et al. Treatment of unresectable hepatocellular carcinoma with use of 90Y microspheres: safety, tumor response and survival. J Vasc Interv Radiol 2005; 16: Lau WY, Ho S, Leung TWT, et al. Selective internal radiation therapy for nonresectable hepatocellular carcinoma with intrarterial infusion of 90yttrium microspheres. Int J Radiat Oncol Biol Phys 1998; 40: Salem R, Lewandowski RJ, Mulcahy MF, et al. Radioembolization for hepatocellular carcinoma using yttrium-90 microspheres: a comprehensive report of long-term outcomes. Gastroenterology 2010; 138: Iñarrairaegui M, Martinez-Cuesta A, Rodríguez M, et al. Analysis of prognostic factors after yttrium-90 radioembolization of advanced hepatocellular carcinoma. Int J Radiat Oncol Biol Phys 2010 (in press); /j.ijrobp Trevisani F, D Intino PE, Caraceni P, et al. Etiologic factors and clinical presentation of hepatocellular carcinoma. Differences between cirrhotic and noncirrhotic Italian patients. Cancer 1995; 7: Calvet X, Bruix J, Ginés P, et al. Prognostic factors of hepatocellular carcinoma in the west: a multivariate analysis in 206 patients. Hepatology 1990; 12: Llovet JM, Bustamante J, Castells A, et al. Natural history of untreated nonsurgical hepatocellular carcinoma: rationale for the design and evaluation of therapeutic trials. Hepatology 1999; 29: Bruix J, Sherman M. Practice Guidelines Committee, American Association for the Study of Liver Diseases. Management of hepatocellular carcinoma. Hepatology 2005; 42: Kennedy AS, Nutting C, Coldwell D, Gaiser J, Drachenberg C. Pathologic response and microdosimetry of 90Y microspheres in man: review of four explanted whole livers. Int J Radiat Oncol Phys 2004; 60: Salem R, Thurston KG. Radioembolization with 90yttrium microspheres: a state-of-the-art brachytherapy treatment for primary and secondary liver malignancies: Part 2: special topics. J Vasc Interv Radiol 2006; 17: Kennedy A, Nag S, Salem R, et al. Recommendations for radioembolization of hepatic malignancies using yttrium-90 microsphere brachytherapy: a consensus panel report from the radioembolization brachytherapy oncology consortium. Int J Radiat Oncol Biol Phys 2007; 68: Salem R, Lewandowski R, Roberts C, et al. Use of yttrium-90 glass microspheres (TheraSphere) for the treatment of unresectable hepatocellular carcinoma in patients with portal vein thrombosis. J Vasc Interv Radiol 2004; 15: Sato K, Lewandowski RJ, Bui JT, et al. Treatment of unresectable primary and metastatic liver cancer with yttrium-90 microspheres (TheraSphere): assessment of hepatic arterial embolization. Cardiovasc Intervent Radiol 2006; 29: Kulik L, Carr BI, Mulcahy MF, et al. Safety and efficacy of 90Y radiotherapy for hepatocellular carcinoma with and without portal vein thrombosis. Hepatology 2008; 47: Bilbao J, De Martino A, De Luis E, et al. Biocompatibility, inflammatory response and recanalization characteristics of non-radioactive resin microspheres: histological findings. Cardiovasc Intervent Radiol 2009; 32: Sangro B, Bilbao JI, Boan J, et al. Radioembolization using 90Y-resin microspheres for patients with advanced hepatocellular carcinoma. Int

8 1212 Radioembolization with 90 Y Microspheres in HCC and PV Thrombosis August 2010 JVIR J Radiat Oncol Biol Phys 2006; 66: Ueno S, Tanabe G, Sako K, et al. Discrimination value of the new western prognostic system (CLIP score) for hepatocellular carcinoma in 662 Japanese patients. Hepatology 2001; 34: Cancer Therapy Evaluation Program, Common Terminology Criteria for Adverse Events, version 3.0, March 31, Bethesda, MD: National Institutes of Health, Therasse P, Arbuck SG, Eisenhauer EA, et al. Special article: new guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000; 92: Llovet JM, Real MI, Montaña X, et al. Arterial embolization or chemoembolization versus symptomatic treatment in patients with unresectable carcinoma: a randomized controlled trial. Lancet 2002; 359: Lo CM, Ngan H, Tso WK, et al. Randomized controlled trial of transarterial lipiodol chemoembolization for unresectable hepatocellular carcinoma. Hepatology 2002; 35: Sherman M, Mazzaferro V, Amadori D, et al, on behalf of the SHARP investigators study group. Efficacy and safety of sorafenib in patients with advanced hepatocellular carcinoma and vascular invasion or extrahepatic spread: a subanalysis from the SHARP trial (abstr.). J Clin Oncol 2008; 26 (Suppl):4584.