RAPID COMMUNICATION | Hepatology, VOL. 74, NO.

5, 2021 

Yttrium-­90 Radioembolization for the

Treatment of Solitary, Unresectable HCC:
The LEGACY Study
Riad Salem ,1 Guy E. Johnson,2 Edward Kim,3 Ahsun Riaz,1 Vivian Bishay,3 Eveline Boucher,4 Kirk Fowers,4
Robert Lewandowski,1 and Siddharth A. Padia 5

(11 of 162) of patients, respectively, and as primary treatment for

SEE EDITORIAL ON PAGE 2333 all others. Median follow-­ up time was 29.9 months by reverse
Kaplan-­Meier. ORR (best response) was 88.3% (CI: 82.4-­ 92.4),
BACKGROUND AND AIMS: Locoregional therapies, in- with 62.2% (CI: 54.1-­ 69.8) exhibiting a DoR ≥ 6 months.
cluding yttrium-­ 90 radioembolization, play an important Three-­year overall survival was 86.6% for all patients and 92.8%
role in the treatment of unresectable HCC. The aim of the for those neoadjuvant patients with resected or transplanted liver.
LEGACY (Local radioEmbolization using Glass Microspheres
CONCLUSIONS: In this multicenter study of radioemboli-
for the Assessment of Tumor Control with Y-­ 90) study was
zation, clinical meaningful response rates and prolonged DoR
to evaluate objective response rate (ORR) and duration of
were observed in the treatment of unresectable, solitary HCC
response (DoR) in patients with solitary unresectable HCC
≤ 8 cm. (Hepatology 2021;74:2342-2352).
treated with yttrium-­
90 glass microspheres.

P
APPROACH AND RESULTS: LEGACY is a multicenter,
single-­arm, retrospective study conducted at three sites that in- atients with solitary HCC have several curative
cluded all eligible, consecutive patients with HCC treated with treatment options, including liver transplanta-
radioembolization between 2014 and 2017. Eligibility criteria tion (LT), surgical resection, and thermal abla-
included solitary HCC ≤ 8 cm, Child-­ Pugh A cirrhosis, and tion; however, many patients are not candidates due to
Eastern Cooperative Oncology Group performance status 0-­ 1. tumor size, location, or comorbidities. Although LT
Primary endpoints were ORR and DoR based on modified may be the most ideal option, many patients require
Response Evaluation Criteria in Solid Tumors in the treated area
bridging due to prolonged wait times, or downstaging
(localized), as evaluated by blinded, independent, central review.
Radioembolization was performed with intent of ablative-­ level
to achieve Milan criteria. Surgical resection is consid-
dosimetry in a selective fashion when possible. Overall survival ered curative in patients with solitary tumor, normal
was evaluated using Kaplan-­ Meier and multivariate Cox propor- bilirubin, and absence of portal hypertension; many
tional hazards. Among the 162 patients included, 60.5% were patients do not meet these criteria. Thermal ablation
Eastern Cooperative Oncology Group 0, and the median tumor is also considered potentially curative, but is limited to
size was 2.7 cm (range: 1-­ 8) according to blinded, independent, small tumors in favorable locations.(1)
central review. Radioembolization served as neoadjuvant therapy Although radioembolization has traditionally been
for transplantation or resection in 21.0% (34 of 162) and 6.8%
used to treat advanced HCC, recent refinements in

Abbreviations: AE, adverse event; BCLC, Barcelona Clinic Liver Cancer; BICR, blinded, independent, central review; CR, complete response;
DoR, duration of response; ECOG, Eastern Cooperative Oncology Group; LEGACY, Local radioEmbolization using Glass Microspheres for the
Assessment of Tumor Control with Y-­90; LT, liver transplantation; mRECIST, modif ied Response Evaluation Criteria in Solid Tumors; NE, not
evaluable; ORR, objective response rate; OS, overall survival; PD, progressive disease; PFS, progression-­f ree survival; PR, partial response; RECIST,
Response Evaluation Criteria in Solid Tumors; SAE, serious adverse event; TTP, time-­to-­progression; 90Y, yttrium-­90.
Received February 24, 2021; accepted March 16, 2021.
Additional Supporting Information may be found at onlinelibrary.wiley.com/doi/10.1002/hep.31819/suppinfo.
Supported by Boston Scientif ic Corporation.
© 2021 The Authors. Hepatology published by Wiley Periodicals LLC on behalf of American Association for the Study of Liver Diseases. This is an
open access article under the terms of the Creative Commons Attribution-­NonCommercial License, which permits use, distribution and reproduction in
any medium, provided the original work is properly cited and is not used for commercial purposes.

2342
Hepatology, Vol. 74, No. 5, 2021 SALEM ET AL.

technique have shown promising response rates in soli- underwent radioembolization using 90Y glass micro-
tary HCC.(2) Rather than using a conventional strategy spheres between January 1, 2014, and December 31,
of lobar yttrium-­90 (90Y) infusion, selective radioembo- 2017, were evaluated for study enrollment. Inclusion
lization of the tumor-­bearing hepatic segment, termed criteria included age ≥ 18 years, a solitary tumor with
“radiation segmentectomy,” spares most of the nonin- largest diameter ≤ 8 cm, treatment with lobar or selective
volved hepatic parenchyma. With a segmental treatment, hepatic radioembolization, Child-­Pugh A cirrhosis, and
higher radiation doses can be safely delivered, translating Eastern Cooperative Oncology Group (ECOG) perfor-
to robust and consistent response rates.(3-­7) The dura- mance score 0-­1. Exclusion criteria included prior LT or
tion of response (DoR), however, a concept thoroughly resection, prior locoregional or systemic therapy, vascular
investigated with immunotherapies, is understudied in invasion, extrahepatic metastases, clinically significant
radioembolization. DoR takes into account objective ascites, or hepatic encephalopathy. The Food and Drug
response and provides information relating to the time Administration (FDA) agreed with key parameters of
component and durability of the response, variables of the LEGACY protocol, which served as the basis for
significant interest to patients and clinicians. Recently, Premarket Approval (PMA) of TheraSphere (Boston
DoR has served as a key parameter leading to regulatory Scientific Corp., Marlborough, MA) for local tumor
approval of several immunotherapy agents. control of unresectable, solitary HCC < 8 cm.
The purpose of this retrospective, multicenter study
was to assess the response rates and DoR following
treatment with 90Y glass microspheres in patients with
TREATMENT
unresectable, solitary HCC. All patients underwent radioembolization using
TheraSphere glass microspheres. This was performed
according to standard practices, which included a
Materials and Methods mapping angiogram with lung shunt fraction determi-
nation followed by 90Y microsphere infusion. During
LEGACY (Local radioEmbolization using Glass the mapping procedure, selective hepatic angiography
Microspheres for the Assessment of Tumor Control and C-­arm CT were performed to determine tumor
with Y-­90) is a retrospective, single-­
arm, multicenter arterial supply. Technetium macroaggregated albumin
HCC study conducted across three US sites. The was infused into the tumor-­bearing lobe, and the lung
study was approved by each site’s institutional review shunt fraction was determined. Perfused volume was
board. Consecutively treated patients with HCC who measured retrospectively using either the treatment

View this article online at wileyonlinelibrary.com.

DOI 10.1002/hep.31819
Potential conflict of interest: Dr. Boucher is employed and owns stock in Boston Scientif ic. Dr. Bishay consults for Boston Scientif ic. Dr. Padia
consults for Boston Scientific and Bristol-­Myers Squibb. He received grants from Varian. Dr. Fowers is employed and owns stock in Boston Scientif ic.
Dr. Riaz consults for Boston Scientif ic. Dr. Salem consults for Eisai, Boston Scientif ic, Sirtex, AstraZeneca, Genentech, Cook, Bard, and QED.
Dr. Kim consults, advises, and is on the speakers’ bureau for Boston Scientif ic. He advises Sirtex. Dr. Johnson consults for Boston Scientif ic, Genentech,
and AstraZeneca. Dr. Lewandowski consults for Boston Scientif ic.

ARTICLE INFORMATION:
From the 1 Department of Radiology, Northwestern Feinberg School of Medicine, Chicago, IL; 2 Department of Radiology, University
of Washington, Seattle, WA; 3 Department of Radiology, Mount Sinai Health System, New York, NY; 4 Boston Scientific Corporation,
Marlborough, MA; 5 Department of Radiological Sciences, University of California Los Angeles Medical Center, Los Angeles, CA.

ADDRESS CORRESPONDENCE AND REPRINT REQUESTS TO:

Riad Salem, M.D., M.B.A. Chicago, IL 60611
Department of Radiology E-­mail: [email protected]
Northwestern Feinberg School of Medicine Tel.: +1-­312-­695-­6371
676 N. St. Clair, Suite 800

2343
SALEM ET AL. Hepatology, November 2021

cone beam CT or post–­Y90 single-­proton emission independent reviewer. To compensate for the limita-
computerized tomography/CT. The absorbed dose to tions inherent in a retrospective analysis, the BICR
the perfused volume was then calculated based on the reads were performed blinded to dosing data, clinical
infused Y90 activity and the perfused volume. This is outcomes, and subsequent HCC treatments.
standard unicompartment dosimetry and is consistent
with the package insert.
OUTCOMES
DATA COLLECTION There were co-­ primary endpoints. In agreement
with the FDA, the study was considered successful if
Data were abstracted from medical records by study both of the following criteria were met: a lower limit
team members at each site and entered into the elec- of the 95% CI for confirmed objective response rate
tronic Clinical Report Form within the main study (ORR) by localized mRECIST > 40% and, DoR
database; sites were monitored on-­site periodically to by localized mRECIST > 6 months for ≥ 60% of
ensure accuracy of data entry. Collected patient data responders. Best and confirmed ORRs were reported,
included demographics, medical history, disease char- the latter declared when a subsequent scan ≥ 4 weeks
acteristics (etiology, Barcelona Clinic Liver Cancer later confirmed the initial response.(9) Secondary end-
[BCLC] stage, Child-­Pugh status), clinical assessments points included ORR/DoR by mRECIST/RECIST
(ECOG performance status, ascites, encephalopathy), 1.1, time to progression (TTP), progression-­free sur-
laboratory assessments (bilirubin, albumin, interna- vival (PFS), and overall survival (OS).(10)
tional normalized ratio, alpha-­fetoprotein, hematology
panel [white blood cell count, hemoglobin, platelet
STATISTICAL ANALYSIS
count]), and tumor assessment and liver volume esti-
mation based on diagnostic imaging. Serious adverse For the hypothesized 50% localized mRECIST ORR
events (SAEs) and nonserious adverse events (AEs) with 90Y, 100 patients would be required to achieve a
within 60 days following treatment were recorded; only CI of 40%-­60%; it would narrow to 43%-­57% with 200
those possibly related to treatment or procedure were patients based on the method of Wilson.(11) Hence, a
collected thereafter (through 12 months). AE severity minimum sample size of 100 was targeted. Baseline
was graded according to the National Cancer Institute’s data were analyzed using descriptive statistics. ORR was
Common Terminology Criteria for Adverse Events summarized as counts and percentages, with CIs. DoR
version 5.0.(8) Whether the SAEs and AEs were possi- was computed as the time between the first localized
bly or probably related to device use or administration tumor response and progressive disease (PD) observed.
was determined by each site investigator. If PD was not observed, the DoR was computed using
the date of the last imaging assessment before any fur-
RESPONSE ASSESSMENT ther treatment was administered within the treatment
area. DoR was summarized using descriptive statistics,
Imaging follow-­up was performed per standard of and time-­ to-­
endpoint analyses were performed using
care guidelines every 2-­3 months. Radiologic images Kaplan-­Meier.(12) Incidence of AEs and SAEs were
were reviewed and evaluated using localized mod- reported as counts and percentages and coded according
ified Response Evaluation Criteria in Solid Tumors to the Medical Dictionary for Regulatory Activities.(13)
(mRECIST) and RECIST 1.1 criteria by a blinded,
independent, central review (BICR). Localized mRE-
CIST was defined as the response/progression within
the radioembolization-­treated region. mRECIST and
Results
RECIST assessed the treated segment for response,
BASELINE CHARACTERISTICS
but also included new nodules and extrahepatic
metastases as progression. The BICR consisted of two A total of 162 patients were enrolled (Fig. 1).
independent radiologists who assessed responses on Patient characteristics are listed in Table 1. The
follow-­up imaging, blinded to visit sequence for the median age was 66 (range: 21-­90); most were male
primary read. Discordance was adjudicated by a third (75.9%, 123 of 162), afflicted hepatitis C (69.1%, 112

2344
Hepatology, Vol. 74, No. 5, 2021 SALEM ET AL.

FIG. 1. Flow diagram of patient study status.

of 162), and BCLC A (60.5%, 98 of 162). Cirrhosis months (CI: 3.5-­4.1) (Table 3 and Fig. 2). There were
was present in 92.0% (149 of 162). On the basis of no local progressors by BICR. Median DoR for con-
cancer-­related symptoms/pain, 39.5% were ECOG 1 firmed response was 11.8 months, with 76.1% exhibit-
and staged as BCLC C. The median tumor size was ing DoR ≥ 6 months. Figure 3 shows a swimmer’s plot
2.7 cm (range: 1.0-­8.1 cm), according to BICR. detailing imaging/response/transplant/resection status,
and last follow-­up, by confirmed response. Figure 4
and Supporting Fig. S1 shows a spider plot detailing
TREATMENT CHARACTERISTICS
percent change from baseline in target lesion by local-
Treatment intent for initial radioembolization ized mRECIST (confirmed response).
included definitive tumor treatment (66.7%, 108 of
162), bridging to transplantation or surgery (24.7%, 40 mRECIST
of 162), and other/unknown (8.6%, 14 of 162) (Table 2).
Most patients received selective infusions (95.7%, 155 The best ORR and confirmed ORRs were 86.4%
of 162); 1.9% (3 of 162) received lobar segmental infu- (CI: 80.3-­90.9) and 68.5% (CI: 61.0-­75.2), respec-
sions; and 2.5% (4 of 162) received mixed treatments. tively. The summary statistics median DoR for
The median absorbed dose to the treated liver volume confirmed response was 10.6 months, with 74.8%
was 410.1 Gy (interquartile range: 199.7, 797.7). exhibiting DoR ≥ 6 months.

ORR AND DoR RECIST 1.1

The best ORR and confirmed ORRs were 63.0%
Localized mRECIST (CI: 55.3-­70.0) and 46.3% (CI: 38.8-­54.0), respec-
The best ORR and confirmed ORRs were 88.3% tively. The summary statistics median DoR for con-
(CI: 82.4-­92.4) and 72.2% (CI: 64.9-­78.5), respectively, firmed response was 10.6 months, with 72% exhibiting
with a median time-­to-­best confirmed response of 3.9 DoR ≥ 6 months.

2345
SALEM ET AL. Hepatology, November 2021

TABLE 1. Baseline Characteristics TABLE 2. Y-­90 Treatment Characteristics

Treated Population Treated Population
(n = 162), n (%) (n = 162), n (%)

Age 18-­64 69 (42.6%) Treatment approach and Radiation 95 (58.6%)

65-­74 64 (39.5%) goal segmentectomy
≥75 29 (17.9%) Radiation lobectomy 4 (2.5%)
Gender Male 123 (75.9%) Downsizing to surgery 4 (2.5%)
Female 39 (24.1%) Bridge to LT 36 (22.2%)
Race White 80 (49.4%) Treat to local tumor 9 (5.6%)
control
Black or African-­American 16 (9.9%)
Other 1 (0.6%)
Asian 13 (8.0%)
Unknown 13 (8.0%)
Native American, Alaska Native 2 (1.2%)
Lung shunt fraction mean 4.5, (3.4),
Native Hawaiian, Pacific Islander 2 (1.2%)
(SD), median [IQR] 4.0 [2.3, 5.]
Not reported 49 (30.2%)
Total number of vials 1 118 (72.8%)
HCC etiology HBV 15 (9.3)
administered (first 2 38 (23.5%)
HCV 112 (69.1%) treatment)
3 6 (3.7%)
NASH 23 (14.2%)
Type of infusion Selective 155 (95.7%)
Autoimmune disease 3 (1.9%)
Lobar 3 (1.9%)
Alcohol 48 (29.6%)
Mixed 4 (2.5%)
Unknown 4 (2.5%)
Absorbed dose to treated 578.6 (540.1),
Other 1 (0.6%)
liver volume (Gy), mean 410 [199.7, 797.7]
Tumor size <3 cm 100 (61.7%) (SD), median [IQR]
3-­5 cm 50 (30.9%) (n = 155)
>5-­8 cm 9 (5.6%) Number of 90Y treatments 1 130 (80.2%)
Missing 3 (1.9%) ≥2 32 (19.8%)
BCLC/ECOG BCLC A (ECOG 0) 98 (60.5%)
score BCLC C (ECOG 1) 64 (39.5%) Abbreviation: IQR, interquartile range.
Child-­Pugh A5 108 (66.7%)
score A6 54 (33.3%)
and 76.7%, respectively (Fig. 5B). Median PFS was
BCLC A patients exhibited a best ORR of 89.8% 40.7 (CI: 36.0, not evaluable [NE]) for both mRE-
(CI: 82.2-­94.4), with 65.9% (CI: 55.5-­75.0) exhibit- CIST and RECIST.
ing a DoR of ≥ 6 months. BCLC C patients exhibited
a best ORR of 85.9% (CI: 75.4-­92.4), with 56.4% OS
(CI: 43.3-­68.6) exhibiting a DoR of ≥ 6 months.
Median OS was reached (57.9; 95% CI: NE,
NE). For the entire group, OS was 94.8% (CI:
TTP AND PFS 89.5-­97.5) and 86.6% (CI: 78.2-­92.0) at 24 and 36
Median TTP by BICR was not reached. months, respectively (Fig. 5C). Thirteen patients
had >36 months of follow-­up, up to 52 months. The
45 patients who went on to LT or resection exhib-
Localized mRECIST ited an OS of 100.0% (CI: 100.0-­100.0) and 92.8%
At 24 months, the local progression rate was 0, and (CI: 74.2-­98.2) at 24 and 36 months, respectively
PFS rate was 93.9%. Median PFS was 57.9% (CI: (Fig. 5C).
40.7, 57.9).
ECOG and Milan Criteria
mRECIST/RECIST 1.1
Using a complete case analysis, performance status
A total of 84.1% (CI: 72.6-­ 91.0) and 82.0% was maintained or improved in 69 of 77 (89.6%), 41
(CI:71.6-­
88.9) experienced no progression at 24 of 47 (87.2%), and 14 of 17 (82.4%) of patients at
months, respectively (Fig. 5A). PFS rates were 78.8% 6, 12 and 24 months, respectively. Milan criteria was

2346
Hepatology, Vol. 74, No. 5, 2021 SALEM ET AL.

TABLE 3. Response to Y-­90 Treatment

Localized mRECIST, n (%) mRECIST, n (%) RECIST 1.1, n (%)

ORR, confirmed response, n (%) [95% CI] 117 (72.2%) [64.9%, 78.5%] 111 (68.5%) [61.0%, 75.2%] 75 (46.3%) [38.8%, 54.0%]
ORR, best response, n (%) [95% CI] 143 (88.3%) [82.4%, 92.4%] 140 (86.4%) [80.3%, 90.9%] 102 (63.0%) [55.3%, 70.0%]
Best overall response
CR 136 (84%) 133 (82.1%) 13 (8.0%)
PR 7 (4.3%) 7 (4.3%) 89 (54.9%)
Stable disease 0 0 38 (23.5%)
PD 0 3 (1.9%) 3 (1.9%)
NE 19 (11.7%) 19 (11.7%) 19 (11.7%)
No imaging assessments after day 46 5 (3.1%) 5 (3.1%) 5 (3.1%)
No imaging assessments after day 46 due to LT or 9 (5.6%) 9 (5.6%) 9 (5.6%)
resection
Other reasons 5 (3.1%) 5 (3.1%) 5 (3.1%)
DoR* in months, mean (SD), median 15.1 (11.2), 11.8 14.0 (10.5), 10.6 13.8 (10.4), 10.6
DoR* ≥ 6 months, n (%) [95% CI] 89 (76.1%) [67.6%, 82.9%] 83 (74.8%) [66.0%, 81.9%] 54 (72%) [61.0%, 80.9%]

*DoR based the number of confirmed responders by BICR (n = 117 for localized mRECIST, n = 111 for mRECIST, and n = 75 for
RECIST.
Best Percent Change from Baseline in Target

CR
-20 PR
Lesion Size

SD
PD
-40
NE
Liver Transplant/
-60 Resection

-80

-100

Evaluable Patients (N=143)

FIG. 2. Waterfall plot of greatest decrease in target lesion size and best overall tumor response by localized mRECIST among evaluable
patients in the LEGACY study (n = 143). Abbreviations: CR, complete response; NE, not evaluable; PR, partial response; SD, stable
disease.

maintained in 116 of 122 (95.1%), 77 of 86 (89.5%), disorders (n = 7); fatigue (n = 1); pyrexia (n = 1);
and 29 of 34 (85.3%) at 6, 12 and 24 months, hepatobiliary disorders (ascites [n = 3], gallblad-
respectively. der obstruction [n = 1], and portal vein thrombosis
[n = 1]); infection (n = 4); increase in bilirubin (n = 3)
and international normalized ratio (n = 2); decrease in
Safety Outcomes
lymphocyte count (n = 12), platelets (n = 1), and white
Grade 3 events occurred in 31 of 162 patients blood cells (n = 1); hypoalbuminemia (n = 1); mus-
(19.1%), including anemia (n = 1); gastrointestinal culoskeletal and connective tissue disorders (n = 2);

2347
SALEM ET AL. Hepatology, November 2021

1st image within 9 months

1st image > 9 months

Ongoing Response

Complete Response

Partial Response

Death
Individual Responders

Last known alive

Liver transplant/resection

0 6 12 18 24 30 36 42 48 54 60
Time after TheraSphere treatment (months)

FIG. 3. Swimmer plot of TTR and DoR by localized mRECIST (confirmed response).

0
CR
Percent Change from Baseline In Target Lesion (%)

PR
-20
CR

PR
-40
NE

-60

-80

-100

0 1 2 3 4 5 6 7 8 9 10
Time after Y-90 Treatment (Months)

FIG. 4. Detailed spider plot of percent change from baseline in target lesion by localized mRECIST (confirmed response).

2348
Hepatology, Vol. 74, No. 5, 2021 SALEM ET AL.

Proportion without progression

1.0
0.9
0.8
0.7
0.6
0.5 Localized mRECIST
RECIST 1.1
0.4 mRECIST

0.3 N At Risk
0.2 162 114 76 45 32 20 12 6 2 0

0.1 162 110 67 38 26 15 9 3 2 0

162 111 69 40 27 15 9 3 2 0
0.0

0 6 12 18 24 30 36 42 48 54
Kaplan-Meier curve
Time (months)

B
1.0
Proportion without progression

0.9
0.8
0.7
0.6
0.5 Localized mRECIST
RECIST 1.1
0.4 mRECIST

0.3 N At Risk
0.2 162 114 78 48 35 23 15 7 3 1 0

0.1 162 110 69 41 29 18 11 3 2 0 0

162 111 71 43 30 18 11 3 2 0 0
0.0

0 6 12 18 24 30 36 42 48 54 60
Kaplan-Meier curve
Time (months)

C 1.0
0.9
0.8
Proportion Alive

0.7
0.6
+Censored
0.5 Transplant/Resection - bridging/downstaging TheraSphere
All Patients
0.4 TheraSphere as Primary Therapy
0.3
N At Risk
0.2 45 45 45 40 34 26 17 10 5 3 0

0.1 162 158 151 118 93 73 51 29 14 7 0

0.0 117 113 106 78 59 47 34 19 9 4 0

0 6 12 18 24 30 36 42 48 54 60
Kaplan-Meier curve
Time (months)

FIG. 5. Kaplan-­Meier analysis of TTP (A), PFS (B), and OS (C) by transplantation/resection status.

2349
SALEM ET AL. Hepatology, November 2021

nervous system disorder (n = 1); dyspnea (n = 1); and noted to be durable, translating to a prolonged clinical
pulmonary embolism (n = 1). Note that the number patient benefit.
of AEs adds up to > 31 as patients could have > 1 AE. PFS by localized mRECIST was 93.9% at 2 years.
Grade 4 (worsened platelet count) and grade 5 (cere- Local tumor recurrence rates were low and competi-
brovascular accident) AEs occurred in 1 patient each; tive to outcomes of thermal ablation.(2,20-­22) Prolonged
neither were related to radioembolization. Most AEs TTP in the treated tumor area is critical for patients
resolved during the course of the study period. SAEs being bridged to LT, given the mandated 6-­month
occurred in 9.9% of patients; 5.6% were at least pos- waiting period.(20,23-­26) In LEGACY, the high rates
sibly related to radioembolization. No patient expe- of tumor response achieved using radioemboliza-
rienced radiation-­induced liver disease. There were tion allowed bridging to LT and/or resection in 45
19 deaths (11.7%), 12 of which were attributable to patients. The remaining patients who did not receive
nontarget disease progression, and none were related a LT/resection exhibited OS similar to other curative-­
to treatment. There was one death in the transplant intent treatments.(2,27,28)
group (multisystem organ failure at 3 years), and one Advancements in the understanding of adminis-
in the resection group (HCC recurrence at 3 years). tration technique, optimal dosing, and the segmental
approach has led to changes in practice with radio-
embolization. The LEGACY study included patients
Discussion treated using the strategy of high-­dose radiation deliv-
ery to the tumor, with the goal of improved tumor
Over the last few years, there has been an increasing response; median absorbed dose to the treated liver
role for immunotherapy in the management of HCC, volume was 410 Gy. In fact, in a subset dose-­pathology
with several studies demonstrating antitumoral activ- correlation of resection/transplantation, LEGACY
ity translating to ORR and prolonged DoR. While patients exhibited complete pathological necrosis when
these observations have mostly been reported for sys- their dose exceeded 400 Gy to the tumor-­bearing tis-
temic therapies, there is a dearth of such data in the sue, establishing this as the new “threshold dose” for
locoregional therapy (LRT) space. Radioembolization an ablative effect.(21) This approach is supported by
with 90Y is an LRT applied to HCC across all stages, a phase 2 study, in which investigators determined a
including the curative setting in BCLC A, downstag- significant difference in tumor dose of responders ver-
ing and bridging in the transplant patient, and pallia- sus nonresponders (490 Gy vs. 275 Gy).(29) Another
tion in BCLC B/C.(2,14-­16) LEGACY is a study that prospective trial using photon emission tomography
investigates ORR/DoR in the homogenous patient to determine tumor dose after radioembolization also
population as defined by solitary HCC < 8 cm with showed that responders had a higher median dose to
preserved performance status. The study was consid- the tumor compared with nonresponders (225 gray vs.
ered positive, surpassed the clinically meaningful co-­ 83 gray).(30) Finally, in DOSISPHERE, a prospective
primary endpoint thresholds, and establishing 90Y as randomized trial, the median OS survival was more
clinically efficacious in this patient population.(17,18) than doubled, from 10.7 to 26.6 months, when per-
The ORR observed suggests that focal tumors sonalized multicompartment dosimetry was applied.
amenable to catheterization respond well to arte- The amalgamation of these findings further points to
rially delivered 90Y radiation. Although localized personalized and optimized dosimetry representing
mRECIST and mRECIST best responses replicate the next area of clinical research in 90Y.
prior less-­controlled investigations, the strength of Limitations include the retrospective design, lack of
the observation is reinforced with the confirmatory a control arm, and limited sample size compared with
scan.(19) LEGACY is the first to report on 90Y con- systemic therapy trials. Strengths include the multi-
firmatory response as reported by a BICR. The local center nature, long-­term, comprehensive follow-​­up, and
control of 100% by mRECIST also imparts radiation strict independent monitoring/auditing. Interestingly,
segmentectomy the status of “ablative treatment,” with the outcomes were similar across the study institu-
progressive disease manifesting solely as new distant tions. Also, all images were reviewed using a BICR,
lesions or metastases. Furthermore, in conjunction blinded to treatment information. LEGACY eval-
with an elevated response rate, those responses were uated tumor response using three different criteria

2350
Hepatology, Vol. 74, No. 5, 2021 SALEM ET AL.

(localized mRECIST, mRECIST, and RECIST 1.1), Author Contributions: All authors performed all of
permitting cross-­ comparison and providing evalu- those tasks.
ation of local tumor control, and liver and systemic
extent of HCC following 90Y. Localized mRECIST REFERENCES
was the designated primary endpoint, given that Y90
1) Galle PR, Forner A, Llovet JM, Mazzaferro V, Piscaglia F, Raoul
is a local treatment. Eighteen percent of the patients JL, et al. EASL Clinical Practice Guidelines management of he-
enrolled were >75 years of age, in accordance with the patocellular carcinoma. J Hepatol 2018;69:182-­236.
2) Lewandowski RJ, Gabr A, Abouchaleh N, Ali R, Al Asadi
contemporary focus on geriatric oncology and inclu- A, Mora RA, et al. Radiation segmentectomy: potential cu-
sion of elderly patients in clinical trials.(31) The focus rative therapy for early hepatocellular carcinoma. Radiology
on DoR is of interest, given its established role and 2018;287:1050-­1058.
3) Padia SA, Kwan SW, Roudsari B, Monsky WL, Coveler A, Harris
purported benefit in immuno-­ oncology therapies. WP. Superselective yttrium-­90 radioembolization for hepatocel-
The applicability of a loco-­regional therapy such as lular carcinoma yields high response rates with minimal toxicity.
90
Y in BCLC C is of particular interest in the con- J Vasc Interv Radiol 2014;25:1067-­1073.
4) Padia SA, Johnson GE, Horton KJ, Ingraham CR, Kogut MJ,
text of liver-­limited disease. OS is a critical endpoint Kwan S, et al. Segmental Yttrium-­90 radioembolization versus
that contextualizes the entire patient cohort and was segmental chemoembolization for localized hepatocellular carci-
reported in accordance with oncology standards. noma: results of a single-­center, retrospective, propensity score-­
matched study. J Vasc Interv Radiol 2017;28:777-­785.e771.
Although OS could have been considered the primary 5) Biederman DM, Titano JJ, Korff RA, Fischman AM, Patel RS,
endpoint, it is well-­known that BCLC-­A studies are Nowakowski FS, et al. Radiation segmentectomy versus selective
challenging, given the long natural history and effec- chemoembolization in the treatment of early-­stage hepatocellular
carcinoma. J Vasc Interv Radiol 2018;29:30-­37.e32.
tive post-­progression treatments; surrogate endpoints 6) Biederman DM, Titano JJ, Bishay VL, Durrani RJ, Dayan E,
are therefore integral to the advancement of HCC Tabori N, et al. Radiation segmentectomy versus TACE com-
therapies in early disease. Finally, the co-­ primary bined with microwave ablation for unresectable solitary hepato-
cellular carcinoma up to 3 cm: a propensity score matching study.
endpoints exceeded the hypothesis and generated Radiology 2017;283:895-­905.
compelling data that are sufficient for the first PMA 7) Vouche M, Habib A, Ward TJ, Kim E, Kulik L, Ganger D, et al.
Unresectable solitary hepatocellular carcinoma not amenable to
of its kind, highlighting a successful collaboration
radiofrequency ablation: multicenter radiology-­ pathology cor-
among investigators, industry, and regulatory bodies relation and survival of radiation segmentectomy. Hepatology
to advance the field of HCC. 2014;60:192-­201.
8) . Common Terminology Criteria for Adverse Events (CTCAE),
In conclusion, the multicenter LEGACY study Version 5. Bethesda, MD: US Department of Health and Human
demonstrates that radioembolization provides a strong Services; 2018.
and durable response in solitary unresectable HCC, 9) Salem R, Lewandowski RJ, Gates VL, Nutting CW, Murthy R,
Rose SC, et al. Research reporting standards for radioembolization
with ORR and DoR observations that are significant of hepatic malignancies. J Vasc Interv Radiol 2011;22:265-­278.
and clinically meaningful. Additionally, OS and PFS 10) Lencioni R, Llovet JM. Modified RECIST (mRECIST) assess-
compared favorably between patients receiving radi- ment for hepatocellular carcinoma. In: Seminars in Liver Disease.
New York, NY: Thieme Medical Publishers; 2010:52-­60.
oembolization as definitive treatment and those who 11) Wilson E. Calculating a confidence interval of a proportion. J Am
underwent curative therapies. Radioembolization Stat Assoc 1927;22:209-­212.
for solitary HCC is safe and efficacious when used 12) Kaplan EL, Meier P. Nonparametric estimation from incomplete
observations. J Am Stat Assoc 1958;53:457-­481.
either as neoadjuvant to transplant or resection, or as 13) Brown EG, Wood L, Wood S. The medical dictionary for regula-
a stand-­alone treatment. tory activities (MedDRA). Drug Saf 1999;20:109-­117.
14) Lewandowski RJ, Kulik LM, Riaz A, Senthilnathan S, Mulcahy
Acknowledgment: The authors thank Binal Patel for MF, Ryu RK, et al. A comparative analysis of transarterial down-
staging for hepatocellular carcinoma: chemoembolization versus
the statistical support, Evelyn Schnuerer for the sci- radioembolization. Am J Transplant 2009;9:1920-­1928.
entific expertise, Joanne MacFie for clinical trial 15) Salem R, Gabr A, Riaz A, Mora R, Ali R, Abecassis M, et al.
management, and Alexandra J. Greenberg-­ Worisek, Institutional decision to adopt Y90 as primary treatment for hepa-
tocellular carcinoma informed by a 1,000-­patient 15-­year experi-
Ph.D., M.P.H., for medical writing assistance (all from
ence. Hepatology 2019;68:1429-­1440.
Boston Scientific Corp.). They also acknowledge the 16) Abouchaleh N, Gabr A, Ali R, Al Asadi A, Mora RA, Kallini JR,
efforts of research coordinators Carlene del Castillo et al. (90)Y radioembolization for locally advanced hepatocellular
(Northwestern), Ellen Weiss (Mount Sinai), and carcinoma with portal vein thrombosis: long-­term outcomes in a
185-­patient cohort. J Nucl Med 2018;59:1042-­1048.
Ann Wilson (University of Washington), in addition 17) Yamashita T, Kudo M, Ikeda K, Izumi N, Tateishi R, Ikeda M,
to the efforts of Ahmed Gabr, M.D., research fellow et al. REFLECT-­a phase 3 trial comparing efficacy and safety of
(Northwestern).

2351
SALEM ET AL. Hepatology, November 2021

lenvatinib to sorafenib for the treatment of unresectable hepato- 25) Mehta N, Heimbach J, Lee D, Dodge JL, Harnois D, Burns J,
cellular carcinoma: an analysis of Japanese subset. J Gastroenterol et al. Wait time of less than 6 and greater than 18 months predicts
2020;55:113-­122. hepatocellular carcinoma recurrence after liver transplantation:
18) Yau T, Kang YK, Kim TY, El-­Khoueiry AB, Santoro A, Sangro proposing a wait time “sweet spot”. Transplantation 2017;101:​
B, et al. Efficacy and safety of nivolumab plus ipilimumab in pa- 2071-­2078.
tients with advanced hepatocellular carcinoma previously treated 26) Mehta N, Yao FY. What are the optimal liver transplanta-
with sorafenib: the CheckMate 040 randomized clinical trial. tion criteria for hepatocellular carcinoma? Clin Liv Dis
JAMA Oncol 2020;6:e204564. https://doi.org/10.1001/jamao​ 2019;13:20-­25.
ncol.2020.4564. Online ahead of print. 27) Ahn KS, Kang KJ. Appropriate treatment modality for soli-
19) Memon K, Kulik L, Lewandowski RJ, Wang E, Riaz A, Ryu RK, tary small hepatocellular carcinoma: radiofrequency ablation
et al. Radiographic response to locoregional therapy in hepatocel- vs. resection vs. transplantation? Clin Mol Hepatol 2019;25:​
lular carcinoma predicts patient survival times. Gastroenterology 354-­359.
2011;141:526-­535.e522. 28) Eilard MS, Naredi P, Helmersson M, Hemmingsson O, Isaksson
20) Gabr A, Kulik L, Mouli S, Riaz A, Ali R, Desai K, et al. Liver B, Lindell G, et al. Survival and prognostic factors after transplan-
transplantation following Yttrium-­ 90 radioembolization: 15-­ tation, resection and ablation in a national cohort of early hepato-
year experience in 207-­ patient cohort. Hepatology 2021;73:​ cellular carcinoma. HPB 2021;23:394-­403.
998-­1010. 29) Mazzaferro V, Sposito C, Bhoori S, Romito R, Chiesa C,
21) Gabr A, Riaz A, Johnson GE, Kim E, Padia S, Lewandowski Morosi C, et al. Yttrium-­90 radioembolization for intermediate-­
RJ, et al. Correlation of Y90-­absorbed radiation dose to patho- advanced hepatocellular carcinoma: a phase 2 study. Hepatology
logical necrosis in hepatocellular carcinoma: confirmatory mul- 2013;57:1826-­1837.
ticenter analysis in 45 explants. Eur J Nucl Med Mol Imaging 30) Chan KT, Alessio AM, Johnson GE, Vaidya S, Kwan SW, Monsky
2021;48:580-­583. W, et al. Prospective trial using internal pair-­production positron
22) Gabr A, Polineni P, Mouli SK, Riaz A, Lewandowski RJ, Salem R. emission tomography to establish the Yttrium-­90 radioemboliza-
Neoadjuvant radiation lobectomy as an alternative to portal vein em- tion dose required for response of hepatocellular carcinoma. Int J
bolization in hepatocellular carcinoma. Semin Nucl Med 2019;49:197-­ Radiat Oncol Biol Phys 2018;101:358-­365.
203. https://doi.org/10.1053/j.semnu​clmed.2019.01.009. 31) Swaminathan D, Swaminathan V. Geriatric oncology: problems
23) Roberts JP, Venook A, Kerlan R, Yao F. Hepatocellular carci- with under-­treatment within this population. Cancer Biol Med
noma: ablate and wait versus rapid transplantation. Liver Transpl 2015;12:275-­283.
2010;16:925-­929.
24) Halazun KJ, Patzer RE, Rana AA, Verna EC, Griesemer AD,
Parsons RF, et al. Standing the test of time: outcomes of a de-
cade of prioritizing patients with hepatocellular carcinoma, re- Supporting Information
sults of the UNOS natural geographic experiment. Hepatology
2014;60:1957-­1962. Additional Supporting Information may be found at
onlinelibrary.wiley.com/doi/10.1002/hep.31819/suppinfo.

2352