Single PET Scan Research Articles

Task- Versus Amphetamine-Induced Displacement of High-Affinity D2/3 Receptor Ligands

TO THE EDITOR: We have read with great interest a recent article by Ceccarini et al. ([1][1]). In their study, the authors evaluated the sensitivity of a single PET scan with the high-affinity dopamine D2/3 receptor ligand 18F-fallypride to reward-induced dopamine release. Ceccarini et al. concluded

Verification of a semi-automated MRI-guided technique for non-invasive determination of the arterial input function in 15O-labeled gaseous PET

A semi-automated MR-guided technique has been evaluated for non-invasive estimation of cerebral metabolic rate of oxygen (CMRO2) using the sequential administration of 15O oxygen (O2) and 15O carbon dioxide (CO2) during a single PET scan. Two mathematical models, which assess the arterial input function (AIF) from time-activity curves (TAC) in the internal carotid artery region, were tested, namely one with a simple correction for the recovery coefficient (RC) and another with corrections for RC and spillover from surrounding tissues. RC was determined from MRA and black-blood image. RC was also determined from C15O blood volume images as a reference. RC agreed between MR-based and C15O-PET based methods, suggesting validity of MR-based methods. Area-under-the-curve (AUC) of the early portion of estimated AIF agreed with that of measured AIF in both models. AUC of the delayed phase of estimated AIF was largely overestimated in the first model, but was sufficiently improved by the spillover correction implemented in the second model.

Need for Standardization of 18F-FDG PET/CT for Treatment Response Assessments

Many factors affect standardized uptake values (SUVs) in (18)F-FDG PET/CT. The use of the SUV from a single PET scan in multicenter studies requires the standardization of (18)F-FDG PET/CT procedures. In the context of treatment response assessments (repeated PET scans), many factors may seem to have minor effects on percentage changes in SUVs, provided that imaging procedures are executed in a consistent manner for each subject. However, the use of (18)F-FDG PET/CT in a nonstandardized manner will result in unknown biases and reproducibilities of SUVs and SUV-based response measures. This article provides an overview of the need for standardization in relation to the specific use of SUVs and SUV changes in studies of treatment response assessments.

Simultaneous in vivo measurements of receptor density and affinity using [11C]flumazenil and positron emission tomography: Comparison of full saturation and steady state methods

The binding of PET radiotracer [(11)C]flumazenil to the GABA(A) receptors is described by the receptor density (B(max)) and binding affinity (K(D)). The estimation of B(max) and K(D) is usually based on Scatchard analysis including at least two PET scans at steady state of various specific activities. Recently, a novel full saturation method to estimate both B(max) and K(D) was proposed, in which a saturating dose of flumazenil is given to cover a wide range of different receptor occupancies within a single scan. The aim of the present study was a direct comparison of steady state and full saturation methods for determining B(max) and K(D) of [(11)C]flumazenil in the same group of male Sprague-Dawley rats. Fourteen rats underwent 3 consecutive [(11)C]flumazenil scans of 30 min duration each. A tracer dose was injected at the start of the first scan. Prior to the second scan the tracer was mixed with 5, 20, 100 or 500 μg unlabelled (cold) flumazenil to cover a wide range of receptor occupancies during the scan. The third scan was performed during a constant intravenous infusion of unlabelled flumazenil, resulting in ~50% GABA(A) receptor occupancy. The first and third scans were part of the steady state method, whilst the second scan was performed according to the full saturation method. For both methods, B(max) and K(D) were then derived by compartmental modelling. Both methods yielded similar B(max) and K(D) estimates. The full saturation method yielded B(max) values of 37 ± 5.8 ng · mL(-1) and K(D) values of 7.6 ± 2.0 ng · mL(-1), whilst the steady state method yielded B(max) values of 33 ± 5.4 ng · mL(-1) and K(D) values of 7.1 ± 0.8 ng · mL(-1). The main advantage of the full saturation method is that B(max) and K(D) can be obtained from a single PET scan.

Role of PET/CT in the detection of liver metastases from colorectal cancer

The aim of this study was to compare the diagnostic accuracy of 2-[fluorine-18] fluoro-2-deoxy-D-glucose positron emission tomography (18F-FDG-PET) and computed tomography (CT) with PET/CT in the detection of liver metastases during tumour staging in patients suffering from colorectal carcinoma for the purposes of correct surgical planning and follow-up. A total of 467 patients underwent a PET/CT scan using an iodinated contrast medium. We compared images obtained by the single PET scan, the single CT scan and by the fusion of the two procedures (PET/CT). The final diagnosis was obtained by histological examination and/or by the follow-up of all patients, including those who did not undergo surgery or biopsy. The PET scan had 94.05% sensitivity, 91.60% specificity and 93.36% accuracy; the CT scan had 91.07% sensitivity, 95.42% specificity and 92.29% accuracy. The combined procedures (PET/CT) had the following values: sensitivity 97.92%, specificity 97.71% and accuracy 97.86%. This study indicates that PET/CT is very useful in staging and restaging patients suffering from colorectal cancer. It was particularly useful when recurrences could not be visualised either clinically or by imaging despite increasing tumour markers, as it guaranteed an earlier diagnosis. PET/CT not only provides high diagnostic performance in terms of sensitivity and specificity, enabling modification of patient treatment, but it is also a unique, high-profile procedure that can produce cost savings.

Optimized 124I PET Dosimetry Protocol for Radioiodine Therapy of Differentiated Thyroid Cancer

Iodine kinetics and lesion dose per administered 131I activity (LDpA) of differentiated thyroid cancer metastases were determined using 124I PET. These data were analyzed to derive an optimized dosimetry protocol. We evaluated the time-activity-concentration curves of 37 lesions in 17 patients who had undergone thyroidectomies. LDpA determination involved 124I PET images acquired at 4, 24, 48, 72, and 96 h after intake of a capsule containing 20-40 MBq of 124I. A combination of a linear and a monoexponential or a monoexponential function only parameterized the time-activity-concentration curves. The LDpAs, calculated using data from all 5 PET time points, served as reference. The lesions were classified into 3 groups, according to potential for cure with 131I therapy: low (< or =5 Gy GBq(-1); n = 14), medium (between 5 and 10 Gy GBq(-1); n = 9), or high LDpAs (>10 Gy GBq(-1); n = 14). Using the reference approach, the differences in the empiric kinetic parameters within the LDpA groups were evaluated. The reference LDpAs were compared with those derived from only 2, 3, or 4 PET data points and from 1 adapted 2-point approach. Lin's concordance correlation coefficient (rho c) and the mean absolute percentage deviation in LDpAs were used to assess agreement between simplified and reference approaches. The effective 124I half-life, linear activity-concentration rate (alpha), and 24-h activity concentration (CpA) (the latter 2 per administered 124I activity) differed significantly among the LDpA groups (P < 0.05). LDpAs correlated with 24-h CpAs (r = 0.94, P < 0.001). Using the 4-, 24-, and 96-h measurements, a rho c value of greater than or equal to 0.90 was found, and the mean absolute percentage deviation was less than or equal to 16%. Similar statistical values were obtained for the adapted approach, which was based on 24- and 96-h PET data points only. Lesion classification into LDpA groups was feasible using a single PET scan at approximately 24 h. Because of the highly variable kinetics, 1 additional measurement at approximately 96 h was needed to obtain a sufficiently reliable LDpA estimate. The adapted 24-96-h approach appears to be the optimal 124I protocol and is a reliable simplification of the 5-point protocol.

Evaluation of rapid dual-tracer 62Cu-PTSM + 62Cu-ATSM PET in dogs with spontaneously occurring tumors

We are developing methods for imaging multiple PET tracers in a single scan with staggered injections, where imaging measures for each tracer are separated and recovered using differences in tracer kinetics and radioactive decay. In this work, signal separation performance for rapid dual-tracer 62Cu-PTSM (blood flow) + 62Cu-ATSM (hypoxia) tumor imaging was evaluated in a large animal model. Four dogs with pre-existing tumors received a series of dynamic PET scans with 62Cu-PTSM and 62Cu-ATSM, permitting evaluation of a rapid dual-tracer protocol designed by previous simulation work. Several imaging measures were computed from the dual-tracer data and compared with those from separate, single-tracer imaging. Static imaging measures (e.g. SUV) for each tracer were accurately recovered from dual-tracer data. The wash-in (k1) and wash-out (k2) rate parameters for both tracers were likewise well recovered (r = 0.87–0.99), but k3 was not accurately recovered for PTSM (r = 0.19) and moderately well recovered for ATSM (r = 0.70). Some degree of bias was noted, however, which may potentially be overcome through further refinement of the signal separation algorithms. This work demonstrates that complementary information regarding tumor blood flow and hypoxia can be acquired by a single dual-tracer PET scan, and also that the signal separation procedure works effectively for real physiologic data with realistic levels of kinetic model mismatch. Rapid multi-tracer PET has the potential to improve tumor assessment for image-guide therapy and monitoring, and further investigation with these and other tracers is warranted.

Role of (18) F-FDG PET scanning in detecting osseous metastases in non-small cell lung cancer (NSCLC) in a community setting: A retrospective data analysis

17067 Background: Metastasis from a distant site is the most common neoplastic process involving bone. Accurate staging of NSCLC is critical for effective management. Bone scans have been traditionally used for diagnosing bone metastases in NSCLC. Recently, PET scanning has been shown to be effective in detecting bone metastases in several studies. In NSCLC, using a single PET scan to evaluate both the primary tumor and bone metastases might be more cost-effective and if the accuracy of PET scanning for detection of bone metastases can be established, the need for bone scanning may be obviated. We performed a retrospective comparison between PET scan and bone scan in NSCLC patients diagnosed and treated in a community hospital setting. Methods: 50 patients who were diagnosed with NSCLC between January 1, 2002 and December 31, 2004 and who had both PET scan and Bone scan done within 3 months of each other were included in the study. 47 patients had both investigations within 40 days duration. Data regarding all the lesions found on bone and PET scan were collected. Results: The median age of the patients was 70.5 years. Total of 20 patients were diagnosed with bone metastases by one or the other test. Of these 15 were identified with bone scan with a sensitivity of 75% (95% Confidence interval CI 56.2%–97.1%). 17 patients were identified by PET scan with a sensitivity of 85% (95% CI 45.5%–92%), p-value -0.48 compared to bone scan. 3 patients who had positive bone scan had a negative PET scan. 5 patients who had positive PET scan had a negative bone scan. Conclusions: PET scan appears to be as good as bone scan in detecting bone metastases. Unless very high clinical suspicion exists regarding bone metastases, negative PET-Scan for bone metastases should obviate the need for bone scan. [Table: see text] No significant financial relationships to disclose.

Generation of the Acquisition-Specific NEC (AS-NEC) Curves to Optimize the Injected Dose in 3D$^18$F-FDG Whole Body PET Studies

Aim of this work was the implementation and validation, for the Discovery-ST PET/CT (GE Medical Systems) system, of the acquisition-specific noise equivalent counts (AS-NEC) method to establish the amount of tracer to be injected in 3D $^18$ F-FDG whole body (WB) PET studies to achieve the peak of the NEC (NEC-p) at the acquisition time. The AS-NEC method uses prompts, delayed events and detector dead-time of a single reference PET scan to calculate the full shape of the NEC curve. The method was implemented using a 3D decay series of the 70 cm NEMA 2001 (line source in a 20 cm diameter solid polyethylene cylinder) phantom and validated with the cylindrical NEMA 1994 (diameter, 20 cm; length, 20 cm) and NEMA 2001 IEC body phantoms. The NEC curves generated by the single frames of the phantom series, using the AS-NEC method, well correlated with the experimental NEC curves proving the validity of the method and the possible application to clinical studies. The AS-NEC model was then retrospectively applied on 40 3D $^18$ F-FDG WB studies in a range of body mass index (BMI) between 16 and 30 (kg/m $^2$ ) (6 under-weight (uw), 18 normal-weight (nw), 16 over-weight (ow)). For each acquisition frame of each patient study, the activity at the acquisition time, corresponding to the NEC-p was identified on the NEC curves. Furthermore, as the NEC curves show a region around the NEC-p with small variations (nearly a plateau), the values of radioactivity corresponding to a reduction of 1%, 3% and 5% with respect to NEC-p were also calculated to assess a possible reduction of the doses to be injected in clinical studies. The results show that the average activities at the acquisition time corresponding to the NEC-p were comparable for the three BMI classes: 336.7 MBq $( sd=22.2), 329.3$ MBq $( sd=33.3), 344.1$ MBq $( sd=48.1)$ for uw, nw and ow, respectively. Therefore, the total average NEC-p activity for the three BMI classes was 336.7 MBq $( sd=40.7)$ . The mean values of the radioactivity at a reduction of 1%, 3% and 5% with respect to the NEC-p were: 284.9 MBq $( sd=40.7), 247.9$ MBq $( sd=33.3)$ and 225.7 MBq $( sd=29.6)$ respectively. These results indicate the possibility to use, for the Discovery-ST, a single injection protocol of 448 MBq (for the range of BMI here considered) to have an activity at the acquisition time (after 45 min of uptake) of 336.7 MBq (NEC-p). Nevertheless, the plateau near the NEC-p suggests the possibility to reduce significantly the dose to be injected in clinical studies down to about 330 MBq, while preserving suitable NEC performance ( $-3$ %) with respect to the NEC-p. This result was supported by image quality assessment performed on reconstructed images of the NEMA 2001 IEC body phantom.

A rapid CBF/CMRO2 measurement with a single PET scan with dual-tracer/integration technique in human

CBF/CMRO2 may be quantified using PET with 15O-tracers, but the conventional three-step technique (3S) requires a long study period, attributed to need for separate acquisition for three radioactive tracers for CBF (H215O or C15O2), CMRO2 (15O2), and CBV (C15O). Simultaneous determination of CBF, CMRO2 and CBV from a single dynamic 15O2 scan is known to suffer from statistical uncertainty. We recently developed a rapid technique that allows sequential administration of two of three tracers (for CBF and CMRO2) in a single PET scan, and generates quantitative CBF/CMRO2 images with a dual-tracer integration method (DTI). This study asked whether this technique is feasible in normal human subjects and how accurate quantitative values it provides in comparison with the 3S and non-linear fitting method (NLM). Eight normal subjects (24.2+-2.6 y.o) participated in this study. The protocol of the study was approved by the local ethical committee. All subjects received three emission scans, i.e., a 4-min scan after short bolus administration of C15O, a 9.5-min scan with dual administration of C15O2 followed by 15O2, and a 9-min scan with 15O2 followed by H215O. With correction for blood volume using the first C15O scan data, CBF/CMRO2 images were generated by 3S using the first part of the second (CO2) and the third scan (O2), whereas those by DTI method were calculated using both part of PET data in each of the second or the third scan. The calculation used a mathematical formula derived from a single-tissue compartment model including a correction for blood volume and an assumed partition coefficient for water. Regional quantitative values of CBF/CMRO2 in the DTI was compared with those by 3S and NLM. We also compared by voxel-by-voxel CBF or CMRO2 images between the second and the third DTI scan to test the regional difference. Global CBF or CMRO2 values for the DTI did not differ from those by the 3S, and between the second and third DTI scans. The cortical regional values of the DTI neither differed significantly from those of 3S and NLM. In contrast, CMRO2 values of the white matter region in both 3S and DTI differed significantly from NLM, as was predicted from a simulation study for the error in the assumed partition coefficient. Voxel-based comparison of CBF images between the second and the third DTI scan showed significant difference at brain regions around nasal cavity, and at the somatosensory areas that may come from the difference in methodology (inhalation or venous-injection of CBF tracer) and physiology (sensation of intravenous injection), respectively. These were also seen when compared CMRO2 images, but at a lower height threshold. These results suggest that this DTI technique generates CBF/CMRO2 images with a considerably shortened scan period, which are comparable to those by 3S, in terms of quantitative accuracy and image quality. The technique may be of use for clinical studies in patients particularly with acute stroke, and also for investigation of metabolic coupling during neuronal activation or pharmacological challenges in human.

Is oxygen insufficiency compensated during acute ischemia? - A PET study in an ischemic model of non-human primates

Insufficient delivery of oxygen to tissue is the central core of ischemia. A large number of observations in rodents revealed that compensatory responses against ischemia occurs at cellular or biochemical level, however, there is a lack of in vivo evidences of biological responses to fulfil oxygen demand. Among them, vascular response to maintain the delivery (e.g. vasodilatation) and transport efficiency of metabolic substrates from blood to tissue assumed to be fundamental (Hayashi et al. 2003). Here, we show using PET whether there is response in vasculature or in the diffusion efficiency of oxygen in an acute stroke model in non-human primates. We used four adult male macaque monkeys (macaca fascicularis, body weight = 5 kg). The study was approved by the local committee of animal experiments. Experiments were performed under general anesthesia using propofol. A single PET scan with dual tracer/integration (DTI) method for CBF/CMRO2 and a PET scan for CBV were repeatedly performed before and 360 min after the artificial embolization of intracerebral artery. Embolization was produced by injecting autologous blood clot through a thin catheter placed at middle cerebral artery or internal carotid artery. Ischemic of interest (ROI) were defined by searching voxels with the highest values in the oxygen extraction fraction (OEF) images at the time point of 5 min or 180 min after embolization. CBF, CMRO2 and CBV values were also obtained by placing the same ROI to the corresponding images. CBV values were used for the assessment of vascular response, whereas effective oxygen diffusibility (EOD) that was calculated from measured CBF, OEF and arterial haemoglobin content (Hb) and oxygen saturation fraction (Sa) was used as a marker of efficiency of oxygen transport from blood to brain tissues (Hayashi et al. 2003). The values were normalized by ROI values in cerebellum for each image and by the values at the time point of pre-embolization for each subject to minimize measurement- and subject-specific variability. The mirror contralateral to the ROI were used for control. Ischemic (n = 7) showed decrease in CBF (-30%, -32% at 5 min and 180 min, respectively) and CMRO2 (-18%, -10%) and increase in OEF (+18%, +26%) compared to the control. CBV values showed no change (+1%, +2%) whereas EOD decreased (-12%, -11%). When the regions were divided in two categories by the final tissue outcome (non-infarcted or infarcted), significant decreases in CBF, CMRO2 and EOD were found in the of final infracted, but no difference in CBV were seen between the two categories. Similar results were also obtained in the other time points (30, 60, 120, 210 and 360 min post embolization). These results reveal that in acute ischemic there is little response in vasculature and that change in diffusion efficiency of oxygen does not act as a compensatory response rather passively depends on the metabolic demand, although oxygen extraction fraction is increased. The findings indicate that brain tolerance for oxygen insufficiency is not so large that oxygen metabolism during ischemia correlates final tissue outcome.

Rapid CBF/CMRO2 measurement in a single PET scan with dual tracer administration

Rapid CBF/CMRO2 measurement in a single PET scan with dual tracer administration

Feasibility of deriving CBF, OEF and CMRO2 from a single dynamic PET scan using a short inhalation of oxygen-15 labelled molecular oxygen

Feasibility of deriving CBF, OEF and CMRO₂ from a single dynamic PET scan using a short inhalation of oxygen-15 labelled molecular oxygen

Event-Related Potentials Correlate with Task-Dependent Glucose Metabolism

Cognitive processing is associated with brain electrical activity that is reflected in event-related potentials (ERP). ERP during a target detection task, and regional cerebral glucose metabolism (CMRglc) measured simultaneously, may be influenced by the same neurophysiologic processes. We tested the hypothesis that ERP factors could be directly correlated with CMRglc to derive functional brain maps of brain activity at 120, 160, 200, 280, and 400 ms following stimulus presentation in a target detection task. We controlled for the potential confounding effects of age, sex, and task accuracy, and correlate target-related and nontarget-related ERP separately. Increases and decreases in CMRglc at each time point were identified with statistical parametric mapping (P < 0.001, uncorrected). The 120- and 160-ms maps were the same for target and nontarget processing, while maps for 280 and 400 ms clearly distinguished between targets and nontargets. Extrinsic (early) cognitive processes that depend mainly on stimulus characteristics show less variation based on stimulus meaning (i.e., letter vs shape; target vs nontarget) than later (intrinsic) cognitive processes. These early effects are lateralized to the left hemisphere, for negative ERP factors, and positive ERP–PET correlations are more likely than negative ERP-PET correlations. Thus, brain areas related to task processing impact both ERP and CMRglc measures, suggesting a shared neurophysiologic mechanism for negative ERP factors and increased CMRglc. Direct statistical analysis of these two measures using statistical parametric mapping provides high spatial and temporal resolution in multisubject experiments, while requiring only a single 18FDG PET scan per subject.

Encoding activity with PET

With the emergence of single-trial and event-related functional magnetic resonance imaging as a viable and accessible imaging tool for many cognitive neuroscientists, the future of PET blood flow activation studies seems uncertain. However, a number of older PET techniques continue to yield important findings that might tell us as much about the relationship between brain and behaviour as state-of-the-art methods such as event-related fMRI. In one recent paper the role of the hippocampus and the amygdala in long-term memory was examined using PET of regional cerebral glucose1. Eight normal subjects listened to lists of words during the 32 minutes following injection of 18flurodeoxyglucose, a glucose analog tracer used to determine regional glucose metabolic rate. Glucose activity was measured during a single PET scan and the results correlated with performance on an incidental free-recall test given 24 h later. A highly significant relationship was found between activity in the left hippocampal region and word recall, but no correlation was found between activity in the amygdala and recall. This result concurs fully with recent studies using a similar behavioural approach, and event-related fMRI. One important difference in this study, however, was that recall was correlated with activity measured at the encoding stage, rather than at retrieval. PET techniques are thus still providing new information that complements findings from fMRI approaches.