Mathematical complexities in radionuclide metabolic modeling: A review of ordinary differential equation kinetics solvers in biokinetic modeling

Biokinetic models have been employed in internal dosimetry to model the human body's time-dependent retention and excretion of radionuclides. Consequently, biokinetic models have become instrumental in modeling the body burden from biological processes from internalized radionuclides for prospective and retrospective dose assessment. Solutions to biokinetic equations have been modelled as a system of coupled ordinary differential equations (ODEs) representing the time-dependent distribution of materials deposited within the body. In parallel, several solving mathematical algorithms were developed for solving general kinetic problems, upon which biokinetic solution tools were constructed. This paper provides a comprehensive review of mathematical solving methods adopted by some known internal dose computer codes for modeling the distribution and dosimetry for internal emitters, highlighting the mathematical frameworks, their capabilities, and their limitations. Further discussion details the mathematical underpinnings of biokinetic solutions in a unique approach paralleling advancements in internal dosimetry with capabilities in available mathematical solvers in computational systems. A survey of ODE forms, methods, and solvers, including state-of-the-art solvers specifically in Python programming language, was conducted to highlight modern capabilities for advancing the utilization of modern toolkits in internal dosimetry. This review is the first of its kind, which provides a comprehensive analysis of biokinetic solving methods and base knowledge for understanding the computational demands, schemes, and implementations for biokinetic modeling, which can be leveraged for an expedited radiation dose assessment.

Absorbed dose coefficients for pediatric differentiated thyroid cancer patients undergoing radioiodine therapy

The escalating incidence of differentiated thyroid cancer (DTC) in pediatric patients and the resultant growing use of radioactive iodine (RAI) reinforce the need to evaluate radiation exposure to normal tissues and radiation-induced health risks in pediatric patients undergoing RAI therapy. In the current study, we calculated absorbed dose coefficients (i.e. absorbed dose per unit activity administered, mGy MBq-1) specific for pediatric patients with localized DTC undergoing RAI therapy following total thyroidectomy for use in epidemiological studies. We first modified previously-published biokinetic models for adult thyroid cancer patients to achieve a reasonable agreement with iodine biokinetics observed in pediatric patients or design principles addressed in the International Commission on Radiological Protection (ICRP) reference age-specific biokinetic models. We then combined the biokinetic models in conjunction withSvalues derived from ICRP reference pediatric voxel phantoms. The absorbed dose coefficients for pediatric patients were overall greater than those for adults with a ratio (pediatric/adult) up to 11.6 and rapidly decreased with increasing age. The sensitivity analysis showed that the renal clearance rate andSvalues may have the greatest impact on the absorbed dose coefficients with the rank correlation coefficients ranging from -0.53 to -0.82 (negative correlations) and from 0.51 to 0.80 (positive correlations), respectively. The results of the current study may be utilized in clinical or epidemiological studies to estimate organ-specific radiation absorbed doses and radiation-associated health risks among pediatric thyroid cancer patients.

Impact of deep learning-based multiorgan segmentation methods on patient-specific internal dosimetry in PET/CT imaging: A comparative study

PURPOSE

Accurate and fast multiorgan segmentation is essential in image-based internal dosimetry in nuclear medicine. While conventional manual PET image segmentation is widely used, it suffers from both being time-consuming as well as subject to human error. This study exploited 2D and 3D deep learning (DL) models. Key organs in the trunk of the body were segmented and then used as a reference for networks.

METHODS

The pre-trained p2p-U-Net-GAN and HighRes3D architectures were fine-tuned with PET-only images as inputs. Additionally, the HighRes3D model was alternatively trained with PET/CT images. Evaluation metrics such as sensitivity (SEN), specificity (SPC), intersection over union (IoU), and Dice scores were considered to assess the performance of the networks. The impact of DL-assisted PET image segmentation methods was further assessed using the Monte Carlo (MC)-derived S-values to be used for internal dosimetry.

RESULTS

A fair comparison with manual low-dose CT-aided segmentation of the PET images was also conducted. Although both 2D and 3D models performed well, the HighRes3D offers superior performance with Dice scores higher than 0.90. Key evaluation metrics such as SEN, SPC, and IoU vary between 0.89-0.93, 0.98-0.99, and 0.87-0.89 intervals, respectively, indicating the encouraging performance of the models. The percentage differences between the manual and DL segmentation methods in the calculated S-values varied between 0.1% and 6% with a maximum attributed to the stomach.

CONCLUSION

The findings prove while the incorporation of anatomical information provided by the CT data offers superior performance in terms of Dice score, the performance of HighRes3D remains comparable without the extra CT channel. It is concluded that both proposed DL-based methods provide automated and fast segmentation of whole-body PET/CT images with promising evaluation metrics. Between them, the HighRes3D is more pronounced by providing better performance and can therefore be the method of choice for 18F-FDG-PET image segmentation.

Mathematical solutions in internal dose assessment: A comparison of Python-based differential equation solvers in biokinetic modeling

In biokinetic modeling systems employed for radiation protection, biological retention and excretion have been modeled as a series of discretized compartments representing the organs and tissues of the human body. Fractional retention and excretion in these organ and tissue systems have been mathematically governed by a series of coupled first-order ordinary differential equations (ODEs). The coupled ODE systems comprising the biokinetic models are usually stiff due to the severe difference between rapid and slow transfers between compartments. In this study, the capabilities of solving a complex coupled system of ODEs for biokinetic modeling were evaluated by comparing different Python programming language solvers and solving methods with the motivation of establishing a framework that enables multi-level analysis. The stability of the solvers was analyzed to select the best performers for solving the biokinetic problems. A Python-based linear algebraic method was also explored to examine how the numerical methods deviated from an analytical or semi-analytical method. Results demonstrated that customized implicit methods resulted in an enhanced stable solution for the inhaled60Co (Type M) and131I (Type F) exposure scenarios for the inhalation pathway of the International Commission on Radiological Protection (ICRP) Publication 130 Human Respiratory Tract Model (HRTM). The customized implementation of the Python-based implicit solvers resulted in approximately consistent solutions with the Python-based matrix exponential method (expm). The differences generally observed between the implicit solvers andexpmare attributable to numerical precision and the order of numerical approximation of the numerical solvers. This study provides the first analysis of a list of Python ODE solvers and methods by comparing their usage for solving biokinetic models using the ICRP Publication 130 HRTM and provides a framework for the selection of the most appropriate ODE solvers and methods in Python language to implement for modeling the distribution of internal radioactivity.

Impact of Single-Time-Point Estimates of 177Lu-PRRT Absorbed Doses on Patient Management: Validation of a Trained Multiple-Linear-Regression Model in 159 Patients and 477 Therapy Cycles

Dosimetry after 177Lu-DOTATATE peptide receptor radionuclide therapy (PRRT) enables estimation of radiation doses absorbed by normal organs and target lesions. This process is time-consuming and requires multiple posttreatment studies on several subsequent days. In a previous study, we described a newly developed multiple-linear-regression model to predict absorbed doses (ADs) from a single-time-point (STP) posttreatment study acquired 168 h after the first infusion and 24 h after the following ones, with similar results to the standard multiple-time-point (MTP) protocol. The present study aimed to validate this model in a large patient cohort and to assess whether STP dosimetry affects patient management decisions compared with our MTP protocol. Methods: Quantitative 177Lu-DOTATATE SPECT/CT post-PRRT data from 159 consecutive patients (172 therapies, 477 therapy cycles) were retrospectively analyzed. ADs obtained from an STP model were compared with those obtained using an MTP model. We evaluated the impact of the STP model on the decision on whether PRRT should be stopped because of an expected kidney AD exceeding the safety threshold. We hypothesized that patient management based on the STP model does not differ from that based on the MTP model in at least 90% of the cases. Results: There was no difference in management decisions between the MTP and STP models in 170 of 172 therapies (98.8%). A Fisher χ2 test for combined probabilities produced a composite P value of 0.0003. Mean cumulative AD relative differences between the STP and MTP models were 0.8% ± 8.0%, -7.7% ± 4.8%, 0.0% ± 11.4%, -2.8% ± 6.3%, and -2.1% ± 18.4% for kidneys, bone marrow, liver, spleen, and tumors, respectively (Pearson r = 0.99 for all), for patients who underwent 4 therapy cycles. Similar results were obtained with fewer therapy cycles. Conclusion: Estimated radiation ADs and patient management decisions were similar with the STP and MTP models. The STP model can simplify the dosimetry process while also reducing scanner and staff time and improving patient comfort.

Deep Learning-Based Delayed PET Image Synthesis from Corresponding Early Scanned PET for Dosimetry Uptake Estimation

The acquisition of in vivo radiopharmaceutical distribution through imaging is time-consuming due to dosimetry, which requires the subject to be scanned at several time points post-injection. This study aimed to generate delayed positron emission tomography images from early images using a deep-learning-based image generation model to mitigate the time cost and inconvenience. Eighteen healthy participants were recruited and injected with [18F]Fluorodeoxyglucose. A paired image-to-image translation model, based on a generative adversarial network (GAN), was used as the generation model. The standardized uptake value (SUV) mean of the generated image of each organ was compared with that of the ground-truth. The least square GAN and perceptual loss combinations displayed the best performance. As the uptake time of the early image became closer to that of the ground-truth image, the translation performance improved. The SUV mean values of the nominated organs were estimated reasonably accurately for the muscle, heart, liver, and spleen. The results demonstrate that the image-to-image translation deep learning model is applicable for the generation of a functional image from another functional image acquired from normal subjects, including predictions of organ-wise activity for specific normal organs.

18F-PSMA-1007 salivary gland dosimetry: comparison between different methods for dose calculation and assessment of inter- and intra-patient variability

OBJECTIVE

Simplified calculation approaches and geometries are usually adopted for salivary
glands (SGs) dosimetry. Our aims were i) to compare different dosimetry methods to calculate
SGs absorbed doses (ADs) following [18F]-PSMA-1007 injection, and ii) to assess the AD
variation across patients and single SG components.
Approach: Five patients with prostate cancer underwent sequential PET/CT acquisitions of
the head and neck, 0.5, 2 and 4 hours after [18F]-PSMA-1007 injection. Parotid and
submandibular glands were segmented on CT to derive SGs volumes and masses, while PET
images were used to derive Time-Integrated Activity Coefficients. Average ADs to single SG
components or total SG (tSG) were calculated with the following methods: i) direct Monte
Carlo simulation with GATE/GEANT4 considering radioactivity in the entire PET/CT field-of-view (MC) or in the SGs only (MCsgo); ii) spherical model (SM) of OLINDA/EXM 2.1,
adopting either patient-specific or standard ICRP89 organ masses (SMstd); iii) ellipsoidal
model (EM); iv) MIRD approach with organ S-factors from OLINDA/EXM 2.1 and OpenDose
collaboration, with or without contribution from cross irradiation originating outside the SGs.
The maximum percent AD difference across SG components (δmax) and across patients (∆max)
were calculated.
Main results: Compared to MC, ADs to single SG components were significantly
underestimated by all methods (average relative differences ranging between -11.9% and -30.5%). δmax values were never below 25%. The highest δmax (=702%) was obtained with
SMstd. Concerning tSG, results within 10% of the MC were obtained only if cross-irradiation
from the remainder of the body or from the remainder of the head was accounted for. The ∆max
ranged between 58% and 78% across patients.
Significance: Simple geometrical models for SG dosimetry considerably underestimated ADs
compared to MC, particularly if neglecting cross-irradiation from neighboring regions. Specific masses of single SG components should always be considered given their large intra- and inter-patient variability.

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.

Estimation of Human Absorbed Dose of 188Re-Hynic-Bombesin Based on Biodistribution Data in Rats

BACKGROUND

HYNIC-Bombesin (BBN) is a potential peptide for targeted radionuclide therapy in gastrin-releasing peptide receptor (GRPr)-positive malignancies. The ¹⁸⁸Re-HYNICBBN is a promising radiopharmaceutical for use in prostate cancer therapy.

OBJECTIVE

The aim of this study was to estimate the absorbed dose due to ¹⁸⁸Re-HYNIC-BBN radio-complex in human organs based on bio-distribution data of rats.

METHODS

In this research, using bio-distribution data of ¹⁸⁸Re-HYNIC-BBN in rats, its radiation absorbed dose of the adult human was calculated for different organs based on the MIRD dose calculation method.

RESULTS

A considerable equivalent dose amount of ¹⁸⁸Re-Hynic-BBN (0.093 mGy/MBq) was accumulated in the prostate. Moreover, all other tissues except for the kidneys and pancreas approximately received insignificant absorbed doses.

CONCLUSION

Since the acceptable absorbed dose for the complex was observed in the prostate, ¹⁸⁸Re-Hynic-Bombesin can be regarded as a new potential agent for prostate cancer therapy.

An analytic model to calculate voxel s-values for177Lu

OBJECTIVE

¹⁷⁷Lu is one of the most employed isotopes in targeted radionuclide therapies and theranostics, and 3D internal dosimetry for such procedures has great importance. Voxel S-Values (VSVs) approach is widely used for this purpose, but VSVs are available for a limited number of voxel dimensions. The aim of this work is to develop an analytic model for the calculation of 177Lu-VSVs in any cubic voxelized geometry of practical interest.

APPROACH

Monte Carlo (MC) simulations were implemented with the toolkit GAMOS to evaluate VSVs in voxelized geometries of soft tissue from a source of¹⁷⁷Lu homogeneously distributed in the central voxel. Nine geometric setups, containing 15×15×15 cubic voxels of sideslranging from 2 mm to 6 mm, in steps of 0.5 mm, were considered. For eachl, the VSVs computed as a function of the "normalized radius",R_n= R/l(withR= distance from the center of the source voxel), were fitted with a parametric function. The dependencies of the parameters as a function oflwere then fitted with appropriate functions, in order to implement the model for deducing¹⁷⁷Lu-VSVs for anylwithin the aforementioned range.

MAIN RESULTS

The MC-derived VSVs were satisfactorily compared with literature data for validation, and the VSVs computed with the analytic model agree with the MC ones within 2% forR_n≤2 and within 6% forR_n>2.

SIGNIFICANCE

The proposed model enables the easy and fast calculation, with a simple spreadsheet, of¹⁷⁷Lu-VSVs in any cubic voxelized geometry of practical interest, avoiding the necessity of implementingad-hocMC simulations to estimate VSVs for specific voxel dimensions not available in literature data.

A feasibility study of the therapeutic application of a mixture of 67/64 Cu radioisotopes produced by cyclotrons with proton irradiation

PURPOSE

64 Cu and 67 Cu radioisotopes have nuclear characteristics suitable for nuclear medicine applications. The production of 64 Cu is already well established. However, the production of 67 Cu in quantities suitable to conduct clinical trials is more challenging as it leads to the coproduction of other Cu isotopes, in particular 64 Cu. The aim of this study is to investigate the possibility of using a CuCl2 solution with a mixture of 67/64 Cu radioisotopes for therapeutic purposes, providing an alternative solution for the cyclotron production problem.

METHODS

Copper radioisotopes activities were calculated by considering proton beam irradiation of the following targets: (i) 70 Zn in the energy range 70-45 MeV; (ii) 68 Zn in the energy range 70-35 MeV; (iii) a combination of 70 Zn (70-55 MeV) and 68 Zn (55-35 MeV). The contribution of each copper radioisotope to the human-absorbed dose was estimated with OLINDA/EXM software using the biokinetic model for CuCl2 published by ICRP 53. The total absorbed dose generated by the 67/64 CuCl2 mixture, obtained through different production routes, was calculated at different times after the end of the bombardment (EOB). A simple spherical model was used to simulate tumors of different sizes containing uniformly distributed 67/64 Cu mixture and to calculate the absorbed dose of self-irradiation. The biological damage produced by 67 Cu and 64 Cu was also evaluated through cellular dosimetry and cell surviving fraction assessment using the MIRDcell code, considering two prostate cancer cell lines with different radiosensitivity.

RESULTS

The absorbed dose to healthy organs and the effective dose (ED) per unit of administered activity of 67 CuCl2 are higher than those of 64 CuCl2 . Absorbed dose values per unit of administered activity of 67/64 CuCl2 mixture increase with time after the EOB because the amount of 67 Cu in the mixture increases. Survival data showed that the biological damage caused per each decay of 67 Cu is greater than that of 64 Cu, assuming that radionuclides remain accumulated in the cell cytoplasm. Sphere model calculations demonstrated that 64 Cu administered activity must be about five times higher than that of 67 Cu to obtain the same absorbed dose for tumor mass between 0.01 and 10 g and about 10 times higher for very small spheres. Consequently, the 64 CuCl2 -absorbed dose to healthy organs will reach higher values than those of 67 CuCl2 . The supplemental activity of the 67/64 CuCl2 mixture, required to get the same tumor-absorbed dose produced by 67 CuCl2 , triggers a dose increment (DI) in healthy organs. The waiting time post-EOB necessary to keep this DI below 10% (t10% ) depends on the irradiation methods employed for the production of the 67/64 CuCl2 mixture.

CONCLUSIONS

A mixture of cyclotron produced 67/64 Cu radioisotopes proved to be an alternative solution for the therapeutic use of CuCl2 with minimal DI to healthy organs compared with pure 67 Cu. Irradiation of a 70 Zn+68 Zn target in the 70-35 MeV proton energy range for 185 h appears to be the best option from among all the production routes investigated, as it gives the maximum amount of activity, the shortest t10% (10 h), and less than 1% of 61 Cu and 60 Cu impurities.

Biological and internal dosimetry for radiation medicine: current status and future perspectives

The International Atomic Energy Agency (IAEA) and Hiroshima International Council for Health Care of the Radiation-Exposed (HICARE) jointly organized two relevant workshops in Hiroshima, Japan, i.e. a Training Meeting 'Biodosimetry in the 21st century' (BIODOSE-21) on 10-14 June 2013 and a Workshop on 'Biological and internal dosimetry: recent advance and clinical applications' which took place between 17 and 21 February 2020. The main objective of the first meeting was to develop the ability of biodosimetry laboratories to use mature and novel techniques in biological dosimetry for the estimation of radiation doses received by individuals and populations. This meeting had a special focus on the Asia-Pacific region and was connected with the then on-going IAEA Coordinated Research Project (CRP) E35008 'Strengthening of "Biological dosimetry" in IAEA Member States: Improvement of current techniques and intensification of collaboration and networking among the different institutes' (2012-17). The meeting was attended by 25 participants, which included 11 lecturers. The 14 trainees for this meeting came from India, Indonesia, Japan, Malaysia, Philippines, Republic of Korea, Singapore, Thailand and Vietnam. During the meeting 13 lectures by HICARE and IAEA invited lecturers were delivered besides eight research reports presented by the IAEA CRP E35008 network centers from the Asia-Pacific region. Two laboratory exercises were also undertaken, one each at Hiroshima University and the Radiation Effects Research Foundation (RERF). The second training workshop aimed to discuss with the participants the use of mature and novel techniques in biological and internal dosimetry for the estimation of radiation effects by accidental, environmental and medical exposures. The workshop was attended by 19 participants from Indonesia, Jordan, Oman, Philippines, Singapore, Syrian Arab Republic, Thailand, UAE, USA and Yemen. The main outcome of both meetings was a review of the state-of-the-art of biodosimetry and internal dosimetry and their future perspectives in medical management. This report highlights the learning outcome of two meetings for the benefit of all stake-holders in the field of biological and internal dosimetry.

Pre- and post-treatment image-based dosimetry in90Y-microsphere radioembolization using the TOPAS Monte Carlo toolkit

Objective. To evaluate the pre-treatment and post-treatment imaging-based dosimetry of patients treated with 90Y-microspheres, including accurate estimations of dose to tumor, healthy liver and lung. To do so, the Monte Carlo (MC) TOPAS platform is in this work extended towards its utilization in radionuclide therapy.Approach. Five patients treated at the Massachusetts General Hospital were selected for this study. All patients had data for both pre-treatment SPECT-CT imaging using 99mTc-MAA as a surrogate of the 90Y-microspheres treatment and SPECT-CT imaging immediately after the 90Y activity administration. Pre- and post-treatment doses were computed with TOPAS using the SPECT images to localize the source positions and the CT images to account for tissue inhomoegeneities. We compared our results with analytical calculations following the voxel-based MIRD scheme.Main results. TOPAS results largely agreed with the MIRD-based calculations in soft tissue regions: the average difference in mean dose to the liver was 0.14 Gy GBq^-1(2.6%). However, dose distributions in the lung differed considerably: absolute differences in mean doses to the lung ranged from 1.2 to 6.3 Gy GBq^-1and relative differences from 153% to 231%. We also found large differences in the intra-hepatic dose distributions between pre- and post-treatment imaging, but only limited differences in the pulmonary dose.Significance. Doses to lung were found to be higher using TOPAS with respect to analytical calculations which may significantly underestimate dose to the lung, suggesting the use of MC methods for 90Y dosimetry. According to our results, pre-treatment imaging may still be representative of dose to lung in these treatments.

Myelosuppression in patients treated with 177Lutetium-lilotomab satetraxetan can be predicted with absorbed dose to the red marrow as the only variable

BACKGROUND

The aim of this study was to investigate dosimetry data and clinical variables to predict hematological toxicity in non-Hodgkin lymphoma (NHL) patients treated with [¹⁷⁷Lutetium]Lu-lilotomab satetraxetan.

MATERIAL AND METHODS

A total of 17 patients treated with [¹⁷⁷Lu]Lu-lilotomab satetraxetan in a first-in-human phase 1/2a study were included. Absorbed dose to the red marrow was explored using SPECT/CT-imaging of the lumbar vertebrae L2-L4 over multiple time points. Percentage reduction of thrombocytes and neutrophils at nadir compared to baseline (PBN) and time to nadir (TTN) were chosen as indicators of myelosuppression and included as dependent variables. Two models were applied in the analysis, a multivariate linear model and a sigmoidal description of toxicity as a function of absorbed dose. A total of 10 independent patient variables were investigated in the multivariate analysis.

RESULTS

Absorbed dose to the red marrow ranged from 1 to 4 Gy. Absorbed dose to the red marrow was found to be the only significant variable for PBN for both thrombocytes and neutrophils. The sigmoid function gave similar results in terms of accuracy when compared to the linear model.

CONCLUSION

Myelosuppression in the form of thrombocytopenia and neutropenia in patients treated with [¹⁷⁷Lu]Lu-lilotomab satetraxetan can be predicted from the SPECT/CT-derived absorbed dose estimate to the red marrow.

Quantitative Characterization of Population-Wide Tissue- and Metabolite-Specific Variability in Perchloroethylene Toxicokinetics in Male Mice

Quantification of interindividual variability is a continuing challenge in risk assessment, particularly for compounds with complex metabolism and multi-organ toxicity. Toxicokinetic variability for perchloroethylene (perc) was previously characterized across 3 mouse strains and in 1 mouse strain with various degrees of liver steatosis. To further characterize the role of genetic variability in toxicokinetics of perc, we applied Bayesian population physiologically based pharmacokinetic (PBPK) modeling to the data on perc and metabolites in blood/plasma and tissues of male mice from 45 inbred strains from the Collaborative Cross (CC) mouse population. After identifying the most influential PBPK parameters based on global sensitivity analysis, we fit the model with a hierarchical Bayesian population analysis using Markov chain Monte Carlo simulation. We found that the data from 3 commonly used strains were not representative of the full range of variability in perc and metabolite blood/plasma and tissue concentrations across the CC population. Using interstrain variability as a surrogate for human interindividual variability, we calculated dose-dependent, chemical-, and tissue-specific toxicokinetic variability factors (TKVFs) as candidate science-based replacements for the default uncertainty factor for human toxicokinetic variability of 100.5. We found that toxicokinetic variability factors for glutathione conjugation metabolites of perc showed the greatest variability, often exceeding the default, whereas those for oxidative metabolites and perc itself were generally less than the default. Overall, we demonstrate how a combination of a population-based mouse model such as the CC with Bayesian population PBPK modeling can reduce uncertainty in human toxicokinetic variability and increase accuracy and precision in quantitative risk assessment.

Quantification of internal dosimetry in PET patients II: Individualized Monte Carlo-based dosimetry for [18F]fluorocholine PET

Purpose

To obtain individualized internal doses with a Monte Carlo (MC) method in patients undergoing diagnostic [18F]FCH‐PET studies and to compare such doses with the MIRD method calculations.

Methods

A patient cohort of 17 males were imaged after intravenous administration of a mean [18F]FCH activity of 244.3 MBq. The resulting PET/CT images were processed in order to generate individualized input source and geometry files for dose computation with the MC tool GATE. The resulting dose estimates were studied and compared to the MIRD method with two different computational phantoms. Mass correction of the S‐factors was applied when possible. Potential sources of uncertainty were closely examined: the effect of partial body images, urinary bladder emptying, and biokinetic modeling.

Results

Large differences in doses between our methodology and the MIRD method were found, generally in the range ±25%, and up to ±120% for some cases. The mass scaling showed improvements, especially for non‐walled and high‐uptake tissues. Simulations of the urinary bladder emptying showed negligible effects on doses to other organs, with the exception of the prostate. Dosimetry based on partial PET/CT images (excluding the legs) resulted in an overestimation of mean doses to bone, skin, and remaining tissues, and minor differences in other organs/tissues. Estimated uncertainties associated with the biokinetics of FCH introduce variations of cumulated activities in the range of ±10% in the high‐uptake organs.

Conclusions

The MC methodology allows for a higher degree of dosimetry individualization than the MIRD methodology, which in some cases leads to important differences in dose values. Dosimetry of FCH‐PET based on a single partial PET study seems viable due to the particular biokinetics of FCH, even though some correction factors may need to be applied to estimate mean skin/bone doses.

Applications of artificial intelligence in nuclear medicine image generation

Recently, the application of artificial intelligence (AI) in medical imaging (including nuclear medicine imaging) has rapidly developed. Most AI applications in nuclear medicine imaging have focused on the diagnosis, treatment monitoring, and correlation analyses with pathology or specific gene mutation. It can also be used for image generation to shorten the time of image acquisition, reduce the dose of injected tracer, and enhance image quality. This work provides an overview of the application of AI in image generation for single-photon emission computed tomography (SPECT) and positron emission tomography (PET) either without or with anatomical information [CT or magnetic resonance imaging (MRI)]. This review focused on four aspects, including imaging physics, image reconstruction, image postprocessing, and internal dosimetry. AI application in generating attenuation map, estimating scatter events, boosting image quality, and predicting internal dose map is summarized and discussed.

Patient-specific Extravasation Dosimetry Using Uptake Probe Measurements

ABSTRACT

Extravasation is a common problem in radiopharmaceutical administration and can result in significant radiation dose to underlying tissue and skin. The resulting radiation effects are rarely studied and should be more fully evaluated to guide patient care and meet regulatory obligations. The purpose of this work was to show that a dedicated radiopharmaceutical injection monitoring system can help clinicians characterize extravasations for calculating tissue and skin doses. We employed a commercially available radiopharmaceutical injection monitoring system to identify suspected extravasation of 18F-fluorodeoxyglucose and 99mTc-methylene diphosphonate in 26 patients and to characterize their rates of biological clearance. We calculated the self-dose to infiltrated tissue using Monte Carlo simulation and standard MIRD dosimetry methods, and we used VARSKIN software to calculate the shallow dose equivalent to the epithelial basal-cell layer of overlying skin. For 26 patients, injection-site count rate data were used to characterize extravasation clearance. For each, the absorbed dose was calculated using representative tissue geometries. Resulting tissue-absorbed doses ranged from 0.6 to 11.2 Gy, and the shallow dose equivalent to a 10 cm2 area of adjacent skin in these patients ranged from about 0.1 to 5.4 Sv. Extravasated injections of radiopharmaceuticals can result in unintentional doses that exceed well-established radiation protection and regulatory limits; they should be identified and characterized. An external injection monitoring system may help to promptly identify and characterize extravasations and improve dosimetry calculations. Patient-specific characterization can help clinicians determine extravasation severity and whether the patient should be followed for adverse tissue reactions that may present later in time.

[Radiological Technology for Targeted Radionuclide Therapy]

Targeted radioisotope therapy (TRT) is a radiotherapy using radioisotope or drug incorporating it and has been used as a treatment for selectively irradiating cancer cells. In recent years, interest in TRT has increased due to improvements in radionuclide production technology, development of new drugs and imaging modalities, and improvements in radiation technology. In order to enhance the effect of TRT, measurement of individual radiation doses to tumor tissue and organs at risk is important using highly quantitative nuclear medicine images. In this paper, we present a review of literature on optimization of TRT, which is a new research area from the perspective of radiation technology.

Uncertainty quantification of bioassay functions for the internal dosimetry of radioiodine

Bioassay functions, which are provided by the International Commission on Radiological Protection, are used to estimate the intake activity of radionuclides; however, they include considerable uncertainties in terms of the internal dosimetry for a particular individual. During a practical internal dose assessment, the uncertainty in the bioassay function is generally not introduced because of the difficulty in quantification. Therefore, to clarify the existence of uncertainty in the bioassay function and provide dosimetrists with an insight into this uncertainty, this study attempted to quantify the uncertainty in the thyroid retention function used for radioiodine exposure. The uncertainty was quantified using a probabilistic estimation of the thyroid retention function through the propagation of the distribution of biokinetic parameters by the Monte Carlo simulation technique. The uncertainties in the thyroid retention function, expressed in terms of the scattering factor, were in the ranges of 1.55-1.60 and 1.40-1.50 for within 24 h and after 24 h, respectively. In addition, the thyroid retention function within 24 h was compared with actual measurement data to confirm the uncertainty due to the use of first-order kinetics in the biokinetic model calculation. Significantly higher thyroid uptakes (by a factor of 1.9) were observed in the actual measurements. This study indicates that consideration of the uncertainty in the thyroid retention function can avoid a significant over- and under-estimation of the internal dose, particularly when a high dose is predicted.

Red Marrow Absorbed Dose Calculation in Thyroid Cancer Patient Using a Simplified Excel Spreadsheet

Objectives:

Absorbed dose to red marrow (D_rm) can be calculated using blood dosimetry. However, this method is laborious and invasive. Therefore, image-based dosimetry is the method of choice. Nonetheless, the commercial software is expensive. The goal of this work was to develop a simplified excel spreadsheet for image-based radioiodine red marrow dosimetry.

Methods:

The serial whole-body images (acquired at 2^nd, 6^th, 24^th, 48^th, and 72^th hours) of 29 patients from the routine pretherapeutic dosimetry protocol were retrospectively reanalyzed. The commercial OLINDA/EXM image-based dosimetry software was used to calculate the whole-body time-integrated activity coefficient (TIAC_WB) and D_rm [in terms of absorbed dose coefficient (d_rm)]. For the simplified excel spreadsheet, the whole-body count was obtained from the vendor-supplied software. Then, the TIAC_WB was computed by a fitting time-activity curve using an Excel function. S factor was taken from other publications and scaled according to the patient-specific mass. A comparison of the TIAC_WB and d_rm from both methods was done using a non-inferiority test using a paired t-test or the Wilcoxon signed-rank test.

Results:

The TIAC_WB showed no significant difference between both methods (p=0.243). The calculated D_rm from a simplified Excel spreadsheet was assumed to be statistically non-inferior to the commercial OLINDA/EXM image-based dosimetry software with the non-inferiority margin of 0.02 (p<0.05).

Conclusion:

The dose assessment from a simplified Excel spreadsheet is feasible and relatively low cost compared to the commercial OLINDA/EXM image-based dosimetry software.

Ratios of internal doses deposited in different organs to the whole body when such organ is adopted as source of 18F-fluorodeoxyglucose, a Monte Carlo Geant4 study on a male medical internal radiation dose phantom

In the present study, the last stable version of Monte Carlo Geant4 code known as Geant4.10.3 has been used for measuring internal dose ratios to the whole body for about 40 organs. This, by performing a Monte Carlo model of 18F-fluorodeoxyglucose (18F-FDG) inside different organs of medical internal radiation dose male phantom, mimics a human male adult of 70 kg. A dedicated Geant4 user code has been developed in the top of one offered by Geant4 Monte Carlo toolkit and so-called human phantom. Several Monte Carlo simulations have been carried out, and in each of them, we have taken up such organ as source of 18F-FDG with a small amount of radioactivity, evenly distributed across its volume, and we measure ratios of absorbed doses deposited in organs to the whole body. The results have shown that there are radiation dose contributions from surrounding organs and their gravities are so variable; some organs have near-local character; thus, almost all radiations are locally deposited, which generally do not affect surrounding ones mainly including adrenals, thyroid, clavicles, thymus, testes, bladder, pancreas, scapula and upper spine; whereas, it is not the case for many other organs in which radiation doses are deposited outside of their parent volumes. In addition, absorbed doses in some organs that have high-tissue weighting factors, namely colon, lungs, stomach, bladder, thyroid, and liver are seriously affected by radioactivity of surrounding muscle organs, the gravity of such affectation is mainly growth when a patient is identified as having hyperglycemia or undergoing a hard physical activity.

Quantification of internal dosimetry in PET patients: individualized Monte Carlo vs generic phantom-based calculations

PURPOSE

The purpose of this work is to calculate individualized dose distributions in patients undergoing 18 F-FDG PET/CT studies through a methodology based on full Monte Carlo (MC) simulations and PET/CT patient images, and to compare such values with those obtained by employing nonindividualized phantom-based methods.

METHODS

We developed a MC-based methodology for individualized internal dose calculations, which relies on CT images (for organ segmentation and dose deposition), PET images (for organ segmentation and distributions of activities), and a biokinetic model (which works with information provided by PET and CT images) to obtain cumulated activities. The software vGATE version 8.1. was employed to carry out the Monte Carlo calculations. We also calculated deposited doses with nonindividualized phantom-based methods (Cristy-Eckerman, Stabin, and ICRP-133).

RESULTS

Median MC-calculated dose/activity values are within 0.01-0.03 mGy/MBq for most organs, with higher doses delivered especially to the bladder wall, major vessels, and brain (medians of 0.058, 0.060, 0.066 mGy/MBq, respectively). Comparison with values obtained with nonindividualized phantom-based methods has shown important differences in many cases (ranging from -80% to + 260%). These differences are significant (p < 0.05) for several organs/tissues, namely, remaining tissues, adrenals, bladder wall, bones, upper large intestine, heart, pancreas, skin, and stomach wall.

CONCLUSIONS

The methodology presented in this work is a viable and useful method to calculate internal dose distributions in patients undergoing medical procedures involving radiopharmaceuticals, individually, with higher accuracy than phantom-based methods, fulfilling the guidelines provided by the European Council directive 2013/59/Euratom.

Bioassay and alpha spectrometry in indirect monitoring of Spanish workers exposed to enriched uranium

Workers at risk of exposure to uranium compounds should be monitored and their internal exposure quantified in terms of committed effective dose E(50) in mSv. In vitro bioassay methods can quantify uranium in urine and faeces at low activity levels. Alpha spectrometry (AS) is the most common method used for monitoring alpha-emitting radionuclides in internal dosimetry services. It provides isotopic information and low minimum detectable activity (MDA) values (≤0.50 mBq per sample). This study reports the results of a five-year monitoring of workers exposed to uranium at a Spanish Juzbado facility, which produces nuclear fuel elements enriched with up to 5 % of 235U. Monitoring included about 100 workers per year, most of whom had worked at the facility for more than 10 years before the individual monitoring programme was established. We analysed nearly 550 samples of more than 200 workers over five years. The obtained results indicate that workers were adequately protected from uranium exposure through inhalation and had an acceptably low chronic intake at the facility.

Pretreatment CLR 124 Positron Emission Tomography Accurately Predicts CLR 131 Three-Dimensional Dosimetry in a Triple-Negative Breast Cancer Patient

INTRODUCTION

CLR1404 is a theranostic molecular agent that can be radiolabeled with ¹²⁴I (CLR 124) for positron emission tomography (PET) imaging, or ¹³¹I (CLR 131) for single-photon emission computed tomography (SPECT) imaging and targeted radionuclide therapy. This pilot study evaluated a pretreatment dosimetry methodology in a triple-negative breast cancer patient who was uniquely enrolled in both a CLR 124 PET imaging clinical trial and a CLR 131 therapeutic dose escalation clinical trial.

MATERIALS AND METHODS

Three-dimensional PET/CT images were acquired at 1, 3, 24, 48, and 120 h postinjection of 178 MBq CLR 124. One month later, pretherapy 2D whole-body planar images were acquired at 0.25, 5, 24, 48, and 144 h postinjection of 370 MBq CLR 131. Following the therapeutic administration of 1990 MBq CLR 131, 3D SPECT/CT images were acquired at 74, 147, 334, and 505 h postinjection. The therapeutic CLR 131 voxel-level absorbed dose was estimated from PET (RAPID PET) and SPECT (RAPID SPECT) images using a Geant4-based Monte Carlo dosimetry platform called RAPID (Radiopharmaceutical Assessment Platform for Internal Dosimetry), and region of interest (ROI) mean doses were also estimated using the OLINDA/EXM software based on PET (OLINDA PET), SPECT (OLINDA SPECT), and planar (OLINDA planar) images.

RESULTS

The RAPID PET and OLINDA PET tracer-predicted ROI mean doses correlated well (m ≥ 0.631, R² ≥ 0.694, p ≤ 0.01) with both the RAPID SPECT and OLINDA SPECT therapeutic mean doses. The 2D planar images did not have any significant correlations. The ROI mean doses differed by -4% to -43% between RAPID and OLINDA/EXM, and by -19% to 29% between PET and SPECT. The 3D dose distributions and dose volume histograms calculated with RAPID were similar for the PET/CT and SPECT/CT.

CONCLUSIONS

This pilot study demonstrated that CLR 124 pretreatment PET images can be used to predict CLR 131 3D therapeutic dosimetry better than CLR 131 2D planar images. In addition, unlike OLINDA/EXM, Monte Carlo dosimetry methods were capable of accurately predicting dose heterogeneity, which is important for predicting dose-response relationships and clinical outcomes.

Technical Note: Virtual CT for reducing CT dose in targeted radionuclide therapy dosimetry

PURPOSE

Previously we have shown that using sequential CT images is superior to sequential SPECT for nonrigid registration in three-dimensional (3D) targeted radionuclide therapy (TRT) dosimetry. However, sequential CTs are often not available due to radiation concerns. In this paper, we propose a virtual CT (vCT) method for attenuation and scatter correction, image registration, and segmentation for improved dosimetric accuracy with single CT acquisition.

METHODS

We used a population of nine XCAT phantoms with different In-111 Zevalin biokinetics and anatomical variations for the simulations. An analytical projector was used to simulate sequential SPECT/CT acquisitions for a medium energy general purpose collimator at 1, 12, 24, 72, and 144 h postinjection, modeling attenuation, scatter, and geometric collimator-detector response. The corresponding sequential attenuation maps of the phantoms served as real CT (rCT) images. For vCT generation, we investigated three registration methods, that is, (a) SPECT to SPECT; (b) SPECT to CT, and (c) CT to SPECT, and the optimal time point for single CT acquisition. Difference images and average normalized mean square errors (NMSE) were calculated between different vCTs and their corresponding rCTs. Absorbed dose and dose-volume histograms (DVHs) for critical organs were computed for the rCT, optimized vCT, and conventional single CT (1CT) protocols, respectively, for dosimetric analyses.

RESULTS

For vCT generation, SPECT to SPECT registration with a single CT acquired at the first time point shows the smallest difference and NMSE. For organ absorbed doses, the results of vCT were similar to those of rCT and were superior to 1CT, that is, -0.24 ± 1.56% vs -0.49 ± 1.76% vs -6.37 ± 5.63% for the liver, -1.05 ± 2.89% vs -0.69 ± 2.74% vs -4.87 ± 4.35% for kidneys, respectively. The results of DVHs also showed improvement for all organs using vCTs as compared to the conventional 1CT protocol.

CONCLUSION

The optimized vCT method can effectively increase the TRT dosimetric results if there is only a single CT available in the sequential imaging protocol, reducing the substantial increase in radiation burden from repeated CT scans.

Development and Validation of RAPID: A Patient-Specific Monte Carlo Three-Dimensional Internal Dosimetry Platform

This work describes the development and validation of a patient-specific Monte Carlo internal dosimetry platform called RAPID (Radiopharmaceutical Assessment Platform for Internal Dosimetry). RAPID utilizes serial PET/CT or SPECT/CT images to calculate voxelized three-dimensional (3D) internal dose distributions with the Monte Carlo code Geant4. RAPID's dosimetry calculations were benchmarked against previously published S-values and specific absorbed fractions (SAFs) calculated for monoenergetic photon and electron sources within the Zubal phantom and for S-values calculated for a variety of radionuclides within spherical tumor phantoms with sizes ranging from 1 to 1000 g. The majority of the S-values and SAFs calculated in the Zubal Phantom were within 5% of the previously published values with the exception of a few 10 keV photon SAFs that agreed within 10%, and one value within 16%. The S-values calculated in the spherical tumor phantoms agreed within 2% for ¹⁷⁷Lu, ¹³¹I, ¹²⁵I, ¹⁸F, and ⁶⁴Cu, within 3.5% for ²¹¹At and ²¹³Bi, within 6.5% for ¹⁵³Sm, ¹¹¹In, ⁸⁹Zr, and ²²³Ra, and within 9% for ⁹⁰Y, ⁶⁸Ga, and ¹²⁴I. In conclusion, RAPID is capable of calculating accurate internal dosimetry at the voxel-level for a wide variety of radionuclides and could be a useful tool for calculating patient-specific 3D dose distributions.

Murine-specific Internal Dosimetry for Preclinical Investigations of Imaging and Therapeutic Agents

There is a growing need to estimate the absorbed dose to small animals from preclinical investigations involving diagnostic and therapeutic radiopharmaceuticals. This paper introduces a Monte Carlo-based dosimetry platform called RAPID, which is capable of calculating murine-specific three-dimensional (3D) dose distributions. A comparison is performed between absorbed doses calculated with RAPID and absorbed doses calculated in a commonly used reference mouse phantom called MOBY. Four test mice containing different xenografts underwent serial PET/CT imaging using a novel diagnostic therapy (theranostic) agent NM404, which can be labeled with I for imaging or I for therapy. Using the PET/CT data, 3D dose distributions from I-NM404 were calculated in the mice using RAPID. Mean organ doses in these four test mice were compared to mean organ doses derived by using two previously published I S-values datasets in MOBY. In addition, mean tumor doses calculated in RAPID were compared to mean organ doses derived from unit density spheres. Large differences were identified between mean organ doses calculated in the test mice using RAPID and those derived in the MOBY phantom. Mean absorbed dose percent errors in organs ranged between 0.3% and 333%. Overall, mass scaling improved agreement between MOBY phantom calculations and RAPID, where percent errors were all less than 26%, with the exception of the lung in which percent errors reached values of 48%. Percent errors in mean tumor doses in the test mice and unit density spheres were less pronounced but still ranged between 8% and 23%. This work demonstrates the limitations of using pre-computed S-values in computational phantoms to predict organ doses in small animals from theranostic procedures. RAPID can generate accurate 3D dose distributions in small animals and in turn offer much greater insight on the ability of a given theranostic agent to image and treat diseases.

Quantitative Imaging for Targeted Radionuclide Therapy Dosimetry - Technical Review

Targeted radionuclide therapy (TRT) is a promising technique for cancer therapy. However, in order to deliver the required dose to the tumor, minimize potential toxicity in normal organs, as well as monitor therapeutic effects, it is important to assess the individualized internal dosimetry based on patient-specific data. Advanced imaging techniques, especially radionuclide imaging, can be used to determine the spatial distribution of administered tracers for calculating the organ-absorbed dose. While planar scintigraphy is still the mainstream imaging method, SPECT, PET and bremsstrahlung imaging have promising properties to improve accuracy in quantification. This article reviews the basic principles of TRT and discusses the latest development in radionuclide imaging techniques for different theranostic agents, with emphasis on their potential to improve personalized TRT dosimetry.

GHSI EMERGENCY RADIONUCLIDE BIOASSAY LABORATORY NETWORK - SUMMARY OF THE SECOND EXERCISE

The Global Health Security Initiative (GHSI) established a laboratory network within the GHSI community to develop collective surge capacity for radionuclide bioassay in response to a radiological or nuclear emergency as a means of enhancing response capability, health outcomes and community resilience. GHSI partners conducted an exercise in collaboration with the WHO Radiation Emergency Medical Preparedness and Assistance Network and the IAEA Response and Assistance Network, to test the participating laboratories (18) for their capabilities in in vitro assay of biological samples, using a urine sample spiked with multiple high-risk radionuclides (90Sr, 106Ru, 137Cs, and 239Pu). Laboratories were required to submit their reports within 72 h following receipt of the sample, using a pre-formatted template, on the procedures, methods and techniques used to identify and quantify the radionuclides in the sample, as well as the bioassay results with a 95% confidence interval. All of the participating laboratories identified and measured all or some of the radionuclides in the sample. However, gaps were identified in both the procedures used to assay multiple radionuclides in one sample, as well as in the methods or techniques used to assay specific radionuclides in urine. Two-third of the participating laboratories had difficulties in determining all the radionuclides in the sample. Results from this exercise indicate that challenges remain with respect to ensuring that results are delivered in a timely, consistent and reliable manner to support medical interventions. Laboratories within the networks are encouraged to work together to develop and maintain collective capabilities and capacity for emergency bioassay, which is an important component of radiation emergency response.

Specific Absorbed Fractions of Internal Photon and Electron Emitters in a Human Voxel-based Phantom: A Monte Carlo Study

The specific absorbed fraction (SAF) of energy is an essential element of internal dose assessment. Here reported a set of SAFs calculated for selected organs of a human voxel-based phantom. The Monte Carlo transport code GATE version 6.1 was used to simulate monoenergetic photons and electrons with energies ranging from 10 keV to 2 MeV. The particles were emitted from three source organs: kidneys, liver, and spleen. SAFs were calculated for three target regions in the body (kidneys, liver, and spleen) and compared with the results obtained using the MCNP4B and GATE/GEANT4 Monte Carlo codes. For most photon energies, the self-irradiation is higher, and the cross-irradiation is lower in the GATE results compared to the MCNP4B. The results show generally good agreement for photons and high-energy electrons with discrepancies within − 2% ±3%. Nevertheless, significant differences were found for cross-irradiation of photons of lower energy and electrons of higher energy due to statistical uncertainties larger than 10%. The comparisons of the SAF values for the human voxel phantom do not show significant differences, and the results also demonstrated the usefulness and applicability of GATE Monte Carlo package for voxel level dose calculations in nonuniform media. The present SAFs calculation for the Zubal voxel phantom is validated by the intercomparison of the results obtained by other Monte Carlo codes.

GHSI EMERGENCY RADIONUCLIDE BIOASSAY LABORATORY NETWORK: SUMMARY OF A RECENT EXERCISE

The Global Health Security Initiative (GHSI) established a laboratory network within the GHSI community to develop their collective surge capacity for radionuclide bioassay in response to a radiological or nuclear emergency. A recent exercise was conducted to test the participating laboratories for their capabilities in screening and in vitro assay of biological samples, performing internal dose assessment and providing advice on medical intervention, if necessary, using a urine sample spiked with a single radionuclide, 241Am. The laboratories were required to submit their reports according to the exercise schedule and using pre-formatted templates. Generally, the participating laboratories were found to be capable with respect to rapidly screening samples for radionuclide contamination, measuring the radionuclide in the samples, assessing the intake and radiation dose, and providing advice on medical intervention. However, gaps in bioassay measurement and dose assessment have been identified. The network may take steps to ensure that procedures and practices within this network be harmonised and a follow-up exercise be organised on a larger scale, with potential participation of laboratories from the networks coordinated by the International Atomic Energy Agency and the World Health Organization.

Uncertainty Quantification in Internal Dose Calculations for Seven Selected Radiopharmaceuticals

Dose coefficients of radiopharmaceuticals have been published by the International Commission on Radiological Protection (ICRP) and the MIRD Committee but without information concerning uncertainties. The uncertainty information of dose coefficients is important, for example, to compare alternative diagnostic methods and choose the method that causes the lowest patient exposure with appropriate and comparable diagnostic quality. For the study presented here, an uncertainty analysis method was developed and used to calculate the uncertainty of the internal doses of 7 common radiopharmaceuticals.

METHODS

On the basis of the generalized schema of dose calculation recommended by the ICRP and MIRD Committee, an analysis based on propagation of uncertainty was developed and applied for 7 radiopharmaceuticals. The method takes into account the uncertainties contributed from pharmacokinetic models and the so-called S values derived from several voxel computational phantoms previously developed at Helmholtz Zentrum München. Random and Latin hypercube sampling techniques were used to sample parameters of pharmacokinetic models and S values, and the uncertainties of absorbed doses and effective doses were calculated.

RESULTS

The uncertainty factors (square root of the ratio between 97.5th and 2.5th percentiles) for organ-absorbed doses are in the range of 1.1-3.3. Uncertainty values of effective doses are lower in comparison to absorbed doses, the maximum value being approximately 1.4. The ICRP reference values showed a deviation comparable to the effective dose calculated in this study.

CONCLUSION

A general statistical method was developed for calculating the uncertainty of absorbed doses and effective doses for 7 radiopharmaceuticals. The dose uncertainties can be used to further identify the most important parameters in the dose calculation and provide reliable dose coefficients for risk analysis of the patients in nuclear medicine.

Prospective Study of (68)Ga-NOTA-NFB: Radiation Dosimetry in Healthy Volunteers and First Application in Glioma Patients

Purpose: The chemokine receptor CXCR4 is overexpressed in various types of human cancers. As a specific imaging agent of CXCR4, ⁶⁸Ga-NOTA-NFB was investigated in this study to assess its safety, biodistribution and dosimetry properties in healthy volunteers, and to preliminarily evaluate its application in glioma patients.

Methods: Six healthy volunteers underwent whole-body PET scans at 0, 0.5, 1, 2 and 3 h after ⁶⁸Ga-NOTA-NFB injection (mean dose, 182.4 ± 3.7 MBq (4.93 ± 0.10 mCi)). For time-activity curve calculations, 1 mL blood samples were obtained at 1, 3, 5, 10, 30, 60, 90, 120, 150 and 180 min after the injection. The estimated radiation doses were calculated by OLINDA/EXM software. Eight patients with glioma were enrolled and underwent both ⁶⁸Ga-NOTA-NFB and ¹⁸F-FDG PET/CT scans before surgery. The expression of CXCR4 on the resected brain tumor tissues was determined by immunohistochemical staining.

Results:⁶⁸Ga-NOTA-NFB was safe and well tolerated by all subjects. A rapid activity clearance from the blood circulation was observed. The organs with the highest absorbed doses were spleen (193.8 ± 32.5 μSv/MBq) and liver (119.3 ± 25.0 μSv/MBq). The mean effective dose was 25.4 ± 6.1 μSv/MBq. The maximum standardized uptake values (SUV_max) and the maximum target to non-target ratios (T/NT_max) of ⁶⁸Ga-NOTA-NFB PET/CT in glioma tissues were 4.11 ± 2.90 (range, 0.45-8.21) and 9.21 ± 8.75 (range, 3.66-24.88), respectively, while those of ¹⁸F-FDG PET/CT were 7.34 ± 2.90 (range, 3.50-12.27) and 0.86 ± 0.41 (range, 0.35-1.59). The histopathological staining confirmed that CXCR4 was overexpressed on resected tumor tissues with prominent ⁶⁸Ga-NOTA-NFB uptake.

Conclusion: With a favorable radiation dosimetry profile, ⁶⁸Ga-NOTA-NFB is safe for clinical imaging. Compared to ¹⁸F-FDG PET/CT, ⁶⁸Ga-NOTA-NFB PET/CT is more sensitive in detecting glioma and could have potential in diagnosing and treatment planning for CXCR4 positive patients.

A recycling model of the biokinetics of systemic tellurium

PURPOSE

To develop a compartmental model of the systemic biokinetics of tellurium required for calculating the internal dose and interpreting bioassay measurements after incorporation of radioactive tellurium.

MATERIALS AND METHODS

The compartmental model for tellurium was developed with the software SAAM II v. 2.0 (©The Epsilon Group, Charlottesville, Virginia, USA). Model parameters were determined on the basis of published retention and excretion data in humans and animals.

RESULTS

The model consists of two blood compartments, one compartment each for liver, kidneys, thyroid, four compartments for bone tissues and a generic compartment for the soft tissues. The model predicts a rapid urinary excretion of systemic tellurium: 45% in the first 24 h and 84% after 50 d. Faecal excretion amounts to 0.4% after 3 d and 9% after 50 d. Whole body retention is 55% after one day, and 2.8% after 100 d. These values as well as the retained fractions in the single organs are reasonably consistent with the available human and animal data (studies with swine and guinea pigs).

CONCLUSIONS

The proposed model gives a realistic description of the available biokinetic data for tellurium and will be adopted by the International Commission on Radiological Protection for applications in internal dosimetry.

Development of a 9-months pregnant hybrid phantom and its internal dosimetry for thyroid agents

As a consequence of fetal radiosensitivity, the estimation of internal dose received by a fetus from radiopharmaceuticals applied to the mother is often important in nuclear medicine. A new 9-months pregnant phantom based on magnetic resonance (MR) images tied to the International Commission on Radiological Protection (ICRP) reference voxel phantom has been developed. Maternal and fetal organs were segmented from a set of pelvic MR images of a 9-months pregnant subject using 3D-DOCTOR(TM) and then imported into the 3D modeling software package Rhinoceros(TM) for combining with the adult female ICRP voxel phantom and further modeling. Next, the phantom organs were rescaled to match with reference masses described in ICRP Publications. The internal anatomy of previous pregnant phantom models had been limited to the fetal brain and skeleton only, but the fetus model developed in this study incorporates 20 different organs. The current reference phantom has been developed for application in comprehensive dosimetric study in nuclear medicine. The internal dosimetry calculations were performed for thyroid agents using the Monte Carlo transport method. Biokinetic data for these radiopharmaceuticals were used to estimate cumulated activity during pregnancy and maternal and fetal organ doses at seven different maximum thyroid uptake levels. Calculating the dose distribution was also presented in a sagittal view of the pregnant model utilizing the mesh tally function. The comparisons showed, in general, an overestimation of the absorbed dose to the fetus and an underestimation of the fetal thyroid dose in previous studies compared with the values based on the current hybrid phantom.

Internal dosimetry estimates using voxelized reference phantoms for thyroid agents

This work presents internal dosimetry estimates for diagnostic procedures performed for thyroid disorders by relevant radiopharmaceuticals. The organ doses for (131)Iodine, (123)Iodine and (99m)Tc incorporated into the body were calculated for the International Commission on Radiological Protection (ICRP) reference voxel phantoms using the Monte Carlo transport method. A comparison between different thyroid uptakes of iodine in the range of 0-55% was made, and the effect of various techniques for administration of (99m)Tc on organ doses was studied. To investigate the necessity of calculating organ dose from all source regions, the major source organ and its contribution to total dose were specified for each target organ. Moreover, we compared effective dose in ICRP voxel phantoms with that in stylized phantoms. In our method, we directly calculated the organ dose without using the S values or SAFs, as is commonly done. Hence, a distribution of the absorbed dose to entire tissues was obtained. The chord length distributions (CLDs) were also computed for the selected source-target pairs to make comparison across the genders. The results showed that the S values for radionuclides in the thyroid are not sufficient for calculating the organ doses, especially for (123)I and (99m)Tc. The thyroid and its neighboring organs receive a greater dose as thyroid uptake increases. Our comparisons also revealed an underestimation of organ doses reported for the stylized phantoms compared with the values based on the ICRP voxel phantoms in the uptake range of 5-55%, and an overestimation of absorbed dose by up to 2-fold for Iodine administration using blocking agent and for (99m)Tc incorporation.

Biodistribution and radiation dosimetry of 18F-CP-18, a potential apoptosis imaging agent, as determined from PET/CT scans in healthy volunteers

(18)F-CP-18, or (18S,21S,24S,27S,30S)-27-(2-carboxyethyl)-21-(carboxymethyl)-30-((2S,3R,4R,5R,6S)-6-((2-(4-(3-F18-fluoropropyl)-1H-1,2,3-triazol-1-yl)acetamido)methyl)-3,4,5-trihydroxytetrahydro-2H-pyran-2-carboxamido)-24-isopropyl-18-methyl-17,20,23,26,29-pentaoxo-4,7,10,13-tetraoxa-16,19,22,25,28-pentaazadotriacontane-1,32-dioic acid, is being evaluated as a tissue apoptosis marker for PET imaging. The purpose of this study was to determine the biodistribution and estimate the normal-organ radiation-absorbed doses and effective dose from (18)F-CP-18.

METHODS

Successive whole-body PET/CT scans were obtained at approximately 7, 45, 90, 130, and 170 min after intravenous injection of (18)F-CP-18 in 7 healthy human volunteers. Blood samples and urine were collected between the PET/CT scans, and the biostability of (18)F-CP-18 was assessed using high-performance liquid chromatography. The PET scans were analyzed to determine the radiotracer uptake in different organs. OLINDA/EXM software was used to calculate human radiation doses based on the biodistribution of the tracer.

RESULTS

(18)F-CP-18 was 54% intact in human blood at 135 min after injection. The tracer cleared rapidly from the blood pool with a half-life of approximately 30 min. Relatively high (18)F-CP-18 uptake was observed in the kidneys and bladder, with diffuse uptake in the liver and heart. The mean standardized uptake values (SUVs) in the bladder, kidneys, heart, and liver at around 50 min after injection were approximately 65, 6, 1.5, and 1.5, respectively. The calculated effective dose was 38 ± 4 μSv/MBq, with the urinary bladder wall having the highest absorbed dose at 536 ± 61 μGy/MBq using a 4.8-h bladder-voiding interval for the male phantom. For a 1-h voiding interval, these doses were reduced to 15 ± 2 μSv/MBq and 142 ± 15 μGy/MBq, respectively. For a typical injected activity of 555 MBq, the effective dose would be 21.1 ± 2.2 mSv for the 4.8-h interval, reduced to 8.3 ± 1.1 mSv for the 1-h interval.

CONCLUSION

(18)F-CP-18 cleared rapidly through the renal system. The urinary bladder wall received the highest radiation dose and was deemed the critical organ. Both the effective dose and the bladder dose can be reduced by frequent voiding. From the radiation dosimetry perspective, the apoptosis imaging agent (18)F-CP-18 is suitable for human use.

Evaluation and comparison of human absorbed dose of (90)Y-DOTA-Cetuximab in various age groups based on distribution data in rats

The organ radiation-absorbed doses have been evaluated for humans in six age groups and both genders based on animal data. After intravenous administration of ⁹⁰Y-DOTA-Cetuximab to five groups of rats, they were sacrificed at exact time intervals (2, 24, 48, 72, and 96 h) and the percentage of injected dose per gram of each organ was calculated by direct counting from rat data. By using the formulation that Medical Internal Radiation Dose suggests, radiation-absorbed doses for all organs were calculated and extrapolated from rat to human. The total body absorbed dose for all groups was >22 mGy due to pure β-emission of the applied radiopharmaceutical. The effective dose resulting from an intravenously injected activity of 100 MBq is 56.7 mSv for a 60-kg female adult and 60.3 mSv for a 73-kg male adult. The results demonstrated the usefulness of this method for estimation of β-absorbed dose in humans.

Bayesian model selection validates a biokinetic model for zirconium processing in humans

BACKGROUND

In radiation protection, biokinetic models for zirconium processing are of crucial importance in dose estimation and further risk analysis for humans exposed to this radioactive substance. They provide limiting values of detrimental effects and build the basis for applications in internal dosimetry, the prediction for radioactive zirconium retention in various organs as well as retrospective dosimetry. Multi-compartmental models are the tool of choice for simulating the processing of zirconium. Although easily interpretable, determining the exact compartment structure and interaction mechanisms is generally daunting. In the context of observing the dynamics of multiple compartments, Bayesian methods provide efficient tools for model inference and selection.

RESULTS

We are the first to apply a Markov chain Monte Carlo approach to compute Bayes factors for the evaluation of two competing models for zirconium processing in the human body after ingestion. Based on in vivo measurements of human plasma and urine levels we were able to show that a recently published model is superior to the standard model of the International Commission on Radiological Protection. The Bayes factors were estimated by means of the numerically stable thermodynamic integration in combination with a recently developed copula-based Metropolis-Hastings sampler.

CONCLUSIONS

In contrast to the standard model the novel model predicts lower accretion of zirconium in bones. This results in lower levels of noxious doses for exposed individuals. Moreover, the Bayesian approach allows for retrospective dose assessment, including credible intervals for the initially ingested zirconium, in a significantly more reliable fashion than previously possible. All methods presented here are readily applicable to many modeling tasks in systems biology.

Hybrid computational phantoms representing the reference adult male and adult female: construction and applications for retrospective dosimetry

Currently, two classes of computational phantoms have been developed for dosimetry calculation: (1) stylized (or mathematical) and (2) voxel (or tomographic) phantoms describing human anatomy through mathematical surface equations and 3D voxel matrices, respectively. Mathematical surface equations in stylized phantoms are flexible, but the resulting anatomy is not as realistic. Voxel phantoms display far better anatomical realism, but they are limited in terms of their ability to alter organ shape, position, and depth, as well as body posture. A new class of computational phantoms called hybrid phantoms takes advantage of the best features of stylized and voxel phantoms-flexibility and anatomical realism, respectively. In the current study, hybrid computational phantoms representing the adult male and female reference anatomy and anthropometry are presented. These phantoms serve as the starting framework for creating patient or worker sculpted whole-body phantoms for retrospective dose reconstruction. Contours of major organs and tissues were converted or segmented from computed tomography images of a 36-y-old Korean volunteer and a 25-y-old U.S. female patient, respectively, with supplemental high-resolution CT images of the cranium. Polygon mesh models for the major organs and tissues were reconstructed and imported into Rhinoceros™ for non-uniform rational B-spline (NURBS) surface modeling. The resulting NURBS/polygon mesh models representing body contour and internal anatomy were matched to anthropometric data and reference organ mass data provided by Centers for Disease Control and Prevention and International Commission on Radiation Protection, respectively. Finally, two hybrid adult male and female phantoms were completed where a total of eight anthropometric data categories were matched to standard values within 4% and organ volumes matched to ICRP data within 1% with the exception of total skin. The hybrid phantoms were voxelized from the NURBS phantoms at resolutions of 0.158 × 0.158 × 0.158 cm and 0.126 × 0.126 × 0.126 cm for the male and female, respectively. To highlight the flexibility of the hybrid phantoms, graphical displays are given of (1) underweight and overweight adult male phantoms, (2) a sitting position for the adult female phantom, and (3) extraction and higher-resolution voxelization of the small intestine for localized dosimetry of mucosal and stem cell layers. These phantoms are used to model radioactively contaminated individuals and to then assess time-dependent detector count rate thresholds corresponding to 50, 250, and 500 mSv effective dose, as might be needed during in-field radiological triage by first responders or first receivers.