ROC study and SUV threshold using quantitative multi-modal SPECT for bone imaging.

BACKGROUND: We investigated the clinical performance of a quantitative multi-modal SPECT/CT reconstruction platform for yielding radioactivity concentrations of bone imaging with 99mTc-methylene diphosphonate (MDP) or 99mTc-dicarboxypropane diphosphonate (DPD). The novel reconstruction incorporates CT-derived tissue information while preserving the delineation of tissue boundaries. We assessed image-based reader concordance and confidence, and determined lesion classification and SUV thresholds from ROC analysis. METHODS: Seventy-two cancer patients were scanned at three US and two German clinical sites, each contributing two experienced board-certified nuclear medicine physicians as readers. We compared four variants of the reconstructed data resulting from the Flash3D (F3D) and the xSPECT Bone™ (xB) iterative reconstruction methods and presented images to the readers with and without a fused CT, resulting in four combinations. We used an all-or-none approach for inclusion, compiling results only when a reader completed all reads in a subset. After the final read, we conducted a "surrogate truth" reading, presenting all data to each reader. For any remaining discordant lesions, we conducted a consensus read. We next undertook ROC analysis to determine SUV thresholds for differentiating benign and lesional uptake. RESULTS: On a five-point rating scale of image quality, xB was deemed better by almost two points in resolution and one point better in overall acceptance compared to F3D. The absolute agreement of the rendered decision between the nine readers was significantly higher with CT information either inside the reconstruction (xB, xBCT) or simply through image fusion (F3DCT): 0.70 (xBCT), 0.67 (F3DCT), 0.64 (xB), and 0.46 (F3D). The confidence level to characterize the lesion was significantly higher (3.03x w/o CT, 1.32x w/CT) for xB than for F3D. There was high correlation between xB and F3D scores for lesion detection and classification, but lesion detection confidence was 41% higher w/o CT, and 21% higher w/CT for xB compared to F3D. Without CT, xB had 6.6% higher sensitivity, 7.1% higher specificity, and 6.9% greater AUC compared to F3D, and similarly with CT-fusion. The overall SUV-criterion (SUVc) of xB (12) exceeded that for xSPECT Quant™ (xQ; 9), an approach not using the tissue delineation of xB. SUV critical numbers depended on lesion volume and location. For non-joint lesions > 6 ml, the AUC for xQ and xB was 94%, with SUVc > 9.28 (xQ) or > 9.68 (xB); for non-joint lesions ≤ 6 ml, AUCs were 81% (xQ) and 88% (xB), and SUVc > 8.2 (xQ) or > 9.1 (xB). For joint lesions, the AUC was 80% (xQ) and 83% (xB), with SUVc > 8.61 (xQ) or > 13.4 (xB). CONCLUSION: The incorporation of high-resolution CT-based tissue delineation in SPECT reconstruction (xSPECT Bone) provides better resolution and detects smaller lesions (6 ml), and the CT component facilitates lesion characterization. Our approach increases confidence, concordance, and accuracy for readers with a wide range of experience. The xB method retained high reading accuracy, despite the unfamiliar image presentation, having greatest impact for smaller lesions, and better localization of foci relative to bone anatomy. The quantitative assessment yielded an SUV-threshold for sensitively distinguishing benign and malignant lesions. Ongoing efforts shall establish clinically usable protocols and SUV thresholds for decision-making based on quantitative SPECT.

Benzodiazepine-GABAA receptors in idiopathic generalized epilepsy measured with [11C]flumazenil and positron emission tomography.

The neurochemical basis of absence seizures and the mechanism of their suppression by valproate (VPA) are uncertain. We used positron emission tomography (PET) to determine whether an abnormality of [11C]flumazenil binding to benzodiazepine (BZD)-GABAA receptors exists in patients with childhood and juvenile absence epilepsy and to examine the effects of VPA on [11C]flumazenil binding. The regional cerebral volume of distribution (Vd) of [11C]flumazenil in patients not treated with VPA was not different from that in normal controls; Vd was lower in patients treated with VPA, and the number of receptors available for binding was significantly reduced in such patients as compared with normal controls. There was no evidence of a primary abnormality of the BZD-GABAA receptor in childhood and juvenile absence epilepsy (CAE/JAE), but the data suggest that treatment with VPA is associated with a reduction in [11C]flumazenil binding that may be relevant to its mode of action in CAE/JAE.

Benzodiazepine receptor quantification in vivo in humans using [11C]flumazenil and PET: application of the steady-state principle.

Carbon-11-labeled flumazenil combined with positron emission tomography (PET) was used to measure the concentration (Bmax) of the benzodiazepine (Bz) receptor in the brain and its equilibrium dissociation constant (KD) for flumazenil in five normal subjects. The steady-state approach was used injecting the tracer as a bolus of high specific activity. In each subject two studies were carried out. The first study was performed at essentially zero receptor occupancy, the tracer alone study. The second study was performed at a steady-state receptor occupancy of about 50%, achieved by a prolonged constant infusion of nonlabeled ("cold") flumazenil starting 2h before the bolus tracer injection and continuing until the end of scanning period. In this second study the free concentration of unmetabolized flumazenil in plasma water was measured in multiple blood samples. The observed tissue and plasma tracer curves, calibrated in the same units of radioactivity per millimeter, were analyzed in two ways: (a) by the noncompartmental (stochastic) approach making no assumptions regarding number of compartments in the tissue, and (b) by the single-compartment approach assuming rapid exchange (mixing) of tracer between all tissue compartments. The noncompartmental and the compartmental analyses gave essentially the same values for the distribution volume of the tracer, the parameter used for quantitation of the Bz receptor. As the compartmental approach could be applied to a shorter observation period (60 min instead of 120 min) it was preferred. The five subjects had a mean KD value of 12 nM/L of water and Bmax values of the grey matter ranging from 39 +/- 11 in thalamus to 120 +/- 14 nM/L of brain in occipital cortex. Most previous studies have been based on the pseudoequilibrium approach using the brain stem as a receptor-free reference region. This yields practically the same KD but lower Bmax values than the steady-state approach presented here.

Quantification of opiate receptors in two patients with mesiobasal temporal lobe epilepsy, before and after selective amygdalohippocampectomy, using positron emission tomography.

Using PET, reduced glucose metabolism (rCMRglu) and increased binding of the mu opiate receptor ligand 11C-carfentanil have been demonstrated in lateral temporal cortex overlying mesial temporal epileptic foci. Binding of the non-specific opiate receptor ligand 11C-diprenorphine (DPN) to lateral temporal cortex has not shown consistent asymmetries. We measured rCMRglu with 18F-FDG and binding of 11C-diprenorphine in two patients with temporal lobe epilepsy (TLE) before and 5 months after selective amygdalo-hippocampectomy (both patients were seizure-free post-operatively). Pre-operatively, in both patients rCMRglu was decreased in mesial temporal lobe (MTL) (asymmetry index AI = -9.1 and 9.0) ipsilateral to the EEG focus. A more marked reduction was seen in ipsilateral lateral temporal cortex (LTC) (AI = -32.0 and 18.9). DPN binding was reduced in MTL and LTC (AI MTL = -9.3 and 16.2; AI LTC = -8.0 and 5.5) ipsilateral to the focus, but was within 2 SD of the normal range. Post-operatively, the reduction of rCMRglu in LTC was accentuated in one patient and decreased in the other (AI = -23.1 and 45.7) while there was a further reduction of DPN binding in LTC in both patients (AI = -27.8 and 9.8). These preliminary results in only two patients are compatible with downregulation of opiate receptors in LTC after removal of the epileptic focus or post-operative neuronal dysfunction.

Investigation of the opioid system in absence seizures with positron emission tomography.

The neuroanatomical and pathophysiological basis of primary generalised absences is uncertain. Administration of endogenous opioids has been shown to result in absence-like seizures in animal models. Positron emission tomography scans were performed in eight patients with primary generalised epilepsy and eight control subjects. Regional cerebral blood flow was measured interictally with C15O2, after which a 90 minute dynamic study with the opioid-receptor ligand 11C-diprenorphine was performed. Serial absences were precipitated by hyperventilation for 10 minutes, starting 30-40 minutes after injection of diprenorphine. Absences, with generalised spike-wave discharges on the EEG, occurred for between 10% and 51% of the provocation period. No individual (normal or patient) had any interictal focal abnormalities of cerebral blood flow. After provocation of serial absence seizures, there was increased diprenorphine elimination from the association cortex, but not from the thalamus, basal ganglia, or cerebellum, compared with control subjects and patients scanned without provocation of absences. It was possible to simulate the observed increased diprenorphine elimination following seizures in cerebral cortex using a two tissue compartment model, with an estimated 15-41% decrease in the specific tracer uptake rate constant (k3). These results suggest that endogenous opioids are released in the association cortex at the time of serial absences, lead to increased receptor occupancy, and may have an important role in the pathophysiology of generalised absences.