Q.Liver software for the planning of treatment of liver cancer via transarterial radioembolization with yttrium-90 resin microspheres based on single-photon emission computed tomography-computed tomography

Introduction

Liver cancer is one of the most common malignancies in the world, and colorectal cancer and liver cancer rank third and sixth in cancer incidence and second and third in cancer mortality, respectively according to 2022 data on global cancer burden published by the International Agency for Research on Cancer of the World Health Organization (1). The liver is the most common organ where colorectal cancer metastasizes, with an estimated 60% of patients with colorectal cancer ultimately developing liver metastases (2,3). This is mainly due to the direct connection of the portal venous system between the colorectum and the liver (4). In a cohort study of 26,813 patients with colorectal adenocarcinoma, the overall cumulative incidence of colorectal cancer liver metastases was 4.3% [95% confidence interval (CI): 3.8–4.8%] after 1 year, 11.0% (95% CI: 10.3–11.8%) after 3 years, and 12.9% (95% CI: 12.1–13.7%) after 5 years (5). Most deaths from primary liver cancer and colorectal cancer liver metastasis (CRLM) are due to unresectable liver metastases and progression, with one study reporting a mortality rate of 86% for colorectal cancer, with CRLM being the main cause of death (6).

Primary treatment modalities for hepatocellular carcinoma (HCC) include various approaches, such as surgical resection, ablative therapy, transarterial chemoembolization (TACE), percutaneous anhydrous ethanol injection, liver transplantation, and chemotherapy. For patients with unresectable, mid-stage HCC with preserved liver function, TACE is a standard treatment (7,8). Similar to TACE, transarterial radioembolization (TARE), also known as selective internal radiation therapy (SIRT), is a minimally invasive procedure. It involves delivering radioactive microspheres directly into the blood vessels supplying the tumor, where they lodge, emit beta radiation, and cause localized tumor destruction, thereby achieving the goal of tumor treatment (9). Studies have shown that TARE has a higher tumor response rate and better efficacy compared to other locoregional therapies. In the study by Salem et al. (10), among 179 patients with stage A or Barcelona Clinic Liver Cancer (BCLC) staging system HCC, the tumor response rate was similar between TARE and TACE methods (74% vs. 87%; P=0.433), but the median time to progression was longer with TARE than with TACE (>26 vs. 6.8 months; P<0.001). Results from a randomized controlled, prospective phase II clinical trial showed that yttrium-90 radioembolization had a similar safety profile to drug-eluting bead chemoembolization but provided superior tumor control and survival in participants with early and intermediate unresectable HCC (11). A multicenter study spanning the years from 2014 to 2017 found that the response rate and duration of response to yttrium-90 TARE for the treatment of unresectable, solitary HCC smaller than 8 cm were clinically meaningful (12). Another study reported that liver function was higher post yttrium-90 treatment than it was pretreatment, with significant recovery at 24 months (13). Thus, yttrium-90 TARE has been proven to be an effective method for treating patients with unresectable HCC or CRLM (14).

The diagnostic dose of technetium-99m macroaggregated albumin (^99mTc-MAA) injected into the tumor through the hepatic artery can predict the distribution of yttrium-90 microspheres in the tumor and the abnormal shunting of yttrium-90 outside the liver, playing a crucial role in yttrium-90 TARE treatment (15). In clinical yttrium-90 TARE treatment planning, to avoid serious radioactive damage to the lung due to liver-lung shunting, ^99mTc-MAA imaging is used to mimic the distribution of yttrium-90 microspheres and calculate the lung shunt fraction (LSF). Traditionally, the LSF in ^99mTc-MAA planar scintigraphy is calculated using the simple geometric mean method via the definition of the regions of interest (ROIs) in the liver and lung on anterior and posterior planar scintigraphy. However, due to the blurring of the liver dome boundary and the overlapping of the lung basal structure, ^99mTc-MAA planar scintigraphy is not accurate. Additionally, planar scintigraphy does not account for radioactive attenuation or scatter effects (16-19), leading to significant errors in LSF estimation. This problem can be avoided through the application of radioactivity counts in three-dimensional space in single-photon emission computed tomography-computed tomography (SPECT/CT) images. However, previous three-dimensional (3D) methods have been limited by the necessity of liver segmentation, which is time-consuming, tedious, and highly operator dependent.

Artificial intelligence (AI) can not only automate volume calculations, reducing intra- and interobserver variability and improving the accuracy of the result (20), but it can also process images quickly, reducing the time required for segmentation. GE HealthCare’s latest Q.Liver software (GE HealthCare, Chicago, IL, USA) enables a fully automated process for preoperative yttrium-90 TARE planning via AI technology. Q.Liver software offers advanced segmentation and quantification of SPECT/CT data, providing effective image reconstruction with attenuation, scatter corrections, and resolution compensation. Q.Liver simplifies the currently used yttrium-90 TARE treatment planning process by automatically estimating liver and lung volumes from SPECT/CT and 3D radiological calculations. In traditional yttrium-90 TARE clinical practice, liver and lung volumes from CT require inhalation followed by breath-holding, whereas SPECT requires tidal breathing. Whether the difference in liver and lung volumes calculated by the two methods affects the treatment plan for yttrium-90 TARE is unclear.

The primary objective of this study was thus to evaluate the efficacy of the Q.Liver software in planning yttrium-90 TARE for liver cancer based on SPECT/CT imaging and to compare it with that of the conventional method based on planar scintigraphy. The secondary objective was to determine the feasibility of replacing CT-derived liver and lung volumes with SPECT/CT-derived ones. We present this article in accordance with the STROBE reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1471/rc).

Methods

Research participants

Forty-five participants with primary HCC or CRLM were consecutively enrolled from May 2022 to January 2024. Among these patients, 43 cases (37 males and 6 females) had a single lesion, had an age range of 27 to 82 years (53.0±14.0 years), and underwent hepatic radioembolization with resin yttrium-90 resin microspheres (SIR-spheres; SIRTeX Medical, Woburn, MA, USA). The other 2 cases with multiple lesions were not included because it was not possible to retrospectively analyze the exact absorbed dose per lesion. The laboratory and demographic characteristics of the 43 patients with yttrium-90 TARE are shown in Table 1. Clinical data including patient age, gender, height, weight, tumor type, tumor stage, severity of underlying disease, previous treatment history, and serological parameters were obtained from the electronic medical record.

In this study, two methods are used to compare the LSF and yttrium-90 treatment planning for the same patient. The conventional approach for yttrium-90 TARE, referred to in this study as the planar method, involves calculating the LSF using ^99mTc-MAA planar scintigraphy prior to TARE, the liver volume using contrast-enhanced CT, the lung volume using CT, the perfusion volume using cone beam CT or SPECT, and the tumor volume using cone beam CT or contrast-enhanced CT. The partition model is then used to calculate the liver-absorbed dose (D_Liver), lung-absorbed dose (D_Lung), and yttrium-90-prescribed activity in the treatment planning system.

The Q.Liver method involves SPECT/CT imaging after ^99mTc-MAA planar scintigraphy. In this method, liver and lung volumes are calculated using the Q.Liver software based on SPECT/CT imaging, while perfusion and tumor volumes are calculated in the same manner as that in the conventional method. In all 45 patients, liver volumes, lung volumes, and LSF were obtained. In the 43 cases treated with TARE, perfusion volumes, tumor volumes, and yttrium-90 treatment planning were recorded. Moreover, liver volume, lung volume, LSF, prescribed activities of yttrium-90, D_Liver, and D_Lung were compared in pairs for the same patient between the planar and Q.Liver methods.

This study was approved by the Institutional Review Board of the First Affiliated Hospital of Xi’an Jiaotong University (approval No. X-XJTU1AF2025LSYY-436) and conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The requirement for informed consent was waived due to the retrospective nature of the analysis.

Patient preparation

All patients underwent lung CT, and a baseline contrast-enhanced CT scan of the liver within 2 weeks prior to angiography. An arterial catheter was inserted through the femoral artery in each patient with the tip positioned in the artery supplying blood to the tumor. All procedures involved selective catheterization with a microcatheter. The study was performed via digital subtraction angiography and cone beam CT.

Before undergoing CT, cone beam CT, or ^99mTc-MAA planar scintigraphy and SPECT/CT scans, all patients were required to wear clothing without metal objects or zippers, avoid wearing any jewelry, and remove anything that might interfere with the CT imaging, such as eyeglasses, dentures, hairpins or hearing aids. Patients were instructed to hold their breath before the lung CT examination. Patients were also required to have an empty stomach and no iodine allergies for the acquisition of liver-enhanced CT and cone beam CT.

^99mTc-MAA planar scintigraphy and SPECT/CT imaging were performed after cone beam CT, so patient preparation for ^99mTc-MAA imaging was the same as that for the cone beam CT.

Data acquisition, reconstruction, processing or analysis

CT and enhanced CT

Imaging data acquisition

Patients were scanned separately using a LightSpeed VCT scanner (GE HealthCare). For the lung, a CT scan was performed under the following parameters: a scan area from below the thoracic inlet to the level of the bilateral adrenal glands, a tube current of 300 mAs, a tube voltage of 120 kVp, a reconstruction slice thickness of 1 mm, and a scan matrix of 512×512.

For the liver, the CT scan was performed with the following parameters: a scan area extending from the top of the diaphragm to the inferior poles of both kidneys, a scanning field of view of 60–70 cm, a tube current of 300 mAs, a tube voltage of 120 kV, a scan matrix of 512 × 512, a slice thickness ranging from 0.625 to 1.0 mm; a pitch of 0.984, a time per rotation of 0.5 s, a scanning slice thickness of 0.625–1.000 mm, and a slice interval of ≤1 mm. Nonionic contrast medium (iohexol injection) was injected at a total dose of 70–80 mL per body weight (0.9 mL/kg) through an intravenous cannula at a rate of 2.5–3.0 mL/s to improve visualization of the abdominal viscera and vasculature. The scan was started with a delay of 25–30 s after contrast injection for the hepatic arterial phase. The portal venous phase scan was acquired 65–70 s after contrast injection, and delayed phase images were acquired 120–150 s after contrast injection (1).

Data reconstruction and analysis

Images were reconstructed with a slice width of 5 mm in the axial and coronal planes.

For lung recognition and segmentation, the right and left lung volumes of interest (VOIs) were delineated separately by an experienced radiologist (21). A VB-net liver segmentation network within the United Imaging Intelligence’s (Shanghai, China) one-stop research platform (uAI Research Portal; V20230915; https://urp.united-imaging.com) was used for automatic lung recognition and segmentation in CT images. Lung volumes were recorded in cubic centimeters.

For liver recognition and segmentation, liver VOIs were delineated separately by an experienced radiologist (21). A VB-net liver segmentation network within the United Imaging Intelligence’s one-stop research platform (uAI Research Portal; V20230915; https://urp.united-imaging.com/) was used to automatically detect and segment the liver in CT images. Liver volume was recorded in cubic centimeters.

Cone beam CT

Data acquisition

For cone beam CT acquisition, the patient was placed in the supine position with hands on the head, and the area of interest was centered in the C-ray image. After local anesthesia at the puncture site, a cone beam CT scan was routinely performed with the Azurion 7 device (Philips, Amsterdam, the Netherlands) under fluoroscopic guidance, and standard digital subtraction angiography was used to determine the best arterial access. A 12-mL mixture of contrast agent (iohexol injection at 350 mg iodine/mL; Jiangsu Hengrui Pharmaceutical Co., Ltd., Lianyungang, China) mixed with 0.9% sodium chloride in a 2:1 ratio was injected at 0.6 mL/s and a pressure of 800 psi, with a 10-s scan delay through a microcatheter placed in the target artery. A fixed tube voltage of 118 kV and a current-time product of 138 mAs were used.

Data reconstruction and analysis

Perfusion volume and tumor volume were calculated with Smart CT software (Philips), which provides real-time 3D image guidance and segmentation of target blood vessels and lesions.

Planar scintigraphy

Imaging data acquisition

At the end of mapping angiography, approximately 167 MBq of ^99mTc-MAA (Beijing Atom High Tech Co. Ltd., Beijing, China) with radiochemical purity >95% was slowly injected into the target artery via the arterial catheter at a flow rate of approximately 40 mL/hr. After the interventional injection of ^99mTc-MAA, 5-minute anterior and posterior planar scintigraphy of the chest and upper abdomen, as well as whole-body images, was obtained with a Discovery NM/CT 670 pro SPECT/CT device (GE HealthCare) with low-energy high-resolution collimators, a 140 KeV ±10% energy setting, and a 256×256 matrix size.

Data analysis

Planar scintigraphy images were processed with the Xeleris 3.1 workstation (GE HealthCare). The ROIs of the left lung, right lung, and liver were manually delineated in anterior and posterior scintigraphy. To avoid including mediastinal and cardiac radioactivity counts in the lung, the mediastinum and heart were excluded from the lung ROIs.

The geometric mean counts of lung or liver in planar scintigraphy were calculated via Eq. [1], and the LSF of the planar scintigraphy was calculated via Eq. [2].

Geometriccountsofthelungorliver=(Anteriorcounts×Posteriorcounts)[1]

LSF=GeometriccountsofthelungGeometriccountsofthelung+Geometriccountsoftheliver×100(%)[2]

SPECT/CT imaging

Data acquisition

SPECT images of the chest and upper abdomen were acquired 45±15 minutes after injection of ^99mTc-MAA with a Discovery NM/CT 670 pro SPECT/CT device (GE HealthCare) equipped with a low-energy, high-resolution collimator. The parameters included a main energy window of 140 KeV ±10%, a scatter energy window of 120 KeV ±5%, a 128×128 image matrix, and a pixel size of 4.42 mm. SPECT images were obtained in step and shoot mode, with 20 seconds per projection, 64 projections over 360°, and with body contour on. A spiral CT scan was performed after SPECT, with automatic tube current modulation in the longitudinal direction (current range 150–300 mA and noise index 13.28), a tube voltage of 120 kV, a 512×512 image matrix, and a slice thickness of 2.5 mm.

Data analysis

Q.Metrix software in the Xeleris 3.1 workstation (GE HealthCare) was used to reconstruct and process SPECT/CT data. The iterative ordered subset expectation maximization algorithm, with 10 subsets and 2 iterations being used for image reconstruction. The 3D postfilter is a Butterworth filter with a cutoff frequency of 0.48 cycles/cm and a power of 10. CT-based attenuation correction, scatter correction, and resolution recovery were applied.

The LSF based on the SPECT/CT image was calculated via Eq. [3] (see Figure 1). The tumor-nontumor ratio (TNR) and the absorbed dose to the tumor were calculated via Eqs. [4] and [5], respectively (22).

LSF=LungcountsLungcounts+Livercounts×100(%)[3]

TNR=ATumor[GBq]/MTumor[Kg]APNL[GBq]/MPNL[Kg][4]

Figure 1 LSF of the planar method (12.92%) and the Q.Liver method (7.31%). (A) A patient with an LSF of 12.92% from the planar method whose D_Lung calculated by the planar method was 25.94 Gy. (B,C) The same case with an LSF of 7.31% in the Q.Liver method, whose D_Lung calculated by the Q.Liver method was 13.79 Gy. This suggests that the LSF of the planar method can reduce D_Lung in patients treated with yttrium-90 TARE. D_Liver, liver-absorbed dose; D_Tumor, tumor-absorbed dose; LSF, lung shunt fraction; TARE, transarterial radioembolization.

where A_PNL and A_tumor are the activity of the perfused normal liver and tumor, respectively; and M_PNL, M_tumor, and M_Lung are the mass (kg) of the perfused normal liver, tumor and lung, respectively. Lung and liver volumes and radioactive counts from the Q.Liver method were obtained with Q.Liver software in the Xeleris 5 workstation (GE HealthCare), and then the LSF based on SPECT/CT was automatically calculated. The lungs were automatically segmented based upon the CT values and then manually adjusted if necessary. The liver was also automatically segmented based upon the CT image via deep learning technology, which was followed by manual adjustment if necessary.

Dosimetry calculation

Estimation of the yttrium-90-prescribed activity and predicted D_Lung for patients with yttrium-90 TARE were calculated via the partition model formula (18,23). The yttrium-90-prescribed activity to be administered was calculated from the target D_Tumor via Eq. [6] (24,25), and the D_Lung was calculated via Eq. [7] (18).

DTumor[Gy]=TNR×DPNL[GBq][5]

A(GBq)=DTumor[Gy]×(MPNL[Kg]+MTumor[Kg]×TNR)49.67×TNR×(1−LSF)[6]

DLung[Gy]=49.67×Totalamountofinjectedactivity(GBq)MLung[Kg]×LSF[7]

where D_PNL and D_Tumor were the mean absorbed dose of perfused normal liver and tumor, respectively, and A id the yttrium-90-prescribed activity calculated with the partition model.

The yttrium-90 dose conversion factor of 49.67 J/GBq is the energy deposited locally in tissue per unit of activity, as recommended by the National Nuclear Data Center (26) and the SIR-Spheres package insert (27). A lung density of 0.3 g/mL was assumed and multiplied by the lung volume (mL) to obtain M_Lung (17,24).

Comparison of data

First, liver volume and lung volume were compared between the planar and Q.Liver methods. Second, the LSF of the two methods was compared. Third, D_Lung calculated with lung volume based on enhanced CT was compared with D_Lung calculated with lung volume based on Q.Liver via the partition model. Fourth, D_Lung calculated with lung volume based on enhanced CT was compared with that based on Q.Liver. Finally, D_Tumor, yttrium-90-prescribed activity and D_Lung were compared between the planar method and Q.Liver methods with D_Lung being the same between the two methods.

Statistical analysis

Normally distributed continuous variables are described as the mean ± standard deviation, while nonnormally distributed continuous variables are described as the median and interquartile range (IQR). The differences between the two testing methods were analyzed with paired Wilcoxon signed-rank tests. The data were analyzed and visualized via R version 4.1.1 (The R Foundation for Statistical Computing, Vienna, Austria). All statistical analyses were considered statistically significant at P<0.05.

Results

Comparison of liver and lung volumes between planar and Q.Liver methods

The comparison of liver and lung volumes between the planar and Q.Liver methods in 45 cases is shown in Table 2. The lung volume measured by the planar method was higher than that measured by the Q.Liver method (V=928; P=5.8×10⁻⁶), but no significant difference in liver volume was found between the two methods (V=396; P=0.36).

Comparison of LSF between the planar and Q.Liver methods

The comparison of the LSF between the planar and Q.Liver methods for the 43 cases is shown in Figure 2. The LSF in the planar method was higher than that in the Q.Liver method (V=0.36; P=4.80×10⁻⁶), with median values of 6.08% (IQR, 4.91–8.78%) and 3.96% (IQR, 2.30–5.85%), respectively.

Figure 2 Comparison of the LSF between planar and Q.Liver method in 43 cases. The LSF of the Q.Liver method was lower than that of the planar method, suggesting that the use of SPECT/CT LSF can reduce the lung absorption dose in patients treated with yttrium-90 TARE, which may change the treatment plan of yttrium-90 TARE. ^99mTc-MAA, technetium-99m macroaggregated albumin; LSF, lung shunt fraction; SPECT/CT, single-photon emission computed tomography-computed tomography; TARE, transarterial radioembolization.

Comparison of D_Lung calculated via lung volume based on enhanced CT with D_Lung calculated via lung volume based on Q.Liver

When the LSF remained constant, the D_Lung calculated via the lung volume based on Q.Liver was higher than that calculated via the lung volume based on enhanced CT (V=11.00; P=1.25×10⁻¹¹). The median values was 7.37 (IQR, 2.63–14.91) for Q.Liver and 4.67 (IQR, 2.22–14.00) for enhanced CT.

Comparison of yttrium-90-prescribed activity and D_Lung between the planar and Q.Liver methods

At a constant tumor absorbed dose (D_Tumor) per patient, the yttrium-90-prescribed activity was higher in the planar method than in the Q.Liver method (Table 3 and Figure 3). However, the D_Lung was lower in the planar method than in the Q.Liver method, which is inconsistent with the sum of the D_Lung for the two methods. The total yttrium-90-prescribed activity and D_Lung in the planar method were 89.89 GBq and 345.53 Gy, respectively, which were higher than the 87.04 GBq and 221.82 Gy in the Q.Liver method, respectively (Table 3).

Figure 3 Comparison of (A) yttrium-90-prescribed activity and (B) D_Lung between the planar and Q.Liver methods with same D_tumor. (A) The yttrium-90-prescribed activity in the Q.Liver method was lower than that in the planar method. This suggests that using SPECT/CT to calculate LSF can result in the same therapeutic effect as that of the planar method with lower yttrium-90-prescribed activity. (B) The sum of D_Lung in the Q.Liver method was lower than that in the planar method, suggesting that using SPECT/CT to calculate LSF can reduce the sum of D_Lung while maintaining the same D_Tumor, which may reduce the occurrence of side effects such as radiation pneumonia. However, this is inconsistent with the fact that the median of the Q.Liver method was higher than that of the planar method. The main reason for this is that the data in the Q.Liver method were more skewed due to the small sample size. D_Lung, lung absorbed dose; D_Tumor, tumor-absorbed dose; LSF, lung shunt fraction; SPECT/CT, single-photon emission computed tomography-computed tomography.

Comparison of D_Tumor, yttrium-90-prescribed activity, and D_Lung between the planar and Q.Liver methods with D_Lung being constant

When D_Lung remained constant in the treatment planning system, it was recalculated via the partition model formula. The results showed that yttrium-90-prescribed activity, D_Tumor, and D_Liver in the Q.Liver method were higher than those in the planar method (Table 4 and Figure 4). With a D_Liver upper limit of 40 Gy for cirrhosis or 70 Gy for a normal liver serving as the threshold for treatment plan adjustment, 44.19% (19/43) of the treatment plans would be modified in the Q.Liver method due to D_Liver limitations.

Figure 4 Comparison of (A) yttrium-90-prescribed activity, (B) D_Tumor, and (C) D_Liver between the planar and Q.Liver methods with the same D_Lung being maintained. These results show that LSF based on SPECT/CT increases the yttrium-9-prescribed activity, D_tumor, and D_Liver compared to that based on planar scintigraphy when the D_Lung is unchanged in the treatment planning system. D_Liver, liver-absorbed dose; D_Lung, lung-absorbed dose; D_Tumor, tumor-absorbed dose; LSF, lung shunt fraction; SPECT/CT, single-photon emission computed tomography-computed tomography.

Discussion

Surgical resection is effective and can achieve a complete cure for patients with early-stage liver cancer. Unfortunately, about 70% of patients with liver cancer are diagnosed at intermediate or advanced stages and thus miss the opportunity for surgical resection (28). In recent years, yttrium-90 TARE has emerged as a novel method for treating patients with advanced HCC. In current clinical practice, the liver and lung volumes for yttrium-90-prescribed activity in TARE patients are calculated with CT. However, Q.Liver software can rapidly calculate liver volume, lung volume, and LSF based on SPECT/CT. CT acquisition requires inhalation followed by breath-holding, whereas SPECT acquisition requires tidal breathing. The difference in liver and lung volumes calculated by the two methods and whether this affects the treatment plan for yttrium-90 TARE remain unclear. In this study, we sought clarify this issue by using Q.Liver software, which can quickly and easily calculate the volume and radioactivity counts in 3D space for the liver and lungs based on SPECT/CT.

In this study, the liver volume calculated with the Q.Liver method was the same as that calculated with the planar method, suggesting that SPECT/CT and enhanced CT or CT images are well matched for liver volume. Therefore, Q.Liver software can be used instead of CT to calculate liver volume. In our study, the lung volume in the Q.Liver method was smaller than that in the planar method. The main reason for this is that lung scanning with CT or enhanced CT requires inhalation followed by breath-holding, whereas SPECT/CT is performed during tidal breathing. In our study, some lung volumes were higher in the Q.Liver method than in the planar method, which may be related to patients not holding their breath properly during enhanced CT in the planar method.

In yttrium-90 TARE treatment, ^99mTc-MAA planar scintigraphy is the recommended method for determining the LSF. To avoid including mediastinal and cardiac radioactivity counts in the lung, the mediastinum and heart were excluded from the lung ROIs in planar scintigraphy. There have also been studies that have used 3D imaging techniques (18,29) to this end. Some suggest that to account for the movement of liver scatter activity into the lungs during a free-breathing SPECT scan, the lower 2 cm of the left and right lungs should be automatically subtracted from the lung VOIs in SPECT/CT imaging (16,30). Both methods may lead to errors in the LSF calculation. Therefore, there is an urgent need to compare the differences between the two methods of LSF calculation and their impact on treatment planning.

Only a few reports on the use of SPECT/CT for LSF calculation have been conducted thus far. The majority of these studies have reported that 3D SPECT imaging is preferable to 2D planar scintigraphy for LSF calculation. Planar scintigraphy overestimates the calculated LSF during yttrium-90 TARE treatment, whereas SPECT/CT imaging with an appropriate segmentation tool provides more accurate calculations (16,18,31). The findings from the work by Allred et al. (16) and Kao et al. (30) also show that the SPECT/CT-based LSF is lower than the planar scintigraphy-based LSF. The results of our study support this conclusion and provide a basis for calculating yttrium-90-prescribed activity via SPECT/CT-based LSF. It has been shown that the LSF calculation method based on 3D SPECT/CT images have higher accuracy and clinical application potential (29). Registration between static ^99mTc-MAA SPECT and corresponding CTs can feasibly reduce the spatial mismatch and improve dosimetric estimation, with the improvement in LSF being greater than TNR (17,32). Therefore, ^99mTc-MAA planar scintigraphy overestimates the LSF compared to SPECT/CT, providing a basis for using SPECT/CT-based LSF to improve yttrium-90 activity in patients.

The partition model is widely used in clinical practice for yttrium-90 radioembolization of HCC or CRLM. It accounts for differences in yttrium-90 activity concentration within normal liver and subliver treatment regions and uses the TNR determined from ^99mTc-MAA SPECT images, providing a more accurate method for calculating the absorbed dose in different tissue compartments. The partition model can deliver a threshold absorbed dose to the tumor while maintaining safe absorbed doses to the liver and lung (22). The parameters in the partition model include the desired absorbed dose to the tumoral liver compartment, TNR, LSF, liver volume, lung volume, tumor volume, and perfusion volume. Previous studies have shown that planar scintigraphy may overestimate the LSF and in turn, the lung mean absorbed dose (33-38). From Equation 7 and the results of this study, it can be concluded that lung volume affects D_Lung: the smaller the lung volume is, the larger the D_Lung if the LSF remains constant. However, the Q.Liver method decreases both LSF and lung volume, making its effect on D_Lung more complex. As shown in Table 3, the sum of D_Lung in the Q.Liver method was lower than that in the planar method (221.82 vs. 345.53 Gy), suggesting that the Q.Liver method can reduce the sum of D_Lung while the D_Tumor remains constant, potentially reducing the incidence of side effects such as radiation pneumonia. However, this is inconsistent with the fact that the median of the Q.Liver method is higher than that of the planar method, which may be explained by two possible reasons. First, it is likely to be due to the small sample size resulting in skewed data in the Q.Liver method. Second, as reported in a study by Kim et al. (39), D_Lung calculated by SPECT/CT is higher than that calculated by the planar method when the patient’s SPECT LSFs are <4.89% and lower than that calculated by the planar method when the patient’s LSFs are >4.89%. These notions provide directions for further research.

In our study, it was further found that in yttrium-90 TARE treatment, the LSF was overestimated in the planar method as compared to in the Q.Liver method, resulting in an increase in the yttrium-90-prescribed activity and D_Lung in the Q.Liver method. In this situation, the yttrium-90 treatment plan may be adjusted to a lower target dose, potentially resulting in inadequate treatment of tumor tissue. Alternatively, yttrium-90 treatment may be discontinued for patients who can benefit from TARE due to exceeding the lung dosimetry threshold. The European Association of Nuclear Medicine’s standard operational procedure recommends that patients with significant lung shunting visible on planar scintigraphy should undergo additional SPECT/CT scans to ensure a more accurate quantification of the LSF (15). The clinical significance of this study is that it suggests an opportunity for patients who are ineligible for yttrium-90 TARE based on planar scintigraphy due to D_Lung being above 30 Gy. Additionally, the Q.Liver method can reduce the absorbed dose to the lung if the absorbed dose to the tumor tissue remains constant, thus reducing the incidence of radiation pneumonitis.

^99mTc-MAA mapping plays a crucial role in the assessment of LSF and detection of potential extrahepatic shunts before yttrium-90 TARE (40). Radiation-induced pneumonitis is one of the main risks of TARE in patients with liver malignancies. If a significant portion of yttrium-90 crosses a tumor-related arteriovenous shunt and lodges in the lung vasculature, this may lead to radiation-induced pneumonitis, a rare but potentially fatal side effect that can occur 1 to 6 months after yttrium-90 TARE administration (37). Estimation of the LSF with planar scintigraphy is essential to reduce the risk of radiation-induced pneumonitis. Studies have shown that a lung dose >30 Gy per session or >50 Gy cumulative (determined by pretreatment planar ^99mTc-MAA scintigraphy) usually leads to radiation pneumonitis and is a relative contraindication to yttrium-90 TARE (17,41). Therefore, in current clinical practice, ^99mTc-MAA planar scintigraphy is used to assess extrahepatic LSF and thus to ensure that healthy lung tissue receives no more than 30 Gy of absorbed dose per treatment or 50 Gy of cumulative absorbed dose (for TheraSphere, Boston Scientific, Marlborough, USA) (42) or that the procedure results in no more than 20% lung shunting (for SIR-Spheres; Sirtex Medical Inc., Sydney, Australia) (27).

There is a strong correlation between the absorbed dose to the tumor and the therapeutic efficacy of TARE. In a study on the relationship between the absorbed dose to the tumor and survival and response in patients with HCC treated with yttrium-90 TARE, Hermann et al. found that patients with advanced HCC who received an absorbed dose of 100 Gy to the tumor or more had longer overall survival and disease control than did those who received less than 100 Gy (43).

From their an exploratory analysis of 50 cases with PET/CT imaging after yttrium-90 segmentectomy, Sarwar et al. reported similar findings: patients with a tumor-absorbed dose ≥300 Gy had a lower rate of disease progression at 2 years than did those with a dose <300 Gy (18% vs. 61%; P=0.047) (44). In our study, to protect the lung tissue, we left the D_Lung of the planar method unchanged and recalculated the prescribed D_Tumor, D_Liver, and yttrium-90 activity of the Q.Liver method in the treatment planning system using the partitioning model formula. As shown in Table 4 and Figure 4, the D_Tumor and yttrium-90 activities of the Q.Liver method were higher than those of the planar method, suggesting that treatment planning with SPECT/CT-imaged liver volumes, lung volumes, and LSF for yttrium-90 TARE will increase yttrium-90 activity and the tumor-absorbed dose as compared to the currently used conventional method. This can increase the absorbed dose to the tumor as much as possible without increasing the absorbed dose to the lung, which has implications for preventing the development of radiation pneumonitis and improving tumor outcomes. However, it should be noted that this assumption also increases the absorbed dose to the liver. We used the liver absorbed dose safety limit of 40 Gy for cirrhosis and 70 Gy for normal liver as the threshold for treatment plan modification as recommended by the current guidelines (45), and based on this criteria, we estimated that 44.19% (19/43) of treatment plans would be modified due to liver absorbed dose limitations. This finding highlights the potential role of Q.Liver software to optimize treatment planning and improve tumor response while maintaining safe dose limits to critical organs.

In addition to these findings, several methodological considerations related to quantitative SPECT imaging should be acknowledged. Reconstruction-dependent variability, including the number of iterations and subsets, and regularization, along with the implementation of attenuation, scatter, and resolution recovery corrections, may affect activity quantification, particularly in regions with low counts, such as the lungs. Respiratory motion mismatch between free-breathing SPECT and breath-hold CT can also introduce spatial misregistration and uncertainty regarding the boundaries of the VOI, particularly near the liver dome. Although Q.Liver incorporates AI-assisted segmentation, minor manual adjustments are still necessary, which may introduce variability in organ-level dose estimation. Furthermore, lung activity is generally low, and poor counting statistics can introduce additional uncertainty in LSF quantification. Partial volume effects also remain an inherent limitation of SPECT. Although Q.Liver includes PSF-based resolution recovery that partially compensates for PVE, it does not provide full partial-volume correction. This may result in conservative activity estimates in small lesions.

The main limitation of this study is the small sample size, and further prospective studies with larger patient cohorts are needed to confirm these findings and evaluate the clinical outcomes of using SPECT/CT-based LSF calculations in yttrium-90 TARE treatment plans. Additionally, the treatment protocols calculated by the Q.Liver software have not yet been implemented in clinical practice. Finally, voxel-based dosimetry and dose-volume histogram analysis would further enhance the interpretation of our results, but they could not be feasibly performed in this retrospective study. The version of the Q.Liver software used in this study does not yet support voxel-level dose kernels or dose-volume histogram export for MAA imaging, and complete follow-up data were not available for all patients. A full outcome-dose correlation would require prospective acquisition of yttrium-90 PET/CT scans, harmonized clinical endpoints, and institutional review board approval for the collection of additional data. As these elements were beyond the scope of the present methodological comparison, we plan to incorporate them into future prospective studies as software capabilities and institutional workflows evolve.

Conclusions

This study demonstrates that SPECT/CT-based Q.Liver software allows for the simultaneous acquisition of LSF and liver and lung volumes, simplifying current yttrium-90 treatment. Planar scintigraphy overestimates the LSF of ^99mTc-MAA and the lung absorbed dose compared to SPECT/CT imaging in yttrium-90 TARE for patients with HCC or CRLM, resulting in the prescribed tumor dose being lower than it should be. If the lung absorbed dose remains constant in the treatment planning system, using SPECT/CT imaging to calculate the LSF will increase the tumor and liver absorbed doses, leading to changes in 44.19% of treatment plans due to liver absorbed dose limitations. Our findings highlight the importance of using SPECT/CT imaging to quantify LSF and optimize treatment planning of TARE with yttrium-90 resin microspheres for patients with liver cancer.

Acknowledgments

We would like to thank GE Healthcare for providing the Xeleris 5 workstation and Q.Liver software for this study.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1471/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1471/dss

Funding: This study was supported by the National Natural Science Foundation of China (grant No. 82272106), and the Clinical Research of the First Affiliated Hospital of Xi’an Jiaotong University, China (No. XJTU1AF2021CRF-024).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1471/coif). J.X. reports grant from National Natural Science Foundation of China (No. 82272106), Clinical Research Funding from the First Affiliated Hospital of Xi’an Jiaotong University, China (No. XJTU1AF2021CRF-024), and supports from GE Healthcare (Xeleris 5 workstation and Q.Liver software). The other authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was approved by the Ethics Committee of the First Affiliated Hospital of Xi’an Jiaotong University (No. X-XJTU1AF2025LSYY-436) and individual consent was not required for this retrospective analysis. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Global cancer burden growing, amidst mounting need for services. Saudi Med J 2024;45:326-7.
2. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
3. Uhlig J, Lukovic J, Dawson LA, Patel RA, Cavnar MJ, Kim HS. Locoregional Therapies for Colorectal Cancer Liver Metastases: Options Beyond Resection. Am Soc Clin Oncol Educ Book 2021;41:133-46. [Crossref] [PubMed]
4. Valderrama-Treviño AI, Barrera-Mera B, Ceballos-Villalva JC, Montalvo-Javé EE. Hepatic Metastasis from Colorectal Cancer. Euroasian J Hepatogastroenterol 2017;7:166-75. [Crossref] [PubMed]
5. Reboux N, Jooste V, Goungounga J, Robaszkiewicz M, Nousbaum JB, Bouvier AM. Incidence and Survival in Synchronous and Metachronous Liver Metastases From Colorectal Cancer. JAMA Netw Open 2022;5:e2236666. [Crossref] [PubMed]
6. Ballarò C, Quaranta V, Giannelli G. Colorectal Liver Metastasis: Can Cytokines Make the Difference? Cancers (Basel) 2023;15:5359. [Crossref] [PubMed]
7. European Association for the Study of the Liver. EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol. 2018;69:182-236. Erratum in: J Hepatol 2019;70:817. [Crossref] [PubMed]
8. Llovet JM, Villanueva A, Marrero JA, Schwartz M, Meyer T, Galle PR, Lencioni R, Greten TF, Kudo M, Mandrekar SJ, Zhu AX, Finn RS, Roberts LRAASLD Panel of Experts on Trial Design in HCC. Trial Design and Endpoints in Hepatocellular Carcinoma: AASLD Consensus Conference. Hepatology 2021;73:158-91. [Crossref] [PubMed]
9. Kim SP, Cohalan C, Kopek N, Enger SA. A guide to (90)Y radioembolization and its dosimetry. Phys Med 2019;68:132-45. [Crossref] [PubMed]
10. Salem R, Gordon AC, Mouli S, Hickey R, Kallini J, Gabr A, Mulcahy MF, Baker T, Abecassis M, Miller FH, Yaghmai V, Sato K, Desai K, Thornburg B, Benson AB, Rademaker A, Ganger D, Kulik L, Lewandowski RJ. Y90 Radioembolization Significantly Prolongs Time to Progression Compared With Chemoembolization in Patients With Hepatocellular Carcinoma. Gastroenterology 2016;151:1155-1163.e2. [Crossref] [PubMed]
11. Dhondt E, Lambert B, Hermie L, Huyck L, Vanlangenhove P, Geerts A, Verhelst X, Aerts M, Vanlander A, Berrevoet F, Troisi RI, Van Vlierberghe H, Defreyne L. (90)Y Radioembolization versus Drug-eluting Bead Chemoembolization for Unresectable Hepatocellular Carcinoma: Results from the TRACE Phase II Randomized Controlled Trial. Radiology 2022;303:699-710. [Crossref] [PubMed]
12. Salem R, Johnson GE, Kim E, Riaz A, Bishay V, Boucher E, Fowers K, Lewandowski R, Padia SA. Yttrium-90 Radioembolization for the Treatment of Solitary, Unresectable HCC: The LEGACY Study. Hepatology 2021;74:2342-52. [Crossref] [PubMed]
13. Lee HM, Alder L, Nguyen M, Dougherty SC, Qu Y, Thacker LR, Poklepovic A. Long-term outcome analysis of Y90 radioembolization in hepatocellular carcinoma. J Gastrointest Oncol 2023;14:1378-91. [Crossref] [PubMed]
14. Lee RC, Liang PC, Liang HL, Chen YF, Yu CY, Cheng PN, Hung CF, Hsia CY, Lai HC, Ho MC, Cheng YF, Liu YS, Chao Y, Chen CH. Multicenter evaluation of the safety and efficacy of selective internal radiation therapy with yttrium-90 resin microspheres in Taiwan: data from the RESIN registry. J Gastroenterol Hepatol 2024;39:1318-27. [Crossref] [PubMed]
15. Chiesa C, Sjogreen-Gleisner K, Walrand S, Strigari L, Flux G, Gear J, Stokke C, Gabina PM, Bernhardt P, Konijnenberg M. EANM dosimetry committee series on standard operational procedures: a unified methodology for (99m)Tc-MAA pre- and (90)Y peri-therapy dosimetry in liver radioembolization with (90)Y microspheres. EJNMMI Phys 2021;8:77. [Crossref] [PubMed]
16. Allred JD, Niedbala J, Mikell JK, Owen D, Frey KA, Dewaraja YK. The value of (99m)Tc-MAA SPECT/CT for lung shunt estimation in (90)Y radioembolization: a phantom and patient study. EJNMMI Res 2018;8:50. [Crossref] [PubMed]
17. Weber M, Lam M, Chiesa C, Konijnenberg M, Cremonesi M, Flamen P, Gnesin S, Bodei L, Kracmerova T, Luster M, Garin E, Herrmann K. EANM procedure guideline for the treatment of liver cancer and liver metastases with intra-arterial radioactive compounds. Eur J Nucl Med Mol Imaging 2022;49:1682-99. [Crossref] [PubMed]
18. Georgiou MF, Kuker RA, Studenski MT, Ahlman PP, Witte M, Portelance L. Lung shunt fraction calculation using (99m)Tc-MAA SPECT/CT imaging for (90)Y microsphere selective internal radiation therapy of liver tumors. EJNMMI Res 2021;11:96. [Crossref] [PubMed]
19. Spina JC, Hume I, Pelaez A, Peralta O, Quadrelli M, Garcia Monaco R. Expected and Unexpected Imaging Findings after (90)Y Transarterial Radioembolization for Liver Tumors. Radiographics 2019;39:578-95. [Crossref] [PubMed]
20. Wendler T, Kreissl MC, Schemmer B, Rogasch JMM, De Benetti F. Artificial Intelligence-powered automatic volume calculation in medical images - available tools, performance and challenges for nuclear medicine. Nuklearmedizin 2023;62:343-53. [Crossref] [PubMed]
21. Chen W, Zhang T, Xu L, Zhao L, Liu H, Gu LR, Wang DZ, Zhang M. Radiomics Analysis of Contrast-Enhanced CT for Hepatocellular Carcinoma Grading. Front Oncol 2021;11:660509. [Crossref] [PubMed]
22. Marquis H, Ocampo Ramos JC, Carter LM, Zanzonico P, Bolch WE, Laforest R, Kesner AL. MIRD Pamphlet No. 29: MIRDy90-A (90)Y Research Microsphere Dosimetry Tool. J Nucl Med 2024;65:794-802. [Crossref] [PubMed]
23. Ho S, Lau WY, Leung TW, Chan M, Ngar YK, Johnson PJ, Li AK. Partition model for estimating radiation doses from yttrium-90 microspheres in treating hepatic tumours. Eur J Nucl Med 1996;23:947-52. [Crossref] [PubMed]
24. Gulec SA, Mesoloras G, Stabin M. Dosimetric techniques in 90Y-microsphere therapy of liver cancer: The MIRD equations for dose calculations. J Nucl Med 2006;47:1209-11.
25. Levillain H, Bagni O, Deroose CM, Dieudonné A, Gnesin S, Grosser OS, et al. International recommendations for personalised selective internal radiation therapy of primary and metastatic liver diseases with yttrium-90 resin microspheres. Eur J Nucl Med Mol Imaging 2021;48:1570-84. [Crossref] [PubMed]
26. National Nuclear Data Center. Nuclear decay data in the MIRD format. [Accessed 1 February 2024]. Available online: https://www.nndc.bnl.gov/nudat3/mird/
27. Sirtex. SIR-Spheres® 90Y resin microspheres. Published 16 December 2019. [Accessed 1 February 2024]. Available online: https://www.sirtex.com/media/55rprxv2/ssl-us-14-sir-spheres-microspheres-ifu-us.pdf
28. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018;68:394-424. [Crossref] [PubMed]
29. Luu MH, Mai HS, Pham XL, Le QA, Le QK, Walsum TV, Le NH, Franklin D, Le VH, Moelker A, Chu DT, Trung NL. Quantification of liver-Lung shunt fraction on 3D SPECT/CT images for selective internal radiation therapy of liver cancer using CNN-based segmentations and non-rigid registration. Comput Methods Programs Biomed 2023;233:107453. [Crossref] [PubMed]
30. Kao YH, Magsombol BM, Toh Y, Tay KH, Chow PKh, Goh AS, Ng DC. Personalized predictive lung dosimetry by technetium-99m macroaggregated albumin SPECT/CT for yttrium-90 radioembolization. EJNMMI Res 2014;4:33. [Crossref] [PubMed]
31. Gill H, Hiller J. Systematic review of lung shunt fraction quantification comparing SPECT/CT and planar scintigraphy for yttrium 90 radioembolization planning. Clinical and Translational Imaging 2021;9:181-8.
32. Lu Z, Chen G, Jiang H, Sun J, Lin KH, Mok GSP. SPECT and CT misregistration reduction in [99mTc]Tc-MAA SPECT/CT for precision liver radioembolization treatment planning. Eur J Nucl Med Mol Imaging 2023;50:2319-30.
33. Lopez B, Mahvash A, Lam MGEH, Kappadath SC. Calculation of lung mean dose and quantification of error for (90) Y-microsphere radioembolization using (99m) Tc-MAA SPECT/CT and diagnostic chest CT. Med Phys 2019;46:3929-40. [Crossref] [PubMed]
34. Lopez BP, Mahvash A, Long JP, Lam MGEH, Kappadath SC. Factors modulating (99m) Tc-MAA planar lung dosimetry for (90) Y radioembolization. J Appl Clin Med Phys 2022;23:e13734. [Crossref] [PubMed]
35. Kappadath SC, Lopez BP, Salem R, Lam MGEH. Reassessment of the lung dose limits for radioembolization. Nucl Med Commun 2021;42:1064-75. [Crossref] [PubMed]
36. Struycken L, Patel M, Kuo P, Hennemeyer C, Woodhead G, McGregor H. Clinical and Dosimetric Implications of Calculating Lung Shunt Fraction for Hepatic (90)Y Radioembolization Using SPECT/CT Versus Planar Scintigraphy. AJR Am J Roentgenol 2022;218:728-37. [Crossref] [PubMed]
37. Stella M, van Rooij R, Lam MGEH, de Jong HWAM, Braat AJAT. Lung Dose Measured on Postradioembolization (90)Y PET/CT and Incidence of Radiation Pneumonitis. J Nucl Med 2022;63:1075-80. [Crossref] [PubMed]
38. Elsayed M, Cheng B, Xing M, Sethi I, Brandon D, Schuster DM, Bercu Z, Galt J, Barron B, Kokabi N. Comparison of Tc-99m MAA Planar Versus SPECT/CT Imaging for Lung Shunt Fraction Evaluation Prior to Y-90 Radioembolization: Are We Overestimating Lung Shunt Fraction? Cardiovasc Intervent Radiol 2021;44:254-60. [Crossref] [PubMed]
39. Kim TP, Gandhi RT, Tolakanahalli R, Herrera R, Chuong MD, Gutierrez AN, Alvarez D. Establishing Updated Safety Standards for Independent (99)(m)Tc-MAA SPECT/CT Treatment Planning in Radioembolization. Int J Radiat Oncol Biol Phys 2024;119:1285-96. [Crossref] [PubMed]
40. Kunnen B, Dietze MMA, Braat AJAT, Lam MGEH, Viergever MA, de Jong HWAM. Feasibility of imaging (90) Y microspheres at diagnostic activity levels for hepatic radioembolization treatment planning. Med Phys 2020;47:1105-14. [Crossref] [PubMed]
41. Kim HC. Radioembolization for the treatment of hepatocellular carcinoma. Clin Mol Hepatol 2017;23:109-14. [Crossref] [PubMed]
42. Boston Scientific. Instructions for use: TheraSphere™ yttrium-90 glass microspheres. Revised March 2021. [Accessed 1 February 2024]. Available online: https://www.accessdata.fda.gov/cdrh_docs/pdf20/P200029C.pdf
43. Hermann AL, Dieudonné A, Ronot M, Sanchez M, Pereira H, Chatellier G, Garin E, Castera L, Lebtahi R, Vilgrain VSARAH Trial Group. Relationship of Tumor Radiation-absorbed Dose to Survival and Response in Hepatocellular Carcinoma Treated with Transarterial Radioembolization with (90)Y in the SARAH Study. Radiology 2020;296:673-84. [Crossref] [PubMed]
44. Sarwar A, Malik MS, Vo NH, Tsai LL, Tahir MM, Curry MP, Catana AM, Bullock AJ, Parker JA, Eckhoff DE, Nasser IA, Weinstein JL, Ahmed M. Efficacy and Safety of Radiation Segmentectomy with (90)Y Resin Microspheres for Hepatocellular Carcinoma. Radiology 2024;311:e231386. [Crossref] [PubMed]
45. Chiesa C, Mira M, Maccauro M, Romito R, Spreafico C, Sposito C, Bhoori S, Morosi C, Pellizzari S, Negri A, Civelli E, Lanocita R, Camerini T, Bampo C, Carrara M, Seregni E, Marchianò A, Mazzaferro V, Bombardieri E. A dosimetric treatment planning strategy in radioembolization of hepatocarcinoma with 90Y glass microspheres. Q J Nucl Med Mol Imaging 2012;56:503-8.