[en] Complete text of publication follows: Objectives: The total metabolic tumor volume (TMTV) estimated from FDG-PET images is a promising biomarker of the total tumor burden, especially in patients with lymphomas. However, estimating TMTV is challenging and often time consuming when multiple targets, including primary tumors, nodes and metastases, with various metabolic activities, volumes and contrasts have to be segmented. We developed a convenient and user-friendly pipeline for estimating TMTV that is now offered as part of the free LIFEx software (www.lifexsoft.org). Methods: The procedure of TMTV calculation is based on four customizable steps: The first step is a semiautomatic initialization of intermediate to high metabolic activity (typically SUV>2) volumes in which the tumor regions should be identified, yielding a possibly disconnected gross volume GV. Then, tumor volume delineation within this initial GV is refined using one or several customizable criteria in combination, including: 1) absolute SUV threshold (i.e. SUV>2.5), 2) region-to-liver activity threshold, 3) percent SUV_max threshold (i.e., 40%), 4) adaptive threshold accounting for surrounding activity. A third step identifies groups of connected voxels, deletes clusters that are too small in volumes, and derives the resulting TMTV. Last, a final step makes it possible for the user to remove regions with physiological uptake (brain, bladder, etc) using a 1-click operation, or to add regions using a 1-click operation that runs a region growing algorithm. The TMTV value is automatically updated during these operations. All regions can be displayed using axial, sagittal and coronal slices and using a maximum-intensity-projection representation. PET volume and associated regions can also be overlapped with the CT or MR for visual analysis. The whole procedure has been implemented in the LIFEx free software (Windows, MacOS and Linux compatible) thus enabling textural indices calculation from each component of the TMTV. The results are exported in an Excel file and in the form of an image booklet displaying the regions. All regions included in the final TMTV volumes can be saved on disk for subsequent use. Results: The intuitive software interface as well as associated demo videos allow for immediate use of the tool. The calculation of TMTV is performed within 4 minutes from a whole body PET scan. Being easily accessible, the software has already been downloaded by more than 250 users over the world and is currently under evaluation to assess its ability to produce fast and robust estimates of TMTV in the context of multicenter studies of lymphoma patient cohorts

[en] Complete text of publication follows: Purpose: The characterization of tumor heterogeneity using textural features in radiomic analyses of PET images has shown promise to predict patient response or survival. In this context, the goal of this study is to identify for each patient a radiomic numerical twin who has similar radiomic feature values to learn from the numerical twin's history and guide patient management. Here, we test this concept to predict treatment response. Subjects and Methods: 107 patients with newly diagnosed esophageal cancer underwent pre-treatment ¹⁸F-FDG PET scan (data extracted from Ypsilantis et al, PLoS ONE 10(9):e0137036). All patients received a neoadjuvant chemotherapy and were later classified as non-responders (NR=69) or responders (R=38). In each patient, the primary lesion in the baseline scan was segmented using a threshold set to 40% of SUV_max. In each resulting volume of interest, 103 radiomic features were extracted including 85 textural or fractal features and 18 histogram indices. Each lesion was associated with a vector b of biomarkers. We computed the element-wise ratio between the vector b(p) of one patient p and the vector b(i) of each of the other 106 patients i (i=1,P-1). A patient N was identified as the radiomic numerical twin of patient p if the distance of b(p)/b(N) to 1 was the lowest among the P-1 distances. Its response to treatment was then predicted as the one observed in patient N. We evaluated the ability of this approach to predict treatment response when using 2 or 3 biomarkers in b by calculating the Youden index in a leave-one-out validation. We compared the results with logistic regression and support vector machine (SVM) models. Results: When including two biomarkers in b, the best performance using the numerical twin concept was obtained using Kurtosis and Energy with 87% NR lesions and 63.2% R lesions accurately classified (Youden=0.50). With three biomarkers (Kurtosis, Energy and Fractal Dimension mean), Youden index increased to 0.55. With 2 or 3 biomarkers, the logistic regression and SVM models always yielded Youden index less than 0.43. Conclusion: This concept of radiomic numerical twins is validated in esophageal cancer to predict treatment response. We found that lesions with similar radiomic profiles consisting of only 2 to 3 biomarkers had similar response to therapy. The identification of numerical twins could assist patient management in the future, based on the disease evolution in the patients used to identify the numerical twin

[en] Complete text of publication follows: Introduction: [¹⁸F]Fludarabine is a PET radiotracer, the specificity of which has been studied in several lymphoma models (follicular RL7, DOHH2; MM RPMI8226) and under various conditions such as cancerous tissue following therapy and inflammation. The first-in-human study in two types of lymphoid malignancies (DLBCL, CLL) has consolidated the great potential of this innovative tool for lymphoma imaging in oncology [5,6]. The purpose of this study was to investigate whether [¹⁸F] fludarabine-PET can help differentiate between central nervous system (CNS) lymphoma and glioblastoma (GBM), considering also multimodal analyses with [¹⁸F]FDG, MRI and histology. Methods: Nude rats were implanted with human MC116 lymphoma- (n=10) or U87 glioma-cells (n=5). Tumor growth was monitored by MRI (7-Tesla, Bruker), with T2-weighted sequence (RARE) for anatomical features and T1-weighted (EPI) with gadolinium (Gd) enhancement for BBB permeability assessment. For PET investigation (Inveon, Siemens), 11 MBq [¹⁸F]fludarabine or [¹⁸F]FDG were injected via tail vein and PET dynamic images were acquired up to 90 min after radiotracer injection. Paired scans of the same rat with the two [¹⁸F]-labelled radiotracers were investigated. Initial volumes of interest (tumor and healthy contralateral or cerebellum tissue) were manually delineated on T2w image and set on co-registered PET image (Pmod 3.7) and tumour-to-background ratios (TBR) were calculated to semi-quantitatively assess the tracer accumulation in the tumour. Results: In lymphoma model, PET time-activity curves revealed a differential response of [¹⁸F]fludarabine between tumoral and healthy tissues with average TBR varying from 2.45 to 3.16 between 5 to 90 min post-injection. In contrast, [¹⁸F]FDG demonstrated similar uptake profiles for tumoral and normal regions with TBR varying from 0.84 to 1.06 between these two time points. In GBM model, the average TBRs were from 1.67 to 1.07 for [¹⁸F]fludarabine and from 1.08 to 1.65 for [¹⁸F]FDG. Therefore, inter-model comparisons showed significantly divergent responses (p<0.001) of [¹⁸F]fludarabine between lymphoma and GBM, while [¹⁸F]FDG demonstrated considerable overlap (p=0.04) between the groups. Tumour characterisation with histology (based mainly on Hoechst, CD68 and CD79), as well as with MRI were in overall better agreement with [¹⁸F]fludarabine-PET than [¹⁸F]FDG with regard to tumour selectivity. Conclusion: The potential of [¹⁸F]fludarabine-PET to distinguish CNS lymphoma from GBM is quite evident and will be further investigated

[en] Complete text of publication follows: Aim: Hybrid PET-based imaging systems (PET/CT, PET/MR) are promising diagnostic tools for multi-parametric image analysis and personalized medicine. However, the comparability of PET studies is still a challenge and quantitative readings have been shown to vary significantly between different PET/CT systems. With the introduction of PET/MR, similar variations to those found with PET/CT are likely. The aim of this work is to assess differences in image quality parameters between different PET/MR systems as a first step towards a future harmonization strategy for PET/MR. Materials and Methods:In this ongoing study, we collected results from NEMA image-quality (IQ) measurements acquired as part of the acceptance testing of PET/MR systems available in the HYBRID consortium (H2020/MSCA No.764458). To date, this includes data from 1 Philips Ingenuity TF (ToF), 1 Siemens Biograph mMR and 2 GE SIGNA PET/MR systems (ToF). We report mean±SD (min-max), contrast recovery coefficients (CRC) and the range of the coefficients-of-variation (CV) obtained for all spheres of the NEMA phantom images acquired with the different PET/MR systems. In addition, we collected information about routine quality control (QC) procedures followed in the individual centres to identify the needs regarding standardization of QC protocols. Results:Cross-calibration measurements were performed regularly for all evaluated systems. QC procedures include vendor-specific daily QC. Mean CRCs were 43.9±6.6% (36.2-50.0)%, 57.5±3.8% (53.8-62.9)%, 68.8±3.1% (66.2-73.3)% and 77.3±3.4% (80.7-83.0)% for the 10 mm, 13 mm, 17 mm and 22 mm hot sphere, respectively. CRCs for the cold spheres were 78.9±9.1% (87.1-68.1)% and 80.8±12.7% (92.7- 63.6)% for the 28 mm and 37 mm sphere, respectively. CVs varied between (6.1-6.6)%, (4.7-5.7)%, (3.3-5.1)%, (2.7-4.9)%, (2.2-5.0)%, (2.2-5.1)%, for the 10 mm, 13 mm, 17 mm, 22 mm, 28 mm and 37 mm background regions, respectively. Discussion and Conclusion: In PET/MR, CRC variability of hot spheres in the IQ phantom is in the same range as reported values for a single PET/ CT system. CRCs of cold spheres increased substantially in systems with ToF, consistent with a previous PET/MR study. CRCs variabilities across the evaluated PET/MR systems were lower than those found in a recent PET/CT multi-centre study. Given these initial results, a standardization of PET data from different PET/MR systems appears feasible but additional evaluations are needed to confirm this hypothesis and establishing a base for a PET/MR harmonization concept. Efforts on designing novel QC methods for MPI will be addressed in future works

[en] Complete text of publication follows: Glioblastomas (GBMs) are aggressive brain tumors resistant to chemotherapy and radiotherapy. Tumor cells survival is related to an increased concentration of the antioxidant glutathione (GSH) which inhibits the action of Reactive Oxygen Species (ROS) produced by radiotherapy. Accumulation of GSH is due to a surexpression of cystine transporter X_c^-, cystine being the precursor of GSH. The anti-inflammatory drug sulfasalazine has been reported to be an inhibitor of the system X_c^- and we sought to label this compound with carbon-11. PET imaging studies using this new tracer could contribute to elucidate the role of system X_c^- in the efficacy of radiotherapies for GBMs patients. We examined two methods to label sulfasalazine at the carboxylic acid position. The first strategy was the direct carboxylation with [¹¹C]CO₂ of sulfasalazine organo-magnesium precursor. The other alternative to synthesize [¹¹C]sulfasalazine was the coupling reaction of [¹¹C]salicylic acid with a diazonium salt in basic aqueous conditions. All attempts to obtain [¹¹C]sulfasalazine according to the direct carboxylation strategy failed. To circumvent this approach, the coupling reaction of [¹¹C]salicylic acid with the diazonium salt has been performed in presence of NaOH under different conditions (temperature, time, reagents and precursors concentrations). After optimization, [¹¹]sulfasalazine has been obtained with a conversion rate of 40%(based on HPLC profiles). [¹¹C]sulfasalazine was successfully obtained by a two-step radiosynthesis from [¹¹C]salicylic acid then coupling with a diazonium salt. Preclinical studies are in progress to better understand the role of system X_c^- in GBMs

No abstract available

No abstract available

No abstract available

[en] Complete text of publication follows: Aim: At the human blood-brain barrier (BBB), P-glycoprotein (ABCB1) limits brain distribution of diverse drugs. Currently available PET tracers to study cerebral ABCB1 function ((R)-[¹¹C] verapamil, [¹¹C]N-desmethyl-loperamide) are avid ABCB1 substrates with low brain uptake, which possess inadequate sensitivity to detect moderate changes in ABCB1 function at the BBB. Metoclopramide is a weak ABCB1 substrate with sufficiently high passive permeability to cross the BBB. The aim of this study was to assess the influence of ABCB1 on the neuro-pharmacokinetics of [¹¹C]metoclopramide. Materials and Methods: Ten drug free, healthy volunteers (mean age: 26±4 years) underwent 60 min PET scans with [¹¹C]metoclopramide before and during a 2 hour infusion of the ABCB1 inhibitor cyclosporine A (CsA) at a dose of 2.5 mg/kg body weight/hour. Arterial plasma concentrations of [¹¹C]metoclopramide and its radiolabelled metabolites were determined with radio-HPLC to derive a metabolite-corrected arterial input function. Individual, MRI based whole brain grey matter region of interest was used to extract PET activity values. PET and plasma data were modeled employing a reversible 1-tissue-2-rate constant compartmental model. In addition, the elimination rate constant of radioactivity from the brain (kE) was determined by linear regression analysis. Results: In baseline scans, [¹¹C]metoclopramide showed appreciable brain uptake (total volume of distribution VT: 2.11±0.33 mL/cm³), which was approximately 3-fold higher than that of [¹¹C]verapamil. During CsA infusion, there was a significant decrease in radioactivity elimination from the brain (change in kE: -40±20%, p ≤ 0.0001 paired t test). Moreover, K1 was increased by +9±12% (p=0.0533), k2 was decreased by -15±5% (p≤ 0.0001) and VT was increased by +29±17% (p=0.0002). Forty minutes after radiotracer injection, 53±14% and 51±13% of total radioactivity in plasma consisted of unchanged parent tracer in scan 1 and scan 2, respectively. Discussion: We found a significant increase in brain distribution of [¹¹C]metoclopramide under conditions of partial ABCB1 inhibition, which confirms that [¹¹C] metoclopramide is transported by ABCB1 at the human BBB. The same CsA infusion protocol caused in humans an approximately 65% increase in K1 and VT and no change in k2 of [¹¹C] verapamil (Muzi et al. JNM 2009;50:1267-75). In contrast to currently available avid ABCB1 substrate radiotracers, ABCB1 inhibition led to a decrease in radiotracer elimination from the brain, which suggests that ABCB1 exerts a pharmacokinetically different impact on brain distribution of weak as compared with avid ABCB1 substrates. [¹¹C]Metoclopramide may be better suited to detect disease-induced alterations in ABCB1 function at the BBB than avid ABCB1 substrate

[en] Complete text of publication follows: Purpose: In vitro radiolabeling of cells combined with nuclear medicine imaging provides a potential method for in vivo cell trafficking analysis. Besides, the use of beta(+)-emitting radionuclides is of particular interest given the good intrinsic characteristics offered by PET technology. The labeling procedure still raises concerns regarding the high amount of radioactivity typically used which induces a cell cytotoxicity. Accurate dosimetry is required to allow for a better understanding of the toxicity associated to labeling. This work aimed to develop a realistic multi-cellular dosimetry and apply the model to the analysis of 3 radionuclides used for labeling in PET: ¹⁸F, ⁶⁴Cu and ⁶⁸Ga. Methods: A program using Python was developed to generate cell coordinates from a set of initial parameters i.e. cell size, cell density, bounding volume size. A cubic cell arrangement was assumed and the calculation approach was based on an analytical formula using derived-Monte Carlo S factors. Both the contribution to the dose of radioactivity located within the cells and in the culture medium were considered. The mean absorbed dose to cells was evaluated as a function of the cell density and the intra-to-extra cellular activity repartition. The results were compared with those obtained using conventional dosimetry. Dose calculations were also done under realistic conditions of leukocytes labeling with ¹⁸F -FDG based on labeling parameters used in few reported studies. Results: The cellular-to-conventional dose ratios can vary significantly with cell density, reaching important values at low density and approaching 1 at high density. They were also influenced by the activity repartition between the cells and the medium, the highest values occurring when all activity was incorporated in the cells. A significant underestimation of mean dose to the cell by conventional model was particularly observed for ⁶⁴Cu up to a factor of 15 at low density, due to the emission of short-range auger electrons. ⁶⁸Ga and ¹⁸F presented lower discrepancies between cellular and conventional doses considering the emission of larger-range beta(+). The calculations done for labeling conditions taken from literature showed that a same injected activity per cell can result in significantly different absorbed doses. Conclusions: The development of a realistic multi-cellular dosimetry offered a better understanding of how absorbed dose to cell is affected as a function of key labeling parameters. The establishment of the dose-effect correlation for mesenchymal stem cells is under way through functional tests in vitro after labeling with ¹⁸F-FDG and external irradiation for comparison