3D image-based dosimetry for Yttrium-90 radioembolization of hepatocellular carcinoma: Impact of imaging method on absorbed dose estimates

DosePatch: physics-inspired cropping layout for patch-based Monte Carlo simulations to provide fast and accurate internal dosimetry

BACKGROUND

Dosimetry-based personalized therapy was shown to have clinical benefits e.g. in liver selective internal radiation therapy (SIRT). Yet, there is no consensus about its introduction into clinical practice, mainly as Monte Carlo simulations (gold standard for dosimetry) involve massive computation time. We addressed the problem of computation time and tested a patch-based approach for Monte Carlo simulations for internal dosimetry to improve parallelization. We introduce a physics-inspired cropping layout for patch-based MC dosimetry, and compare it to cropping layouts of the literature as well as dosimetry using organ-S-values, and dose kernels, taking whole-body Monte Carlo simulations as ground truth. This was evaluated in five patients receiving Yttrium-90 liver SIRT.

RESULTS

The patch-based Monte Carlo approach yielded the closest results to the ground truth, making it a valid alternative to the conventional approach. Our physics-inspired cropping layout and mosaicking scheme yielded a voxel-wise error of < 2% compared to whole-body Monte Carlo in soft tissue, while requiring only ≈  10% of the time.

CONCLUSIONS

This work demonstrates the feasibility and accuracy of physics-inspired cropping layouts for patch-based Monte Carlo simulations.

Quantitative evaluation of 90Y-PET/CT and 90Y-SPECT/CT-based dosimetry following Yttrium-90 radioembolization

BACKGROUND

Following yttrium-90 radioembolization (90Y-RE), 90Y-PET/CT and 90Y-SPECT/CT imaging provide the means to calculate the voxelized absorbed dose distribution. Given the widespread use of the two imaging modalities and lack of well-established standardized dosimetry protocols for 90Y-RE, there is a clinical need to systematically investigate and evaluate differences in the performance of voxel-based dosimetry between 90Y-PET/CT and 90Y-SPECT/CT.

PURPOSE

To quantitatively analyze and compare 90Y-PET/CT and 90Y-SPECT/CT-based dosimetry following 90Y-RE.

METHODS

90Y-PET/CT and 90Y-SPECT/CT imaging was acquired for 35 patients following 90Y-RE with TheraSphere for the treatment of unresectable hepatocellular carcinoma. Dosimetry was performed using the local deposition method with known activity and the mean dose (Dmean) was calculated for perfused liver volumes (PV), tumors (T), and perfused normal livers (NL). Additionally, the absorbed dose to x% of the volume (Dx, x ∈ $ \in $ [5%, 10%, …, 90%, 95%]) and the volume receiving y Gy (Vy, y ∈ $ \in $ [10 Gy, 20 Gy, …, 190 Gy, 200 Gy]) were calculated for T and NL, respectively. Dose metrics were compared using linear regression, Bland-Altman analysis, and statistical testing.

RESULTS

Both 90Y-SPECT/CT and 90Y-PET/CT-based tumor Dmean were strongly correlated (R2 ≥ 0.90) with Dx, excluding metrics on the extrema. Intra-modality comparisons of various Dx and Vy metrics yielded statistically significant differences (ANOVA, p < 0.001) for both90Y-PET/CT and 90Y-SPECT/CT. Based on statistical testing, only Dx metrics separated by greater than 20%-30% coverage, and only Vy metrics separated by greater than 40-70 Gy, reported significant differences. For PV, there was a strong correlation (R2 ≥ 0.99) between Dmean derived separately from 90Y-PET/CT and 90Y-SPECT/CT imaging. The strength of the correlation was slightly reduced for T and NL with R2 = 0.91 and R2 = 0.95, respectively. For PV, the mean bias ± standard error (SE) and 95% limits of agreement (LOA) between Dmean from the two modalities was effectively zero with -0.8 ± 0.4% (± 2.5%). For T and NL, the mean bias ± SE (± LOA) was -14.5 ± 3.7% (± 24%) and 9.4 ± 4.7% (± 27%), respectively.

CONCLUSION

The strong correlation between Dmean and Dx suggests information from multiple dose metrics (e.g., D70 and Dmean) is largely redundant when establishing dose-response relationships in 90Y-RE. Dmean is highly correlated between 90Y-PET/CT and 90Y-SPECT/CT-based dosimetry, for all liver VOIs. Relative to 90Y-SPECT/CT, 90Y-PET/CT, on average, yielded higher Dmean to tumors (14%) and lower Dmean to perfused normal livers (9%). Absorbed dose differences for perfused liver volumes between 90Y-SPECT/CT and 90Y-PET/CT were negligible.

Role of Flex-Dose Delivery Program in Patients Affected by HCC: Advantages in Management of Tare in Our Experience

Background: Introduced in the latest BCLC 2022, endovascular trans-arterial radioembolization (TARE) has an important role in the treatment of unresectable hepatocellular carcinoma (HCC) as a "bridge" or "downstaging" of disease. The evolution of TARE technology allows a more flexible and personalized target treatment, based on the anatomy and vascular characteristics of each HCC. The flex-dose delivery program is part of this perspective, which allows us to adjust the dose and its radio-embolizing power in relation to the size and type of cancer and to split the therapeutic dose of Y90 in different injections (split-bolus). Methods: From January 2020 to January 2022, we enrolled 19 patients affected by unresectable HCC and candidates for TARE treatment. Thirteen patients completed the treatment following the flex-dose delivery program. Response to treatment was assessed using the mRECIST criteria with CT performed 6 and 9 months after treatment. Two patients did not complete the radiological follow-up and were not included in this retrospective study. The final cohort of this study counts eleven patients. Results: According to mRECIST criteria, six months of follow-up were reported: five cases of complete response (CR, 45.4% of cases), four cases of partial response (PR, 36.4%), and two cases of progression disease (PD, 18.2%). Nine months follow-up reported five cases of complete response (CR, 45.4%), two cases of partial response (PR, 18.2%), and four cases of progression disease (PD, 36.4%). No intra and post-operative complications were described. The average absorbed doses to the hepatic lesion and to the healthy liver tissue were 319 Gy (range 133-447 Gy) and 9.5 Gy (range 2-19 Gy), respectively. Conclusions: The flex-dose delivery program represents a therapeutic protocol capable of "saving" portions of healthy liver parenchyma by designing a "custom-made" treatment for the patient.

Comparison of absorbed doses to the tumoral and non-tumoral liver in HCC patients undergoing 99mTc-MAA and 90Y-microspheres radioembolization

PURPOSE

This study aimed to determine the absorbed doses in the tumoral-liver and non-tumoral liver of hepatocellular carcinoma (HCC) patients undergoing radioembolization with Yttrium-90 (90Y) resin microspheres, and compared with those derived from 99mTc-MAA using the partition model.

METHODS

A total of 42 HCC patients (28 males and 14 females, mean age 65 ± 11.51 years) who received 45 treatment sessions with 90Y-microspheres between 2016 and 2021 were included. Pre-treatment 99mTc-MAA and post-treatment 90Y-bremsstrahlung SPECT/CT were acquired for each patient. Semi-automated segmentation of regions of interest (ROIs) was performed using MIM Encore software to determine the tumor-liver ratio (TLR) encompassing the liver volume, tumoral-liver, and lungs, and verified by both nuclear medicine physician and interventional radiologist. A partition dosimetry model was used to estimate the administered activity of 90Y-microspheres and the absorbed doses to the tumoral-liver and non-tumoral liver. The student's paired t test and Bland-Altman plot were used for the statistical analysis.

RESULTS

The mean TLR values obtained from 99mTc-MAA SPECT/CT and 90Y-bremsstrahlung SPECT/CT were 4.78 ± 3.51 and 2.73 ± 1.18, respectively. The mean planning administered activity of 90Y-microspheres based on 99mTc-MAA SPECT/CT was 1.56 ± 0.80 GBq, while the implanted administered activity was 2.53 ± 1.23 GBq (p value < 0.001). The mean absorbed doses in the tumoral-liver estimated from 99mTc-MAA and 90Y-bremsstrahlung SPECT/CT were 127.44 ± 4.36 Gy and 135.98 ± 6.30 Gy, respectively. The corresponding mean absorbed doses in the non-tumoral liver were 34.61 ± 13.93 Gy and 55.04 ± 16.36 Gy.

CONCLUSION

This study provides evidence that the administered activity of 90Y-microspheres, as estimated from 90Y-bremsstrahlung SPECT/CT, was significantly higher than that estimated from 99mTc-MAA SPECT/CT resulted in increased absorbed doses in both the tumoral-liver and non-tumoral liver. However, 99mTc-MAA SPECT/CT remains a valuable planning tool for predicting the distribution of 90Y-microspheres in liver cancer treatment.

Phantom study for 90Y liver radioembolization dosimetry with a long axial field-of-view PET/CT

PURPOSE

The physical properties of yttrium-90 (90Y) allow for imaging with positron emission tomography/computed tomography (PET/CT). The increased sensitivity of long axial field-of-view (LAFOV) PET/CT scanners possibly allows to overcome the small branching ratio for positron production from 90Y decays and to improve for the post-treatment dosimetry of 90Y of selective internal radiation therapy.

METHODS

For the challenging case of an image quality body phantom, we compare a full Monte Carlo (MC) dose calculation with the results from the two commercial software packages Simplicit90Y and Hermes. The voxel dosimetry module of Hermes relies on the 90Y images taken with a LAFOV PET/CT, while the MC and Simplicit90Y dose calculations are image independent.

RESULTS

The resulting doses from the MC calculation and Simplicit90Y agree well within the error margins. The image-based dose calculation with Hermes, however, consistently underestimates the dose. This is due to the mismatch of the activity distribution in the PET images and the size of the volume of interest. We found that only for the smallest phantom sphere there is a statistically significant dependence of the Hermes dose on the image reconstruction parameters and scan time.

CONCLUSION

Our study shows that Simplicit90Y's local deposition model can provide a reliable dose estimate. On the other hand, the image based dose calculation suffers from the suboptimal reconstruction of the 90Y distribution in small structures.

AAPM Medical Physics Practice Guideline 14.a: Yttrium-90 microsphere radioembolization

Radioembolization using Yttrium-90 (90 Y) microspheres is widely used to treat primary and metastatic liver tumors. The present work provides minimum practice guidelines for establishing and supporting such a program. Medical physicists play a key role in patient and staff safety during these procedures. Products currently available are identified and their properties and suppliers summarized. Appropriateness for use is the domain of the treating physician. Patient work up starts with pre-treatment imaging. First, a mapping study using Technetium-99m (Tc-99m ) is carried out to quantify the lung shunt fraction (LSF) and to characterize the vascular supply of the liver. An MRI, CT, or a PET-CT scan is used to obtain information on the tumor burden. The tumor volume, LSF, tumor histology, and other pertinent patient characteristics are used to decide the type and quantity of 90 Y to be ordered. On the day of treatment, the appropriate dose is assayed using a dose calibrator with a calibration traceable to a national standard. In the treatment suite, the care team led by an interventional radiologist delivers the dose using real-time image guidance. The treatment suite is posted as a radioactive area during the procedure and staff wear radiation dosimeters. The treatment room, patient, and staff are surveyed post-procedure. The dose delivered to the patient is determined from the ratio of pre-treatment and residual waste exposure rate measurements. Establishing such a treatment modality is a major undertaking requiring an institutional radioactive materials license amendment complying with appropriate federal and state radiation regulations and appropriate staff training commensurate with their respective role and function in the planning and delivery of the procedure. Training, documentation, and areas for potential failure modes are identified and guidance is provided to ameliorate them.

A machine learning-based model for a dose point kernel calculation

PURPOSE

Absorbed dose calculation by kernel convolution requires the prior determination of dose point kernels (DPK). This study reports on the design, implementation, and test of a multi-target regressor approach to generate the DPKs for monoenergetic sources and a model to obtain DPKs for beta emitters.

METHODS

DPK for monoenergetic electron sources were calculated using the FLUKA Monte Carlo (MC) code for many materials of clinical interest and initial energies ranging from 10 to 3000 keV. Regressor Chains (RC) with three different coefficients regularization/shrinkage models were used as base regressors. Electron monoenergetic scaled DPKs (sDPKs) were used to assess the corresponding sDPKs for beta emitters typically used in nuclear medicine, which were compared against reference published data. Finally, the beta emitters sDPK were applied to a patient-specific case calculating the Voxel Dose Kernel (VDK) for a hepatic radioembolization treatment with [Formula: see text]Y.

RESULTS

The three trained machine learning models demonstrated a promising capacity to predict the sDPK for both monoenergetic emissions and beta emitters of clinical interest attaining differences lower than [Formula: see text] in the mean average percentage error (MAPE) as compared with previous studies. Furthermore, differences lower than [Formula: see text] were obtained for the absorbed dose in patient-specific dosimetry comparing against full stochastic MC calculations.

CONCLUSION

An ML model was developed to assess dosimetry calculations in nuclear medicine. The implemented approach has shown the capacity to accurately predict the sDPK for monoenergetic beta sources in a wide range of energy in different materials. The ML model to calculate the sDPK for beta-emitting radionuclides allowed to obtain VDK useful to achieve reliable patient-specific absorbed dose distributions required short computation times.

Lesion Quantification Accuracy of Digital 90Y PET Imaging in the Context of Dosimetry in Systemic Fibroblast Activation Protein Inhibitor Radionuclide Therapy

Therapy with ⁹⁰Y-labeled fibroblast activation protein inhibitors (⁹⁰Y-FAPIs) was recently introduced as a novel treatment concept for patients with solid tumors. Lesion and organ-at-risk dosimetry is part of assessing treatment efficacy and safety and requires reliable quantification of tissue uptake. As ⁹⁰Y quantification is limited by the low internal positron-electron pair conversion rate, the increased effective sensitivity of digital silicon photomultiplier-based PET/CT systems might increase quantification accuracy and, consequently, allow for dosimetry in ⁹⁰Y-FAPI therapy. The aim of this study was to explore the conditions for reliable lesion image quantification in ⁹⁰Y-FAPI radionuclide therapy using a digital PET/CT system. Methods: Two tumor phantoms were filled with ⁹⁰Y solution using different sphere activity concentrations and a constant signal-to-background ratio of 40. The minimum detectable activity concentration was determined, and its dependence on acquisition time (15 vs. 30 min per bed position) and smoothing levels (all-pass vs. 5-mm gaussian filter) was investigated. Quantification accuracy was evaluated at various activity concentrations to estimate the minimum quantifiable activity concentration using contour-based and oversized volume-of-interest-based quantification approaches. A ±20% deviation range between image-derived and true activity concentrations was regarded as acceptable. Tumor dosimetry for 3 patients treated with ⁹⁰Y-FAPI is presented to project the phantom results to clinical scenarios. Results: For a lesion size of 40 mm and a clinical acquisition time of 15 min, both minimum detectable and minimum quantifiable activity concentrations were 0.12 MBq/mL. For lesion sizes of greater than or equal to 30 mm, accurate quantification was feasible for detectable lesions. Only for the smallest 10-mm sphere, the minimum detectable and minimum quantifiable activity concentrations differ substantially (0.43 vs. 1.97 MBq/mL). No notable differences between the 2 quantification approaches were observed. For the investigated tumors, absorbed dose estimates with reliable accuracy were achievable. Conclusion: For lesion sizes and activity concentrations that are expected to be observed in patients treated with ⁹⁰Y-FAPI, quantification with reasonable accuracy is possible. Further dosimetry studies are needed to thoroughly investigate the efficacy and safety of ⁹⁰Y-FAPI therapy.

A Machine Learning based model for a Dose Point Kernel calculation

PURPOSE

Absorbed dose calculation by kernel convolution requires the prior determination of dose point kernels (DPK). This study shows applications of machine learning to generate the DPKs for monoenergetic sources and a model to obtain DPKs for beta emitters.

METHODS

DPK for monoenergetic electron sources were calculated using the FLUKA Monte Carlo (MC) code for many materials of clinical interest and initial energies ranging from 10 to 3000 keV. Three machine learning (ML) algorithms were trained using the MC DPKs. Electron monoenergetic scaled DPKs (sDPKs) were used to assess the corresponding sDPKs for beta emitters typically used in nuclear medicine, which were compared against reference published data. Finally, the ML sDPK approach was applied to a patient-specific case calculating the dose voxel kernels (DVK) for a hepatic radioembolization treatment with \(^{90}\)Y.

RESULTS

The three trained machine learning models demonstrated a promising capacity to predict the sDPK for both monoenergetic emissions and beta emitters of clinical interest attaining differences lower than \(10%\) in the mean average percentage error (MAPE) as compared with previous studies. Furthermore, differences lower than \(7 %\) were obtained for the absorbed dose in patient-specific dosimetry comparing against full stochastic MC calculations.

CONCLUSION

An ML model was developed to assess dosimetry calculations in nuclear medicine. The implemented approach has shown the capacity to accurately predict the sDPK for monoenergetic beta sources in a wide range of energy in different materials. The ML model to calculate the sDPK for beta-emitting radionuclides allowed to obtain VDK useful to achieve reliable patient-specific absorbed dose distributions required remarkable short computation times.

Utilizing 3D Slicer to incorporate tomographic images into GATE Monte Carlo simulation for personalized dosimetry in yttrium-90 radioembolization

PURPOSE

Monte Carlo (MC) simulation is an important technique that can help design advanced and challenging experimental setups. GATE (Geant4 application for tomographic emission) is a useful simulation toolkit for applications in nuclear medicine. Transarterial radioembolization is a treatment for liver cancer, where microspheres embedded with yttrium-90 (⁹⁰ Y) are administered intra-arterially to the tumor. Personalized dosimetry for this treatment may provide higher dosimetry accuracy compared to the conventional partition model (PM) calculation. However, incorporation of three-dimensional tomographic input data into MC simulation is an intricate process. In this article, 3D Slicer, free and open-source software, was utilized for the incorporation of patient tomographic images into GATE to demonstrate the feasibility of personalized dosimetry in hepatic radioembolization with ⁹⁰ Y.

METHODS

In this article, the steps involved in importing, segmenting, and registering tomographic images using 3D Slicer were thoroughly described, before importing them into GATE for MC simulation. The absorbed doses estimated using GATE were then compared with that of PM. SlicerRT, a 3D Slicer extension, was then used to visualize the isodose from the MC simulation.

RESULTS

A workflow diagram consisting of all the steps taken in the utilization of 3D Slicer for personalized dosimetry in ⁹⁰ Y radioembolization has been presented in this article. In comparison to the MC simulation, the absorbed doses to the tumor and normal liver were overestimated by PM by 105.55% and 20.23%, respectively, whereas for lungs, the absorbed dose estimated by PM was underestimated by 25.32%. These values were supported by the isodose distribution obtained via SlicerRT, suggesting the presence of beta particles outside the volumes of interest. These findings demonstrate the importance of personalized dosimetry for a more accurate absorbed dose estimation compared to PM.

CONCLUSION

The methodology provided in this study can assist users (especially students or researchers who are new to MC simulation) in navigating intricate steps required in the importation of tomographic data for MC simulation. These steps can also be utilized for other radiation therapy related applications, not necessarily limited to internal dosimetry.

Towards personalised dosimetry in patients with liver malignancy treated with 90Y-SIRT using in vivo-driven radiobiological parameters

BACKGROUND

The prediction of response is one of the major challenges in radiation-based therapies. Although the selection of accurate linear-quadratic model parameters is essential for the estimation of radiation response and treatment outcome, there is a limited knowledge about these radiobiological parameters for liver tumours using radionuclide treatments.

METHODS

The "clinical radiobiological" parameters ([Formula: see text], [Formula: see text], [Formula: see text], [Formula: see text]) for twenty-five patients were derived using the generalised linear-quadratic model, the diagnostic ([¹⁸F] FDG PET/CT) and therapeutic ([⁹⁰Y]-SIR-Spheres PET/CT) images to compute the biological effective dose and tumour control probability (TCP) for each patient.

RESULTS

It was estimated that the values for [Formula: see text] and [Formula: see text] parameters range in ≈ 0.001-1 Gy^-1 and ≈ 1-49 Gy, respectively. We have demonstrated that the time factors, [Formula: see text], [Formula: see text] and [Formula: see text] are the key parameters when evaluating liver malignancy lesional response to [⁹⁰Y]SIR-Spheres treatment. Patients with cholangiocarcinoma have been shown to have the longest average [Formula: see text] (≈ 236 ± 67 d), highest TCP (≈ 53 ± 17%) and total liver lesion glycolysis response ([Formula: see text] ≈ 64%), while patients with metastatic colorectal cancer tumours have the shortest average [Formula: see text] (≈ 129 ± 19 d), lowest TCP (≈ 28 ± 13%) and [Formula: see text] ≈ 8%, respectively.

CONCLUSIONS

Tumours with shorter [Formula: see text] have shown a shorter [Formula: see text] and thus poorer TCP and [Formula: see text]. Therefore, these results suggest for such tumours the [⁹⁰Y]SIR-Spheres will be only effective at higher initial dose rate (e.g.  > 50 Gy/day).

Impact of contouring methods on pre-treatment and post-treatment dosimetry for the prediction of tumor control and survival in HCC patients treated with selective internal radiation therapy

INTRODUCTION

The aim of this study was to evaluate the impact of the contouring methods on dose metrics and their predictive value on tumor control and survival, in both situations of pre-treatment and post-treatment dosimetry, for patients with advanced HCC treated with SIRT.

METHODS

Forty-eight patients who underwent SIRT between 2012 and 2020 were retrospectively included in this study. Target volumes were delineated using two methods: MRI-based contours manually drawn by a radiologist and then registered on SPECT/CT and PET/CT via deformable registration (Pre-C_MRI and Post-C_MRI), ^99mTc-MAA-SPECT and ⁹⁰Y-microspheres-PET 10% threshold contouring (Pre-C_SPECT and Post-C_PET). The mean absorbed dose (Dm) and the minimal absorbed dose delivered to 70% of the tumor volume (D70) were evaluated with both contouring methods; the tumor-to-normal liver uptake ratio (TNR) was evaluated with MRI-based contours only. Tumor response was assessed using the mRECIST criteria on the follow-up MRIs.

RESULTS

No significant differences were found for Dm and TNR between pre- and post-treatment. TNR evaluated with radiologic contours (Pre-C_MRI and Post-C_MRI) were predictive of tumor control at 6 months on pre- and post-treatment dosimetry (OR 5.9 and 7.1, respectively; p = 0.02 and 0.01). All dose metrics determined with both methods were predictive of overall survival (OS) on pre-treatment dosimetry, but only Dm with MRI-based contours was predictive of OS on post-treatment images with a median of 23 months for patients with a supramedian Dm versus 14 months for the others (p = 0.04).

CONCLUSION

In advanced HCC treated with SIRT, Dm and TNR determined with radiologic contours were predictive of tumor control and OS. This study shows that a rigorous clinical workflow (radiologic contours + registration on scintigraphic images) is feasible and should be prospectively considered for improving therapeutic strategy.