Radioembolization With Selective Internal Radiation
Microspheres for Neuroendocrine Liver Metastases
Julie King, MPH1
Richard Quinn, MB, BSc2
Derek M. Glenn, MBBS3
Julia Janssen, BSc2
Denise Tong, MBBS1
Winston Liaw, MBBS4
David L. Morris, MD, PhD1
1Department of Surgery, University of New South
Wales, St. George Hospital, Sydney, New South
2Department Nuclear Medicine, University of
New South Wales, St. George Hospital, Sydney,
New South Wales, Australia.
3Department of Radiology, University of New
South Wales, St. George Hospital, Sydney, New
South Wales, Australia.
4Department of Oncology, University of New
South Wales, St. George Hospital, Sydney, New
South Wales, Australia.
BACKGROUND. There are limited effective treatment options available and a poor
5-year survival for patients with inoperable neuroendocrine liver metastases
(NETLMs). In this study, the authors prospectively assessed the safety and effi-
cacy of treatment with yttrium 90 (90Y) radioactive microspheres for patients
with unresectable NETLMs.
Spheres]) were administered through a temporarily placed percutaneous hepatic
90Y resin microspheres (selective internal radiation [SIR-
artery catheter concomitantly with a 7-day systemic infusion of 5-fluorouracil to
patients with progressive, unresectable NETLMs. Patients were monitored pro-
spectively, and the response to treatment was measured by using cancer markers
and tumor size on computed tomography imaging studies.
RESULTS. Thirty-four patients (22 men) with a mean age 61 years (range, 32-
79 years) who had unresectable NETLMs were treated between December 2003
and December 2005. The mean (?standard error) follow-up was 35.2 ? 3.2
months. The site of the primary neuroendocrine tumor was the bronchus in
1 patient, the medullary thyroid in 2 patients, gastrointestinal in 15 patients, the
pancreas in 8 patients, and of unknown origin in 8 patients. The tumors were
classified as vipoma (1 tumor), somatostatinoma (1 tumor), glucagonoma
(2 tumors), large cell (3 tumors), carcinoid (25 tumors), and of unknown origin
(2 tumors). Complications after90Y radioembolization included abdominal pain,
which was mild to severe; nausea and fever; and lethargy that lasted from 1 week
to 1 month. Two patients developed biopsy-proven radiation gastritis, 1 patient
developed a duodenal ulcer, and there was 1 early death from liver dysfunction
and pneumonia. Subjective changes from recorded baseline hormone symptoms
were reported every 3 months. Symptomatic responses were observed in 18 of
33 patients (55%) at 3 months and in 16 of 32 patients (50%) at 6 months. Radi-
ologic liver responses were observed in 50% of patients and included 6 (18%)
complete responses and 11 (32%) partial responses, and the mean overall survival
was 29.4 ? 3.4 months). In patients who had evaluable chromogranin A (CgA)
marker levels, there was a fall in CgA marker levels after90Y radioembolization in
19 patients (26%) at 1 month, in 19 patients (41%) at 3 months, in 15 patients
(43%) at 6 months, in 11 patients (42%) at 12 months, in 8 patients (38%) at
24 months, and in 3 patients (46%) at 30 months.
CONCLUSIONS. In this open study of 34 patients, the results demonstrated that
90Y resin microspheres can achieve relatively long-term
responses in some patients with nonresectable NETLMs. Cancer 2008;113:921–9.
? 2008 American Cancer Society.
KEYWORDS: selective internal radiation spheres, radioembolization, unresectable
neuroendocrine tumors, liver metastases, yttrium 90.
pancreatic neuroendocrine tumor (NET) has improved over the
lthough the recognition and treatment of primary gastroentero-
Sirtex Medical (Sydney, Australia) provided the
selective internal radiation spheres free of charge
for this study but provided no other financial sup-
port. Erin McKay, Senior Physicist, Nuclear Medi-
emission computed tomography and tomographic
imaging to analyze target-to-nontarget uptake
Address for reprints: David L. Morris, MD, PhD,
Department of Surgery, St. George Hospital, Uni-
versity of New South Wales, Sydney NSW 2217,
Australia; Fax: (011) 61-2-9350 3997; E-mail:
Received November 26, 2007; revision received
April 14, 2008; accepted April 15, 2008.
ª2008 American Cancer Society
Published online 10 July 2008 in Wiley InterScience (www.interscience.wiley.com).
last decades, liver
(NETLMs) are common.1NETLMs frequently are re-
sponsible for symptoms because of hormone secre-
tion, pressure on structures, or replacement of liver.
Resection and ablation can be associated with long-
term survival,2,3but the treatments seldom are cura-
tive, and the 5-year survival rate for patients who
have unresected LMs is between 25% and 50%.1Ab-
lative therapies (radiofrequency, laser therapy, or
cryotherapy) are limited to the small proportion of
patients with few tumors.4,5,6
achieves modest response rates of limited duration,
is better for pancreatic NETs compared with meta-
static carcinoid tumors,7and frequently is accompa-
nied by significant toxicity.8The Eastern Cooperative
Oncology Group’s randomized study of doxorubicin
with fluorouracil or streptozocin with fluorouracil
followed by dacarbazine at disease progression in
patients with metastatic carcinoid demonstrated an
8.2% response rate but significant treatment-related
toxicity.9Although chemoembolization can achieve a
response in up to 79% of patients, the effects usually
are of short duration, and a survival benefit has not
Patients with NETLMs can be affected variously
by carcinoid syndrome or by endocrine symptoms
that seriously can reduce quality of life. Somatostatin
analogues can ameliorate hormone symptoms, but it
is not clear that they have altered the survival of
patients with metastatic carcinoid.12
Radioembolization with yttrium 90 (90Y) has
been used for well over a decade to treat patients
with nonresectable LMs from primary and secondary
liver cancersand had
results.13-18Yttrium 90 is a pure b emitter that has
average range of penetration that is 5 mm with the
standard dose of 2 gigabecquerel (GBq) contained in
50 million resin microspheres. In 7 patients with
nonresectable hepatocellular carcinoma, Lau et al
reported a 50% tumor response in 27% of patients
who received90Y.15In patients with colorectal LMs, a
decrease in serum carcinoembryonic (CEA) levels
was achieved in >90% of patients, and 82% of
patients had some decrease in tumor size as meas-
ured by computed tomography (CT) imaging.13,14
It is recognized that NETLMs are hypervascular
tumors, and this characteristic has been used to
obtain therapeutic benefit with hepatic artery liga-
tion, intermittent occlusion, and embolization treat-
ments.20In this, we prospectively assessed the safety
and efficacy of90Y radioactive resin microspheres in
patients with unresectable NETLMs.
MATERIALS AND METHODS
Thirty-four patients with NETLMs were treated with
90Y microspheres (selective internal radiation spheres
[SIR-Spheres]; Sirtex, Sydney, Australia) between De-
cember 2003 and December 2005. All patients who
were treated were enrolled in a prospective study
with Human Research Ethics Committee approval
(South East Sydney Area Health Service Ethics Com-
mittee Approval No. 03/173; D.L.M.), and the study
was conducted at St. George Hospital, Sydney.
All patients who were considered for study entry
were current, single-institution patients with progres-
sive, unresectable NET. Inclusion criteria stipulated
that all patients had radiologically proven LMs from
primary neuroendocrine origin with NETLMs that
were not amenable to curative surgical resection; CT
evidence of demonstrated liver disease progression
within the last 6 months; a patent right or left portal
vein; adequate hematologic, renal, and hepatic func-
tion; no portal hypertension; prior surgery, ablation,
or chemotherapy forLMs
informed consent; a World Health Organization per-
formance status from 0 to 2; aged 18 years to
85 years; and an expected survival >3 months.
radioembolization was permitted for patients with
extrahepatic disease (EHD). All 34 patients who were
considered for entry fulfilled the study criteria for90Y
after clinical assessment by the chief investigator and
a multidisciplinary team.
Patients had a liver CT studies obtained and
were reviewed clinically within 10 days before
treatment. A hepatic angiogram with 99m techne-
MAA) scintigraph also was obtained to demonstrate
any aberrant hepatic anatomy, distribution of iso-
tope within the liver, and the percentage of pulmo-
nary shunting. High-percentage shunting can cause
<20% arteriovenous lung shunting on the
MAA scan and a tumor/normal liver uptake ratio
>2:1 were eligible. The tomographic regions were
determined as regions of interest around the tumor
compared with normal parenchyma, and the tumor
volume/normal liver area was assessed by the nu-
clear medicine department, which coregistered the
pre-90Y CT liver scan and the99mTc-MAA scan. The
90Y GBq dose was adjusted for tumor volume and
lung shunt fraction. The systemic 5-fluorouracil
dose was based on patient body surface area, and a
(225 mg/m2) was administered as a radiosensitizer
concurrently for 7 days through a peripherally
inserted central catheter the day before
922CANCERSeptember 1, 2008 / Volume 113 / Number 5
embolization. The baseline serum marker CgA level
The90Y resin microspheres (29-35 microns in di-
ameter with a half-life of 64 hours) were injected
over 10 minutes through a temporary hepatic artery
catheter that was placed percutaneously through the
femoral or brachial artery. Whole liver treatments
were administered for bilobar NETLMs with a di-
vided dose of 2 separate injections (two-thirds to the
right lobe and one-third to the left lobe of liver). The
catheter was placed into the right hepatic artery,
proximal to the junction with the gastroduodenal ar-
tery, and a two-thirds dose was infused over 5 min-
utes. Then, the catheter was withdrawn partially and
repositioned into the left hepatic artery, again avoid-
ing the gastroduodenal junction for infusion of the
remaining one-third90Y dose.
Aberrant vessels were not embolized before radio-
embolization. The dose of
(between 1.4 GBq and 2.6 GBq) was titrated to the
extent of disease (according to User Manual guide-
lines). All patients had prophylaxis against carcinoid
crisis. An H2 antagonist was taken for 1 month after
90Y with oral analgesia as required. All patients were
hospitalized overnight. Blood and biochemical levels
were obtained weekly for 4 weeks then every 3 months,
and tumor marker levels and CT scans were obtained
monthly and then every 3 months.
Response to treatment was reported as the
change from baseline measurement in tumor volume
Response Evaluation Criteria in Solid Tumors was
used to measure liver response and to assess overall
The distribution of
liver was imaged the day after treatment by measuring
the Bremsstrahlung radiation emitted as the90Y b rays
decelerated within the tissue. Single-photon emission
CT imaging was performed, and tomographic recon-
struction was achieved by using an iterative algo-
rithm22that incorporated corrections for attenuation
losses and resolution recovery. The reconstructed
images were analyzed by classifying voxels into
3 groups—liver, tumor, and other—and the total counts
in liver and tumor regions were measured. By assuming
that all of the administered activity was trapped in the
liver/tumor system, it was possible to calculate the ini-
tial activity distribution and, hence, to estimate the
radiation-absorbed dose to both liver and tumor by
using the method of voxel dose kernel convolution.23
90Y resin microspheres
after treatment. The
90Y microspheres within the
SPSS 14.0 for Windows software (SPSS Inc., Chicago,
Ill) was used for data management and statistical
analysis. A P value of .05 was considered significant
and was used with the Kaplan-Meier product method
and the log-rank test.
There were 34 patients (22 men), the mean patient
age was 61 years (range, 32-79 years), and the mean
follow-up (?standard error [SE]) after
spheres was 35.2 ? 3.2 months. The primary NET
site was the bronchus in 1 patient, the stomach in
1 patient, the medullary thyroid in 2 patients, the
small bowel in 11 patients, the colon/rectum in 3
patients, the pancreas in 8 patients, and of unknown
origin in 8 patients. The primary NETs were classified
as vipoma in 1 patient, somatostatinoma in 1 patient,
glucagonoma in 2 patients, large cell in 3 patients,
carcinoid in 25 patients, and unknown in 2 patients.
All patients had histologic evidence of NET, and prior
treatments included liver surgery in 10 patients, sys-
temic chemotherapy in 5 patients, and thyroid sur-
gery in 2 patients. Twenty-five patients were receiving
a somatostatin analogue at study entry.
The mean (?SE) interval from the diagnosis of pri-
mary NET to90Y treatment was 55.9 ? 11.8 months,
and the mean interval from diagnosis of NETLMs to
treatment with SIR therapy (SIRT) was 36.61 ? 6.72
months. At study entry, 20 patients (59%) had histolo-
gic or documented CT evidence of EHD (Tables 1, 2).
The estimated percentage shunting to the lungs on
99mTc-MAA scans was 6.3% (SE, ?0.7%). The mean
dose of90Y was 1.99 GBq (SE, ?0.6 GBq; range, 0.92-
2.80 GBq). One patient had a 40% reduced dose of SIR-
Spheres because of 17% lung shunting.
pain and vomiting, which were alleviated by intrave-
nous narcotics and antiemetics. All patients reported
some degree of mild-to-severe abdominal pain from
1 week to 1 month after treatment. Variable symp-
toms of nausea, lethargy, anorexia, and fever per-
sisted for 1 week to 1 month for all patients and
generally required nonopioid analgesia.
demonstrated on endoscopy/biopsy at 1 month,
2 months, and 6 months after treatment because of
malperfusion of90Y microspheres.24,25Ulceration was
self-limiting in 2 patients who remained well and
asymptomatic after90Y treatment with stable disease
(SD) at 44 months and with a complete response
(CR) at 27 months on CT imaging. The other patient,
whose gastritis persisted, died at 12 months with
90Y treatment, 4 patients experienced severe
Radioembolization With SIR Spheres/King et al923
progressive liver disease. In those 3 patients, we had
scans, but no aberrant anatomy was evident before
90Y treatment in the 2 patients who developed radia-
tion gastritis or in the 1 patient who had a duodenal
Two patients developed jaundice at 2 weeks that
resolved in 1 patient. The second patient, who died
at 1 month after90Y treatment, presented 7 months
previously with abdominal pain only. The patient
was diagnosed with large cell NET of unknown pri-
mary site, 50% bilobar hepatic metastases, ascites,
and a large portacaval lymph node (3.7 3 3.9 cm)
within gall bladder lumen. There were no adverse
events at the time of90Y infusion with two-thirds of
the dose delivered to the right lobe and one-third of
the dose delivered to the left lobe. The patient had
jaundice and raised bilirubin at 2 days after90Y infu-
sion. Endoscopy revealed normal duodenoscopy but
a short, possibly inflammatory stricture proximal to
the hepatic confluence. A stent was inserted, and the
jaundice resolved. The patient who lived alone and
had longstanding cerebral ischemia was discharged
to a nursing home for recuperation, where she
increasingly became confused, developed pneumo-
nia, and died. A CT scan at 3 weeks after
ment revealed decreased liver tumor bulk.
Retreatment and Differential Response in Liver Lobes
The routine study treatment was90Y infusion to the
whole liver. Three patients had an initial partial
response (PR)/SD on CT scans in the right liver but
had progressive disease (PD) in the left lobe only
and were retreated at 2 months, 7 months, and
8 months. All patients had progressive liver disease,
and 1 patient died. No chemotherapy was given to
any patient who was retreated with90Y radioemboli-
zation. An additional 4 patients who had sympto-
matic, bilobar PD had a second whole liver90Y dose
at 17 months, 18 months, 22 months, and 26 months.
Two patients with extensive liver and bone metasta-
ses died, and the third patient, who had progressive
liver disease at 15 months, had SD at 11 months after
38 years with an unknown primary carcinoid who
underwent liver resection in 2000) received90Y treat-
ment in December 2003 and had a PR/SD for
18 months until June 2005, when a new liver lesion
developed, which was resected at 19 months after
90Y treatment. The patient developed symptomatic,
90Y treatment. The fourth patient (aged
Characteristics of Patients the With Best Liver Response to Yttrium-90 Radioembolization by Response Evaluation Criteria in Solid Tumors
in Liver Primary Site
Yttrium 90 Estimated
Tumor Dose, GyCgA Fall, %
CT indicates computed tomography; SIR, selective internal radiation; GBq, gigabecquerel; Gy, grays; CgA chromogranin A; CR, complete response; LR, liver resection; 1, positive; PR, partial response; IV, systemic
chemotherapy; SD, stable disease.
924CANCERSeptember 1, 2008 / Volume 113 / Number 5
progressive liver disease at 38 months and received a
carcinoid with the same histology as the liver lesion)
was resected in August 2007. At 5 years after the first
radiotherapy and radiolabeled octreotide for PD in
the liver and bone.
Difficulty in the administration of
was recorded in 2 patients because of vessel spasm
during the procedure. Both patients, who had similar
celiac axis anomalies documented on screening
treatments to the right lobe only and the left lobe
only 1 week apart. The first patient, who had multi-
focal EHD, had a liver PR at 6 months but died at
12 months with overall PD. The second patient, who
had no prior EHD, had an overall CR at 27 months.
90Y infusion. A left eye/orbital mass (NET/
The mean survival (?SE) was 27.6 ? 2.3 months.
Fourteen patients (41%) died from progressive meta-
static disease at 1 to 28 months, and the mean sur-
vival was 14.6 ? 2.2
remained alive with a mean survival of 36.7 ? 1.8
months, and 20 patients
A subjective hormone response to90Y radioemboliza-
tion was observed in this study. Twenty-four patients
(71%) had symptoms of diarrhea, flushing, or rash at
the baseline assessment. Eighteen of 33 patients
(55%) at 3 months and 16 of 32 patients (50%) at
6 months who previously were symptomatic reported
improvement. No patient reported de novo hormone
symptoms. Two patients with EHD had increased tri-
cuspid disease at 29 months after90Y treatment.
Table 1, and data on hepatic progression are shown
in Table 2. In the liver, a CR was observed in
6 patients, 11 patients had a PR, 5 patients had SD,
and 11 patients had PD.
for patientswhodemonstrateda hepatic
90Y radioembolization are shown in
Tumor Marker Response
To assess the CgA marker response to90Y radioembo-
lization, 33 of 34 patients who had baseline CgA and
serial data available were divided into 2 groups:
18 patients with EHD at study entry and 15 patients
without EHD at study entry. The CgA values were
categorized as no change from baseline, <25%
decrease, ?0% decrease, >25% decrease, and ?50%
increase. In the non-EHD group, all the patients had
a decrease in CgA levels. Four patients (27%) had a
decrease <25%, 9 patients (60%) had a decrease
<50%, and 1 patient each (7%) had an increase
>25% and >50% in CgA levels.
In the EHD group, there were 3 patients (17%)
without CgA changes from baseline, 4 patients (22%)
had a decrease <25%, 6 patients (33%) had a
decrease <50%, and 5 patients (28%) had an increase
>50% in CgA levels.
For the EHD group (n 5 18 patients), the mean
CgA nadir (?SE) was 328.4 ? 109.7 U/L; and, for
the non-EHD group (n 5 15 patients), the mean CgA
Characteristics of Patients Without Liver Response to Yttrium-90 Radioembolization by Response Evaluation Criteria in Solid Tumors Criteria
CT Response in
Liver Primary Site
Replacement Follow-up, mo
Yttrium 90 Estimated
Tumor Dose, GyCgA Fall, %
Died at 1 mo
CT indicates computed tomography; SIR, selective internal radiation; GBq, gigabecquerel; Gy, grays; CgA chromogranin A; PD, progressive disease; IV, systemic chemotherapy; 1, positive; LR, liver resection,
Radioembolization With SIR Spheres/King et al925
nadir was 146.2 ? 50.2 U/L. For the EHD group, the
mean time to CgA nadir was 5.1 ? 1.5 months; and,
for the non-EHD group, the mean time to CgA nadir
was 8.6 ? 2.4 months.
Factors Affecting Response
Tumor volume replacement,90Y dose, and estimated
90Y dose to tumor are shown in Tables 1 and 2 for
responders and nonresponders. The most important
finding of this study was that 12 patients (35%)
remained alive for mean (?SE) of 33.3 ? 2.3 months
with no recurrence of liver disease after treatment
with Y.90In those 12 patients, there was no obvious
contributing factor evident that could account for
the good to excellent response to90Y, except perhaps
for less hepatic volume replacement. Radiologic CRs
in the liver achieved by 3 patients after90Y treatment
are illustrated in Figures 1 through 3.
There was a trend toward a difference in the
mean percentage (?SE) of hepatic tumor replace-
ment in patients who had a radiologic responses to
90Y treatment: The hepatic tumor placement was
26.8% ? 9.8%
28.6% ? 4.6% in 11 patients who had a PR, and
25% ? 6.7% in 5 patients who had SD compared
with 41.2% ? 4% in 12 patients who had PD. Two of
the patients who had a CR had >50% hepatic repla-
cement. There was no difference in the mean GBq
dose (?SE) of microspheres delivered, which was
1.8 GBq for patients who had a CR, 1.9 GBq for
patients who had a PR, 2.1 GBq for patients who had
SD, and 2 GBq for patients who had PD. The mean
(?SE) to the liver tumors was 40 ? 10.9 MBq for
patients who had a CR, 58.5 ? 7.2 MBq for patients
who had a PR, 70.2 ? 11.9 MBq for patients who had
SD, and 45.6 ? 5.5 MBq for patients who had PD.
90Y megabecquerel (MBq) dose delivered
FIGURE 1. Computed tomography scans before (Top) and 27 months after
(Bottom)90Y radioembolization in a patient who had a complete response in
FIGURE 2. Computed tomography scans before (Top) and 36 months after
(Bottom)90Y radioembolization in a patient who had a complete response in
926CANCER September 1, 2008 / Volume 113 / Number 5
response and the maximum percentage decrease in
CgA (CR, 50.6% ? 22.9%; PR, 33.6% ? 21.3%; SD,
57.6% ? 31.3%; and PD, 29.4% ? 26.4%).
was no clear relation between radiologic
The results of the current study demonstrate that
some patients can achieve a CR in their NETLMs after
treatment with90Y microspheres for ?2 years and that
an additional proportion of patients may attain a
response of SD. To date, the long-term duration of
response to90Y is not clear, but it is most encouraging;
because patients who achieve a CR in the liver survive
from 26 months to 48 months after radioembolization.
It is questionable whether any other therapy pre-
viously has achieved such useful results in patients
with inoperable disease. Although the morbidity asso-
ciated with90Y radioembolization is well known,24,25a
CR assessed by CT scan has been described in only 2
of >500 patients who underwent by hepatic arterial
embolization or hepatic arterial chemoembolization
in the last 20 years (Table 3).20,26-38Two other patients
reportedly had a CR, but that assessment was on the
basis of angiographic assessment, and not CT assess-
ment. Four other CRs were reported in 2003 by Loewe
et al, who used a different and experimental technique
of permanent hepatic artery occlusion with cyanoa-
crylate.34Although we accept that a CR occur can after
hepatic arterial embolization alone, it is uncommon;
whereas we observed a CR in 6 of 34 patients after90Y
treatment. Radioembolization/hepatic arterial emboli-
zation combined or chemotherapy postradioemboli-
previously in patients with NETLMs, but this may offer
an additional benefit. A report indicated the effective-
ness of90Y radioembolization, even in patients with
refractory metastatic disease who were treated with
previous nonradioactive embolization procedures.39
In the current study, vessel spasm was noted
during the90Y infusion in 2 patients who had variant
angiogram/99mTc-MAA scan. Both patients received
separate right lobe and left lobe
1 week apart. In this study, we did not embolize
aberrant vessels prophylactically, and we do not
know whether doing so could have optimized deliv-
ery and/or decreased symptoms. It was reported in
2006 that 45% of 68 patients who had LMs treated by
SIRT had abnormal arterial anatomy.40
review advised that, because of the large degree of
variant hepatic arterial anatomy, aberrant vessels
should be embolized prophylactically at CT angiogra-
phy before the
single procedure no longer is advised.41It is feasible
that more focused delivery of radiation to a smaller
area of the liver or access to previously unattainable
vessels can be attained now with microcatheter tech-
nology. This may decrease damage to normal liver
and can take advantage of tumor vascularity.40We
had re-examined the angiograph/99mTc-MAA scans
of the 3 patients who had developed radiation gastri-
tis and the 1 duodenal ulcer, but we observed no evi-
dence of aberrant anatomy before90Y treatment.
The symptomatic responses observed in 55% of
patients and the greater CgA reduction in responding
patients also support our belief that90Y is an active
therapy. The finding of an observed response in 1
side of the liver and disease progression in the other
with a response attained after a second dose admi-
nistered through the missed vessel (usually the
accessory left hepatic artery from the left gastric ar-
tery) is good evidence of a biologic effect.
99mTc-MAA scan for patients who
90Y, and whole liver radioembolization at a
FIGURE 3. Computed tomography scans before (Top) and 36 months after
(Bottom)90Y radioembolization in a patient who had a complete response in
Radioembolization With SIR Spheres/King et al927
We can only speculate regarding why some
patientsachieve such excellent
others do not respond. In this study, we examined
the percentage of liver tumor replacement, the type
of NET, and the90Y dose delivered. We observed no
difference between the responders and nonrespon-
ders with respect to either the90Y dose delivered to
the liver (P 5.3) or the estimated dose taken up by
the tumors (P 5.8). We did not study the tumor ki-
netic index in this work, and we did not attempt to
measure the vascularity of tumors. There may be a
relation between99mTc-MAA uptake and the disease
response,42but we did not quantify the
uptake ratio in this study. Pancreatic NETs appear to
be a little more likely to respond than other NETs.
The distribution of90Y microspheres assessed by
nuclear imaging to estimate the radiation dose to tu-
mor and liver (target to nontarget-ratio [TNT]) can
only be estimated. This is because it is known that dif-
ferent tumor classes demonstrate different TNT ratios,
even within the same patient; there is no method to
assess small individual tumors; the distribution of90Y
microspheres is fixed after administration, and there
is no proof that all radioactivity drawn up is adminis-
tered into the liver/tumor and that none escapes else-
where or remains in the delivery apparatus.
We chose to use the pre-existing protocols invol-
ving systemic 5-fluorouracil at the time of
sion. Whether this is necessary or not in patients
with NETs is speculative.
These very good results suggest that long-term
control of unresectable NETLMs by90Y microspheres
is possible. Response of at least some magnitude
may be expected in a significant population of
patients. The favorable clinical response to radioem-
bolization and the effect on long-term survival are
limited by the presence and extent of significant
EHD. Preliminary results indicate significant hepatic
tumor response and amelioration of symptoms with
low treatment-related toxicity.
1.Delcore R, Friesen SR. Gastrointestinal neuroendocrine
tumors. J Am Coll Surg. 2004;178:188-211.
2. Chamberlain RS, Canes D, Brown KT, et al. Hepatic neuro-
endocrine metastases: does intervention alter outcomes? J
Am Coll Surg. 2000;190:432-445.
3. Hellman P, Ladjevardi S, Skogseid B, et al. Radiofrequency
tissue ablation using cooled tip for liver metastases of en-
docrine tumors. World J Surg. 2002;26:1052-1056.
4.Florman J, Toure B, Kim L, et al. Liver transplantation for neu-
roendocrine tumours. J Gastrointest Surg. 2004;8:208-212.
5.Touzios JG, Kiely JM, Pitt SC, et al. Neuroendocrine hepatic
metastases: does aggressive management improve survival?
Ann Surg. 2005;241:775-783.
6.Cozzi PJ, Englund R, Morris DL. Cryotherapy treatment of
patients with hepatic metastases from neuroendocrine
tumors. Cancer. 1995;76:501-509.
7.Rougier P, Mitry E. Chemotherapy in the treatment of neu-
roendocrine malignant tumors. Digestion. 2000;62(suppl 1):
8. Rivera E, Ajani JA. Doxorubicin, streptozotocin and 5-fluor-
ouracil chemotherapy for patients with metastatic islet-cell
carcinoma. Oncology. 1998;21:316-338.
9.Sun W, Lipsitz S, Catalano P , Maillard JA, Haller DA. Phase II/
III study ofdoxorubicin
with streptozotocin with fluorouracil or dacarbazine in the
treatment of advanced carcinoid tumors: Eastern Cooperative
with fluorouracil compared
Clinical Studies of Neuroendocrine Tumor Liver Metastases
ReferenceHAE/HACE CR, No. PR, No.
Mean No. of
22 HAE, CYA*
Median response, 14 mo; CR
Response duration: HAE, 4 mo; HACE, 18 mo
Median response, 17 mo
Carcinoid, islet cell
HAE indicates hepatic arterial embolization; HACE, hepatic arterial chemoembolization; CR, complete response; PR, partial response; CYA, cyanoacrylate; NR, not reached.
*Permanent embolization with CYA.
928CANCERSeptember 1, 2008 / Volume 113 / Number 5
Oncology Group Study E1281. J Clin Oncol. 2005;23:4897-
10. Bajetta E, Ferrari L, Procopio G, et al. Efficacy of a chemo-
therapy combination for the treatment of metastatic neu-
roendocrine tumours. Ann Oncol. 2002;13:614-621.
11. Ruszniewski P, Malka D. Hepatic arterial chemoemboliza-
tion in the management of advanced digestive endocrine
tumors. Digestion. 2000;62(suppl 1):79-83.
12. Oberg K, Kvols L, Caplin M, et al. Consensus report on the
use of somatostatin analogs for the management of neuro-
endocrine tumors of the gastroenteropancreatic system.
Ann Oncol. 2004;15:966-973.
13. Gray B, van Hazel G, Hope M, et al. Randomised trial of
SIR-Spheres1 plus chemotherapy versus chemotherapy
alone for treating patients with liver metastases from pri-
mary large bowel cancer. Ann Oncol. 2001;12:1711-1720.
14. Stubbs RS, Cannan RJ, Mitchell AW. Selective internal
radiation therapy (SIRT) with yttrium microspheres for
extensive colorectal liver metastases. Hepatogastroenterology.
15. Lau Wy, Ho S, Leung TW, et al. Selective internal radiation
therapy for nonresectable hepatocellular carcinoma with
intra-arterial infusion of 90yttrium microspheres. Int J
Radiat Oncol Biol Phys. 1998;40:583-592.
16. Murthy R, Xiong H, Nunez R, et al. Yttrium 90 microspheres
for the treatment of unresectable colorectal hepatic metasta-
ses after failure of multiple chemotherapy regimens: prelimi-
nary results. J Vasc Interv Radiol. 2005;16:937-945.
17. Lim LC, Gibbs P, Yip D, et al. A prospective evaluation of
treatment with selective internal radiation therapy (SIR-
spheres) in patients with unresectable liver metastases
from colorectal cancer previously treated with 5-FU based
chemotherapy. BMC Cancer. 2005;5:132.
18. Kennedy AS, Coldwell D, Nutting C, et al. Resin 90Y-micro-
sphere brachytherapy for unresectable colorectal liver me-
tastases: modern USA experience. Int J Radiat Oncol Biol
19. Kennedy AS, Liu D, Dezarn AW, et al. Resin
sphere brachytherapy for unresectable neuroendocrine he-
patic metastases. Am J Oncol Rev. 2006;4(suppl):1-8.
20. Ruszniewski P, Rougier P, Roche A, et al. Hepatic arterial
chemoembolization in patients with liver metastases of
neuroendocrine tumors. A prospective phase II study in 24
patients. Cancer. 1993;71:2624-2630.
21. Duffaud F, Therasse P. New guidelines to evaluate the
response to treatment in solid tumors. Bull Cancer. 2000;
22. Hudson HM, Larkin RS. Accelerated image reconstruction
using ordered subsets of projection data. IEEE Trans Nucl
23. Bolch WE, Bouchet LG, Robertson JS, et al. MIRD pam-
phlet 17: the dosimetry of nonuniform activity distribu-
tions-radionuclide S values at the voxel level. J Nucl Med.
24. Yip D, Allen R, Ashton C, Jain S. Radiation-induced ulceration
of the stomach secondary to hepatic embolization with radi-
oactive yttrium microspheres in the treatment of metastatic
colon cancer. J Gastroenterol Hepatol. 2004;19:347-349.
25. Maclean F, Ng T, Morris D, Millar E, Kench J. Gastric
ulceration complicating yttrium90microsphere therapy of
hepatic metastases. Presented as a poster at the Australian
Division of the International Academy of Pathology. 31st
Annual Scientific Meeting, Sydney, Australia, June 2–4,
26. Carrasco CH, Chuang VP , Wallace S. Apudomas metastatic
to the liver: treatment by hepatic artery embolization.
27. Ajani JA, Carrasco CH, Charnsangavej C, et al. Islet cell
tumors metastatic to the liver: effective palliation by se-
quential hepatic artery embolization. Ann Intern Med. 1988;
28. Hajarizadeh H, Ivancev K, Mueller CR, et al. Effective palli-
ative treatment of metastatic carcinoid tumors with intra-
arterial chemotherapy/chemoembolization combined with
octreotide acetate. Am J Surg. 1992;163:479-483.
29. Therasse E, Breittmayer F, Roche A, et al. Transcatheter
chemoembolization of progressive carcinoid liver metasta-
sis. Radiology. 1993;189:541-547.
30. Mavligit GM, Pollock RE, Evans HL, et al. Durable hepatic
tumor regression after arterial chemoembolization-infusion
in patients with islet cell carcinoma of the pancreas meta-
static to the liver. Cancer. 1993;72:375-380.
31. Moertel CG, Johnson CM, McKusick MA, et al. The man-
agement of patients with advanced carcinoid tumors and
islet cell carcinomas. Ann Intern Med. 1994;120:302-309.
32. Eriksson BK, Larsson EG, Skogseid BM, Lofberg AM, Lore-
lius LE, Oberg KE. Liver embolizations of patients with ma-
lignant neuroendocrine gastrointestinal tumors. Cancer.
33. Gupta S, Yao JC, Ahrar K, et al. Hepatic artery embolization
and chemoembolization for treatment of patients with
metastatic carcinoid tumors: the M.D. Anderson experi-
ence. Cancer J. 2003;9:261-267.
34. Loewe C, Schindl M, Cejna M, Niederle B, Lammer J,
Thurnher S. Permanent transarterial embolization of neu-
roendocrine metastases of the liver using cyanoacrylate
and lipiodol: assessment of mid- and long-term results.
AJR Am J Roentgenol. 2003;180:1379-1384.
35. Roche A, Girish BV, De Baere T, et al. Trans-catheter arterial
chemoembolization as first-line treatment for hepatic metas-
tases from endocrine tumors. Eur Radiol. 2003;13:136-140.
36. Gupta S, Johnson MM, Murthy R, et al. Hepatic arterial
embolization and chemoembolization for the treatment of
patients with metastatic neuroendocrine tumors. Cancer.
37. Osborne DA, Zervos EE, Strosberg J, et al. Improved out-
come with cytoreduction versus embolization for sympto-
matic hepatic metastases of carcinoid and neuroendocrine
tumors. Ann Surg Oncol. 2006;13:572-581.
38. Ho AS, Picus J, Darcy MD, et al. Long-term outcome after
chemoembolization and embolization of hepatic meta-
static lesions from neuroendocrine tumors. AJR Am J
39. Murthy R, Kamat P, Nunez R, et al. Yttrium-90 microsphere
radioembolotherapy of hepatic metastatic neuroendocrine
carcinomas after hepatic arterial embolization. J Vasc
Interv Radiol. 2008;19:145-151.
40. Bester L, Hollands M. Interventional radiological approaches
to improving outcomes of selective internal radiation ther-
apy: HP009. ANZ J Surg. 2006;76(suppl 1): A36-A37.
41. Lewandowski RJ, Sato KT, Atassi B, et al. Radioembolization
with90Y microspheres: angiographic and technical consid-
erations. Cardiovasc Intervent Radiol. 2007;30:571-592.
42. Dhabuwala A, Lamerton P, Stubbs RS. Relationship of 99m
technetium-labeled macroaggregated albumin (99mTc-MAA)
uptake by colorectal liver metastases to response after selec-
tive internal radiation therapy (SIRT). BMC Nuclear Med.
Radioembolization With SIR Spheres/King et al929