Y-90 Radioembolization Dosimetry & Safety

Y-90 radioembolization, also called selective internal radiation therapy (SIRT) or transarterial radioembolization (TARE), is a liver-directed therapy that delivers millions of yttrium-90-loaded microspheres through the hepatic artery so they lodge in the tumor microvasculature and irradiate the tumor from within. Because Y-90 is a short-range beta emitter, the therapeutic dose is deposited within a few millimeters of each sphere, concentrating dose in the tumor and sparing most of the normal liver — but that same physics makes the dosimetry, the lung shunt evaluation, and the radiation-safety program unusually specific to this procedure.

This guide walks through the decay physics that make Y-90 well suited to SIRT, the two dominant microsphere products, the pre-treatment Tc-99m-MAA mapping workup, the three dosimetry methods a physicist must understand (body surface area, MIRD mono-compartment, and the partition model), the lung-dose constraint, and the regulatory framework under 10 CFR 35.1000. DRPS provides this analysis as part of its PET/CT and nuclear medicine physics support across Florida, Maryland, Virginia, Washington DC, California, and Nevada.

Introduction

Y-90 radioembolization combines an interventional-radiology delivery technique with internal radionuclide dosimetry, so a defensible treatment depends as much on the physics of dose calculation as on the angiography. Unlike external-beam liver radiotherapy, where the dose is planned and delivered by a controllable machine, SIRT delivers dose from a source that is permanently implanted inside the liver and decays over days. The physicist cannot dial the dose after injection; the prescription must get the activity right before the spheres are delivered.

Y-90 microspheres treat both primary liver cancer (hepatocellular carcinoma, HCC) and metastatic disease, most commonly metastatic colorectal cancer. Two products dominate United States practice: TheraSphere, a glass microsphere, and SIR-Spheres, a resin microsphere. They are not interchangeable from a physics standpoint — they differ in activity per sphere, the number of spheres injected, specific activity, embolic load, and the dosimetry method historically used to prescribe activity. ^{1, 5}

The central tension in modern SIRT is the same one that runs through all of radiopharmaceutical therapy dosimetry: simple, population-based dosing methods are easy to implement but ignore the patient-specific tumor and shunt physiology, whereas individualized dosimetry is more demanding but — in at least one randomized trial — improves outcomes. This article explains how the dosimetry actually works, then situates it within the radiation-safety and regulatory requirements every program must meet.

Topic Explanation

What is selective internal radiation therapy?

SIRT is intra-arterial brachytherapy: radioactive microspheres are infused into the hepatic artery and trapped in the arteriolar bed feeding the tumor, where they deliver a permanently implanted internal radiation dose. The biological rationale is vascular. Normal liver parenchyma derives most of its blood supply from the portal vein, while liver tumors are supplied predominantly by the hepatic artery. Microspheres infused arterially therefore preferentially lodge in and around the tumor, producing a higher tumor dose than normal-liver dose for the same injected activity. ^{1, 5}

The microspheres are roughly 20–60 micrometers in diameter — small enough to lodge in the tumor microvasculature, large enough not to pass freely into the venous circulation. They are essentially permanent and non-biodegradable; the therapeutic effect comes entirely from the radiation, not from a chemotherapeutic payload. Because the implant is permanent, the time-integrated activity is fixed by physical decay alone once the spheres are in place — there is no biological clearance to model, which simplifies the dosimetry relative to a circulating radiopharmaceutical.

Why Y-90 is well suited to liver radioembolization

Yttrium-90 is essentially a pure beta-minus emitter with a short tissue range and a convenient 64-hour half-life, which concentrates a high absorbed dose over a few millimeters and delivers the great majority of that dose within about two weeks. ³ Its key decay parameters:

Maximum beta energy of approximately 2.28 MeV, with a mean beta energy near 0.93 MeV. ³
Physical half-life of approximately 64.05 hours (about 2.67 days). ³
Mean soft-tissue range of roughly 2.5 mm and a maximum range of about 11 mm. ¹
No primary gamma emission — Y-90 decays to stable zirconium-90. The imageable signal comes from secondary bremsstrahlung and a tiny internal-pair-production branch (on the order of 32 positron pairs per million decays) that enables Y-90 PET. ¹

The short range is the whole point: a 2.5 mm mean range means dose deposited by a sphere lodged in tumor barely reaches into adjacent normal parenchyma, so a steep dose gradient protects untreated liver. The 64-hour half-life means roughly 94% of the total dose is delivered within about two physical half-lives (about 5.3 days) and effectively all of it within two weeks, which also governs how long the patient and any contaminated materials remain a radiation-safety concern. For background on where Y-90 sits among other therapy and PET radionuclides, see Understanding Common Isotopes in PET & Radiopharmaceutical Therapy.

Mapping, shunting, and the pre-treatment workup

Before any Y-90 is delivered, a pre-treatment hepatic angiogram and a Tc-99m-MAA simulation scan are performed to map the arterial supply, estimate the lung shunt fraction, and detect extrahepatic deposition. Technetium-99m macroaggregated albumin particles are similar in size to the microspheres and, injected at the planned catheter position, approximate where the Y-90 spheres will go. Planar imaging and SPECT/CT then quantify how much activity would reach the lungs (the lung shunt) and confirm that none would deposit in the gastrointestinal tract, where it could cause radiation ulceration. ^{1, 5} The MAA scan is also the imaging basis for partition-model dosimetry, because it provides a patient-specific estimate of tumor-to-normal-liver uptake.

Key Technical Principles

The three dosimetry methods

Three dosimetry methods are used to prescribe Y-90 activity — body surface area (BSA), MIRD mono-compartment, and the partition model — and they differ fundamentally in how much patient-specific physiology they incorporate. The BSA method is empirical and product-specific to resin spheres; the MIRD mono-compartment method assumes the activity is uniformly distributed in a single target mass; and the partition model splits the liver into tumor, normal-liver, and lung compartments for the most individualized estimate. ^{1, 5}

Method	Core inputs	How it prescribes	Strengths	Limitations
BSA (body surface area)	Patient body surface area; tumor involvement fraction of the liver	Empirical formula scaling activity to BSA and tumor burden; historically used for resin SIR-Spheres	Simple; fast; embeds an implicit safety margin from clinical experience	Not biologically based; BSA is a weak surrogate for liver volume; can under-dose the tumor ^{1, 5}
MIRD mono-compartment	Target mass M (kg); activity A (GBq) assumed retained in M	$D = 49.67 \times A / M$ for the whole perfused volume; historically used for glass TheraSphere	Transparent; tied directly to decay physics; easy to audit	Assumes uniform distribution; gives an average dose that ignores the tumor-to-normal contrast ^{1, 5}
Partition model	Tumor mass, normal-liver mass, lung mass; tumor-to-normal (T/N) ratio and lung shunt from Tc-99m-MAA	Solves simultaneously for tumor, normal-liver, and lung doses, constraining normal-liver and lung to tolerance	Most individualized; separates tumor and organ-at-risk doses; supports personalized prescriptions	Demands segmentation and quantitative MAA imaging; sensitive to T/N and volume measurement error ^{1, 5}

The trend in the literature is decisively toward partition-model and voxel-based personalized dosimetry, because the simpler methods deliver an average dose that can substantially under-treat the tumor while a partition calculation reveals how much tumor dose is actually available within normal-liver and lung tolerance. ^{2, 6}

The MIRD mono-compartment dose equation

For a single compartment, the Y-90 absorbed dose follows the standard relationship $D_{liver} [Gy] = 49.67 \times A [GBq] / M [kg]$ , where A is the activity assumed retained in the target and M is the target mass. ¹ The 49.67 factor is not arbitrary — it follows directly from Y-90's decay physics under the assumption that all of the beta energy is absorbed locally in the target tissue.

Starting from the mean energy emitted per decay and the physical half-life, the total energy deposited per unit administered activity is:

D = \frac{E ˉ _{β} \cdot N _{decays}}{M}

where $\overset{ˉ}{E}_{β} \approx 0.9337$ MeV is the mean beta energy per decay and $N_{decays} = A_{0} / λ = A_{0} \cdot T_{1/2} / ln 2$ is the total number of decays from complete physical decay of the initial activity $A_{0}$ . Carrying the unit conversions (MeV to joules, GBq to decays per second, the 64.05-hour half-life, and 1 Gy = 1 J/kg) yields the working form: ¹

D_{liver} [Gy] = 49.67 \times \frac{A [ GBq ]}{M [ kg ]}

Worked example. Suppose a perfused liver target has a mass of $M = 1.8$ kg and the prescription calls for an average dose of 120 Gy to that volume (a typical glass-sphere target). Rearranging:

A = \frac{D \cdot M}{49.67} = \frac{120 \times 1.8}{49.67} \approx 4.35 GBq

So about 4.35 GBq of Y-90 retained uniformly in a 1.8 kg perfused volume delivers roughly 120 Gy. Conversely, if 3.0 GBq is delivered to that same 1.8 kg volume, the average dose is:

D = 49.67 \times \frac{3.0}{1.8} \approx 82.8 Gy

The equation makes the central caveat obvious: this is a uniform-distribution average. The real spheres do not distribute uniformly — they cluster preferentially in tumor — so the true tumor dose is higher and the true normal-liver dose is lower than this single number suggests. That is precisely why the partition model exists. (Clinical prescription targets such as 120 Gy or the personalized targets discussed below are program- and product-specific; confirm against your protocol and the device labeling. )

Lung shunt fraction and lung dose

The lung shunt fraction is the proportion of injected activity that bypasses the liver and reaches the lungs, computed from the Tc-99m-MAA scan as lung counts divided by the sum of lung and liver counts. ⁴ In equation form:

LSF = \frac{C _{lung}}{C _{lung} + C _{liver}}

where $C_{lung}$ and $C_{liver}$ are the decay-corrected counts in lung and liver regions of interest on the MAA images. Because the lungs are an organ at risk that cannot tolerate the same dose as tumor, the resulting lung dose is estimated and used to limit or contraindicate treatment. A common form for the lung absorbed dose treats the shunted activity as deposited in the lung mass:

D_{lung} [Gy] = 49.67 \times \frac{LSF \times A [ GBq ]}{M _{lung} [ kg ]}

Worked example. Suppose the planned injected activity is $A = 3.0$ GBq, the MAA scan gives an LSF of 12%, and the lung mass is taken as 1.0 kg. The activity reaching the lungs is $0.12 \times 3.0 = 0.36$ GBq, so:

D_{lung} = 49.67 \times \frac{0.36}{1.0} \approx 17.9 Gy

That estimate (about 18 Gy for a single treatment) sits below the lung-dose limits commonly cited in the SIRT literature — frequently quoted as on the order of 30 Gy per single administration and roughly 50 Gy cumulative to avoid radiation pneumonitis — but the exact numeric thresholds and the assumed lung mass vary by product label and guideline, so they must be confirmed against the device labeling and the institution's protocol.

Activity decay between assay and administration

Because Y-90 decays with a 64.05-hour half-life, the activity must be decay-corrected from the calibration (assay) time to the administration time using the standard exponential decay law. ³

A (t) = A_{0} e^{- λ t}, λ = \frac{ln 2}{T _{1/2}} = \frac{0.693}{64.05 h} \approx 0.01082 h^{- 1}

Worked example. A vial is assayed at 5.00 GBq at the reference time. If it is administered 18 hours later, the activity at administration is:

A (18) = 5.00 \times e^{- 0.01082 \times 18} = 5.00 \times e^{- 0.1948} \approx 5.00 \times 0.823 = 4.11 GBq

About 0.89 GBq has decayed away in those 18 hours, which is why the prescribed activity, the assayed activity, and the residual activity left in the vial and delivery line all have to be reconciled against a single reference time. Verifying the delivered activity — by assaying the vial before and the residual waste after — is part of the dose-calibrator quality program; see our guide to dose calibrator quality control.

Clinical Impact

TheraSphere versus SIR-Spheres

The two approved Y-90 products achieve the same therapeutic goal with very different physical microspheres: glass spheres carry far more activity per sphere and impose a lighter embolic load, while resin spheres carry less activity per sphere and deliver many more particles. This difference drives the dosimetry method, the number of spheres injected, and the practical handling.

Property	TheraSphere (glass)	SIR-Spheres (resin)
Sphere material	Yttrium-aluminosilicate glass	Resin with bound Y-90
Activity per sphere (at calibration)	~2,500 Bq (high specific activity)	~50 Bq (lower specific activity)
Number of spheres per treatment	~1.2–8 million (fewer spheres)	~40–80 million (more spheres)
Embolic load	Lower (fewer, higher-activity spheres)	Higher (more spheres)
Historical dosimetry method	MIRD mono-compartment	Body surface area (BSA)
US regulatory pathway	Humanitarian Device Exemption (HDE)	Premarket Approval (PMA)

The activity-per-sphere and sphere-count figures above are representative literature values and vary with calibration time and product version; confirm exact specifications against current device labeling.

The practical upshot for a physics program: a glass-sphere case delivers a large dose with relatively few, high-activity spheres and a modest embolic effect, whereas a resin-sphere case relies on a larger embolic particle burden, which influences infusion technique and the risk of stasis. The dosimetry method historically paired with each product reflects this history, although partition-model and personalized dosimetry are increasingly applied to both. ^{2, 5}

Does personalized dosimetry change outcomes?

Unlike most of radiopharmaceutical therapy, Y-90 radioembolization has randomized evidence that personalized dosimetry improves outcomes — the DOSISPHERE-01 trial. According to PubMed, in DOSISPHERE-01 (a randomized, multicenter, open-label phase 2 trial in locally advanced HCC), patients treated with personalized dosimetry targeting at least 205 Gy to the index lesion had a significantly higher objective response rate than those treated with standard dosimetry (120 ± 20 Gy to the perfused lobe) — 71% versus 36% (p = 0.0074), with glass microspheres. ⁶ This is a genuinely positive personalization result, and it is the reason the field has moved toward tumor-absorbed-dose-driven prescriptions rather than fixed single-compartment targets.

That said, DOSISPHERE-01 was a phase 2 trial of about 60 patients in a specific HCC population, and the lobar/index-lesion dose targets it used should not be transplanted uncritically onto every patient, product, or tumor type. The EANM and the international working-group recommendations frame personalized dosimetry as the direction of travel while emphasizing multidisciplinary tumor-board decision-making and product-specific dose thresholds. ^{1, 2} The contrast with Lu-177 theranostics — where no randomized trial has yet shown a personalization benefit — is instructive: SIRT is one of the few radiopharmaceutical therapies where the dose-response evidence for individualization is randomized rather than retrospective.

Post-treatment verification imaging

Because Y-90 has no primary gamma ray, post-treatment confirmation of where the spheres actually landed relies on bremsstrahlung SPECT or, more quantitatively, on Y-90 PET that exploits the tiny internal-pair-production branch. ¹ Bremsstrahlung SPECT is widely available and confirms gross distribution, but its energy spectrum is continuous and broad, which limits quantitative accuracy. Y-90 PET, despite the extremely low positron yield, provides better spatial resolution and quantification and is increasingly used to compute the delivered tumor and normal-liver doses for comparison against the plan. Verifying delivered dose closes the loop between the predicted (MAA-based) and achieved dose distribution, which is the foundation of any dose-response analysis.

Practical Optimization Tips

For a facility standing up or refining a Y-90 radioembolization program:

1. Treat the Tc-99m-MAA scan as a dosimetry input, not just a safety check

The MAA scan does double duty: it screens for lung shunting and extrahepatic deposition and it supplies the tumor-to-normal ratio that the partition model needs. Acquire SPECT/CT, not planar alone, so the tumor and normal-liver volumes can be segmented on the co-registered CT.

2. Match the dosimetry method to the product and the goal

A single-compartment MIRD calculation is adequate for a first-pass average-dose estimate, but use the partition model when the clinical intent is to maximize tumor dose within normal-liver and lung tolerance. Document which method was used and why.

3. Decay-correct everything to one reference time

Reconcile the prescribed activity, the assayed activity, the residual in the vial and tubing, and the LSF-adjusted lung activity against a single reference time using $A (t) = A_{0} e^{- λ t}$ . A mismatch here is a direct dose error.

4. Assay and survey the delivery hardware

The vial, delivery box, tubing, and any residual are radioactive waste. Assay the residual to compute the actually delivered activity, survey the delivery area, and handle waste per your license. Post-procedure contamination practices are covered in nuclear medicine decontamination best practices.

5. Verify with post-treatment imaging

Acquire bremsstrahlung SPECT or Y-90 PET after delivery to confirm distribution, document any unexpected extrahepatic activity, and — where the program is mature — compute delivered tumor and organ doses to build an internal dose-response dataset.

Common pitfalls

Treating BSA or single-compartment dose as the tumor dose — it is a volume average that under-reports tumor dose and over-reports normal-liver dose.
Ignoring the embolic difference between products — resin and glass spheres infuse differently and are not physically interchangeable.
Mis-timing the decay correction so the delivered activity does not match the prescription.
Skipping residual-activity assay, which leaves the delivered dose unknown.
Copying lung-dose or prescription thresholds across products without confirming them against the specific device labeling.

Regulatory Considerations

In the United States, Y-90 microspheres are byproduct material regulated under 10 CFR Part 35, and because they are neither a listed sealed source nor a listed unsealed therapy drug, they are administered under 10 CFR 35.1000 ("other medical uses of byproduct material") — which requires a signed written directive and an authorized user for every administration. ⁷ The radiation safety program must address activity assay, contamination control, waste handling, and patient release, and the facility's license (or Agreement State equivalent) must specifically authorize the procedure.

Written directive and authorized user. Under 10 CFR 35.1000 and the associated NRC licensing guidance, each Y-90 microsphere administration requires a written directive signed by an authorized user before the treatment, identifying the radionuclide, activity, and treatment site. Training and experience requirements for the authorized user are product- and license-specific. ^{7, 8}
Device-specific licensing. TheraSphere and SIR-Spheres are licensed through the NRC's "other medical use" pathway with product-specific guidance, and the manufacturers provide training programs that licensees reference in their applications. NRC NUREG-1556 Volume 9 is the umbrella guidance for medical-use licenses. ⁸
Patient release. Because Y-90 is a beta emitter retained permanently in the liver, external dose rates to others are low, and most patients are released under 10 CFR 35.75 using the 5 mSv (0.5 rem) dose-to-others criterion. Release instructions still address body-fluid precautions and the (rare) scenario of contamination.
Jurisdiction. Florida, Maryland, Virginia, California, and Nevada administer their own Agreement State programs that adopt the 10 CFR Part 20 dose limits and equivalent Part 35 medical-use rules; Washington DC is regulated directly by the NRC. A program operating across states must confirm the specific requirements of each regulator.

The shielding and contamination side of beta-emitter therapy — including why a pure beta emitter like Y-90 needs low-Z (for example acrylic) shielding to minimize bremsstrahlung rather than lead — is discussed in our RPT shielding guide for Lu-177, Ra-223, and Ac-225. Programs building out therapy services should coordinate the dosimetry, assay, and release workflow with medical physics consulting and radiation safety officer support.

Frequently Asked Questions (FAQs)

What is Y-90 radioembolization (SIRT)?

Y-90 radioembolization, also called selective internal radiation therapy (SIRT) or transarterial radioembolization (TARE), is a liver-directed therapy in which millions of yttrium-90-loaded microspheres are delivered through the hepatic artery. The spheres lodge in the tumor microvasculature and irradiate the tumor from within using Y-90's short-range beta emission, sparing most normal liver.

How is the Y-90 absorbed dose to the liver calculated?

The MIRD mono-compartment method uses the standard relationship D[Gy] = 49.67 x A[GBq] / M[kg], where A is the activity assumed to be retained in the target mass M. The 49.67 factor follows from Y-90's mean beta energy and 64.05-hour half-life, assuming complete local absorption of the beta energy in the target tissue.

What is the difference between TheraSphere and SIR-Spheres?

TheraSphere is a glass microsphere with very high activity per sphere and relatively few spheres injected; it is dosed by the MIRD mono-compartment method. SIR-Spheres is a resin microsphere with lower activity per sphere and far more spheres injected; it has historically been dosed by the body-surface-area (BSA) method. The two products differ in specific activity, number of spheres, and embolic load.

Why is a Tc-99m-MAA scan done before Y-90 treatment?

Technetium-99m macroaggregated albumin (Tc-99m-MAA) is injected during pre-treatment angiography to simulate microsphere distribution. The resulting SPECT/CT and planar images estimate the lung shunt fraction, detect extrahepatic (for example gastrointestinal) deposition, and support partition-model dosimetry by characterizing tumor-to-normal-liver uptake.

What is the lung shunt fraction and why does it matter?

The lung shunt fraction (LSF) is the proportion of injected activity that bypasses the liver and reaches the lungs through arteriovenous shunts. It is calculated from the Tc-99m-MAA scan as lung counts divided by the sum of lung and liver counts. A high LSF can deliver an unacceptable lung dose, so activity is reduced or treatment withheld above program-defined thresholds.

Is Y-90 a pure beta emitter, and how is treatment imaged afterward?

Y-90 is essentially a pure beta emitter with no primary gamma ray, but it produces bremsstrahlung as the betas decelerate and a very small internal-pair-production branch that yields positrons. Post-treatment distribution can therefore be imaged with bremsstrahlung SPECT or, more quantitatively, with Y-90 PET.

How is Y-90 radioembolization regulated in the United States?

Y-90 microspheres are byproduct material regulated under 10 CFR Part 35, and because they are not a sealed source or a listed unsealed therapy they are administered under 10 CFR 35.1000 (other medical uses). Each administration requires a signed written directive and an authorized user, with the radiation safety program covering activity assay, contamination control, and patient release.

Key Takeaways

Y-90 is a short-range pure beta emitter (E_max ~2.28 MeV, E_mean ~0.93 MeV, T½ ~64.05 h, mean tissue range ~2.5 mm), which concentrates dose within a few millimeters of each sphere and delivers nearly all of it within about two weeks.
The MIRD mono-compartment dose is $D = 49.67 \times A / M$ , a uniform-distribution average derived from Y-90's decay physics; the real tumor dose is higher and the normal-liver dose lower because spheres cluster in tumor.
Three dosimetry methods exist — BSA (empirical, resin), MIRD mono-compartment (glass), and the partition model (tumor/normal-liver/lung) — and the field is moving toward partition-based, personalized dosimetry.
The Tc-99m-MAA scan drives the lung shunt fraction (LSF = lung counts / (lung + liver counts)) and the partition model; lung dose is limited (commonly cited near 30 Gy single / 50 Gy cumulative — confirm against labeling) to avoid radiation pneumonitis.
DOSISPHERE-01 showed personalized dosimetry improves response (71% vs 36% objective response for ≥205 Gy index-lesion vs 120 ± 20 Gy lobar dosimetry with glass spheres), making SIRT one of the few RPTs with randomized evidence for individualization.
Y-90 microspheres are administered under 10 CFR 35.1000, requiring a signed written directive and an authorized user, with assay, contamination control, and patient release governed by Part 35 and the facility license.

Conclusion

Y-90 radioembolization is a procedure where the physics is not an afterthought to the angiography — it is half of the treatment. The decay characteristics of Y-90 (short beta range, no primary gamma, 64-hour half-life) explain why the dose is concentrated in tumor and why the procedure is imaged the way it is. The dosimetry method chosen — BSA, MIRD mono-compartment, or partition — determines whether the prescription reflects a crude volume average or a patient-specific tumor dose, and the randomized DOSISPHERE-01 evidence now argues for the more individualized end of that spectrum. Around all of it sits a radiation-safety and 10 CFR 35.1000 framework that demands accurate activity assay, lung-shunt evaluation, written directives, and verified delivery.

For a medical physicist or RSO, the practical posture is to build a rigorous, well-documented dosimetry and assay workflow, prefer partition-model thinking when the clinical intent is tumor-dose maximization, and confirm every product-specific threshold against current device labeling rather than borrowing numbers across products. A program that calculates dose carefully today is positioned to adopt fully personalized, verified-delivery dosimetry as the standard of care continues to shift in that direction.

How DRPS Can Help

Diagnostic Radiation Physics Services supports nuclear medicine and interventional programs offering Y-90 radioembolization with dosimetry method selection and review, Tc-99m-MAA lung-shunt and partition-model analysis, dose-calibrator and assay quality control, post-treatment bremsstrahlung SPECT and Y-90 PET verification, contamination-control and waste guidance, patient-release evaluation, and 10 CFR 35.1000 written-directive and authorized-user program support aligned with NRC or Agreement State requirements. Whether a facility is launching a SIRT program or refining an existing one, DRPS helps translate the radioembolization physics literature into a defensible, documented clinical process.

DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware. To discuss a Y-90 program, contact DRPS or learn more about our PET/CT and nuclear medicine physics and radiation safety officer services.

Related Resources

References

Dezarn WA, Cessna JT, DeWerd LA, et al. Recommendations of the American Association of Physicists in Medicine on dosimetry, imaging, and quality assurance procedures for 90Y microsphere brachytherapy in the treatment of hepatic malignancies (AAPM Task Group 144). Med Phys. 2011;38(8):4824-4845.
Weber M, Lam M, Chiesa C, et al. EANM procedure guideline for the treatment of liver cancer and liver metastases with intra-arterial radioactive compounds. Eur J Nucl Med Mol Imaging. 2022;49(5):1682-1699. doi:10.1007/s00259-021-05600-z. doi.org
National Nuclear Data Center, Brookhaven National Laboratory. NuDat / decay data for Yttrium-90 (E_max ~2.28 MeV; T½ = 64.05 h; decay to stable Zr-90). nndc.bnl.gov
Kallini JR, Gabr A, Hickey R, et al. Indicators of lung shunt fraction determined by technetium-99m macroaggregated albumin in patients with hepatocellular carcinoma. Cardiovasc Intervent Radiol. 2017;40(8):1213-1222. doi:10.1007/s00270-017-1619-z. doi.org
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 (REBOC). Int J Radiat Oncol Biol Phys. 2007;68(1):13-23. doi:10.1016/j.ijrobp.2006.11.060. doi.org
Garin E, Tselikas L, Guiu B, et al. Personalised versus standard dosimetry approach of selective internal radiation therapy in patients with locally advanced hepatocellular carcinoma (DOSISPHERE-01): a randomised, multicentre, open-label phase 2 trial. Lancet Gastroenterol Hepatol. 2021;6(1):17-29. doi:10.1016/S2468-1253(20)30290-9. doi.org
Salem R, Padia SA, Lam M, et al. Clinical, dosimetric, and reporting considerations for Y-90 glass microspheres in hepatocellular carcinoma: updated 2022 recommendations from an international multidisciplinary working group. Eur J Nucl Med Mol Imaging. 2023;50(2):328-343. doi:10.1007/s00259-022-05956-w. doi.org
U.S. Nuclear Regulatory Commission. 10 CFR Part 35 — Medical Use of Byproduct Material (including §35.1000 Other medical uses, §35.40 Written directives, and §35.75 Release of patients). nrc.gov
U.S. Nuclear Regulatory Commission. NUREG-1556, Volume 9, Rev. 3 — Consolidated Guidance About Materials Licenses: Program-Specific Guidance About Medical Use Licenses. nrc.gov
U.S. Food and Drug Administration. TheraSphere Yttrium-90 Glass Microspheres — Humanitarian Device Exemption (H980006), Instructions for Use. fda.gov
U.S. Food and Drug Administration. SIR-Spheres Yttrium-90 Resin Microspheres — Premarket Approval (P990065), Instructions for Use. accessdata.fda.gov