Abstract
Synergistic effect of radiotherapy and immunotherapy for the treatment of hepatocellular carcinoma (HCC) has been reported. This phase I/IIa pilot trial evaluated preliminary efficacy and safety of combination of radioembolization with yttrium-90 microspheres (Y90-radioembolization) and durvalumab in patients with locally advanced unresectable HCC.
Patients with Child-Pugh score ≤ 7 and locally advanced HCC, defined as Barcelona Clinic Liver Cancer (BCLC) stage B HCC or BCLC-C disease without extrahepatic metastases, received Y90-radioembolization followed by intravenous durvalumab 1,500 mg 7 to 14 days after Y90-radioembolization and every 4 weeks thereafter. Primary endpoint was time to progression (TTP) assessed by modified RECIST (mRECIST). Secondary endpoints included overall survival (OS), progression-free survival (PFS), objective response rate (ORR) determined by mRECIST, and safety.
All 24 patients enrolled received Y90-radioembolization and 23 received at least one dose of durvalumab. Median follow-up duration was 19.0 months (range, 2.2–24.2). Median TTP was 15.2 months [95% confidence interval (CI), 6.1–not estimated]. Median OS was not reached and 18-month OS rate was 58.3% (95% CI, 36.4–75.0). Median PFS was 6.9 months (95% CI, 5.4–15.2). Seven (29.2%) patients had a complete response and 13 (54.2%) had a partial response; ORR was 83.3% (95% CI, 62.6–95.3). Eleven (47.8%) patients experienced any-grade treatment-related adverse events. Two (8.7%) patients had grade 3 treatment-related adverse events (neutropenia and fever). None experienced any treatment-related serious adverse events.
In patients with locally advanced unresectable HCC, the combination of Y90-radioembolization and durvalumab demonstrated promising efficacy and safety, warranting further evaluation in large-scale controlled trials.
Immune checkpoint inhibitors have emerged as the first-line standard treatment option for advanced hepatocellular carcinoma (HCC). As the evidence on immune activation after radiotherapy for cancer has been gathered, the combination of radioembolization with yttrium-90 microspheres (Y90-radioembolization) and immune checkpoint blockade for the treatment of HCC warrants further investigation. This phase I/IIa pilot trial including 24 patients with locally advanced unresectable HCC found that the combination treatment with Y90-radioembolization and durvalumab was clinically feasible and had encouraging efficacy. Tolerability of this combination was favorable. These findings highlight that the combination treatment with Y90-radioembolization and durvalumab warrants further investigation in large-scale randomized controlled trials.
Introduction
Hepatocellular carcinoma (HCC), the most common primary liver cancer, develops primarily in patients with chronic liver diseases (1, 2). Most patients with HCC present with an unresectable disease at initial diagnosis, rendering locoregional or systemic therapies the mainstay of treatment in these patients (1–4). Transarterial chemoembolization (TACE) is considered as the standard of care for patients at an intermediate-stage HCC [Barcelona Clinic Liver Cancer (BCLC) stage B; refs. 1–3]. Transarterial radioembolization with yttrium-90 microspheres (Y90-radioembolization), another type of transarterial therapy, had a longer time to progression (TTP) than TACE in patients with BCLC-A or -B and a higher response rate and a better safety profile than sorafenib in patients with locally advanced HCC, but failed to show an overall survival (OS) benefit (5–7).
Several multi-tyrosine kinase inhibitors, including sorafenib and lenvatinib, have been adopted for the treatment of unresectable HCC for many years (8, 9). Recently, immune checkpoint inhibitors, including cytotoxic T lymphocyte-associated antigen 4 (CTLA-4), programmed cell death 1 (PD-1), and programmed cell death ligand 1 (PD-L1) inhibitors, have been investigated in patients with unresectable or advanced HCC (10–14). Compared with sorafenib, atezolizumab (anti–PD-L1) plus bevacizumab (anti-vascular endothelial growth factor) demonstrated significantly improved overall and progression-free survival (PFS) outcomes (10, 11). In the phase III HIMALAYA trial, combination of a single high priming dose of tremelimumab (anti–CTLA-4) and durvalumab (anti–PD-L1) showed better OS than sorafenib, and durvalumab monotherapy demonstrated noninferiority to sorafenib for OS (12).
Because the preclinical and clinical evidence on immune activation following radiotherapy or locoregional therapy has been gathered, the need for strategic clinical trials to evaluate the efficacy and safety of locoregional therapy–immunotherapy combinations has emerged (15–23). Thus, we conducted a single-arm, open-label, phase I/IIa pilot trial to investigate the preliminary efficacy and safety of combination treatment with Y90-radioembolization and Durvalumab in patients with locally advanced unresectable HCC (SOLID).
Patients and Methods
Patients
Eligible patients were 19 years of age or older with unresectable HCC confirmed histologically and/or diagnosed radiologically. Patients had locally advanced HCC, which was defined as BCLC-B or -C disease without extrahepatic metastases, Child-Pugh score ≤ 7, Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1, and at least one untreated target lesion per modified RECIST (mRECIST; ref. 24). Prior locoregional treatment or systemic therapy other than immunotherapy was permitted. Patients with main portal vein invasion were eligible. Chronic infection with hepatitis B virus or hepatitis C virus was allowed. Prior to enrollment, a diagnostic angiography and a technetium-99 m (99mTc) macroaggregated albumin scan were performed to ensure eligibility for Y90-radioembolization. On the basis of the 99mTc macro-aggregated albumin scan, patients in whom radiation exposure to the lungs was expected to exceed 30 Gy for a single infusion of Y90 microspheres were excluded. A complete list of eligibility criteria is provided in Sections 4.1 and 4.2 of the protocol in the Supplementary Materials.
All patients provided written informed consent, and the trial protocol was approved by the institutional review board of Seoul National University Hospital (IRB number: H-1906–042–1039). The Good Clinical Practice guidelines of the International Conference on Harmonization and the principles of the Declaration of Helsinki were followed. This trial is registered with ClinicalTrials.gov (ClinicalTrials.gov identifier: NCT04124991) and has been completed.
Study design and procedures
SOLID was a single-center, single-arm, open-label, phase I/IIa pilot trial conducted at the Seoul National University Hospital, Seoul, Republic of Korea to evaluate the preliminary efficacy and safety of Y90-radioembolization followed by durvalumab in patients with locally advanced unresectable HCC. All patients who fulfilled the eligibility criteria were enrolled and underwent transarterial radioembolization with Y90 glass microspheres (TheraSphere; Boston Scientific, Marlborough, MA). The dose of Y90 glass microspheres administered was determined by an interventional radiologist and a nuclear medicine physician based on pretreatment angiography and the 99mTc macro-aggregated albumin scan results (25). At least 30% of the total functional liver volume was planned to be spared from radiation after Y90-radioembolization.
The dosimetry target for patients undergoing lobar treatment was to deliver 100 to 150 Gy to the perfused liver lobe. In patients suitable for selective treatment, the dosimetry target was increased up to 360 Gy (triple the standard dose of 120 Gy) based on the experience of the interventional radiologist, while not exceeding 30 Gy to the lungs (26). For patients with tumors to be treated confined to one or two segments, the dosimetry target was to deliver 400 to 800 Gy to the target tissue for radiation segmentectomy (27). For patients with bilobar disease, if selective treatments were feasible, bilobar tumors were treated in a single session of Y90-radioembolization. Sequential treatment was performed at 1-month intervals for patients with bilobar disease for whom selective treatments were not feasible. To prevent radiation-induced liver disease in these patients undergoing sequential treatment, a standard dose of 100 to 150 Gy was delivered to the dominant lobe with the largest tumor burden, but a reduced dose of 30 to 100 Gy was delivered to the contralateral lobe. On-demand Y90-radioembolization was allowed to be repeated up to two more times at the investigator's discretion during the study period.
Patients received 1,500 mg of durvalumab intravenously 7 to 14 days after Y90-radioembolization. Thereafter, patients received intravenous durvalumab 1,500 mg every 4 weeks (±3 days) until confirmed progression, occurrence of unacceptable toxicity, withdrawal of consent, or other discontinuation criteria were met. Interruption of durvalumab was allowed, but dose reduction was not permitted. Considering potential overlapping toxicity, if additional Y90-radioembolization is performed, the interval between the Y90-radioembolization procedure and durvalumab administration should be at least 1 week.
Tumors were assessed by investigators using mRECIST at baseline (≤28 days before Y90-radioembolization), 8 (±2) weeks after Y90-radioembolization, and every 8 (±1) weeks thereafter until confirmed tumor progression, treatment discontinuation, or death. All images in the study were reviewed by at least one experienced radiologist independently. Safety assessments, including physical examination, vital signs, and ECGs and laboratory tests (if clinically indicated), were performed throughout the treatment period until 30 days after Y90-radioembolization or 30 days after the last dose of durvalumab. Adverse events were monitored and graded according to the NCI Common Terminology Criteria for Adverse Events, version 5.0. The relationship between adverse events and treatment was assessed by investigators.
Outcomes
The primary endpoint was TTP, defined as time from Y90-radioembolization to the date of confirmed progression per mRECIST. Secondary endpoints were OS, defined as time from Y90-radioembolization to the date of death from any cause; OS rates at 6, 12, and 18 months; PFS, defined as time from Y90-radioembolization to the date of confirmed progression per mRECIST or death from any cause; objective response rate (ORR), defined as the proportion of patients achieving a confirmed complete or partial response according to mRECIST; disease control rate, defined as the proportion of patients achieving a confirmed complete or partial response or stable disease according to mRECIST; time to response, defined as time from Y90-radioembolization to the date of first documented response (complete or partial response per mRECIST); duration of response, defined as time from the first documented date of a confirmed complete or partial response until confirmed tumor progression or death from any cause; and safety and tolerability.
Statistical analysis
The sample size for this phase I/IIa pilot trial was not based on formal statistical computation. The study planned to recruit 24 patients, and the trial had a safety run-in cohort of initial 6 patients. If more than 2 patients treated in the safety run-in cohort were to develop treatment-related adverse events leading to treatment discontinuation prior to the second dose of durvalumab, the trial would be stopped early. Otherwise, an additional 18 patients would be enrolled.
Efficacy endpoints were analyzed in the intent-to-treat population, defined as all patients who were enrolled. The Kaplan–Meier method was used to estimate the survival distributions for time-to-event endpoints. Median TTP, OS, PFS, time to response, duration of response, and OS rates at 6, 12, and 18 months were reported with the corresponding two-sided 95% confidence intervals (CI). Safety was assessed in the per-protocol population, defined as all patients who underwent Y90-radioembolization and at least one dose of durvalumab, and presented using descriptive statistics. All statistical analyses were performed using SAS software, version 9.4 (SAS Institute, Cary, NC).
Data availability
The datasets generated and/or analyzed within the SOLID trial are not publicly available due to information that could compromise patient privacy or consent. However, they are available upon reasonable request from the corresponding author, Yoon Jun Kim.
Results
Patients and treatment
From June 12, 2020 to December 22, 2020, 28 patients were screened, of whom 24 patients were enrolled (Fig. 1). All 24 enrolled patients underwent Y90-radioembolization (the intent-to-treat population); 23 of these 24 patients received at least one dose of durvalumab (the per-protocol population). Because 1 patient had a poor performance status 10 days after Y90-radioembolization, the first dose of durvalumab was delayed. On 3-week follow-up, the patient deteriorated to have a Child-Pugh score of 8 (hypoalbuminemia and ascites) and was deemed to be ineligible for durvalumab treatment at the investigator's discretion.
Baseline characteristics of 24 enrolled patients are provided in Table 1. The median age of the patients was 63 years [interquartile range (IQR), 58.5 to 73] and 21 (87.5%) patients were male. Twenty-one (87.5%) patients had a Child-Pugh score of 5, and three (12.5%) had a Child-Pugh score of 6. Eight (33.3%) patients had BCLC-B disease, and 16 (66.7%) had BCLC-C disease. Macrovascular invasion was observed in 15 (62.5%) patients at baseline. Nine (37.5%) patients had unilobar disease, and 15 (62.5%) had bilobar disease. Thirty procedures of Y90-radioembolization were performed; 18 patients (9 with unilobar disease and 9 with bilobar disease) were treated in a single session, and 6 patients received sequential treatment at 1-month intervals. Nine of the 15 patients who had bilobar disease at baseline underwent a single session of selective treatment; 6 patients received sequential treatment. In the first session of Y90-radioembolization, the median prescribed activity of Y90 microspheres was 4.2 GBq (range, 1.9 to 9.4), corresponding to the median dose of 191 Gy (range, 108 to 712) delivered to the target tissue. In the 6 patients who underwent sequential treatment, the median dose delivered to the target tissue in the second session was 69 Gy (range, 40 to 121).
Characteristic . | N = 24 . |
---|---|
Age, median (IQR; range), years | 63 (58.5 to 73; 32 to 80) |
Sex | |
Male | 21 (87.5) |
Female | 3 (12.5) |
ECOG performance status | |
0 | 20 (83.3) |
1 | 4 (16.7) |
Etiology | |
Hepatitis B | 15 (62.5) |
Hepatitis C | 2 (8.3) |
Alcohol | 4 (16.7) |
Hepatitis C and alcohol | 1 (4.2) |
Other | 2 (8.3) |
Child-Pugh class/score | |
A/5 | 21 (87.5) |
A/6 | 3 (12.5) |
BCLC stage | |
B | 8 (33.3) |
C | 16 (66.7) |
Macrovascular invasion | |
No | 9 (37.5) |
Yes | 15 (62.5) |
LN metastasis | |
No | 23 (95.8) |
Yes | 1 (4.2) |
Tumor distribution | |
Unilobar | 9 (37.5) |
Bilobar | 15 (62.5) |
AFP | |
<400 ng/mL | 14 (58.3) |
≥400 ng/mL | 10 (41.7) |
Prior treatment | |
Liver resection | 1 (4.2) |
TACE | 3 (12.5) |
Y90-radioembolization | 1 (4.2) |
Number of liver lesions, median (IQR; range), n | 3 (2 to 3; 1 to 5) |
Size of largest liver lesion, median (IQR; range), mm | 76 (51.5 to 113; 32 to 159) |
Characteristic . | N = 24 . |
---|---|
Age, median (IQR; range), years | 63 (58.5 to 73; 32 to 80) |
Sex | |
Male | 21 (87.5) |
Female | 3 (12.5) |
ECOG performance status | |
0 | 20 (83.3) |
1 | 4 (16.7) |
Etiology | |
Hepatitis B | 15 (62.5) |
Hepatitis C | 2 (8.3) |
Alcohol | 4 (16.7) |
Hepatitis C and alcohol | 1 (4.2) |
Other | 2 (8.3) |
Child-Pugh class/score | |
A/5 | 21 (87.5) |
A/6 | 3 (12.5) |
BCLC stage | |
B | 8 (33.3) |
C | 16 (66.7) |
Macrovascular invasion | |
No | 9 (37.5) |
Yes | 15 (62.5) |
LN metastasis | |
No | 23 (95.8) |
Yes | 1 (4.2) |
Tumor distribution | |
Unilobar | 9 (37.5) |
Bilobar | 15 (62.5) |
AFP | |
<400 ng/mL | 14 (58.3) |
≥400 ng/mL | 10 (41.7) |
Prior treatment | |
Liver resection | 1 (4.2) |
TACE | 3 (12.5) |
Y90-radioembolization | 1 (4.2) |
Number of liver lesions, median (IQR; range), n | 3 (2 to 3; 1 to 5) |
Size of largest liver lesion, median (IQR; range), mm | 76 (51.5 to 113; 32 to 159) |
Note: Data are expressed as number (%) of patients unless indicated otherwise.
Abbreviations: AFP, alpha-fetoprotein; BCLC, Barcelona Clinic Liver Cancer; ECOG, Eastern Cooperative Oncology Group; IQR, interquartile range; LN, lymph node; TACE, transarterial chemoembolization; Y90-radioembolization, transarterial radioembolization with yttrium-90 microspheres.
At the data cutoff date (July 6, 2022), the median duration of follow-up was 19.0 months (IQR, 5.8 to 22.5; range, 2.2 to 24.2). The median number of doses of durvalumab received was 7 (IQR, 4 to 17.5).
Efficacy
During the follow-up period, 11 of 24 patients in the intent-to-treat population experienced disease progression. The median TTP was 15.2 months (95% CI, 6.1–not estimated; Fig. 2A). Eleven patients died and the median OS was not reached (Fig. 2B). The OS rates at 6, 12, and 18 months were 75% (95% CI, 52.6–87.9), 58.3% (95% CI, 36.4–75.0), and 58.3% (95% CI, 36.4–75.0), respectively. Nineteen patients progressed or died: 11 patients progressed and 8 patients died before confirmation of tumor progression. The median PFS was 6.9 months (95% CI, 5.4–15.2; Fig. 2C). The PFS rates at 6, 12, and 18 months were 58.3% (95% CI, 36.5–75.0), 37.5% (95% CI, 19.0–56.0), and 20.8% (95% CI, 7.6–38.5), respectively. Corresponding data in the per-protocol population are shown in Supplementary Fig. S1A–S1C.
Seven (29.2%) patients had a complete response, 13 (54.2%) patients had a partial response, and 2 (8.3%) patients had a stable disease according to mRECIST (Table 2). Two (8.3%) patients died before the first tumor response assessment and were not evaluable: 1 patient deteriorated after Y90-radioembolization and did not receive durvalumab treatment, whereas the other developed hepatic failure 2 weeks after the second dose of durvalumab. Although tumor response could not be assessed, the cause of death in these patients was presumed to be disease progression. Confirmed ORR was 83.3% (95% CI, 62.6–95.3) and disease control rate was 91.7% (95% CI, 73.0–99.0). The median time to response was 2.3 months (95% CI, 2.2–2.4). Of the 20 patients who had an objective response, 15 patients progressed or died until data cutoff; the median duration of response was 7.2 months (95% CI, 3.2–13.0). The ORR and disease control rate in the per-protocol population are provided in Supplementary Table S1. The best response of target lesions of individual patients compared with baseline is illustrated in Fig. 3A; 11 patients showed disappearance of arterial enhancement in all target lesions. Changes in the sum of target lesions over time from baseline are plotted in Fig. 3B.
Parameter . | N = 24 . |
---|---|
Objective response, no. (%) | |
Complete response | 7 (29.2) |
Partial response | 13 (54.2) |
Stable disease | 2 (8.3) |
Progressive disease | 0 (0) |
Not evaluable | 2 (8.3) |
ORR (95% CI) | 83.3% (62.6–95.3) |
Disease control rate (95% CI) | 91.7% (73.0–99.0) |
Parameter . | N = 24 . |
---|---|
Objective response, no. (%) | |
Complete response | 7 (29.2) |
Partial response | 13 (54.2) |
Stable disease | 2 (8.3) |
Progressive disease | 0 (0) |
Not evaluable | 2 (8.3) |
ORR (95% CI) | 83.3% (62.6–95.3) |
Disease control rate (95% CI) | 91.7% (73.0–99.0) |
Abbreviations: CI, confidence interval; mRECIST, modified RECIST; ORR, objective response rate.
The median TTP was 7.7 months (95% CI, 5.7–not estimated) among patients who were at BCLC stage B and 15.2 months (95% CI, 5.9–not estimated) for patients who were at BCLC stage C (Supplementary Fig. S2A). The median OS among patients who were at BCLC stage B was not reached, whereas it was 9.5 months (95% CI, 4.8–not estimated) for patients who were at BCLC stage C (Supplementary Fig. S2B). The OS rate at 12 months was 87.5% (95% CI, 38.7–98.1) for patients who were at BCLC stage B and 43.8% (95% CI, 19.8–65.6) for those who were at BCLC stage C. The median PFS was 6.9 months (95% CI, 4.4–not estimated) among patients who were at BCLC stage B and 7.7 months (95% CI, 4.4–15.2) for patients who were at BCLC stage C (Supplementary Fig. S2C). The PFS rate at 12 months was 37.5% (95% CI, 8.7–67.4) for patients who were at BCLC stage B and 37.5% (95% CI, 15.4–59.8) for those who were at BCLC stage C. Corresponding data in the per-protocol population are shown in Supplementary Fig. S3A–S3C. The ORR was 87.5% (95% CI, 47.4–99.7) for 8 patients with BCLC-B disease and 81.3% (95% CI, 54.4–96.0) for 16 patients with BCLC-C disease (Supplementary Table S2). The disease control rate was 87.5% (95% CI, 47.4–99.7) for patients with BCLC-B disease and 93.8% (95% CI, 69.8–99.8) for those with BCLC-C disease. The ORRs and disease control rates according to BCLC stage in the per-protocol population are provided in Supplementary Table S3.
The median TTP was 15.2 months (95% CI, 5.8–not estimated) among patients with hepatitis B and 11.4 months (95% CI, 5.7–not estimated) for patients with other etiologies (Supplementary Fig. S4A). The median OS was not reached for both subgroups (Supplementary Fig. S4B). The OS rate at 12 months was 53.3% (95% CI, 26.3–74.4) for patients with hepatitis B and 66.7% (95% CI, 28.2–87.8) for those with other etiologies. The median PFS was 6.1 months (95% CI, 4.1–15.2) among patients with hepatitis B and 7.7 months (95% CI, 4.4–not estimated) for patients with other etiologies (Supplementary Fig. S4C). The PFS rate at 12 months was 40.0% (95% CI, 16.5–62.8) for patients with hepatitis B and 33.3% (95% CI, 7.8–62.3) for those with other etiologies. Corresponding data in the per-protocol population are shown in Supplementary Fig. S5A–S5C. The ORR was 86.7% (95% CI, 59.5–98.3) for 15 patients with hepatitis B and 77.8% (95% CI, 40.0–97.2) for 9 patients with other etiologies (Supplementary Table S4). The disease control rate was 93.3% (95% CI, 68.1–99.8) for patients with hepatitis B and 89.9% (95% CI, 51.8–99.7) for those with other etiologies. The ORRs and disease control rates according to etiology in the per-protocol population are provided in Supplementary Table S5.
Safety
Safety analyses included 23 patients who underwent Y90-radioembolization and at least one dose of durvalumab (the per-protocol population). Eleven (47.8%) patients experienced treatment-related adverse events of any grade (Table 3). The most common any-grade treatment-related adverse event was hyperkalemia [grade 2, 2 (8.7%) patients]. Two (8.7%) patients had grade 3 treatment-related adverse events: one developed a grade 3 neutropenia, and the other had a grade 3 fever. Neutropenia, which was deemed to be related to durvalumab, was transient and asymptomatic. Fever, which was related to Y90-radioembolization, was also transient. One (4.3%) patient developed a grade 1 pneumonitis, which was deemed to be related to durvalumab, and resumed durvalumab after receiving glucocorticosteroids. No treatment-related serious adverse event was reported. Adverse events of any grade regardless of attribution were reported by 21 (91.3%) patients (Supplementary Table S6). The most common grade 3 or 4 event was bilirubin elevation [grade 3, 2 (8.7%) patients; grade 4, 1 (4.3%) patient]. Serious adverse events occurred in 7 (30.4%) patients. One patient developed a grade 5 event (aortic dissection), which was considered to be unrelated to treatment. The observed safety profile in the intent-to-treat population is provided in Supplementary Tables S7 and S8.
. | N = 23 . | ||||
---|---|---|---|---|---|
Event . | Grade 1 . | Grade 2 . | Grade 3 . | Grade 4 . | Grade 5 . |
Treatment-related adverse events, no. (%) | |||||
Chills | 0 | 1 (4.3) | 0 | 0 | 0 |
Fever | 0 | 0 | 1 (4.3) | 0 | 0 |
Hyperkalemia | 0 | 2 (8.7) | 0 | 0 | 0 |
Nausea | 1 (4.3) | 0 | 0 | 0 | 0 |
Neutrophil count decreased | 0 | 0 | 1 (4.3) | 0 | 0 |
Palmar-plantar erythrodysesthesia syndrome | 0 | 1 (4.3) | 0 | 0 | 0 |
Pneumonitis | 1 (4.3) | 0 | 0 | 0 | 0 |
Rash | 1 (4.3) | 0 | 0 | 0 | 0 |
Urticaria | 1 (4.3) | 1 (4.3) | 0 | 0 | 0 |
Treatment-related serious adverse events, no. | 0 | 0 | 0 | 0 | 0 |
. | N = 23 . | ||||
---|---|---|---|---|---|
Event . | Grade 1 . | Grade 2 . | Grade 3 . | Grade 4 . | Grade 5 . |
Treatment-related adverse events, no. (%) | |||||
Chills | 0 | 1 (4.3) | 0 | 0 | 0 |
Fever | 0 | 0 | 1 (4.3) | 0 | 0 |
Hyperkalemia | 0 | 2 (8.7) | 0 | 0 | 0 |
Nausea | 1 (4.3) | 0 | 0 | 0 | 0 |
Neutrophil count decreased | 0 | 0 | 1 (4.3) | 0 | 0 |
Palmar-plantar erythrodysesthesia syndrome | 0 | 1 (4.3) | 0 | 0 | 0 |
Pneumonitis | 1 (4.3) | 0 | 0 | 0 | 0 |
Rash | 1 (4.3) | 0 | 0 | 0 | 0 |
Urticaria | 1 (4.3) | 1 (4.3) | 0 | 0 | 0 |
Treatment-related serious adverse events, no. | 0 | 0 | 0 | 0 | 0 |
Note: Data are expressed as number (%).
Discussion
In SOLID, a phase I/IIa pilot trial, the combination of Y90-radioembolization and durvalumab showed encouraging efficacy with a favorable safety profile in patients with locally advanced unresectable HCC. The median TTP was 15.2 months and the median OS was not reached after a median follow-up of 19.0 months. The median PFS was 6.9 months. The objective response (83%) and disease control rates (92%) observed in this study were clinically meaningful. In addition, while up to 50% of patients experienced any-grade treatment-related adverse events during the study, none developed any treatment-related serious adverse events.
SOLID is the first study to evaluate the combination of Y90-radioembolization with an anti–PD-L1 agent in patients with HCC. Y90-radioembolization has been reported to possess definite advantages, such as a longer TTP, higher response rate, and more favorable tolerability profile compared to TACE or sorafenib in patients with BCLC-B or -C HCC; however, it failed to improve OS (5–7). In the phase II SORAMIC trial, Y90-radioembolization in addition to sorafenib was compared with sorafenib monotherapy in patients with advanced HCC (28). Although the combination of Y90-radioembolization with sorafenib elicited potential survival benefits in several subgroups (younger patients ≤ 65 years, non-cirrhotic patients, and patients with HCC of non-alcoholic etiology), it also failed to meet the primary endpoint of OS improvement over sorafenib alone. Thereafter, attempts have been made to combine Y90-radioembolization with immune checkpoint inhibitors that have proven antitumor activity against HCC with an acceptable safety profile. Combination therapy with Y90-radioembolization and anti–PD-1 nivolumab was tested recently in two phase II trials, CA 209–678 and NASIR-HCC (19, 20). In the two studies including patients with unresectable HCC at BCLC stage A–C, the median TTP was 5.6–8.8 months, and the ORR determined by RECIST version 1.1 was reported to be 30.6% to 41.5%. These findings suggest that addition of an immune checkpoint inhibitor to Y90-radioembolization may be an efficacious treatment strategy for the treatment of unresectable HCC. Because the response was assessed using mRECIST in our study, direct comparison of response rates in our study with those of other studies is challenging. However, considering that the ORR by mRECIST was 41.7% in the CA 209–678 trial, the ORR of 83.3% observed in our study is noteworthy.
No obvious signs of synergistic toxicity of Y90-radioembolization in combination with durvalumab were noted in our study. Treatment-related adverse events of any grade were reported by 48% of patients, of which grade 3 or higher treatment-related adverse events were reported by only 2 (9%) patients. Previous clinical trials comparing Y90-radioembolization and sorafenib reported that treatment-related adverse events of any grade occurred in 60% to 77% of patients receiving Y90-radioembolization and grade 3 or higher treatment-related adverse events in 28% to 41%. The low incidence of adverse events related to treatment in our study might be attributed to the better patient selection for Y90-radioembolization and the more selective and personalized approach currently used to deliver Y90 microspheres to tumors (29, 30). In the HIMALAYA trial, treatment-related adverse events of any grade were reported by 52% of patients receiving durvalumab monotherapy, with grade 3 or 4 treatment-related adverse events reported by 13% of patients. Serious adverse events were observed in 30% of patients in the durvalumab monotherapy group; treatment-related serious adverse events were observed in 8% of patients. In our study, serious adverse events occurred in 30% of patients, as in the HIMALAYA trial; however, none were deemed related to treatment. This disparity may be partly explained by the difference in the proportion of patients at an advanced-stage (BCLC stage C) HCC. Therefore, these results suggest that the combination treatment with Y90-radioembolization and durvalumab has a favorable safety profile without any apparent additive or synergistic toxicity.
On the basis of the tolerability of the combination of Y90-radioembolization with immune checkpoint inhibitors, efficacy could be further improved through future study designs that select appropriate patient candidates and deliver a sufficient dose of radiation to tumors. In the CA 209–678 trial evaluating the combination of Y90-radioembolization and nivolumab, the ORR among patients with extrahepatic spread was reported to be 8% in contrast to the ORR of 44% among patients without extrahepatic spread (19). Moreover, among patients without extrahepatic spread, encouraging efficacy (e.g., objective response and disease control rates of 64% and 73%, respectively) was verified even in patients with BCLC-C disease. These results were consistent with the objective response and disease control rates of patients at BCLC stage C in our study excluding patients with extrahepatic metastases. However, among patients who were at BCLC stage C without extrahepatic spread, the median OS and PFS in our study (9.5 months and 7.7 months, respectively) was shorter than those in the CA 209–678 trial (18.2 months and 27.6 months, respectively; ref. 19). Moreover, the median TTP among patients with BCLC-C disease (15.2 months) was longer than that among patients with BCLC-B disease (7.7 months) in our study. This finding might be attributed to the fact that a significant number of patients with BCLC-C disease died before confirmation of tumor progression. These findings can be explained by a higher proportion of patients with macrovascular invasion of HCC [15 (62.5%) patients] and a significant number of patients with Vp3-Vp4 portal vein tumor thrombosis [5 (20.8%) patients; ref. 31]. Collectively, patients with unresectable HCC without extrahepatic spread may be good candidates for further investigation of the combination of Y90-radioembolization and durvalumab, and the optimal indications for this treatment strategy need to be established.
The main limitation of our study was that it was a single-arm study with a small sample size of 24 patients (Supplementary Table S9). Because safety data of combination treatment with Y90-radioembolization and immune checkpoint inhibitors were not available at the phase of clinical trial design, we planned to conduct a pilot study without a control group. On the basis of signs of activity against HCC and a favorable safety profile obtained from the current study, future large-scale controlled studies are warranted. Another major weakness of SOLID was that it was a single-center study in which a highly experienced interventional radiologist performed Y90 radioembolization. A standard dose delivered to the target tissue is considered to be 100 to 150 Gy by single-compartment dosimetry, and the threshold of dose to the tumor that can elicit a favorable response was reported to be 205 Gy by multicompartment dosimetry (32). However, the dose delivered to the target tissue (median, 191 Gy; range 108–712) was higher than the standard dose, and selective treatment was performed in our study, which was thought to contribute to the high response rate and minimal adverse events. Therefore, concerns about the reproducibility of the study results being limited have been raised.
In conclusion, the SOLID trial showed that the combination treatment with Y90-radioembolization and durvalumab was efficacious and tolerable in patients with locally advanced unresectable HCC. The results from our phase I/IIa pilot study underscore the need for further evaluation of this treatment strategy in a randomized controlled trial.
Authors' Disclosures
Y.B. Lee reports grants from Yuhan Pharmaceuticals outside the submitted work. J.H. Yoon reports grants from Roche, AstraZeneca, and Hanmi Pharmaceuticals outside the submitted work. Y.J. Kim reports grants from AstraZeneca and Boston Scientific during the conduct of the study; grants from Yuhan Pharmaceuticals and personal fees from Bayer HealthCare Pharmaceuticals and MSD Korea outside the submitted work. No disclosures were reported by the other authors.
Authors' Contributions
Y.B. Lee: Conceptualization, resources, data curation, formal analysis, investigation, visualization, writing–original draft. J.Y. Nam: Resources, data curation, investigation, writing–review and editing. E.J. Cho: Resources, data curation, investigation, writing–review and editing. J.H. Lee: Resources, data curation, investigation, writing–review and editing. S.J. Yu: Resources, data curation, investigation, writing–review and editing. H.C. Kim: Resources, data curation, formal analysis, validation, investigation, writing–review and editing. J.C. Paeng: Resources, data curation, investigation, writing–review and editing. J.H. Yoon: Resources, data curation, investigation, writing–review and editing. Y.J. Kim: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, investigation, writing–review and editing.
Acknowledgments
The SOLID trial was an investigator-initiated study funded by AstraZeneca and Boston Scientific, who also provided the study drugs. We thank all patients, their families and caregivers, and all investigators for contribution to this trial. We also thank the Medical Research Collaborating Center (MRCC) at Seoul National University for assistance in the statistical analyses.
The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).