Showing papers in "EJNMMI Physics in 2018"

A deep learning approach for 18 F-FDG PET attenuation correction

Abstract: To develop and evaluate the feasibility of a data-driven deep learning approach (deepAC) for positron-emission tomography (PET) image attenuation correction without anatomical imaging. A PET attenuation correction pipeline was developed utilizing deep learning to generate continuously valued pseudo-computed tomography (CT) images from uncorrected 18F-fluorodeoxyglucose (18F-FDG) PET images. A deep convolutional encoder-decoder network was trained to identify tissue contrast in volumetric uncorrected PET images co-registered to CT data. A set of 100 retrospective 3D FDG PET head images was used to train the model. The model was evaluated in another 28 patients by comparing the generated pseudo-CT to the acquired CT using Dice coefficient and mean absolute error (MAE) and finally by comparing reconstructed PET images using the pseudo-CT and acquired CT for attenuation correction. Paired-sample t tests were used for statistical analysis to compare PET reconstruction error using deepAC with CT-based attenuation correction. deepAC produced pseudo-CTs with Dice coefficients of 0.80 ± 0.02 for air, 0.94 ± 0.01 for soft tissue, and 0.75 ± 0.03 for bone and MAE of 111 ± 16 HU relative to the PET/CT dataset. deepAC provides quantitatively accurate 18F-FDG PET results with average errors of less than 1% in most brain regions. We have developed an automated approach (deepAC) that allows generation of a continuously valued pseudo-CT from a single 18F-FDG non-attenuation-corrected (NAC) PET image and evaluated it in PET/CT brain imaging.

The physics of radioembolization

Abstract: Radioembolization is an established treatment for chemoresistant and unresectable liver cancers. Currently, treatment planning is often based on semi-empirical methods, which yield acceptable toxicity profiles and have enabled the large-scale application in a palliative setting. However, recently, five large randomized controlled trials using resin microspheres failed to demonstrate a significant improvement in either progression-free survival or overall survival in both hepatocellular carcinoma and metastatic colorectal cancer. One reason for this might be that the activity prescription methods used in these studies are suboptimal for many patients. In this review, the current dosimetric methods and their caveats are evaluated. Furthermore, the current state-of-the-art of image-guided dosimetry and advanced radiobiological modeling is reviewed from a physics’ perspective. The current literature is explored for the observation of robust dose-response relationships followed by an overview of recent advancements in quantitative image reconstruction in relation to image-guided dosimetry. This review is concluded with a discussion on areas where further research is necessary in order to arrive at a personalized treatment method that provides optimal tumor control and is clinically feasible.

Initial experience with a SiPM-based PET/CT scanner: influence of acquisition time on image quality.

Abstract: A newly introduced PET/CT scanner (Discovery Meaningful Insights—DMI, GE Healthcare) includes the silicon photomultiplier (SiPM) with time-of-flight (TOF) technology first used in the GE SIGNA PET/MRI. In this study, we investigated the impact of various acquisition times on image quality using this SiPM-based PET/CT. We reviewed data from 58 participants with cancer who were scanned using the DMI PET/CT scanner. The administered dosages ranged 295.3–429.9 MBq (mean ± SD 356.3 ± 37.4) and imaging started at 71–142 min (mean ± SD 101.41 ± 17.52) after administration of the radiopharmaceutical. The patients’ BMI ranged 19.79–46.16 (mean ± SD 26.55 ± 5.53). We retrospectively reconstructed the raw TOF data at 30, 60, 90, and 120 s/bed and at the standard image acquisition time per clinical protocol (180 or 210 s/bed depending on BMI). Each reconstruction was reviewed blindly by two nuclear medicine physicians and scored 1–5 (1—poor, 5—excellent quality). The liver signal-to-noise ratio (SNR) was used as a quantitative measure of image quality. The average scores ± SD of the readers were 2.61 ± 0.83, 3.70 ± 0.92, 4.36 ± 0.82, 4.82 ± 0.39, and 4.91 ± 0.91 for the 30, 60, 90, and 120 s/bed and at standard acquisition time, respectively. Inter-reader agreement on image quality assessment was good, with a weighted kappa of 0.80 (95% CI 0.72–0.81). In the evaluation of the effects of time per bed acquisition on semi-quantitative measurements, we found that the only time point significantly different from the standard time were 30 and 60 s (both with P 25, images can be acquired as fast as 90 s/bed using the SiPM PET/CT and still result in very good image quality (average score > 4).

Impact of a Bayesian penalized likelihood reconstruction algorithm on image quality in novel digital PET/CT: clinical implications for the assessment of lung tumors

Abstract: The aim of this study was to evaluate and compare PET image reconstruction algorithms on novel digital silicon photomultiplier PET/CT in patients with newly diagnosed and histopathologically confirmed lung cancer. A total of 45 patients undergoing 18F-FDG PET/CT for initial lung cancer staging were included. PET images were reconstructed using ordered subset expectation maximization (OSEM) with time-of-flight and point spread function modelling as well as Bayesian penalized likelihood reconstruction algorithm (BSREM) with different β-values yielding a total of 7 datasets per patient. Subjective and objective image assessment with all image datasets was carried out, including subgroup analyses for patients with high dose (> 2.0 MBq/kg) and low dose (≤ 2.0 MBq/kg) of 18F-FDG injection regimen. Subjective image quality ratings were significantly different among all different reconstruction algorithms as well as among BSREM using different β-values only (both p < 0.001). BSREM with a β-value of 600 was assigned the highest score for general image quality, image sharpness, and lesion conspicuity. BSREM reconstructions resulted in higher SUVmax of lung tumors compared to OSEM of up to + 28.0% (p < 0.001). BSREM reconstruction resulted in higher signal-/ and contrast-to-background ratios of lung tumor and higher signal-/ and contrast-to-noise ratio compared to OSEM up to a β-value of 800. Lower β-values (BSREM450) resulted in the best image quality for high dose 18F-FDG injections, whereas higher β-values (BSREM600) lead to the best image quality in low dose 18F-FDG PET/CT (p < 0.05). BSREM reconstruction algorithm used in digital detector PET leads to significant increases of lung tumor SUVmax, signal-to-background ratio, and signal-to-noise ratio, which translates into a higher image quality, tumor conspicuity, and image sharpness.

Feasibility of simplifying renal dosimetry in 177 Lu peptide receptor radionuclide therapy

Abstract: Recently, 177Lu-dotatate therapy for neuroendocrine tumours has received regulatory approval. Dosimetry can be used to optimize treatment on an individual basis, but there is no international consensus as to how it should be done. The aim of this study is to determine a feasible and accurate dosimetry method to guide individualized peptide receptor radionuclide therapy (PRRT) for patients with neuroendocrine tumours. As part of a clinical trial on 177Lu-dotatate therapy, renal dosimetry was performed for all patients in each treatment cycle, using a hybrid planar-SPECT/CT method. In the present study, we use the image data acquired from 22 patients and 119 cycles and define a set of alternative treatment planning strategies, each representing a simplification in terms of image acquisition and dosimetric calculations. The results from the simplified strategies are compared to the results from the protocol-prescribed hybrid planar-SPECT/CT-based method by analysing differences both in per-cycle and total cumulative absorbed dose (AD) analyses. In general, the SPECT-based methods gave results that were largely consistent with the protocol-specified hybrid method, both in the per-cycle and cumulative AD analyses. Notably, performing one SPECT/CT per cycle at 96 h yielded ADs that were very similar to the protocol method. The methods using planar dosimetry resulted in larger variations, as expected, while giving 4 cycles to all patients resulted in the largest inter-individual differences in cumulative AD. Performing one SPECT/CT at 96 h in every treatment cycle gives sufficiently reliable dosimetric results to base individualized treatment planning on, with a reasonable demand on resources.

Feasibility study of a novel general purpose CZT-based digital SPECT camera: initial clinical results.

Abstract: The performance of a prototype novel digital single-photon emission computed tomography (SPECT) camera with multiple pixelated CZT detectors and high sensitivity collimators (Digital SPECT; Valiance X12 prototype, Molecular Dynamics) was evaluated in various clinical settings. Images obtained in the prototype system were compared to images from an analog camera fitted with high-resolution collimators. Clinical feasibility, image quality, and diagnostic performance of the prototype were evaluated in 36 SPECT studies in 35 patients including bone (n = 21), brain (n = 5), lung perfusion (n = 3), and parathyroid (n = 3) and one study each of sentinel node and labeled white blood cells. Images were graded on a scale of 1–4 for sharpness, contrast, overall quality, and diagnostic confidence. Digital CZT SPECT provided a statistically significant improvement in sharpness and contrast in clinical cases (mean score of 3.79 ± 0.61 vs. 3.26 ± 0.50 and 3.92 ± 0.29 vs. 3.34 ± 0.47 respectively, p < 0.001 for both). Overall image quality was slightly higher for the digital SPECT but not statistically significant (3.74 vs. 3.66). CZT SPECT provided significantly improved image sharpness and contrast compared to the analog system in the clinical settings evaluated. Further studies will evaluate the diagnostic performance of the system in large patient cohorts in additional clinical settings.

Determination of gamma camera calibration factors for quantitation of therapeutic radioisotopes.

Abstract: Camera calibration, which translates reconstructed count map into absolute activity map, is a prerequisite procedure for quantitative SPECT imaging. Both planar and tomographic scans using different phantom geometries have been proposed for the determination of the camera calibration factor (CF). However, there is no consensus on which approach is the best. The aim of this study is to evaluate all these calibration methods, compare their performance, and propose a practical and accurate calibration method for SPECT quantitation of therapeutic radioisotopes. Twenty-one phantom experiments (Siemens Symbia SPECT/CT) and 12 Monte Carlo simulations (GATE v6.1) using three therapy isotopes (131I, 177Lu, and 188Re) have been performed. The following phantom geometries were used: (1) planar scans of point source in air (PS), (2) tomographic scans of insert(s) filled with activity placed in non-radioactive water (HS + CB), (3) tomographic scans of hot insert(s) in radioactive water (HS + WB), and (4) tomographic scans of cylinders uniformly filled with activity (HC). Tomographic data were reconstructed using OSEM with CT-based attenuation correction and triple energy window (TEW) scatter correction, and CF was determined using total counts in the reconstructed image, while for planar scans, the photopeak counts, corrected for scatter and background with TEW, were used. Additionally, for simulated data, CF obtained from primary photons only was analyzed. For phantom experiments, CF obtained from PS and HS + WB agreed to within 6% (below 3% if experiments performed on the same day are considered). However, CF from HS + CB exceeded those from PS by 4–12%. Similar trend was found in simulation studies. Analysis of CFs from primary photons helped us to understand this discrepancy. It was due to underestimation of scatter by the TEW method, further enhanced by attenuation correction. This effect becomes less important when the source is distributed over the entire phantom volume (HS + WB and HC). Camera CF could be determined using planar scans of a point source, provided that the scatter and background contributions are removed, for example using the clinically available TEW method. This approach is simple and yet provides CF with sufficient accuracy (~ 5%) to be used in clinics for radiotracer quantification.

Accuracy and reproducibility of simplified QSPECT dosimetry for personalized 177 Lu-octreotate PRRT.

Abstract: Routine dosimetry is essential for personalized 177Lu-octreotate peptide receptor radionuclide therapy (PRRT) of neuroendocrine tumors (NETs), but practical and robust dosimetry methods are needed for wide clinical adoption. The aim of this study was to assess the accuracy and inter-observer reproducibility of simplified dosimetry protocols based on quantitative single-photon emission computed tomography (QSPECT) with a limited number of scanning time points. We also updated our personalized injected activity (IA) prescription scheme. Seventy-nine NET patients receiving 177Lu-octreotate therapy (with a total of 279 therapy cycles) were included in our study. Three-time-point (3TP; days 0, 1, and 3) QSPECT scanning was performed following each therapy administration. Dosimetry was obtained using small volumes of interest activity concentration sampling for the kidney, the bone marrow and the tumor having the most intense uptake. Accuracy of the simplified dosimetry based on two-time-point (2TP; days 1 and 3, monoexponential fit) or a single-time-point (1TPD3; day 3) scanning was assessed, as well as that of hybrid methods based on 2TP for the first cycle and 1TP (day 1 or 3; 2TP/1TPD1 and 2TP/1TPD3, respectively) or no imaging at all (based on IA only; 2TP/no imaging (NI)) for the subsequent induction cycles. The inter-observer agreement was evaluated for the 3TP, 2TP, and hybrid 2TP/1TPD3 methods using a subset of 60 induction cycles (15 patients). The estimated glomerular filtration rate (eGFR), body size descriptors (weight, body surface area (BSA), lean body weight (LBW)), and products of both were assessed for their ability to predict IA per renal absorbed dose at the first cycle. The 2TP dosimetry estimates correlated highly with those from the 3TP data for all tissues (Spearman r > 0.99, P < 0.0001) with small relative errors between the methods, particularly for the kidney and the tumor, with median relative errors not exceeding 2% and interdecile ranges spanning over less than 6% and 4%, respectively, for the per-cycle and cumulative estimates. For the bone marrow, the errors were slightly greater (median errors < 6%, interdecile ranges < 14%). Overall, the strength of correlations of the absorbed dose estimates from the simplified methods with those from the 3TP scans tended to progressively decrease, and the relative errors to increase, in the following order: 2TP, 2TP/1TPD3, 1TPD3, 2TP/1TPD1, and 2TP/NI. For the tumor, the 2TP/NI scenario was highly inaccurate due to the interference of the therapeutic response. There was an excellent inter-observer agreement between the three observers, in particular for the renal absorbed dose estimated using the 3TP and 2TP methods, with mean errors lesser than 1% and standard deviations of 5% or lower. The eGFR · LBW and eGFR · BSA products best predicted the ratio of IA to the renal dose (GBq/Gy) for the first cycle (Spearman r = 0.41 and 0.39, respectively; P < 0.001). For the first cycle, the personalized IA proportional to eGFR · LBW or eGFR · BSA decreased the range of delivered renal absorbed dose between patients as compared with the fixed IA. For the subsequent cycles, the optimal personalized IA could be determined based on the prior cycle renal GBq/Gy with an error of less than 21% in 90% of patients. A simplified dosimetry protocol based on two-time-point QSPECT scanning on days 1 and 3 post-PRRT provides reproducible and more accurate dose estimates than the techniques relying on a single time point for non-initial or all cycles and results in limited patient inconvenience as compared to protocols involving scanning at later time points. Renal absorbed dose over the 4-cycle induction PRRT course can be standardized by personalizing IA based on the product of eGFR with LBW or BSA for the first cycle and on prior renal dosimetry for the subsequent cycles.

Feasibility and accuracy of single time point imaging for renal dosimetry following 177Lu-DOTATATE ('Lutate') therapy.

Abstract: This study aims to assess both feasibility and accuracy of renal dosimetry imaging protocols in patients receiving Lutate therapy for neuroendocrine tumours (NETs), when data acquisition over multiple days is not possible on all cycles. Patients who had received a full 4 cycles of Lutate therapy with complete imaging at each cycle were included. Imaging consisted of quantitative SPECT/CT of the kidneys at 4, 24 and 96–120 h post injection. Renal absorbed dose was calculated for each data set, and in addition, five alternative methods were explored for comparison. Method 1: a patient average clearance time (t1/2 average) derived from the first half of contributing patient data was used to estimate absorbed dose for subsequent patients based on 4 h imaging alone; method 2: t1/2 average was applied to subsequent patients on 24 h imaging alone; method 3: a patient-specific clearance rate (t1/2 patient) was determined from complete image data of cycle 1 and applied subsequently to remaining cycles using 4 h image data alone; method 4: t1/2 patient was applied to 24 h imaging alone in subsequent cycles; method 5: the 120 h data was estimated on subsequent cycles based on the cycle 1 fraction of injected activity (%IA) at 24 and 120 h. Twenty treatments from 18 patients, resulting in 80 cycles of therapy, were analysed. The measured average renal absorbed dose per cycle of treatment was 0.38 ± 0.19 Gy/GBq when derived from full imaging data. The use of t1/2 average applied to a single time point led to large deviations of dose estimates from true values (on average 59% and 30%, when using 4 h data and 24 h data, respectively). The use of complete image data on cycle 1 and the derivation of t1/2 patient led to improved dose estimates, with an average deviation from true values of 13% and 2% when using 4 h data only and 24 h data only, respectively. The use of a 120 h %IA derived from cycle 1 led to an average deviation from true dose estimates of 14%. In instances where demands on both patients and facilities make multiple time point data acquisition impractical, renal dosimetry is best derived through complete imaging at cycle 1 only followed by a single 24 h imaging time point on subsequent cycles, assuming no significant changes in renal function during the time course of therapy.

Fast GPU-based Monte Carlo code for SPECT/CT reconstructions generates improved 177 Lu images.

Abstract: Full Monte Carlo (MC)-based SPECT reconstructions have a strong potential for correcting for image degrading factors, but the reconstruction times are long. The objective of this study was to develop a highly parallel Monte Carlo code for fast, ordered subset expectation maximum (OSEM) reconstructions of SPECT/CT images. The MC code was written in the Compute Unified Device Architecture language for a computer with four graphics processing units (GPUs) (GeForce GTX Titan X, Nvidia, USA). This enabled simulations of parallel photon emissions from the voxels matrix (1283 or 2563). Each computed tomography (CT) number was converted to attenuation coefficients for photo absorption, coherent scattering, and incoherent scattering. For photon scattering, the deflection angle was determined by the differential scattering cross sections. An angular response function was developed and used to model the accepted angles for photon interaction with the crystal, and a detector scattering kernel was used for modeling the photon scattering in the detector. Predefined energy and spatial resolution kernels for the crystal were used. The MC code was implemented in the OSEM reconstruction of clinical and phantom 177Lu SPECT/CT images. The Jaszczak image quality phantom was used to evaluate the performance of the MC reconstruction in comparison with attenuated corrected (AC) OSEM reconstructions and attenuated corrected OSEM reconstructions with resolution recovery corrections (RRC). The performance of the MC code was 3200 million photons/s. The required number of photons emitted per voxel to obtain a sufficiently low noise level in the simulated image was 200 for a 1283 voxel matrix. With this number of emitted photons/voxel, the MC-based OSEM reconstruction with ten subsets was performed within 20 s/iteration. The images converged after around six iterations. Therefore, the reconstruction time was around 3 min. The activity recovery for the spheres in the Jaszczak phantom was clearly improved with MC-based OSEM reconstruction, e.g., the activity recovery was 88% for the largest sphere, while it was 66% for AC-OSEM and 79% for RRC-OSEM. The GPU-based MC code generated an MC-based SPECT/CT reconstruction within a few minutes, and reconstructed patient images of 177Lu-DOTATATE treatments revealed clearly improved resolution and contrast.

Comparison of radiobiological parameters for 90 Y radionuclide therapy (RNT) and external beam radiotherapy (EBRT) in vitro

Abstract: Dose rate variation is a critical factor affecting radionuclide therapy (RNT) efficacy. Relatively few studies to date have investigated the dose rate effect in RNT. Therefore, the aim of this study was to benchmark 90Y RNT (at different dose rates) against external beam radiotherapy (EBRT) in vitro and compare cell kill responses between the two irradiation processes. Three human colorectal carcinoma (CRC) cell lines (HT29, HCT116, SW48) were exposed to 90Y doses in the ranges 1–10.4 and 6.2–62.3 Gy with initial dose rates of 0.013–0.13 Gy/hr (low dose rate, LDR) and 0.077–0.77 Gy/hr (high dose rate, HDR), respectively. Results were compared to a 6-MV photon beam doses in the range from 1–9 Gy with constant dose rate of 277 Gy/hr. The cell survival parameters from the linear quadratic (LQ) model were determined. Additionally, Monte Carlo simulations were performed to calculate the average dose, dose rate and the number of hits in the cell nucleus. For the HT29 cell line, which was the most radioresistant, the α/β ratio was found to be ≈ 31 for HDR–90Y and ≈ 3.5 for EBRT. LDR–90Y resulting in insignificant cell death compared to HDR–90Y and EBRT. Simulation results also showed for LDR–90Y, for doses ≲ 3 Gy, the average number of hits per cell nucleus is ≲ 2 indicating insufficiently delivered lethal dose. For 90Y doses $\gtrsim $ 3 Gy the number of hits per nucleus decreases rapidly and falls below ≈ 2 after ≈ 5 days of incubation time. Therefore, our results demonstrate that LDR–90Y is radiobiologically less effective than EBRT. However, HDR–90Y at ≈ 56 Gy was found to be radiobiologically as effective as acute ≈ 8 Gy EBRT. These results demonstrate that the efficacy of RNT is dependent on the initial dose rate at which radiation is delivered. Therefore, for a relatively long half-life radionuclide such as 90Y, a higher initial activity is required to achieve an outcome as effective as EBRT.

Impact of PET reconstruction protocols on quantification of lesions that fulfil the PERCIST lesion inclusion criteria

Abstract: The aim of this study was to compare liver and oncologic lesion standardized uptake values (SUV) obtained through two different reconstruction protocols, GE’s newest clinical lesion detection protocol (Q.Clear) and the EANM Research Ltd (EARL) harmonization protocol, and to assess the clinical relevance of potential differences and possible implications for daily clinical practice using the PERCIST lesional inclusion criteria. NEMA phantom recovery coefficients (RC) and SUV normalized for lean body mass (LBM), referred to as SUV normalized for LBM (SUL), of liver and lesion volumes of interest were compared between the two reconstruction protocols. Head-to-toe PET/CT examinations and raw data from 64 patients were retrospectively retrieved. PET image reconstruction was carried out twice: once optimized for quantification, complying with EARL accreditation requirements, and once optimized for lesion detection, according to GE’s Q.Clear reconstruction settings. The two reconstruction protocols showed different NEMA phantom RC values for different sphere sizes. Q.Clear values were always highest and exceeded the EARL accreditation maximum for smaller spheres. Comparison of liver SULmean showed a statistically significant but clinically irrelevant difference between both protocols. Comparison of lesion SULpeak and SULmax showed a statistically significant, and clinically relevant, difference of 1.64 and 4.57, respectively. For treatment response assessment using PERCIST criteria, the harmonization reconstruction protocol should be used as the lesion detection reconstruction protocol using resolution recovery systematically overestimates true SUL values.

Deadtime effects in quantification of 177Lu activity for radionuclide therapy.

Abstract: The aim of this study was to investigate the deadtime (DT) effects that are present in 177Lu images acquired after radionuclide therapy injection, assess differences in DT based on the full spectrum and the photopeak-only measurements, and design a method to correct for the deadtime losses. A Siemens SymbiaT SPECT/CT camera with a medium energy collimator was used. A 295-mL bottle was placed off-center inside a large cylinder filled with water, and 177Lu activity was sequentially added up to a maximum of 9.12 GBq. The true count rates vs. observed count rates were plotted and fitted to the DT paralyzable model. This analysis was performed using counts recorded in the full spectrum and in other energy windows. The DT correction factors were calculated using the percentage difference between the true and the observed count rates. The DT values of 5.99 ± 0.02 μs, 4.60 ± 0.052 μs, and 0.19 ± 0.18 μs were obtained for the primary photons (PP) recorded in the 113- and 208-keV photopeaks and for the full spectrum, respectively. For the investigated range of count rates, the DT correction factors of up to 23% were observed for PP corresponding to the 113-keV photopeak, while for the 208-keV photopeak values of up to 20% were obtained. These values were almost three times higher than the deadtime correction factors derived from the full spectrum. The paralyzable model showed to be appropriate for the investigated range of counts, which were five to six times higher than those observed in the patient post-therapy imaging. Our results suggest that the deadtime corrections should be based on count losses in the scatter-corrected photopeak window and not on the deadtime determined from the full spectrum. Finally, a general procedure that can be followed to correct patient images for deadtime is presented.

Inter-comparison of quantitative imaging of lutetium-177 (177Lu) in European hospitals

Abstract: This inter-comparison exercise was performed to demonstrate the variability of quantitative SPECT/CT imaging for lutetium-177 (177Lu) in current clinical practice. Our aim was to assess the feasibility of using international inter-comparison exercises as a means to ensure consistency between clinical sites whilst enabling the sites to use their own choice of quantitative imaging protocols, specific to their systems. Dual-compartment concentric spherical sources of accurately known activity concentrations were prepared and sent to seven European clinical sites. The site staff were not aware of the true volumes or activity within the sources—they performed SPECT/CT imaging of the source, positioned within a water-filled phantom, using their own choice of parameters and reported their estimate of the activities within the source. The volumes reported by the participants for the inner section of the source were all within 29% of the true value and within 60% of the true value for the outer section. The activities reported by the participants for the inner section of the source were all within 20% of the true value, whilst those reported for the outer section were up to 83% different to the true value. A variety of calibration and segmentation methods were used by the participants for this exercise which demonstrated the variability of quantitative imaging across clinical sites. This paper presents a method to assess consistency between sites using different calibration and segmentation methods.

Feasibility of PET/CT system performance harmonisation for quantitative multicentre 89Zr studies

Abstract: The aim of this study was to investigate the variability in quantitative performance and feasibility of quantitative harmonisation in 89Zr PET/CT imaging. Eight EANM EARL-accredited (Kaalep A et al., Eur J Nucl Med Mol Imaging 45:412–22, 2018) PET/CT systems were investigated using phantom acquisitions of uniform and NEMA NU2-2007 body phantoms. The phantoms were filled according to EANM EARL guidelines for [18F]FDG, but [18F]FDG solution was replaced by a 89Zr calibration mixture. For each system, standard uptake value (SUV) accuracy and recovery coefficients (RC) using SUVmean, SUVmax and SUVpeak metrics were determined. All eight investigated systems demonstrated similarly shaped RC curves, and five of them exhibited closely aligning recoveries when SUV bias correction was applied. From the evaluated metrics, SUVpeak was found to be least sensitive to noise and reconstruction differences among different systems. Harmonisation of PET/CT scanners for quantitative 89Zr studies is feasible when proper scanner-dose calibrator cross-calibration and harmonised image reconstruction procedures are followed. An accreditation programme for PET/CT scanners would facilitate multicentre 89Zr quantitative studies.

Evaluation of zero-echo-time attenuation correction for integrated PET/MR brain imaging-comparison to head atlas and 68Ge-transmission-based attenuation correction.

Abstract: MRI does not offer a direct method to obtain attenuation correction maps as its predecessors (stand-alone PET and PET/CT), and bone visualisation is particularly challenging. Recently, zero-echo-time (ZTE) was suggested for MR-based attenuation correction (AC). The aim of this work was to evaluate ZTE- and atlas-AC by comparison to 68Ge-transmission scan-based AC. Nine patients underwent brain PET/MR and stand-alone PET scanning using the dopamine transporter ligand 11C-PE2I. For each of them, two AC maps were obtained from the MR images: an atlas-based, obtained from T1-weighted LAVA-FLEX imaging with cortical bone inserted using a CT-based atlas, and an AC map generated from proton-density-weighted ZTE images. Stand-alone PET 68Ge-transmission AC map was used as gold standard. PET images were reconstructed using the three AC methods and standardised uptake value (SUV) values for the striatal, limbic and cortical regions, as well as the cerebellum (VOIs) were compared. SUV ratio (SUVR) values normalised for the cerebellum were also assessed. Bias, precision and agreement were calculated; statistical significance was evaluated using Wilcoxon matched-pairs signed-rank test. Both ZTE- and atlas-AC showed a similar bias of 6–8% in SUV values across the regions. Correlation coefficients with 68Ge-AC were consistently high for ZTE-AC (r 0.99 for all regions), whereas they were lower for atlas-AC, varying from 0.99 in the striatum to 0.88 in the posterior cortical regions. SUVR showed an overall bias of 2.9 and 0.5% for atlas-AC and ZTE-AC, respectively. Correlations with 68Ge-AC were higher for ZTE-AC, varying from 0.99 in the striatum to 0.96 in the limbic regions, compared to atlas-AC (0.99 striatum to 0.77 posterior cortex). Absolute SUV values showed less variability for ZTE-AC than for atlas-AC when compared to 68Ge-AC, but bias was similar for both methods. This bias is largely caused by higher linear attenuation coefficients in atlas- and ZTE-AC image compared to 68Ge-images. For SUVR, bias was lower when using ZTE-AC than for atlas-AC. ZTE-AC shows to be a more robust technique than atlas-AC in terms of both intra- and inter-patient variability.

Validation of the physiological background correction method for the suppression of the spill-in effect near highly radioactive regions in positron emission tomography

Abstract: Positron emission tomography (PET) imaging has a wide applicability in oncology, cardiology and neurology. However, a major drawback when imaging very active regions such as the bladder is the spill-in effect, leading to inaccurate quantification and obscured visualisation of nearby lesions. Therefore, this study aims at investigating and correcting for the spill-in effect from high-activity regions to the surroundings as a function of activity in the hot region, lesion size and location, system resolution and application of post-filtering using a recently proposed background correction technique. This study involves analytical simulations for the digital XCAT2 phantom and validation acquiring NEMA phantom and patient data with the GE Signa PET/MR scanner. Reconstructions were done using the ordered subset expectation maximisation (OSEM) algorithm. Dedicated point-spread function (OSEM+PSF) and a recently proposed background correction (OSEM+PSF+BC) were incorporated into the reconstruction for spill-in correction. The standardised uptake values (SUV) were compared for all reconstruction algorithms. The simulation study revealed that lesions within 15–20 mm from the hot region were predominantly affected by the spill-in effect, leading to an increased bias and impaired lesion visualisation within the region. For OSEM, lesion SUVmax converged to the true value at low bladder activity, but as activity increased, there was an overestimation as much as 19% for proximal lesions (distance around 15–20 mm from the bladder edge) and 2–4% for distant lesions (distance larger than 20 mm from the bladder edge). As bladder SUV increases, the % SUV change for proximal lesions is about 31% and 6% for SUVmax and SUVmean, respectively, showing that the spill-in effect is more evident for the SUVmax than the SUVmean. Also, the application of post-filtering resulted in up to 65% increment in the spill-in effect around the bladder edges. For proximal lesions, PSF has no major improvement over OSEM because of the spill-in effect, coupled with the blurring effect by post-filtering. Within two voxels around the bladder, the spill-in effect in OSEM is 42% (32%), while for OSEM+PSF, it is 31% (19%), with (and without) post-filtering, respectively. But with OSEM+PSF+BC, the spill-in contribution from the bladder was relatively low (below 5%, either with or without post-filtering). These results were further validated using the NEMA phantom and patient data for which OSEM+PSF+BC showed about 70–80% spill-in reduction around the bladder edges and increased contrast-to-noise ratio up to 36% compared to OSEM and OSEM+PSF reconstructions without post-filtering. The spill-in effect is dependent on the activity in the hot region, lesion size and location, as well as post-filtering; and this is more evident in SUVmax than SUVmean. However, the recently proposed background correction method facilitates stability in quantification and enhances the contrast in lesions with low uptake.

SMART (SiMulAtion and ReconsTruction) PET: an efficient PET simulation-reconstruction tool

Abstract: Positron-emission tomography (PET) simulators are frequently used for development and performance evaluation of segmentation methods or quantitative uptake metrics. To date, most PET simulation tools are based on Monte Carlo simulations, which are computationally demanding. Other analytical simulation tools lack the implementation of time of flight (TOF) or resolution modelling (RM). In this study, a fast and easy-to-use PET simulation-reconstruction package, SiMulAtion and ReconsTruction (SMART)-PET, is developed and validated, which includes both TOF and RM. SMART-PET, its documentation and instructions to calibrate the tool to a specific PET/CT system are available on Zenodo. SMART-PET allows the fast generation of 3D PET images. As input, it requires one image representing the activity distribution and one representing the corresponding CT image/attenuation map. It allows the user to adjust different parameters, such as reconstruction settings (TOF/RM), noise level or scan duration. Furthermore, a random spatial shift can be included, representing patient repositioning. To evaluate the tool, simulated images were compared with real scan data of the NEMA NU 2 image quality phantom. The scan was acquired as a 60-min list-mode scan and reconstructed with and without TOF and/or RM. For every reconstruction setting, ten statistically equivalent images, representing 30, 60, 120 and 300 s scan duration, were generated. Simulated and real-scan data were compared regarding coefficient of variation in the phantom background and activity recovery coefficients (RCs) of the spheres. Furthermore, standard deviation images of each of the ten statistically equivalent images were compared. SMART-PET produces images comparable to actual phantom data. The image characteristics of simulated and real PET images varied in similar ways as function of reconstruction protocols and noise levels. The change in image noise with variation of simulated TOF settings followed the theoretically expected behaviour. RC as function of sphere size agreed within 0.3–11% between simulated and actual phantom data. SMART-PET allows for rapid and easy simulation of PET data. The user can change various acquisition and reconstruction settings (including RM and TOF) and noise levels. The images obtained show similar image characteristics as those seen in actual phantom data.

Quantitative performance of 124 I PET/MR of neck lesions in thyroid cancer patients using 124 I PET/CT as reference.

Abstract: In patients with differentiated thyroid cancer (DTC), serial 124I PET/CT imaging is, for instance, used to assess the absorbed (radiation) dose to lesions. Frequently, the lesions are located in the neck and they are close to or surrounded by different tissue types. In contrast to PET/CT, MR-based attenuation correction in PET/MR may be therefore challenging in the neck region. The aim of this retrospective study was to assess the quantitative performance of 124I PET/MRI of neck lesions by comparing the MR-based and CT-based 124I activity concentrations (ACs). Sixteen DTC patients underwent PET/CT scans at 24 and 120 h after administration of about 25 MBq 124I. Approximately 1 h before or after PET/CT examination, each patient additionally received a 24-h PET/MR scan and sometimes a 120-h PET/MR scan. PET images were reconstructed using the respective attenuation correction approach. Appropriate reconstruction parameters and corrections were used to harmonize the reconstructed PET images to provide, for instance, similar spatial resolution. For each lesion, two types of ACs were ascertained: the maximum AC (max-AC) and an average AC (avg-AC). The avg-AC is the average activity concentration obtained within a spherical volume of interest with a diameter of 7 mm, equaling the PET scanner resolution. For each type of AC, the percentage AC difference between MR-based and CT-based ACs was determined and Lin’s concordance correlation analysis was applied. Quantitative performance was considered acceptable if the standard deviation was ± 25% (precision), and the mean value was within ± 10% (accuracy). The avg-ACs (max-ACs within parentheses) of 74 lesions ranged from 0.20 (0.33) to 657 (733) kBq/mL. Excluding two lesions with ACs of approximately 1 kBq/mL, the mean (median) ± standard deviation (range) was − 4% (− 5%) ± 14% (− 28 to 29%) for the avg-AC and − 9% (− 11%) ± 14% (− 33 to 33%) for the max-AC. Lin’s concordance correlation coefficients were ≥ 0.97, indicating substantial AC agreement. Quantification of lesions in the neck region using 124I PET/MR showed acceptable quantitation performance to 124I PET/CT for AC above 1 kBq/mL. The PET/MRI-based 124I ACs in the neck region can be therefore reliably used in pre-therapy dosimetry planning.

Quantitative analysis of phantom studies of 111In and 68Ga imaging of neuroendocrine tumours.

Abstract: Nuclear medicine imaging of neuroendocrine tumours is performed either by SPECT/CT imaging, using 111In-octreotide or by PET/CT imaging using 68Ga-radiolabelled somatostatin analogs. These imaging techniques will give different image quality and different detection thresholds for tumours, depending on size and activity uptake. The aim was to evaluate the image quality for 111In-SPECT and 68Ga-PET imaging, i.e. the smallest volume possible to visualize for different source-to-background activity ratios. The accuracy of quantification of lesion volume and activity was also investigated to develop an objective evaluation for radionuclide therapy eligibility. The phantom study was performed using the NEMA IEC Body Phantom with six hot spheres having inner diameters of 10, 13, 17, 22, 28, and 37 mm, filled with either 68Ga or 111In with sphere-to-background ratios (SBRs) of no background activity, 5:1, 2.5:1, and 1.25:1. Activity ratios of 1.25:1 and 2.5:1 are clinically found for lesions close to the liver and spleen. Clinical acquisition and reconstruction protocols were applied. Line profiles were drawn to evaluate the smallest detectable volume within a given SBR. Recovery curves based on threshold-based VOIs, threshold-based VOIs adapted to the background and CT-based ROIs were obtained for all SBRs and sphere diameters, allowing for quantification. The 10-mm sphere was not possible to detect in SPECT images. It was detectable in PET images for SBRs of 2.5:1 and higher. In a background corresponding to the activity uptake in the liver, spheres larger than 22–37 mm were detectable in the 111In-SPECT images and spheres larger than 13–22 mm were detectable in the 68Ga-PET images. The maximum activity concentration was accurately quantified for spheres larger than 22 mm in the PET images; however, the quantification was impaired by sphere size and background activity. It was not possible to detect the 10-mm sphere in any of the SPECT images. In a background corresponding to the activity uptake in the liver, spheres larger than approximately 30 mm were visible in the 111In-SPECT images and spheres larger than approximately 17 mm were visible in the 68Ga-PET images. Sphere diameter and background activity strongly affect the possibility of a correct quantification.

Fast quantitative reconstruction with focusing collimators for liver SPECT.

Abstract: Generation of a SPECT scan during procedure may aid in the optimization of treatments as liver radioembolization by offering image-guided dosimetry. This, however, requires both shortened acquisition times and fast quantitative reconstruction. Focusing collimators increase sensitivity and thus may speed up imaging. Monte Carlo-based iterative reconstruction has shown to provide quantitative results for parallel hole collimators but may be slow. The purpose of this work is to develop fast Monte Carlo-based reconstruction for focusing collimators and to evaluate the impact of reconstruction and collimator choice on quantitative accuracy of liver dosimetry by means of simulations. The developed fast Monte Carlo simulator was found to accurately generate projections compared to a full Monte Carlo simulation, providing projections in several seconds instead of several days. Monte Carlo-based scatter correction was superior to other scatter correction methods in describing recovered activity and reached similar noise levels as dual-energy window scatter correction. Although truncation artifacts were present in the cone beam collimator (50 cm), the region inside the field of view (FOV) could be reconstructed without loss of accuracy. Provided the object to image is inside the FOV, the focusing collimator with 50 cm focal distance could retrieve the same noise levels as a parallel hole collimator in 68% of the total scanning time, the multifocal collimator in 73% of the time, and the 100-cm focal distance collimator in 84% of the time. Focusing collimators combined with Monte Carlo-based reconstruction have the ability to enable quantitative imaging of the FOV in a significantly shorter timeframe. The proposed approach to the forward projector will additionally make it possible to reconstruct within minutes. These are crucial steps in moving toward real-time dosimetry during interventions.

Performance evaluation of a new high-sensitivity time-of-flight clinical PET/CT system

Abstract: PoleStar m660 is a newly developed clinical PET/CT system with time-of-flight (TOF) capability. The aim of this study is to characterize the performance of the new system. Spatial resolution, sensitivity, scatter fraction, and noise equivalent count rate (NECR) were measured on the scanner according to the NEMA NU 2-2012 protocol. The timing resolution was measured using a rotating line source that orbited around the center of field of view (CFOV) at a radius of 20 cm. The image quality phantom was also imaged to quantify the percent contrast, percent background variability, and residual error. The impacts of data acquisition time and bed overlap on the PET image quality were also evaluated using phantom and patient studies. The transverse (axial) spatial resolutions were 3.59 mm (3.67), 4.08 mm (4.65), and 5.32 mm (6.48) full width at half maximum (FWHM) at 1 cm, 10 cm, and 20 cm, respectively, off the CFOV. The measured sensitivity was 10.7 cps/kBq at the CFOV and 10.4 cps/kBq at 10 cm off the CFOV. The peak NECR was 216.7 kcps at an activity concentration of 29.1 kBq/ml, and the scatter fraction was 38.2%. An average of 435 ps FWHM timing resolution was measured. For the image quality phantom, the contrast recovery ratios ranged from 33.9 to 76.4%, while the background variability ranged from 4.7 to 2.0%. In the preliminary clinical study, no noticeable difference in the image quality was observed when the scan time for the whole body and brain was reduced to 1 min/bed and 3 min, respectively. The tested 21% bed overlap showed no significant difference in the image quality compared with the default 38% bed overlap setting. The physical performances of the PoleStar m660 PET/CT system showed good sensitivity, count rate performance, and timing resolution. The improved performance could help to reduce the acquisition time and bed overlap in the clinical application without detectable compromise in the image quality.

Performance evaluation of a novel brain-dedicated SPECT system.

Abstract: Single-photon emission computed tomography (SPECT) imaging is an important diagnostic tool for the early detection of the loss of nigrostriatal dopaminergic neurons in Parkinson’s disease (PD) and similar neurodegenerative disorders. Visualization and quantification of dopamine transporter (DAT) binding in the striatum is an established diagnostic tool to detect nigrostriatal dopaminergic degeneration. Given the small size of the striatum, high-resolution imaging is recommended. The InSPira HD system, a novel brain-dedicated SPECT scanner, allows for such detailed information with a spatial resolution down to ~ 3 mm full width at half maximum (FWHM). The current study examines performance of the InSPira HD for DAT imaging, by combining phantom tests from NU 1-2012 and NU 2-2012, and striatal scans. Due to the unique geometry of the InSPira, and fixed acquisition and reconstruction settings, standard National Electrical Manufacturers Association (NEMA) testing is not applicable. Therefore, a combination of NU 1-2012 and NU 2-2012 standards were applied, with modifications to accommodate the InSPira HD. A small Jaszczak phantom with hot spheres and cold rod inserts was used to determine recovery coefficients, contrast, and uniformity. Spatial resolution was evaluated across the field of view (FOV) for point and line sources in air and water. A striatal phantom was used to model DAT imaging. A clinical, a high-resolution, and an experimental research reconstruction method were compared. Acquired SPECT images demonstrated spatial resolution in air of ~ 3 mm in the center in the FOV for the high-resolution reconstruction approach. Spatial resolution in air for the clinical and research reconstruction approach was ~ 6–8 mm in the center of the FOV, which decreased in the transaxial plane with increasing radial distance from the center of the FOV. Reconstructed images of the uniform area of the Jaszczak phantom showed limited variability with a coefficient of variation of 2.6% for the clinical reconstruction and 3.0% for the research reconstruction. The ≥ 6-mm rod group and all spheres were resolved for the clinical and research reconstruction approaches. Recovery coefficients (RCs) for the Jaszczak phantom ranged from 0.49 to 0.89 (sphere diameters between 9.8 and 31.2 mm). RCs for the striatal phantom ranged from 0.50 to 0.55, with linearity of striatal ratios for a range of background concentrations (R = 0.97). Results from the phantom data demonstrated acceptable image quality for the InSPira HD system for DAT SPECT imaging in humans.

System evaluation of automated production and inhalation of 15O-labeled gaseous radiopharmaceuticals for the rapid 15O-oxygen PET examinations.

Abstract: 15O-oxygen inhalation PET is unique in its ability to provide fundamental information regarding cerebral hemodynamics and energy metabolism in man. However, the use of 15O-oxygen has been limited in a clinical environment largely attributed to logistical complexity, in relation to a long study period, and the need to produce and inhale three sets of radiopharmaceuticals. Despite the recent works that enabled shortening of the PET examination period, radiopharmaceutical production has still been a limiting factor. This study was aimed to evaluate a recently developed radiosynthesis/inhalation system that automatically supplies a series of 15O-labeled gaseous radiopharmaceuticals of C15O, 15O2, and C15O2 at short intervals. The system consists of a radiosynthesizer which produces C15O, 15O2, and C15O2; an inhalation controller; and an inhalation/scavenging unit. All three parts are controlled by a common sequencer, enabling automated production and inhalation at intervals less than 4.5 min. The gas inhalation/scavenging unit controls to sequentially supply of qualified radiopharmaceuticals at given radioactivity for given periods at given intervals. The unit also scavenges effectively the non-inhaled radioactive gases. Performance and reproducibility are evaluated. Using an 15O-dedicated cyclotron with deuteron of 3.5 MeV at 40 μA, C15O, 15O2, and C15O2 were sequentially produced at a constant rate of 1400, 2400, and 2000 MBq/min, respectively. Each of radiopharmaceuticals were stably inhaled at < 4.5 min intervals with negligible contamination from the previous supply. The two-hole two-layered face mask with scavenging device minimized the gaseous radioactivity surrounding subject’s face, while maintaining the normocapnia during examination periods. Quantitative assessment of net administration doses could be assessed using a pair of radio-detectors at inlet and scavenging tubes, as 541 ± 149, 320 ± 103, 523 ± 137 MBq corresponding to 2-min supply of 2574 ± 255 MBq for C15O, and 1-min supply of 2220 ± 766 and 1763 ± 174 for 15O2 and C15O2, respectively. The present system allowed for automated production and inhalation of series of 15O-labeled radiopharmaceuticals as required in the rapid 15O-Oxygen PET protocol. The production and inhalation were reproducible and improved logistical complexity, and thus the use of 15O-oxygen might have become practically applicable in clinical environments.

Impact of different approaches to calculation of treatment activities on achieved doses in radioiodine therapy of benign thyroid diseases

Abstract: Radioiodine has been used for the treatment of benign thyroid diseases for over 70 years. However, internationally, there is no common standard for pretherapeutic dosimetry to optimally define the individual therapy activity. Here, we analyze how absorbed tissue doses are influenced by different approaches to pretherapeutic activity calculation of varying complexity. Pretherapeutic determination of treatment activity was retrospectively recalculated in 666 patients who had undergone radioiodine therapy for benign thyroid diseases (Graves’ disease, non-toxic goiter, and uni- and multinodular goiter). Approaches considering none, some, or all of a set of individual factors, including target volume, maximum radioiodine uptake, and effective half-life, were applied. Assuming individually stable radioiodine kinetics, which had been monitored twice a day under therapy, hypothetically achieved tissue doses based on hypothetically administered activities resulting from the different methods of activity calculation were compared to intended target doses. The Marinelli formula yields the smallest deviations of hypothetically achieved doses from intended target doses. Approaches taking individual target volume into consideration perform better than fixed therapy activities, which lead to high variances in achieved doses and high deviations of hypothetically achieved doses from intended target doses. Elaborate pretherapeutic dose planning, taking individual radioiodine uptake, half-life, and target volume into consideration, should be used whenever possible. The use of disease-specific fixed activities cannot be recommended. Deviations of achieved tissue doses from target doses can already be significantly lowered by application of volume-adapted treatment activities if more elaborate means are not available.

Molecular imaging using the theranostic agent 197(m) Hg: phantom measurements and Monte Carlo simulations.

Abstract: Radiomercury 197mHg and 197Hg, henceforth referred to as 197(m)Hg, is a promising theranostic radionuclide endowed with properties that allow diagnostic and therapeutic applications. The aim of this work was to investigate the capabilities of 197(m)Hg for nuclear medicine imaging. Therefore measurements were performed by using a Philips BrightView SPECT camera. Furthermore, Monte Carlo simulations using the GATE software were performed to theoretically explore the imaging contribution from the various gamma and X-ray emissions from 197(m)Hg for a commercial clinical camera with low-energy high-resolution (LEHR) and high-energy general-purpose (HEGP) collimators. We estimated the spatial resolution by using a four-quadrant bar phantom, and we evaluated the planar and tomographic images from an abdominal phantom containing three cylindrical sources of 197(m)Hg solution. A good accordance between measurements and simulations was found for planar and SPECT imaging. Simulations allowed the decomposition of the detected energy spectrum into photon origins. Measurements and simulations for the bar phantom revealed that for the LEHR collimator, the 6-mm pattern could be resolved, whereas for the HEGP collimator, the resolution is about 10 mm. Furthermore, we found that no significant image distortion results from high-energy photons when using the LEHR collimator. We demonstrated the imaging capabilities of 197(m)Hg which is essential both for diagnostic applications and to determine the in vivo biodistribution for dose calculations in therapeutic applications.

Predictive power of the post-treatment scans after the initial or first two courses of [ 177 Lu]-DOTA-TATE

Abstract: The aim of this study was to evaluate the predictive power of the absorbed dose to kidneys after the first course of treatment with [177Lu]-DOTA-TATE for neuroendocrine tumors (NETs) on the cumulative kidney absorbed dose after 3 or 4 cycles of treatment. Post-treatment scans (PTS) are acquired after each cycle of peptide receptor radionuclide therapy (PRRT) with [177Lu]-DOTA-TATE for personalized radiation dosimetry in order to ensure a cumulative absorbed dose to kidneys under a safety threshold of 25 Gy. One hundred eighty-seven patients who completed treatment with [177Lu]-DOTA-TATE and underwent PTS for dosimetry calculation were included in this retrospective study. The correlation between the cumulative absorbed dose to kidneys after the completion of treatment and the absorbed dose after the first cycle(s) was studied. Multilinear regression analysis was done to predict the cumulative absorbed dose to the kidneys of the subsequent cycles, and an algorithm for the follow up of kidney absorbed dose is proposed. Patients whose absorbed dose to kidneys after the first cycle of treatment is below 5.6 Gy can receive four cycles of treatment with a cumulative dose less than 25 Gy (p < 0.1). For the other patients, the cumulative absorbed dose after 3 or 4 cycles of treatment can be predicted after the second cycle of treatment to allow for an early decision regarding the number of cycles that may be given. The follow up of kidney absorbed dose after PRRT can be simplified with the algorithm presented in this study, reducing by one-third the number of post-treatment scans and reducing hospitalization time for more than half of the treatment cycles.

Biodistribution, pharmacokinetics, and organ-level dosimetry for 188Re-AHDD-Lipiodol radioembolization based on quantitative post-treatment SPECT/CT scans

Abstract: Rhenium-188-labelled-Lipiodol radioembolization is a safe and cost-effective treatment for primary liver cancer. In order to determine correlations between treatment doses and patient response to therapy, accurate patient-specific dosimetry is required. Up to date, the reported dosimetry of 188Re-Lipiodol has been based on whole-body (WB) planar imaging only, which has limited quantitative accuracy. The aim of the present study is to determine the in vivo pharmacokinetics, bio-distribution, and organ-level dosimetry of 188Re-AHDD-Lipiodol radioembolization using a combination of post-treatment planar and quantitative SPECT/CT images. Furthermore, based on the analysis of the pharmacokinetic data, a practical and relatively simple imaging and dosimetry method that could be implemented in clinics for 188Re-AHDD-Lipiodol radioembolization is proposed. Thirteen patients with histologically proven hepatocellular carcinoma underwent 188Re-AHDD-Lipiodol radioembolization. A series of 2–3 WB planar images and one SPECT/CT scan were acquired over 48 h after the treatment. The time-integrated activity coefficients (TIACs, also known as residence-times) and absorbed doses of tumors and organs at risk (OARs) were determined using a hybrid WB/SPECT imaging method. Whole-body imaging showed that 188Re-AHDD-Lipiodol accumulated mostly in the tumor and liver tissue but a non-negligible amount of the pharmaceutical was also observed in the stomach, lungs, salivary glands, spleen, kidneys, and urinary bladder. On average, the measured effective half-life of 188Re-AHDD-Lipiodol was 12.5 ± 1.9 h in tumor. The effective half-life in the liver and lungs (the two organs at risk) was 12.6 ± 1.7 h and 12.0 ± 1.9 h, respectively. The presence of 188Re in other organs was probably due to the chemical separation and subsequent release of the free radionuclide from Lipiodol. The average doses per injected activity in the tumor, liver, and lungs were 23.5 ± 40.8 mGy/MBq, 2.12 ± 1.78 mGy/MBq, and 0.11 ± 0.05 mGy/MBq, respectively. The proposed imaging and dosimetry method, consisting of a single SPECT/CT for activity determination followed by 188Re-AHDD-Lipiodol clearance with the liver effective half-life of 12.6 h, resulted in TIACs estimates (and hence, doses) mostly within ± 20% from the reference TIACs (estimated using three WB images and one SPECT/CT). The large inter-patient variability of the absorbed doses in tumors and normal tissue in 188Re-HDD-Lipiodol radioembolization patients emphasizes the importance of patient-specific dosimetry calculations based on quantitative post-treatment SPECT/CT imaging.

Multi institutional quantitative phantom study of yttrium-90 PET in PET/MRI: the MR-QUEST study.

Abstract: Yttrium-90 (90Y) radioembolization involves the intra-arterial delivery of radioactive microspheres to treat hepatic malignancies. Though this therapy involves careful pre-treatment planning and imaging, little is known about the precise location of the microspheres once they are administered. Recently, there has been growing interest post-radioembolization imaging using positron-emission tomography (PET) for quantitative dosimetry and identifying lesions that may benefit from additional salvage therapy. In this study, we aim to measure the inter-center variability of 90Y PET measurements as measured on PET/MRI in preparation for a multi-institutional prospective phase I/II clinical trial. Eight institutions participated in this study and followed a standardized phantom filling and imaging protocol. The NEMA NU2-2012 body phantom was filled with 3 GBq of 90Y chloride solution. The phantom was imaged for 30 min in listmode on a Siemens Biograph mMR non-TOF PET/MRI scanner at five time points across 10 days (0.3–3.0 GBq). Raw PET data were sent to a central site for image reconstruction and data analysis. Images were reconstructed with optimal parameters determined from a previous study. Volumes of interest (VOIs) matching the known sphere diameters were drawn on the vendor-provided attenuation map and propagated to the PET images. Recovery coefficients (RCs) and coefficient of variation of the RCs (COV) were calculated from these VOIs for each sphere size and activity level. Mean RCs ranged from 14.5 to 75.4%, with the lowest mean RC coming from the smallest sphere (10 mm) on the last day of imaging (0.16 MBq/ml) and the highest mean RC coming from the largest sphere (37 mm) on the first day of imaging (2.16 MBq/ml). The smaller spheres tended to exhibit higher COVs. In contrast, the larger spheres tended to exhibit lower COVs. COVs from the 37 mm sphere were 25%. Post-radioembolization dosimetry of lesions or other VOIs ≥ 22 mm in diameter can be consistently obtained (< 25% variability) at a multi-institutional level using PET/MRI for any clinically significant activity for 90Y radioembolization.

Comparison of PET/CT and PET/MR imaging and dosimetry of yttrium-90 ( 90 Y) in patients with unresectable hepatic tumors who have received intra-arterial radioembolization therapy with 90 Y microspheres

Abstract: The aim of our study was to compare 90Y dosimetry obtained from PET/MRI versus PET/CT post-therapy imaging among patients with primary or metastatic hepatic tumors. First, a water-filled Jaszczak phantom containing fillable sphere with 90Y-chloride was acquired on both the PET/CT and PET/MRI systems, in order to check the cross-calibration of the modalities. Following selective internal radiation therapy (SIRT) with 90Y microspheres, 32 patients were imaged on a PET/CT system, immediately followed by a PET/MRI study. Reconstructed images were transferred to a common platform and used to calculate 90Y dosimetry. A Passing-Bablok regression scatter diagram and the Bland and Altman method were used to analyze the difference between the dosimetry values. The phantom study showed that both modalities were calibrated with less than 1% error. The mean liver doses for the 32 subjects calculated from PET/CT and PET/MRI were 51.6 ± 24.7 Gy and 46.5 ± 22.7 Gy, respectively, with a mean difference of 5.1 ± 5.0 Gy. The repeatability coefficient was 9.0 (18.5% of the mean). The Spearman rank correlation coefficient was very high, ρ = 0.97. Although the maximum dose to the liver can be significantly different (up to 40%), mean liver doses from each modalities were relatively close, with a difference of 18.5% or less. The two main contributors to the difference in 90Y dosimetry calculations using PET/CT versus PET/MRI can be attributed to the differences in regions of interest (ROIs) and differences attributed to attenuation correction. Due to the superior soft-tissue contrast of MRI, liver contours are usually better seen than in CT images. However, PET/CT provides better quantification of PET images, due to better attenuation correction. In spite of these differences, our results demonstrate that the dosimetry values obtained from PET/MRI and PET/CT in post-therapy 90Y studies were similar.