Indications
| Indication | Agent | Notes |
|---|---|---|
| HCC — unresectable | TheraSphere or SIR-Spheres | BCLC B/C; bridging to transplant; downstaging; preferred over TACE when portal vein thrombosis present |
| HCC + portal vein thrombosis | TheraSphere (FDA HDE approved) | Major advantage over TACE — Y-90 does not require portal vein flow for safety |
| Colorectal liver metastases | SIR-Spheres (FDA approved) | FDA approval with floxuridine (FUDR) chemotherapy; also used in salvage setting |
| Other hepatic metastases | Either | Neuroendocrine tumors (NET), breast, cholangiocarcinoma — off-label but widely used |
Contraindications
| Type | Contraindication |
|---|---|
| Absolute | LSF >20% (lung dose >30 Gy) on MAA scan · GI shunting without ability to coil-protect target vessels · Child-Pugh C / bilirubin >2 · Tumor burden >50% liver replacement · Life expectancy <3 months |
| Relative | LSF 15–20% — reduce dose, multidisciplinary discussion · Prior external beam radiation to liver · Severely compromised renal function (contrast load) · Uncorrectable coagulopathy (arterial access) |
TheraSphere vs. SIR-Spheres
| Feature | TheraSphere (Glass) | SIR-Spheres (Resin) |
|---|---|---|
| FDA approval | HCC (HDE); HCC + portal vein thrombosis | Colorectal mets with floxuridine (FUDR) |
| Sphere count | ~2.5 million per dose | ~40–60 million per dose |
| Specific activity | Higher per sphere | Lower per sphere |
| Flow dynamics | Less embolic | More embolic (higher sphere count) |
| Dosimetry method | MIRD or partition model | BSA or partition model (target high T:N ratio) |
Hepatic Arterial Anatomy
Standard and Variant Anatomy
Normal: Celiac trunk → common hepatic artery → proper hepatic artery → right + left hepatic arteries. HCC vascularity is 90%+ hepatic arterial (vs. 70% portal for normal liver) — this differential vascularity enables selective tumor delivery.
Replaced right hepatic artery from SMA (~15%): Must be catheterized separately for right lobe treatment. Missing this on SMA angiogram results in an undertreated right lobe and unprotected GI vessels in that territory.
Replaced left hepatic from left gastric (~5%): Must identify and coil-protect — left gastric artery branches can deliver spheres to the stomach.
Dome lesions: May recruit supply from inferior phrenic arteries — include in mapping assessment.
Two-Visit Protocol
The following is a high-level summary. Full mapping angiogram technique, protective embolization decision-making, TheraSphere and SIR-Spheres dosimetry calculations, and Y-90 delivery protocols are available in RadCall Pro.
Visit 1 — Mapping Angiogram
- Coordinate Tc-99m MAA 24–48h ahead — order with nuclear medicine directly; MAA must be ready for injection at time of angiogram
- Celiac + SMA angiogram: AP and oblique views — hepatic bifurcation, variant anatomy, replaced vessels
- Selective hepatic angiogram + CBCT: Map tumor hypervascularity; identify all non-target vessels; plan treatment territory (lobar vs. selective)
- Protective coil embolization: Right gastric artery (#1 GI complication cause) when in treatment field; cystic artery if gallbladder in field; falciform artery; replaced left hepatic if arising from LGA. Note: routine protective coiling is falling out of favor — coil only if significant supply to treatment field.
- MAA injection: Position microcatheter in planned treatment position — identical to Y-90 delivery position. Inject Tc-99m MAA. Patient immediately to nuclear medicine for planar + SPECT imaging.
MAA Scan Interpretation
| LSF (%) | Lung Dose Implication | Action |
|---|---|---|
| <10% | Minimal lung dose | Proceed with standard planned activity |
| 10–15% | Moderate lung dose | Proceed with reduced activity; recalculate with partition model |
| 15–20% | Approaching lung dose limit | Further dose reduction; multidisciplinary team discussion |
| >20% | Exceeds lung threshold | ABSOLUTE CONTRAINDICATION — do not treat |
Also check: no gastric uptake · no duodenal uptake · hepatic distribution matches planned treatment territory. Any extrahepatic GI uptake → return for additional vessel protection before treatment.
LSF calculation: LSF = lung counts ÷ (liver + lung counts) × 100%. Lung dose (Gy) = (LSF/100) × Total activity (GBq) × 50 Gy/GBq. Single treatment lung dose limit: <30 Gy. Cumulative (multiple treatments): <50 Gy.
Visit 2 — Treatment Session (Week 2–4)
- Review MAA scan results and confirm LSF <20%
- Confirm ordered activity and dosimetry with radiation safety
- Re-angiogram to confirm anatomy and that coil-protected vessels remain occluded
- Position microcatheter in identical position to MAA injection — catheter position is the dosimetry delivery location; movement changes distribution
- CBCT to confirm position and tumor coverage before delivery
- Post-treatment Bremsstrahlung SPECT or PET-CT confirms treatment territory
Complications
| Complication | Rate | Prevention / Management |
|---|---|---|
| GI ulceration | 2–4% without protection; <1% with coil protection | Right gastric artery coil embolization; PPI post-procedure |
| Radiation pneumonitis | <1% with LSF <20% | MAA scan screening; dose reduction for elevated LSF; steroids if symptomatic |
| REILD | Rare | Radioembolization-induced liver disease — elevated LFTs, jaundice, liver failure 4–8 weeks post; from excessive dose to non-tumor liver; partition model dosimetry reduces risk |
| Post-radioembolization syndrome | Common | Fatigue, nausea, low-grade fever 1–3 weeks; self-limited; supportive care |
| Biliary injury | Uncommon | Bile duct stenosis or biloma; more common in patients with biliary anatomy alterations |
Post-Procedure Care
After Mapping Visit
- Standard post-arterial access care; minimal radiation precautions (MAA T½ = 6h)
- Communicate LSF result and plan to referring oncologist
- Schedule treatment visit 2–4 weeks later
After Treatment Visit
- Radiation precautions per institutional protocol — Y-90 is a pure beta emitter; close-contact restrictions × 7 days (sleep alone; avoid prolonged close contact with children or pregnant women)
- Anti-emetics, dexamethasone if post-embolization syndrome concern; PPI for gastroprotection
- Fatigue counseling: expected 1–4 weeks post; activity restrictions
- LFTs weekly × 4 weeks: ALP and GGT commonly rise transiently (expected). Significant bilirubin rise beyond 8 weeks → consider REILD
- Imaging follow-up at 4–6 weeks: MRI or CT liver; use mRECIST or LI-RADS TR (not RECIST 1.1 size criteria); restart systemic chemotherapy at ~4 weeks post-treatment
Dosimetry Reference
- BSA method (TheraSphere — older): A (GBq) = [0.0217 × mean absorbed dose (Gy)] × BSA. Target dose: 80–120 Gy for HCC.
- Partition model (preferred — personalized dosimetry): Uses MAA SPECT to calculate liver, tumor, and lung dose separately. Most accurate; increasingly standard of care.
- MIRD / quadrant approach (SIR-Spheres): Activity based on liver volume and tumor involvement percentage.
Evidence Base for Y-90 Radioembolization
Y-90 TARE is an established locoregional therapy for hepatic malignancies. The central paradigm shift of the past decade is personalized dosimetry — achieving ≥205 Gy to tumor dramatically improves outcomes compared to the standard dosimetry used in earlier negative trials, which explains the apparent discrepancy between phase 2 and phase 3 data.
1. Guideline Recommendations
NCCN HCC (v1.2026):[1]
- Y-90 included among arterially directed therapies for liver-confined, unresectable HCC
- Delivery of ≥205 Gy to tumor may be associated with increased OS; >400 Gy to ≤25% liver recommended for CTP-A patients
- Radiation segmentectomy should be considered for anatomically limited disease
- Safe with limited portal vein invasion (segmental/lobar PVT); relatively contraindicated with bilirubin >3 mg/dL or main PVT
- RCTs have shown Y-90 is not superior to sorafenib for advanced HCC (see below for context)
AASLD (2023):[2]
- Y-90 radiation segmentectomy as an alternative to thermal ablation for BCLC-A HCC not candidates for resection, including tumors >3 cm (Level 3, Strong Recommendation)
- Y-90 as an alternative to TACE for BCLC-B HCC (Level 3, Strong Recommendation)
- Personalized dosimetry should be used in all future Y-90 treatments and trials
2. Landmark Trials
| Trial | Design | Key Results |
|---|---|---|
| DOSISPHERE-01[3,4] | Phase 2 RCT (n=60); personalized (≥205 Gy) vs. standard dosimetry (120±20 Gy); locally advanced HCC ≥7 cm | ORR: 71% vs. 36% (p=0.007); OS: 24.8 vs. 10.7 mo (p=0.02); downstaging to surgery: 35% vs. 3.5%. Long-term follow-up (65.8 mo): OS benefit sustained (HR 0.51). Patients downstaged to resection had >50% OS at 5 years |
| LEGACY[5] | Single-arm, 162 pts, CTP-A, solitary HCC ≤8 cm; radiation segmentectomy with glass microspheres | ORR 88.3% (mRECIST); DoR ≥6 mo: 76.1%; 3-yr OS: 86.6% overall, 92.8% in neoadjuvant patients who underwent resection/transplant → FDA approval 2021 |
| RASER[6] | Prospective single-arm; radiation segmentectomy for very early to early unresectable HCC | High rates of complete pathological necrosis and durable local control; validated radiation segmentectomy as curative-intent approach |
3. Y-90 vs. TACE
| Study | Design | Key Results |
|---|---|---|
| Salem et al.[7] | Phase 2 RCT; Y-90 vs. cTACE, BCLC A–B (n=45) | TTP: >26 vs. 6.8 mo (HR 0.12, p=0.001); similar OS |
| TRACE[8] | Phase 2 RCT; Y-90 vs. DEB-TACE, BCLC A–B (n=72) — stopped early for efficacy | TTP: 17.1 vs. 9.5 mo (HR 0.36, p=0.002); OS: 30.2 vs. 15.6 mo (HR 0.48, p=0.006) |
| 2025 Meta-analysis[9] | 6 studies, N≈443 | OS: HR 0.68 (95% CI 0.55–0.86, p=0.0009); PFS: HR 0.54 (p<0.00001) — both favoring Y-90 |
| Núñez et al. (2026 PSM)[10] | Y-90 personalized dosimetry vs. DEB-TACE (n=258) | CR: 71% vs. 33%; ORR: 88% vs. 58%; 1-yr retreatment: 12% vs. 40%; superior results in multifocal disease |
4. Y-90 vs. Sorafenib — Why Three Phase 3 RCTs Were Negative
| Trial | N | Y-90 OS | Sorafenib OS | HR / P |
|---|---|---|---|---|
| SARAH[11] | 459 | 8.0 mo | 9.9 mo | HR 1.15, p=0.18 |
| SIRveNIB[12] | 360 | 8.8 mo | 10.0 mo | HR 1.1, p=0.36 |
| SORAMIC[13] | 424 | 12.1 mo (SIRT + sora) | 11.4 mo | HR 1.01, p=0.95 |
Critical caveats: All three trials used resin microspheres with standard dosimetry — no personalized dosimetric endpoints. 20–29% of patients randomized to Y-90 never received treatment. Post-hoc reanalysis of SARAH: patients receiving >100 Gy to tumor had OS 14.1 vs. 6.1 months.[15] These trials reflect the failure of standard dosimetry, not the failure of Y-90.
5. Treatment Approaches
Radiation Segmentectomy (RS) — ablative delivery to ≤2 hepatic segments:[16,17]
- Recognized as potentially curative for early HCC (BCLC 0/A), solitary tumors ≤8 cm
- Can treat subcapsular, subdiaphragmatic, and peri-cardiac tumors not amenable to thermal ablation
- Resin RS: tumors receiving ≥300 Gy had 17% vs. 61% progression at 2 years (p=0.047)[18]
Radiation Lobectomy (RL) — lobar treatment with dual goals:[19]
- Local tumor control + contralateral lobe hypertrophy (can substitute for portal vein embolization)
- Bridge to resection when future liver remnant is inadequate
Key Dosimetry Thresholds:[1,2,3,21]
| Threshold | Clinical Significance |
|---|---|
| ≥205 Gy to tumor | DOSISPHERE-01 threshold for improved response and OS (glass microspheres) |
| ≥300 Gy to tumor | Associated with reduced disease progression in radiation segmentectomy |
| >400 Gy to ≤25% liver | NCCN recommendation for CTP-A patients |
| >600 Gy | Proposed for HCC with portal vein invasion (median OS 49.5 months)[22] |
| Normal liver ≤40 Gy | Safe limit for non-tumoral liver[21] |
| ≥120 Gy minimum tumor dose | Recommended for resin microspheres in HCC and metastases[21] |
6. Y-90 for Colorectal Liver Metastases
NCCN Colon Cancer Guidelines include Y-90 in three settings: radiation lobectomy as PVE alternative, chemotherapy-refractory disease with predominant hepatic mets, and radiation segmentectomy for small unresectable tumors.
EPOCH Trial (phase 3 RCT; Y-90 + 2nd-line chemo vs. chemo alone, n=428):[24] PFS 8.0 vs. 7.2 mo (HR 0.69, p=0.001); hepatic PFS 9.1 vs. 7.2 mo (HR 0.59, p<0.001); median OS 15.0 mo overall, 17.4 months when used as 2nd-line therapy. Tumor dose ≥120 Gy independently predicted improved OS.[26]
7. Y-90 for Portal Vein Invasion
Study of 48 patients with unilobar HCC + portal vein invasion (CTP-A) treated with glass microspheres:[22] ORR 83% (mRECIST); median OS 47.2 months; tumor dose >586 Gy: OS 49.5 vs. 21.9 mo (p=0.021). Supports Y-90 as frontline treatment for localized PVI with preserved liver function — a setting where TACE is contraindicated.
8. Patient Selection — Key Contraindications
AASLD-specified factors for Y-90 unsuitability:[2] lung shunt >25 Gy single treatment or >30 Gy cumulative; non-target treatment zone including uncorrectable gastric/duodenal branches; main portal vein tumor thrombus (Vp4); ALBI 2–3 with disease beyond segmental zone; bilirubin >2 mg/dL for lobar treatment; Child-Pugh ≥B8; >70–75% liver involvement; ECOG >2; ascites.
Bottom line for MDT: Y-90 is not one treatment — it is several. Radiation segmentectomy (curative intent, early HCC), radiation lobectomy (bridge/FLR augmentation), and lobar TARE (intermediate/advanced HCC) have distinct evidence bases and dosimetric targets. The negative vs. sorafenib trials reflect standard dosimetry, not Y-90 failure. Personalized dosimetry achieving ≥205 Gy to tumor is now the standard expectation per AASLD and NCCN.
References
- National Comprehensive Cancer Network. Hepatocellular Carcinoma (v1.2026). Updated 2026-03-10.
- Singal AG, Llovet JM, Yarchoan M, et al. AASLD Practice Guidance on Prevention, Diagnosis, and Treatment of Hepatocellular Carcinoma. Hepatology. 2023;78(6):1922–1965.
- Garin E, Tselikas L, Guiu B, et al. Personalised Versus Standard Dosimetry Approach of SIRT in Patients With Locally Advanced HCC (DOSISPHERE-01). Lancet Gastroenterol Hepatol. 2021;6(1):17–29.
- Garin E, Tselikas L, Guiu B, et al. Long-Term Overall Survival After SIRT for Locally Advanced HCC: Updated Analysis of DOSISPHERE-01. J Nucl Med. 2024;65(2):264–269.
- Salem R, Johnson GE, Kim E, et al. Yttrium-90 Radioembolization for the Treatment of Solitary, Unresectable HCC: The LEGACY Study. Hepatology. 2021;74(5):2342–2352.
- Kim E, Sher A, Abboud G, et al. Radiation Segmentectomy for Curative Intent of Unresectable Very Early to Early Stage HCC (RASER). Lancet Gastroenterol Hepatol. 2022;7(9):843–850.
- Salem R, Gordon AC, Mouli S, et al. Y90 Radioembolization Significantly Prolongs Time to Progression Compared With Chemoembolization in Patients With HCC. Gastroenterology. 2016;151(6):1155–1163.
- Dhondt E, Lambert B, Hermie L, et al. Y-90 Radioembolization Versus Drug-Eluting Bead Chemoembolization for Unresectable HCC: Results From the TRACE Phase II RCT. Radiology. 2022;303(3):699–710.
- de Alcântara JPTL, Götz GWXDR. Transarterial Radioembolization With Y-90 Versus Conventional TACE for HCC: A Systematic Review and Meta-Analysis. Acad Radiol. 2025;32(11):6739–6750.
- Núñez K, Hasani N, Cronan J, et al. Radioembolization (90Y) Achieves Higher Response Rates and Reduces Progression Risk Compared With DEB-TACE in HCC. Hepatol Commun. 2026;10(5):e0935.
- Vilgrain V, Pereira H, Assenat E, et al. Efficacy and Safety of SIRT With Y-90 Resin Microspheres Compared With Sorafenib in Locally Advanced HCC (SARAH). Lancet Oncol. 2017;18(12):1624–1636.
- Chow PKH, Gandhi M, Tan SB, et al. SIRveNIB: Selective Internal Radiation Therapy Versus Sorafenib in Asia-Pacific Patients With HCC. J Clin Oncol. 2018;36(19):1913–1921.
- Ricke J, Klümpen HJ, Amthauer H, et al. Impact of Combined SIRT and Sorafenib on Survival in Advanced HCC (SORAMIC). J Hepatol. 2019;71(6):1164–1174.
- Thornton LM, Abi-Jaoudeh N, Lim HJ, et al. Combination and Optimal Sequencing of Systemic and Locoregional Therapies in HCC. J Vasc Interv Radiol. 2024;35(6):818–824.
- Lawhn-Heath C, Hope TA, Martinez J, et al. Dosimetry in Radionuclide Therapy: The Clinical Role of Measuring Radiation Dose. Lancet Oncol. 2022;23(2):e75–e87.
- Lewandowski RJ, Serhal M, Padia SA, et al. The Evolving Application of Radiation Segmentectomy for the Treatment of Hepatic Malignancy. Radiology. 2025;316(1):e240333.
- Salem R, Padia SA, Toskich BB, et al. Radiation Segmentectomy for Early Hepatocellular Carcinoma Is Curative. J Hepatol. 2025;82(6):1125–1132.
- Sarwar A, Malik MS, Vo NH, et al. Efficacy and Safety of Radiation Segmentectomy With 90Y Resin Microspheres for HCC. Radiology. 2024;311(2):e231386.
- Badar W, Yu Q, Patel M, Ahmed O. Transarterial Radioembolization for Management of HCC. Oncologist. 2023;:oyad327.
- Yu Q, Khanjyan M, Fidelman N, Pillai A. Contemporary Applications of Y90 for the Treatment of HCC. Hepatol Commun. 2023;7(10):e0288.
- Levillain H, Bagni O, Deroose CM, et al. International Recommendations for Personalised SIRT of Primary and Metastatic Liver Diseases With Y-90 Resin Microspheres. Eur J Nucl Med Mol Imaging. 2021;48(5):1570–1584.
- Choi JW, Suh M, Choi Y, et al. Y-90 Glass Microsphere Radioembolization as Frontline Treatment for HCC With Localized Portal Vein Invasion. Eur Radiol. 2025. doi:10.1007/s00330-025-11882-w.
- National Comprehensive Cancer Network. Colon Cancer. Updated 2026-03-05.
- Mulcahy MF, Mahvash A, Pracht M, et al. Radioembolization With Chemotherapy for Colorectal Liver Metastases: A Randomized, Open-Label, International, Multicenter, Phase III Trial (EPOCH). J Clin Oncol. 2021;39(35):3897–3907.
- Emmons EC, Bishay S, Du L, et al. Survival and Toxicities After Y90 TARE of Metastatic Colorectal Cancer in the RESIN Registry. Radiology. 2022;305(1):228–236.
- Dimopoulos MP, Sotirchos VS, Dunne-Jaffe C, et al. Tumor Absorbed Dose Predicts Survival and Local Tumor Control in Colorectal Liver Metastases Treated With 90Y Radioembolization. Cardiovasc Intervent Radiol. 2025. doi:10.1007/s00270-025-04175-8.
- Yu Q, Wang Y, Ungchusri E, et al. Introducing Y-90 Radioembolization to Atezolizumab and Bevacizumab Regimen for Intermediate and Advanced HCC. J Clin Oncol. 2023;41(Suppl 16):e16231.
- Fite EL, Makary MS. Advances and Emerging Techniques in Y-90 Radioembolization for HCC. Cancers. 2025;17(9):1494.
- Busse NC, Al-Ghazi MSAL, Abi-Jaoudeh N, et al. AAPM Medical Physics Practice Guideline 14.a: Yttrium-90 Microsphere Radioembolization. J Appl Clin Med Phys. 2023;:e14157.