Yttrium-90 (Y-90) transarterial radioembolization (TARE), also known as selective internal radiation therapy (SIRT), has been used for the treatment of primary and secondary liver cancer for decades. This therapy has greatly advanced since the seminal publications on the clinical, technical, and procedural aspects of Y-90 therapy.1-3 Numerous subsequent studies have confirmed the safety and efficacy of Y-90 therapy, resulting in significant growth in its use. Multidisciplinary recognition of the value of Y-90 therapy is highlighted by its inclusion in the National Comprehensive Cancer Network guidelines as a category 2A recommended therapy for colon and rectal cancer liver metastases.4 This, as well as a key institution’s multidisciplinary team’s decision to choose Y-90 as first-line therapy for hepatocellular carcinoma (HCC),5 should lead to continued growth and adoption. This article describes technical and procedural innovations in Y-90 radioembolization and introduces promising next-generation investigational technologies for this well-established and effective therapy.


Y-90 TARE has been historically considered a palliative therapy option for patients with unresectable liver malignancies deemed unsuitable for other locoregional therapies; however, more recent publications have explored the prospect of developing TARE techniques with curative intent. Radiation segmentectomy delivers enough radiation to ablate an entire vascular territory, resulting in radionecrosis of tumor and liver tissue, analogous to surgical resection.6 Administering this ablative dose can thus destroy all tumor in the treated volume and provide a margin of tissue destruction surrounding the tumor akin to surgical resection. Additionally, adjacent satellite lesions not visible on imaging as well as any micro- or macrovascular invasion in the volume can be adequately treated.

Published recommendations for radioembolization now include Y-90 ablation techniques in treatment algorithms,7 and these procedures continue to gain clinical adoption as subsequent studies have demonstrated long time to progression (TTP) and meaningful long-term overall survival (OS) rates for patients with limited tumor burden and preserved hepatic function.8 The PREMIERE trial, a prospective randomized trial of glass microsphere radioembolization versus conventional transarterial chemoembolization (cTACE) with doxorubicin for patients with unresectable, nonablatable (percutaneously) HCC, helped introduce segmental (rather than lobar) radioembolization treatments.9 This study revealed a TTP > 26 months for radioembolization compared with 6.8 months for cTACE. This work supporting the benefits of segmental radioembolization over segmental cTACE was corroborated by a retrospective, propensity-matched analysis10 and has led to more widespread application of segmental radioembolization, especially in the bridge to liver transplant setting. Two separate radiology-pathology correlative studies revealed high rates of complete pathologic necrosis at explantation when the segmental radiation dose based on volume of perfusion exceeded 190 Gy.11,12

Radiation segmentectomy also lends itself to same-day radioembolization—mapping angiography, lung shunt evaluation, and radioembolization performed in the same session.13 This paradigm-changing approach makes radioembolization more competitive to other intra-arterial embolotherapies, is convenient for patients, and helps reduce health care costs.

An atrophy-hypertrophy complex was noted when radioembolization was performed in a lobar fashion for patients with unilobar disease where, over time, the treated hepatic lobe atrophies and there is compensatory hypertrophy of the untreated contralateral hepatic lobe.14 This concept is now being employed to facilitate hepatic resection for select patients presenting with inadequate future liver remnants. As an alternative to portal vein embolization, radiation lobectomy allows for synchronous cancer treatment during the time interval to hypertrophy. A modified radiation lobectomy technique includes a segmental ablation dose (boost) to the targeted tumor with the intent to deliver a complete pathologic response.15


The standard method for dosimetry is the body surface area (BSA) model for resin radiomicrospheres (SIR-Spheres, Sirtex Medical Inc.) and the medical internal radiation dose (MIRD) model for glass radiomicrospheres (TheraSphere, Boston Scientific Corporation). The partition model, which requires a tumor-to-normal (T/N) uptake ratio, is theoretically more accurate and allows for personalized dosimetry because it accounts for tumor avidity. However, the latter assumes uniform distribution of injected activity in the tumor and in the normal liver, which is not the case. Recent advances in dosimetry have been made with the application of dose calculation treatment planning software to radioembolization procedures that allow for voxel-based dosimetry.

A voxel is simply a three-dimensional pixel obtained through positron emission tomography (PET)/CT, single-photon emission CT (SPECT), or SPECT/CT and has a calculated value that can be converted into dose (Gy). There are several dose calculation software systems available including MIM SurePlan LiverY90 (MIM Software Inc.), Planet Dose (DOSIsoft), Simplicit90Y (Mirada Medical/Boston Scientific Corporation), and Rapidsphere (Varian Medical Systems, Inc.). These systems allow for automated volume segmentation, measurement of hepatic volumes and tumor volumes, calculation of lung shunts, multimodal image registration, pre- and postimplantation dose calculation, dosimetry comparison, and measurement of treatment response.

Figure 1. Dose-volume histogram following radioembolization demonstrating radiation dose to volume of three separate tumors and normal hepatic parenchyma. Voxel-based dosimetry is performed on the basis of postprocedure bremsstrahlung SPECT/CT or PET/CT scan.

Courtesy of MIM Software Inc.

Voxel-based dosimetry allows for the creation of a dose-volume histogram, which is the gold standard in radiation oncology for radiation therapy planning and dose measurement (Figure 1). The treatment planning software systems evaluate and measure the distribution of Y-90 radiomicrospheres in different compartments and corresponding doses to different structures. Postimplantation dosimetry with this advanced software provides patient-specific information with respect to actual dose delivered to different tumors, whether there was adequate tumoral coverage, and how much dose was delivered to normal liver tissue. Although technetium-99 macroaggregated albumin (Tc-99m MAA) has limitations, some studies have shown reasonable agreement between Tc-99m MAA SPECT and post–Y-90 PET/CT.16 Therefore, there is the potential to use treatment planning software for preimplantation activity determination. Finally, this technology may be valuable as a decision-making tool in patients undergoing multiple therapies (ie, combination of external beam radiation and Y-90 or multiple Y-90 therapies) to determine cumulative dose to different compartments with the goal of optimizing tumor response and minimizing toxicity to normal liver (Figure 2).

Figure 2. Voxel-based cumulative dosimetry for total dose estimation for multiple treatment planning procedures. Safety of repeat radioembolization, for example, may be determined by using treatment planning software to determine actual dose delivered to both tumor and normal liver from previous therapy. These data may be combined with predictive dosimetry to both compartments on the basis of a Tc-99m MAA SPECT scan to optimize prescribed activity to ensure tumor response while preventing hepatic toxicity.

Courtesy of DOSIsoft.

The primary mechanism of action with TARE is related to β radiation emission from the Y-90 microspheres working in unison to create tumoricidal radiation “clouds” and a “crossfire” effect.17,18 As a result, TARE is driven by the amount and distribution of radioactivity administered and not necessarily the number of microspheres. General recommendations with respect to clinical activity and dose plans have been established by the Radioembolization Brachytherapy Oncology Consortium.19 The conceptual convergence of two techniques, one providing precise and optimized dose to elicit tumor response considered the “conventional” transarterial radioembolization (C-TARE) technique versus the other intended to radioablate the entire targeted vascular territory (including both the tumor and the liver) considered “ablative” transarterial radioembolization (A-TARE) technique, can be resolved by the concepts of segmentation and targeting of tumor versus liver parenchyma. The discussion of segmentation and partition has conceptually improved the understanding of the relationship between particle deposition, activity, and tissue response. The interrelationship between activity, tissue susceptibility, target/tumor volume, and microsphere distribution that dictates the micro- and macroscopic radiation dose has been previously described.20 The intent of C-TARE is to target and optimize the deposition of radioactivity to the tumor while minimizing the amount of radiation (ie, dose) to the native liver parenchyma. This stepwise approach to C-TARE is intended to optimize tumor treatment while minimizing the potential for hepatic decompensation and complications related to excessive nontargeted embolization. With A-TARE, complete ablation and necrosis of the entire vascular territory is desired, dictated by high exposure of radiation to the normal liver in the target region and not limited by optimal dose to the tumor.

With regard to tumor dose thresholds, a study evaluating glass Y-90 C-TARE outcomes for HCC with macrovascular invasion using Tc-99m MAA SPECT–based personalized dosimetry and dose intensification demonstrated improved OS if the tumor dose was > 205 Gy.21 Ideally, this tumor dose threshold should be considered for both C-TARE and A-TARE using glass Y-90 with the awareness to reduce the tumor dose toward 190 Gy with C-TARE if there is risk of toxicity to the normal liver. Resin Y-90 C-TARE has similar outcomes using BSA dosimetry, resulting in a tumor dose threshold > 100 Gy.22 For resin A-TARE, a partition dose > 70 Gy to the target region is desired, inclusive of tumor and normal liver, because 70 Gy is the threshold beyond which destruction of normal liver is anticipated.7 This strategy is used anecdotally by two authors (Drs. Gandhi and Liu) for resin A-TARE applications. Table 1 summarizes these thresholds.

Greater understanding of the concepts of segmentation, partition, and the relationship between particle deposition, activity, and tissue response can be gained through use of the DAVYR (Dosimetry and Activity Visualizer for Yttrium-90 Radioembolization) app (Figure 3). This free app is available for download on iOS and Android platforms and provides a concise tabulated data set of the currently used techniques and additionally includes an A-TARE calculator (“RAD SEG” button). Once the user enters patient and tumor parameters (height, weight, liver volume, target volume, T/N ratio), the application displays calculated dosimetry for MIRD, BSA, and partition and provides a direct comparison between these models. The tabulated data sets clearly illustrate the relationship between administered radioactivity (GBq) and estimated dose in the respective segment (lung, liver, or tumor), which clarifies a key point: extrahepatic toxicity is a result of deposition of microspheres in nontarget organs (stomach, lung, pancreas), whereas hepatic toxicity is a result of excessive radiation to normal liver parenchyma for which the nonexposed parenchyma is unable to adequately compensate. The user can adjust parameters such as desired target tumor dose to understand the resultant effect on normal liver and lung.

Figure 3. C-TARE dosimetry is easily derived from MIRD, BSA, and partition models using the DAVYR app (A). Initial clinical parameters are entered (orange arrows; iPad screen format illustrated), which then determines overall activity (yellow arrow), as well as the individual compartment distribution (green arrow), resulting in estimated dose to target compartments (blue arrow). A-TARE using the same parameters by activating the “RAD SEG” button (green arrow), the app recalculates MIRD target activity for glass (yellow arrow) and for partition (blue arrow) based on the parameters as outlined in Table 1 (B).

To help optimize treatment and tailor to patient requirements, Y-90 manufacturers now offer more treatment options to accommodate clinical need for personalized therapy. Ensuring adequate radiomicrosphere coverage of tumor at an appropriate level of specific activity can optimize tumoricidal dose to the tumor. Users can now choose customized treatments to address varying scenarios, including (1) subsegmental treatment of small volume with a small quantity of radiomicrospheres at standard specific activity; (2) varying volumes with small or high radiomicrosphere quantity at high specific activity for segmentectomy; (3) small, medium, or large quantity at standard activity for general lobar treatment depending on target volume and tumor burden; (4) large quantity at higher specific activity for radiation lobectomy; and (5) numerous intervening combinations to tailor treatment to the specific clinical scenario.

Use of multiple glass Y-90 dose vials at one treatment setting as well as use of second-week, extended-shelf-life doses provide many permutations of quantity and activity and additional flexibility in determining day and time of treatment. Resin Y-90, traditionally provided as a dose vial calibrated to provide 3 GBq on the evening of the treatment day from which the prescribed dose is drawn at the time of administration, is now provided up to 3 days precalibration and up to 1 day after calibration. This results in extensive options in desired activity and microsphere quantity at the time of administration, particularly when including the ability to draw a specific dose from the supplied vial. Interventional radiologists can now provide a tailored dose at essentially any day and time desired. Providing a dose vial well before the calibration date and time also allows resin Y-90 to be offered at higher specific activity for A-TARE applications. With regard to specific activity, the long-held view that glass Y-90 has specific activity of 2,500 Bq/sphere versus resin Y-90 at 50 Bq/sphere should be clarified. These values are realistic at the time of calibration for each product; however, the actual day and time of calibration significantly differs between the two devices, resulting in the marked difference. Instead, the focus should be on specific activity at the time of administration, because that is what is clinically relevant. In fact, with the new resin offerings that provide the ability to treat 3 days before the calibration day, the specific activity per microsphere for resin Y-90 starts to approach mid to late second-week specific activity per microsphere for glass Y-90.


Posttherapy implanted Y-90 activity can be evaluated through SPECT/CT detection of bremsstrahlung emission or through Y-90 positron emission through time-of-flight PET/CT. Although SPECT/CT and PET/CT have greatly contributed to our understanding of dose effect to tumor and normal tissue, there are limitations related to resolution, blind spots, and misregistration.23 Holmium-166 (Ho-166) radiomicrospheres (QuiremSpheres, Terumo Europe), originally developed at Utrecht University, have CE Mark designation in Europe and are an investigational product in the United States. Specifications compared to Y-90 are listed in Table 2.

Although Ho-166 emits a lower level of β energy, has lesser depth of penetration, and a shorter half-life compared with Y-90, it has shown promise in initial clinical trials.24 Additional study is needed to determine whether Ho-166 can provide long-term outcomes that are comparable to Y-90. Ho-166 does provide diagnostic γ emission that can be imaged with SPECT or SPECT/CT postimplantation. In addition to allowing for more accurate assessment of absorbed dose to tumor and detection of nontarget embolization after treatment, a “scout” dose can be used in lieu of Tc-99m MAA for assessment of a lung shunt and for predictive dosimetry at the time of workup. Ho-166 is also paramagnetic, which allows for quantification in target tissue using magnetic resonance–based R2* imaging protocols. However, neither Ho-166 nor currently available Y-90 products are clinically radiopaque, which prevents intraprocedural visualization of microsphere deposition on current C-arm fluoroscopy systems or cone-beam CT (CBCT) used to guide the procedure.

Building on the development of inherently radiopaque bland glass microspheres,25 an investigational Y-90 product (ABK Biomedical Inc.) has been developed in which radiopacity is an integral feature of Y-90 administration. Preclinical testing has confirmed strong x-ray fluoroscopic and CBCT radiopacity (Figure 4). This technology may provide intraprocedural real-time confirmation of tumor targeting. On an investigational basis, dosimetry software algorithms have been modified and utilized to detect, map, and quantify radiopacity on CBCT (Figure 4B). Use of this software and proprietary calibration phantoms may allow for the ability to quantify the mass of microspheres deposited in tumor tissue at time of treatment, and by using average specific activity, the dose delivered to tumor could potentially be calculated as either critical threshold levels or, ultimately, as total dose delivered to tumor. The deposited microspheres could additionally serve as fiducials to guide complementary treatments such as thermal ablation or external beam radiation.

Figure 4. Implanted investigational Y-90 microspheres in swine right renal artery demonstrating strong radiopacity of the implanted microspheres (white arrowheads) (A). No contrast media has been used and all radiopacity visualized is due to microsphere deposition. Microsphere radiopacity can be compared to the radiopaque tip of the microcatheter (white arrowhead). CBCT of swine kidney after investigational radiopaque Y-90 implantation. Modified MIM SurePlan LiverY90 software was used to outline any renal tissue above 210 Hounsfield unit (HU) threshold (red outline) (B). Surrounding normal renal tissue below 210 HU is displayed by the software in comparison to normal renal tissue below the 210 HU threshold (blue fill).

Courtesy of MIM Software and ABK Biomedical Inc.


As our understanding of dosimetry continues to evolve, so have the delivery systems. Recent publications have demonstrated that alteration of inflow dynamics, through decreases in inflow pressure or occlusion, may result in diversion and optimization of particle deposition in the tumor.26 Furthermore, steerable microcatheter platforms allow for multipoint administration without the dangers of multiple microcathethers. Due to the advances in design, new generations of microcatheters allow for not only more precise selection for anatomic optimization of payload delivery but also the ability to alter pressure dynamics to maximize delivery to the tumor (in the case of low vascular capacitance tumors) or maximize exposure to the targeted volume as per A-TARE. Figure 5, Figure 6, Figure 7, and Figure 8 show some of these devices, and their features are summarized in Table 3.

Figure 5. The Occlusafe occlusive microballoon catheter with end-hole balloon allows for complete proximal occlusion during administration. Small internal diameter may present challenges with high-volume particle administration.

Courtesy of Terumo Europe.

Figure 6. The Sniper high-flow microcatheter with end-hole balloon and preshaped tips allow for complete proximal occlusion during administration.

Courtesy of Embolx, Inc.

Figure 7. The Surefire infusion system with unique basket pin-pull deployment system provides both proximal protection/antireflux, with continued systolic perfusion, allowing for antegrade flow and delivery. High flow (requires > 0.054-inch inner diameter guide or base catheter) and standard internal diameter models (0.038-inch compatible).

Courtesy of TriSalus Life Sciences.

Figure 8. The SwiftNinja steerable microcatheter is an actively steerable microcatheter through a ratchet wheel, allowing for selection and repositioning without a wire or removal of device. Limited lengths preclude the transradial approach.

Courtesy of Merit Medical Systems, Inc.


Growing evidence demonstrating positive clinical outcomes with Y-90 radioembolization therapy, the new mindset to achieve cure using Y-90 ablative techniques, recent software and microcatheter developments, and the promising future of investigational imageable radiomicrospheres will further our understanding of Y-90 therapy and ultimately should improve patient outcomes.

1. Salem R, Thurston KG. Radioembolization with 90Yttrium microspheres: a state-of-the-art brachytherapy treatment for primary and secondary liver malignancies. Part 1: technical and methodologic considerations. J Vasc Interv Radiol. 2006;17:1251-1278.

2. Salem R, Thurston KG. Radioembolization with 90yttrium microspheres: a state-of-the-art brachytherapy treatment for primary and secondary liver malignancies. Part 2: special topics. J Vasc Interv Radiol. 2006;17:1425-1439.

3. Lewandowski RJ, Sato KT, Atassi B, et al. Radioembolization with 90Y microspheres: angiographic and technical considerations. Cardiovasc Intervent Radiol. 2007;30:571-592.

4. Benson AB 3rd, D’Angelica MI, Abrams TA, et al. Hepatobiliary cancers, version 2.2014. J Natl Compr Cancer Netw. 2014;12:1152-1182.

5. Salem R, Gabr A, Riaz A, et al. Institutional decision to adopt Y90 as primary treatment for hepatocellular carcinoma informed by a 1,000-patient 15-year experience. Hepatology. 2018;68:1429-1440.

6. Riaz A, Gates VL, Atassi B, et al. Radiation segmentectomy: a novel approach to increase safety and efficacy of radioembolization. Int J Radiat Oncol Biol Phys. 2011;79:163-171.

7. Lau WY, Kennedy AS, Kim YH, et al. Patient selection and activity planning guide for selective internal radiotherapy with yttrium-90 resin microspheres. Int J Radiat Oncol Biol Phys. 2012;82:401-407.

8. Lewandowski RJ, Gabr A, Abouchaleh N, et al. Radiation segmentectomy: potential curative therapy for early hepatocellular carcinoma. Radiology. 2018;287:1050-1058.

9. Salem R, Gordon AC, Mouli S, et al. Y90 radioembolization significantly prolongs time to progression compared with chemoembolization in patients with hepatocellular carcinoma. Gastroenterology. 2016;151:1155-1163.e2.

10. Padia SA, Johnson GE, Horton KJ, et al. Segmental yttrium-90 radioembolization versus segmental chemoembolization for localized hepatocellular carcinoma: results of a single-center, retrospective, propensity score-matched study. J Vasc Interv Radiol. 2017;28:777-785.e1.

11. Vouche M, Habib A, Ward TJ, et al. Unresectable solitary hepatocellular carcinoma not amenable to radiofrequency ablation: multicenter radiology-pathology correlation and survival of radiation segmentectomy. Hepatology. 2014;60:192-201.

12. Ahmed AF, Samreen N, Grajo JR, et al. Angiosomal radiopathologic analysis of transarterial radioembolization for the treatment of hepatocellular carcinoma. Abdom Radiol (NY). 2018;43:1825-1836.

13. Gabr A, Ali R, Al Asadi A, et al. Technical aspects and practical approach toward same-day Y90 radioembolization in the management of hepatocellular carcinoma. Tech Vasc Interv Radiol. 2019;22:93-99.

14. Kim RD, Kim JS, Watanabe G, et al. Liver regeneration and the atrophy-hypertrophy complex. Semin Intervent Radiol. 2008;25:92-103.

15. Gabr A, Polineni P, Mouli SK, et al. Neoadjuvant radiation lobectomy as an alternative to portal vein embolization in hepatocellular carcinoma. Semin Nucl Med. 2019;49:197-203.

16. Garin E, Lenoir L, Rolland Y, et al. Dosimetry based on 99mTc-macroaggregated albumin SPECT/CT accurately predicts tumor response and survival in hepatocellular carcinoma patients treated with 90Y-loaded glass microspheres: preliminary results. J Nucl Med. 2012;53:255-263.

17. Morgan B, Kennedy AS, Lewington V, et al. Intra-arterial brachytherapy of hepatic malignancies: watch the flow. Nat Rev Clin Oncol. 2011;8:115-120.

18. Gulec SA, Sztejnberg ML, Siegel JA, et al. Hepatic structural dosimetry in (90)Y microsphere treatment: a Monte Carlo modeling approach based on lobular microanatomy. J Nucl Med. 2010;51:301-310.

19. Kennedy A, Nag S, Salem R, et al. Recommendations for radioembolization of hepatic malignancies using yttrium-90 microsphere brachytherapy: a consensus panel report from the radioembolization brachytherapy oncology consortium. Int J Radiat Oncol Biol Phys. 2007;68:13-23.

20. Liu D, Westcott M, Garcia-Monaco R, et al. Down and dirty with dosimetry. Endovasc Today. 2016;15:70-74, 76.

21. Garin E, Lenoir L, Edeline J, et al. Boosted selective internal radiation therapy with 90Y-loaded glass microspheres (B-SIRT) for hepatocellular carcinoma patients: a new personalized promising concept. Eur J Nucl Med Mol Imaging. 2013;40:1057-1068.

22. Kao YH, Steinberg JD, Tay YS, et al. Post-radioembolization yttrium-90 PET/CT: part 2—dose-response and tumor predictive dosimetry for resin microspheres. EJNMMI Res. 2013;3:57.

23. Kao YH, Steinberg JD, Tay YS, et al. Post-radioembolization yttrium-90 PET/CT—part 1: diagnostic reporting. EJNMMI Res. 2013;3:56.

24. Prince JF, van den Bosch MAAJ, Nijsen JFW, et al. Efficacy of radioembolization with 166Ho-microspheres in salvage patients with liver metastases: a phase 2 study. J Nucl Med. 2018;59:582-588.

25. Kehoe SK, Langman M, Werner-Zwanziger U, et al. Mixture designs to assess composition-structure-property relationships in SiO2-CaO-ZnO-La2O3-TiO2-MgO-SrO-Na2O glasses: potential materials for embolization. J Biomater Appl. 2013;28:416-433.

26. van den Hoven AF, Prince JF, Samim M, et al. Posttreatment PET-CT-confirmed intrahepatic radioembolization performed without coil embolization, by using the antireflux Surefire Infusion System. Cardiovasc Intervent Radiol. 2014;37:523-528.

Robert Abraham, MD, FRCPC, FSIR
Professor, Department of Diagnostic Radiology
QEII Health Sciences Center
Dalhousie University Medical School
Halifax, Nova Scotia, Canada
Disclosures: Chief Medical Officer of and shareholder in ABK Biomedical Inc.

Robert Lewandowski, MD, FSIR
Professor of Radiology, Medicine, and Surgery
Director of Interventional Oncology
Feinberg School of Medicine
Northwestern University
Chicago, Illinois
Disclosures: Consultant to BTG Plc; advisory board for ABK Biomedical Inc.

Ripal T. Gandhi, MD, FSIR, FSVM
Associate Clinical Professor
Miami Cardiac and Vascular Institute
Miami Cancer Institute
Miami, Florida
Disclosures: Proctor for and consultant to Sirtex Medical; consultant to ABK Biomedical Inc.

David M. Liu, MD, FRCPC, FSIR
Interventional Radiologist, EVA Vein Care
Clinical Professor
University of British Columbia
Vancouver, British Columbia, Canada
Disclosures: Consultant to Sirtex Medical and ABK Biomedical Inc.; Director of Merit Medical’s Board of Directors.