US RESiN Registry for the Study and Evaluation of Patients Treated With SIR-Spheres

Abstract: Background: Yttrium-90 (Y90) is increasingly used to treat both primary and secondary liver tumors that are ineligible for resection. In the United States, Y90 resin microspheres (SIR-Spheres; Sirtex Medical) are FDA approved for the treatment of nonresectable colorectal metastases with adjuvant floxuridine chemotherapy, but they are frequently used off label for other liver-dominant malignancies. Data on its real-world use and associated outcomes are limited. Objective: To describe the Radiation-Emitting SIR-Spheres in Non-resectable Liver Tumor (RESiN) registry, a national, multicenter database that will enroll patients with primary or secondary liver cancer who are scheduled for treatment with Y90 resin microspheres as part of their care plan. Methods: We aim to enroll at least 500 patients per year from 50 academic and community hospitals nationwide. Patients will be included without regard to tumor type or additional therapies and followed longitudinally. Patient demographics, primary and secondary diagnoses, medical history, treatment details, laboratory measures, tumor markers, complications, and outcomes will be entered into the database using a secure online interface. Efficacy outcomes will include time to progression and overall survival, and toxicities will be tracked. For patients with colorectal or neuroendocrine tumors, subgroup analyses will be performed to identify factors that influence outcomes in order to develop personalized treatments. For less common applications, we expect that registry data will help in planning future clinical trials. Conclusions: This registry will provide information on United States treatment patterns and long-term outcomes with Y90 resin microspheres and will guide future research with this therapy, especially for conditions with limited data.

Key words: liver cancer; radioembolization; selective internal radiation therapy (SIRT); registry

______________________________________________________________________

Tumors of the liver are either primary or metastatic, with secondary tumors being more common.^1-5 The incidence of and deaths from primary liver cancers (hepatocellular carcinoma [HCC] and intrahepatic cholangiocarcinoma) have steadily increased over the past 25 years.⁶ In 2016, an estimated 39,000 individuals in the United States will be diagnosed with primary liver cancers, and more than 27,000 will die of these diseases.¹ Colorectal cancer (CRC), neuroendocrine tumors (NETs), and pancreatic adenocarcinoma are among the primary cancers with high rates of metastasis to the liver. Colorectal cancer is the third most common cancer type in the United States,¹ and liver metastasis occurs in approximately 60% of cases; in fact, liver metastases are the major cause of morbidity and mortality in patients with CRC.⁵ Although NETs are rare, with an incidence of about 35/100,000, the rate has been increasing for several decades.⁷ The liver is also a common site of NET metastases, which are associated with poor prognosis.⁸ Similarly, of the approximately 49,000 incidences of pancreatic cancer that occur each year,² more than 60% metastasize to the liver,3 and the presence of hepatic metastases can greatly affect disease progression and survival in patients with pancreatic cancer.⁹ Breast and lung cancers metastasize to the liver less frequently (5% to 35% of cases),⁴ but because these are common cancers, each occurring in more than 200,000 Americans each year,¹ they also contribute to the high incidence of secondary liver tumors.

Most primary and secondary liver tumors are diagnosed at advanced stages, when curative resection or transplantation are no longer an option.^10-12 Furthermore, intrahepatic recurrence is common in those who do undergo liver resection with curative intent.^13,14 Locoregional therapies, such as ablative or transarterial procedures, are preferred over systemic therapy for patients with HCC who are not candidates for curative treatments¹⁵ and are important treatment options for patients with other primary and secondary hepatic tumors. Locoregional therapies are particularly accepted for treatment of certain tumors such as NETs.

Radiation doses of 100 Gy or more are needed to effectively treat liver tumors, but external beam radiation is limited by the sensitivity of the healthy liver to radiation: radiation-induced liver disease can emerge with doses of 50 Gy or less.^12,16,17 In contrast, selective internal radiation therapy (SIRT) with yttrium-90 (Y90)–labeled microspheres can deliver doses of up to 150 Gy to the tumor while sparing the normal liver parenchyma.¹² This is possible because tumors are preferentially, and extensively, supplied by the hepatic artery, whereas normal tissue receives most of its blood supply via the portal vein.¹²

SIR-Spheres Y90 resin microspheres (Sirtex Medical) were approved for use in the United States in 2002 for the treatment of unresectable colorectal metastases in conjunction with adjuvant intrahepatic artery chemotherapy with floxuridine.^18,19 They have also been approved in Australia,²⁰ the European Union (CE Mark), and several other countries with a broader indication for the treatment of patients with nonoperable liver tumors. These resin microspheres are delivered by catheter into the branches of the hepatic artery that feed the tumors. Each microsphere is 20-60 μm in diameter and contains about 50 Bq.^12,18 The small size allows the spheres to penetrate deep into the tumor vasculature, but limits actual embolization;^12,21 the tumoricidal effect is due to the continuous delivery of radiation rather than to an ischemic effect.²² Y90 is a pure beta emitter with a 2.6-day half-life, and the beta particles penetrate a mean of 2.5 mm into the tissue.¹² Thus, the radiation from the microspheres is contained within the body, primarily within the tumor, and decays within 11 days after treatment.¹⁸

Numerous trials have examined the use of Y90 resin microspheres alone or in combination with systemic chemotherapy for the treatment of primary and metastatic liver tumors (reviewed by Lee et al¹² and Khajornjiraphan et al⁴), but most of the clinical trials to date have been small, usually enrolling fewer than 100 patients. In the first large randomized trial of Y90 radioembolization (N=530 in the intent-to-treat population),²³ the addition of Y90 resin microspheres to fluorouracil, leucovorin, and oxaliplatin (FOLFOX6)–based chemotherapy as a first-line therapy in patients with liver-only or liver-predominant CRC metastases significantly delayed progression in the liver (median 20.5 months vs 12.6 months, hazard ratio 0.69, 95% CI .55-.90; P=.002) and increased the complete response rate (6.0% vs 1.9%, P=.02), although progression-free survival (PFS) was not significantly extended when all sites were considered.²³ Overall survival (OS) data from this study will be combined with data from 2 similar, ongoing phase III trials (FOXFIRE²⁴ and FOXFIRE Global [NCT01721954]) and are expected to be reported in 2017.²³ Additional phase III trials are investigating the use of Y90 resin microspheres in conjunction with systemic treatments as maintenance therapy for metastatic CRC (SIR-step, NCT01895257) and comparing Y90 resin microspheres with sorafenib in locally advanced or advanced HCC (SIRveNIB, NCT01135056; SARAH, NCT01482442). Ongoing phase II trials include studies of Y90 resin microspheres in the treatment of HCC (NCT01126645), intrahepatic cholangiocarcinoma (NCT01798147), metastatic CRC (NCT02195011), and metastatic uveal melanoma (NCT01473004, NCT01893099).

However, clinical trials involving resin Y90 will not represent the full range of patients treated with this device in clinical practice, which may include patients ineligible for inclusion in trials or who are treated off-label. Patient registries are gaining acceptance as a means to collect data on care provided to more diverse patient populations over longer periods of time than the typical clinical trial; this information can increase the body of knowledge on safety and effectiveness and produce results that are more directly applicable to clinical practice.^25-27 In addition, because registries do not involve randomization of patients to specific treatment protocols, they can be used to examine physician practices, such as patient selection and their effects on outcomes.²⁸

To better understand how Y90 resin microspheres are used in practice in the United States and to more thoroughly assess the benefits and risks of this therapeutic modality, Vanderbilt University, in collaboration with Sirtex Medical Limited, is establishing the Radiation-Emitting SIR-Spheres in Non-resectable Liver Tumor (RESiN) registry. This registry will be used to prospectively follow outcomes of patients treated with Y90 resin microspheres in both academic- and community-based comprehensive cancer treatment centers across the US, regardless of primary tumor type or prior treatment. The data collected will also be used to examine patient, disease, and treatment characteristics that predict response in subgroups of patients and to design prospective studies in populations currently too small for subgroup analyses.

Patients and Methods

Inclusion Criteria for Patients and Centers

The RESiN registry (https://clinicaltrials.gov/ct2/show/NCT02685631) will collect data on patients ≥18 years of age with primary or secondary liver cancer who are scheduled for a first treatment with Y90 resin microspheres as part of their cancer care at participating centers. The registry will not entail research on the experimental use of the microspheres. 

Academic, tertiary care, and community hospitals will be solicited to participate in the registry. Because we wish to include centers with proven expertise in the technique, only centers with demonstrated consistent volumes of resin Y90 cases will be eligible. To ensure that patients with a variety of demographic characteristics, tumor types, and previous therapies are included, centers across the United States will initially be involved, with a baseline goal of enrolling at least 500 patients per year. No upper limit will be set on the number of patients in the registry.

Assessments and Treatment With Y90 Resin Microspheres

Before receiving treatment with Y90 resin microspheres, all patients will receive a physical examination and radiographic evaluation. Baseline liver functions, complete blood cell count, and, when applicable, tumor markers will be assessed to determine the status of the patient and liver (Table 1). Previous therapies and hepatic interventions will be documented. The decision to treat patients will be made by the treating physician based on the standard of care at each institution.

Following the protocols and procedures of the participating centers, patients will undergo an initial arteriogram before administration of the Y90 resin microspheres to map the hepatic vasculature feeding the targeted tumor(s), to identify possible vessels that supply the stomach or small bowel that should be avoided or embolized and to evaluate the amount of shunting through the tumor(s) to the lungs. The dose will be determined using each site’s standard protocol. The microspheres will be delivered through an arterial catheter in an outpatient procedure.

Data Collection, Ethics, and Security

Participating centers must receive approval from their respective institutional review boards before involvement with the RESiN registry. Each patient will give written informed consent to participate in the registry and can withdraw this consent at any time. In the event that a patient withdraws, the registry, by regulation, will retain the ability to use all information collected prior to the revocation of patient authorization. Withdrawal from the registry will not affect planned treatments with Y90 resin microspheres.

The RESiN registry is a secure, online, longitudinal database using Research Electronic Data Capture (REDCAP). All data will be de-identified, and each participant will be given a unique identification number so that their data and outcomes can be tracked in a longitudinal manner. The confidentiality of patients’ personal data will be protected in accordance with the Health Insurance Portability and Accountability Act of 1996 and national data protection laws, as applicable.

Data on each participant will be entered into the registry by the treating center via a secure online portal and will include information obtained before, during, and after treatment (Table 1, Figure 1). The data entered will include demographic information, primary and secondary diagnoses, medical history related to the diagnosis, treatment details, adverse events, and treatment outcomes. This information will be elicited by structured questions, with answers limited to fixed-choice options where possible. Free-text fields will also be available in which to enter patient notes or to provide additional information. The use of mandatory fields will ensure that critical information, such as type of liver cancer, tumor location, prior treatments, and the area targeted for SIRT is captured for all patients.

Governance and Oversight

Vanderbilt University Medical Center is the primary/coordinating site for this national registry. The Vanderbilt-Ingram Cancer Center Multi-Institutional Coordinating Office will provide ongoing monitoring of the entered data for accuracy and completeness.

Data Analysis and Outcomes

Data analysis will occur at prespecified intervals based on accrual, with less common tumors potentially evaluated following a smaller number of treatments. The analysis will be managed by the Vanderbilt Center for Quantitative Sciences, under the supervision of Dr. Yu Shyr, Director.

Adverse events (AEs), including serious AEs, will be tracked using the Common Terminology Criteria for Adverse Events version 4. We will specifically solicit information about the occurrence and grade of AEs that have previously been associated with the use of Y90 resin microspheres,^12,23,29,30 including abdominal pain; fever; nausea; vomiting; gastrointestinal ulceration; radioembolization-induced liver disease; and radiation pneumonitis, cholecystitis, pancreatitis, and dermatitis, as well as select abnormal laboratory values. Each site will also be responsible for reporting AEs to their local institutional review board in accordance with standard institutional practice.

Efficacy outcomes that will be tracked include PFS, survival without progression in the liver (liver PFS), OS, and the tumor response rate in the liver and at extrahepatic sites. Institutional protocols will be used to determine progressive disease and tumor responses from serial computed tomography or magnetic resonance imaging scans. Progression-free survival will be defined as time from the date of treatment to confirmation of disease progression at any site (or in the liver for liver PFS) or death from any cause if this occurs before disease progression is documented.

For patients with metastatic CRC or NETs, which have relatively high rates of treatment with Y90 resin microspheres, we will perform subgroup analyses to identify patient and disease characteristics associated with favorable outcomes or high rates of toxicity. These analyses will allow us to identify potential markers of response to treatment and to develop personalized approaches to Y90 resin microsphere treatment, including planning trials focused on combination treatments with targeted therapies. For liver tumors that are treated with SIRT less frequently, including metastases from breast and pancreatic cancer and cholangiocarcinoma, we will use the registry data to develop hypotheses and to plan future trials with appropriate power to investigate such associations.

Discussion

Despite the increasing use of SIRT in the US, most clinical trials of this treatment modality have been relatively small,^4,12 and patients treated in the community may differ substantially from those who enroll in trials in terms of primary disease type, treatment history, performance status, or other patient or disease characteristics. Combining patient populations from varied treatment settings nationwide into a single registry will contribute to our understanding of how Y90 resin microspheres are currently being used by the oncology community. Pooling data from these patients will also provide the power to conduct subanalyses of patients, particularly those with metastatic CRC or NETs. Our aim is to identify factors that influence patient outcomes to develop personalized treatment strategies that maximize efficacy and minimize toxicity as much as possible for patients receiving Y90 resin microsphere therapy. In groups with fewer patients, such as patients with metastatic breast or lung cancer, we expect that the data will yield hypothesis-generating observations that can be used to plan future studies.

Patient registries are gaining acceptance by health care providers and regulatory agencies as important tools for collecting and assessing the benefits and risks of therapies in the complex patient populations typically found in the cancer care setting.²⁷ The nonrandomized design of registries enables investigation into factors outside of the realm of clinical trials, such as patient selection and prescribing practices, and registries may therefore generate information that is more relevant to actual clinical practice than the more restricted results of trials.²⁵ For example, the Australian TRACC (Treatment of Recurrent and Advanced Colorectal Cancer) registry was established to prospectively capture data on real-world treatment practices for metastatic CRC. Early results from TRACC identified patient factors that influenced clinician prescribing practices as well as the effect of performance status on chemotherapy selection and subsequent survival outcomes.³¹ More recently, data from TRACC were used to show that first-line therapy with Y90 resin microspheres plus chemotherapy extended OS relative to chemotherapy alone in patients with liver metastases from CRC (24.3 months vs 19.3 months, hazard ratio = 0.73, P=.038).32 Because the RESiN registry is dedicated to treatment with Y90 resin microspheres, we expect that it will yield extensive insight into the use and outcomes of SIRT, well beyond what a single-indication study with a narrowly defined patient population can show.

To limit selection bias, a large number of both academic- and community-based centers geographically distributed across the United States will be included, with the presence of experienced practitioners as the only inclusion criterion.²⁵ Indeed, we expect to include a heterogeneous study population that will be representative of patients treated with Y90 resin microspheres, thereby bridging the gap between the clinical literature and clinical practice. In addition, SIRT registries have been established in the United Kingdom (The British Society of International Radiology’s SIRT registry, https://www.bsir.org/registries/sirt-registry/) and Europe (Cardiovascular and Interventional Radiological Society of Europe [CIRSE] Registry for SIR-Spheres Therapy, https://cirse.org/index.php?pid=1175). The potential therefore exists to compare treatment practices across additional locations in the future, or to pool results from the registries to increase the power to associate patient characteristics with safety and effectiveness outcomes.

The prospective collection of standardized data will improve data quality²⁸ and facilitate comparisons across different patients and factors that might affect outcomes. However, some subgroup analyses may be limited by small sample size. Data from a similar noninterventional registry of patients with HCC treated with sorafenib (GIDEON—Global Investigation of Therapeutic Decisions in Hepatocellular Carcinoma and of Treatment with Sorafenib) were used to investigate the safety of administering transarterial chemoembolization at different times relative to the administration of sorafenib.²⁷ Due to small subgroup sizes, OS could not be reported for all of the groups examined, and the authors interpreted the results cautiously due to likely selection bias in the patients who received each treatment. Nonetheless, the study was able to identify coadministration of transarterial chemoembolization and sorafenib as both a common approach to treatment and one that is safe to use in appropriate patients.³³ When analyzing data from RESiN, we will use preliminary findings from small subgroups to formulate hypotheses that can be tested in future adequately powered studies.

In summary, the RESiN registry will be a resource that will provide the oncology community with a more comprehensive understanding of the benefits and risks of Y90 resin microspheres as a localized treatment for a range of hepatic tumors and patients, including patients who would likely not be included in clinical trials. The registry will also guide future research into Y90 resin microsphere therapy, especially for conditions for which limited data currently exist.

Editor’s note: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no financial relationships or conflicts of interest regarding the content herein.

Acknowledgement: The authors thank Naomi Ruff, PhD, of Eubio Medical Communications LLC for providing medical writing support, funded by Sirtex Medical Inc.

Manuscript received April 13, 2016; manuscript accepted May 4, 2016.

Address for correspondence: Daniel B. Brown, MD, Department of Radiology and Radiologic Sciences, Vanderbilt University, 1161 21st Avenue S., CCC-1118 Medical Center North, Nashville, TN 37232, United States.

Suggested citation: Banovac F, Brown DB. US RESiN registry for the study and evaluation of patients treated with SIR-Spheres. Intervent Oncol 360. 2016;4(6):E101-111.

References

1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66(1):7-30.
2. American Cancer Society. Cancer Facts & Figures 2015. Atlanta: American Cancer Society 2015. 
3. Disibio G, French SW. Metastatic patterns of cancers: results from a large autopsy study. Arch Pathol Lab Med. 2008;132(6):931-939.
4. Khajornjiraphan N, Thu NA, Chow PK. Yttrium-90 microspheres: a review of its emerging clinical indications. Liver Cancer. 2015;4(1):6-15.
5. Memon K, Lewandowski RJ, Mulcahy MF, et al. Radioembolization for neuroendocrine liver metastases: safety, imaging, and long-term outcomes. Int J Radiat Oncol Biol Phys. 2012;83(3):887-894.
6. National Cancer Institute. SEER stat fact sheets: liver and intrahepatic bile duct cancer. https://seer.cancer.gov/statfacts/html/livibd.html. Accessed March 3, 2016.
7. Yao JC, Hassan M, Phan A, et al. One hundred years after “carcinoid”: epidemiology of and prognostic factors for neuroendocrine tumors in 35,825 cases in the United States. J Clin Oncol. 2008;26(18):3063-3072.
8. Saha S, Hoda S, Godfrey R, Sutherland C, Raybon K. Carcinoid tumors of the gastrointestinal tract: a 44-year experience. South Med J. 1989;82(12):1501-1505.
9. Kim HW, Lee JC, Paik KH, Lee YS, Hwang JH, Kim J. Initial metastatic site as a prognostic factor in patients with stage IV pancreatic ductal adenocarcinoma. Medicine (Baltimore). 2015;94(25):e1012.
10. Cao CQ, Yan TD, Bester L, Liauw W, Morris DL. Radioembolization with yttrium microspheres for neuroendocrine tumour liver metastases. Br J Surg. 2010;97(4):537-543.
11. Dhanasekaran R, Limaye A, Cabrera R. Hepatocellular carcinoma: current trends in worldwide epidemiology, risk factors, diagnosis, and therapeutics. Hepat Med. 2012;4:19-37.
12. Lee EW, Thakor AS, Tafti BA, Liu DM. Y90 selective internal radiation therapy. Surg Oncol Clin N Am. 2015;24(1):167-185.
13. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005;42(5):1208-1236.
14. Poon R, Fan S, Wong J. Risk factors, prevention, and management of postoperative recurrence after resection of hepatocellular carcinoma. Ann Surg. 2000;232(1):10-24.
15. National Comprehensive Cancer Network. NCCN Guidelines: Pancreatic Adenocarcinoma, Version 2.2105. 2015. https://www.nccn.org/professionals/physician_gls/pdf/pancreatic.pdf
16. Dawson LA, Normolle D, Balter JM, McGinn CJ, Lawrence TS, Ten Haken RK. Analysis of radiation-induced liver disease using the Lyman NTCP model. Int J Radiat Oncol Biol Phys. 2002;53(4):810-821.
17. Pan CC, Kavanagh BD, Dawson LA, et al. Radiation-associated liver injury. Int J Radiat Oncol Biol Phys. 2010;76(3):S94-S100.
18. SIR-Spheres^® microspheres (Yttrium-90 Microspheres) [package insert]. North Sydney, New South Wales, Australia; Sirtex Medical Limited; 2014.
19. U.S. Food and Drug Administration. U.S. Food and Drug Administration. Recently-Approved Devices: SIR-Spheres^®- P990065. 2002. https://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/DeviceApprovalsandClearances/Recently-ApprovedDevices/ucm083605.htm. Accessed March 11, 2016.
20. Australian Government Department of Health Therapeutic Goods Administration. SIR-Spheres: ARTG ID 149332. 2008. https://www.tga.gov.au/artg/artg-id-149332. 
21. Riaz A, Salem R. Radioembolization for liver tumors. In: WR Jarnagin, ed. Blumgart’s Surgery of Liver, Biliary Tract and Pancreas. 5th ed. Philadelphia, PA: Elsevier; 2012.
22. Bilbao JI, de Martino A, de Luis E, et al. Biocompatibility, inflammatory response, and recannalization characteristics of nonradioactive resin microspheres: histological findings. Cardiovasc Intervent Radiol. 2009;32(4):727-736.
23. van Hazel GA, Heinemann V, Sharma NK, et al. SIRFLOX: Randomized phase III trial comparing first-line mFOLFOX6 (plus or minus bevacizumab) versus mFOLFOX6 (plus or minus bevacizumab) plus selective internal radiation therapy in patients with metastatic colorectal cancer. J Clin Oncol. 2016;4(15):1723-1731.
24. Dutton SJ, Kenealy N, Love SB, Wasan HS, Sharma RA; FOXFIRE Protocol Development Group and the NCRI Colorectal Clinical Study Group. FOXFIRE protocol: an open-label, randomised, phase III trial of 5-fluorouracil, oxaliplatin and folinic acid (OxMdG) with or without interventional Selective Internal Radiation Therapy (SIRT) as first-line treatment for patients with unresectable liver-only or liver-dominant metastatic colorectal cancer. BMC Cancer. 2014;14:497.
25. Gliklich R, Dreyer N, Leavy M, eds. Registries for Evaluating Patient Outcomes: A User’s Guide. 3rd ed. Two volumes. (Prepared by the Outcome DEcIDE Center [Outcome Sciences, Inc., a Quintiles company] under Contract No. 290 2005 00351 TO7.) AHRQ Publication No. 13(14)-EHC111. April 2014. https://www.effectivehealthcare.ahrq.gov/registries-guide-3.cfm. 
26. Malmenäs M, Lowton K, Morin I, et al. Analysis of effectiveness in patient registry data. ISPOR Connections. June 2009. https://www.ispor.org/news/articles/june09/aep.asp. Accessed March 31, 2016.
27. U.S. Food and Drug Administration. Balancing premarket and postmarket data collection for devices subject to premarket approval: guidance for industry and food and drug administration staff. 2015. https://www.fda.gov/downloads/medicaldevices/deviceregulationandguidance/guidancedocuments/ucm393994.pdf. 
28. Gliklich R, Dreyer NA, Leavy MB, Velentgas P, Khurana L. Standards in the conduct of registry studies for patient-centered outcomes research: a guidance document for the patient-centered outcomes research institute. 2012. https://www.pcori.orghttps://s3.amazonaws.com/HMP/hmp_ln/imported/Standards-in-the-Conduct-of-Registry-Studies-for-Patient-Centered-Outcomes-Research1.pdf. 
29. Mosconi C, Cappelli A, Pettinato C, Golfieri R. Radioembolization with yttrium-90 microspheres in hepatocellular carcinoma: role and perspectives. World J Hepatol. 2015;7(5):738-752.
30. Mahnken AH, Pereira PL, de Baere T. Interventional oncologic approaches to liver metastases. Radiology. 2013;266:407-430.
31. Field K, Wong HL, Shapiro J, et al. Developing a national database for metastatic colorectal cancer management: perspectives and challenges. Intern Med J. 2013;43:1224-1231.
32. Australian mCRC patient registry analysis of SIR-Spheres Microspheres utilisation presented at the AGITG annual meeting [press release]. Sirtex; September 3, 2015.
33. Geschwind JF, Gholam P, Goldenberg A, et al. Use of transarterial chemoembolization (TACE) and sorafenib in patients with unresectable hepatocellular carcinoma: US regional analysis of the GIDEON registry. Liver Cancer. 2016;5(1):37-46.