Use of scanner characteristics in iterative image reconstruction for high-resolution positron emission tomography studies of small animals

Reconstruction of input functions from a dynamic PET image with sequential administration of 15O2 and [Formula: see text] for noninvasive and ultra-rapid measurement of CBF, OEF, and CMRO2

CBF, OEF, and CMRO2 images can be quantitatively assessed using PET. Their image calculation requires arterial input functions, which require invasive procedure. The aim of the present study was to develop a non-invasive approach with image-derived input functions (IDIFs) using an image from an ultra-rapid O2 and C15O2 protocol. Our technique consists of using a formula to express the input using tissue curve with rate constants. For multiple tissue curves, the rate constants were estimated so as to minimize the differences of the inputs using the multiple tissue curves. The estimated rates were used to express the inputs and the mean of the estimated inputs was used as an IDIF. The method was tested in human subjects ( n = 24). The estimated IDIFs were well-reproduced against the measured ones. The difference in the calculated CBF, OEF, and CMRO2 values by the two methods was small (<10%) against the invasive method, and the values showed tight correlations ( r = 0.97). The simulation showed errors associated with the assumed parameters were less than ∼10%. Our results demonstrate that IDIFs can be reconstructed from tissue curves, suggesting the possibility of using a non-invasive technique to assess CBF, OEF, and CMRO2.

High-resolution dynamic imaging and quantitative analysis of lung cancer xenografts in nude mice using clinical PET/CT

Considering the general application of dedicated small-animal positron emission tomography/computed tomography is limited, an acceptable alternative in many situations might be clinical PET/CT. To estimate the feasibility of using clinical PET/CT with [F-18]-fluoro-2-deoxy-D-glucose for high-resolution dynamic imaging and quantitative analysis of cancer xenografts in nude mice. Dynamic clinical PET/CT scans were performed on xenografts for 60 min after injection with [F-18]-fluoro-2-deoxy-D-glucose. Scans were reconstructed with or without SharpIR method in two phases. And mice were sacrificed to extracting major organs and tumors, using ex vivo γ-counting as a reference. Strikingly, we observed that the image quality and the correlation between the all quantitive data from clinical PET/CT and the ex vivo counting was better with the SharpIR reconstructions than without. Our data demonstrate that clinical PET/CT scanner with SharpIR reconstruction is a valuable tool for imaging small animals in preclinical cancer research, offering dynamic imaging parameters, good image quality and accurate data quatification.

Tumor quantification in clinical positron emission tomography

Positron emission tomography (PET) is used extensively in clinical oncology for tumor detection, staging and therapy response assessment. Quantitative measurements of tumor uptake, usually in the form of standardized uptake values (SUVs), have enhanced or replaced qualitative interpretation. In this paper we review the current status of tumor quantification methods and their applications to clinical oncology. Factors that impede quantitative assessment and limit its accuracy and reproducibility are summarized, with special emphasis on SUV analysis. We describe current efforts to improve the accuracy of tumor uptake measurements, characterize overall metabolic tumor burden and heterogeneity of tumor uptake, and account for the effects of image noise. We also summarize recent developments in PET instrumentation and image reconstruction and their impact on tumor quantification. Finally, we offer our assessment of the current development needs in PET tumor quantification, including practical techniques for fully quantitative, pharmacokinetic measurements.

P-glycoprotein function at the blood-brain barrier: effects of age and gender

PURPOSE

P-glycoprotein (Pgp) is an efflux transporter involved in transport of several compounds across the blood-brain barrier (BBB). Loss of Pgp function with increasing age may be involved in the development of age-related disorders, but this may differ between males and females. Pgp function can be quantified in vivo using (R)-[(11)C]verapamil and positron emission tomography. The purpose of this study was to assess global and regional effects of both age and gender on BBB Pgp function.

PROCEDURES

Thirty-five healthy men and women in three different age groups were included. Sixty minutes dynamic (R)-[(11)C]verapamil scans with metabolite-corrected arterial plasma input curves were acquired. Grey matter time-activity curves were fitted to a validated constrained two-tissue compartment plasma input model, providing the volume of distribution (V (T)) of (R)-[(11)C]verapamil as outcome measure.

RESULTS

Increased V (T) of (R)-[(11)C]verapamil with aging was found in several large brain regions in men. Young and elderly women showed comparable V (T) values. Young women had higher V (T) compared with young men.

CONCLUSIONS

Decreased BBB Pgp is found with aging; however, effects of age on BBB Pgp function differ between men and women.

Cerebral blood flow with [15O]water PET studies using an image-derived input function and MR-defined carotid centerlines

Full quantitative analysis of brain PET data requires knowledge of the arterial input function into the brain. Such data are normally acquired by arterial sampling with corrections for delay and dispersion to account for the distant sampling site. Several attempts have been made to extract an image-derived input function (IDIF) directly from the internal carotid arteries that supply the brain and are often visible in brain PET images. We have devised a method of delineating the internal carotids in co-registered magnetic resonance (MR) images using the level-set method and applying the segmentations to PET images using a novel centerline approach. Centerlines of the segmented carotids were modeled as cubic splines and re-registered in PET images summed over the early portion of the scan. Using information from the anatomical center of the vessel should minimize partial volume and spillover effects. Centerline time-activity curves were taken as the mean of the values for points along the centerline interpolated from neighboring voxels. A scale factor correction was derived from calculation of cerebral blood flow (CBF) using gold standard arterial blood measurements. We have applied the method to human subject data from multiple injections of [(15)O]water on the HRRT. The method was assessed by calculating the area under the curve (AUC) of the IDIF and the CBF, and comparing these to values computed using the gold standard arterial input curve. The average ratio of IDIF to arterial AUC (apparent recovery coefficient: aRC) across 9 subjects with multiple (n = 69) injections was 0.49 ± 0.09 at 0-30 s post tracer arrival, 0.45 ± 0.09 at 30-60 s, and 0.46 ± 0.09 at 60-90 s. Gray and white matter CBF values were 61.4 ± 11.0 and 15.6 ± 3.0 mL/min/100 g tissue using sampled blood data. Using IDIF centerlines scaled by the average aRC over each subjects' injections, gray and white matter CBF values were 61.3 ± 13.5 and 15.5 ± 3.4 mL/min/100 g tissue. Using global average aRC values, the means were unchanged, and intersubject variability was noticeably reduced. This MR-based centerline method with local re-registration to [(15)O]water PET yields a consistent IDIF over multiple injections in the same subject, thus permitting the absolute quantification of CBF without arterial input function measurements.

Resistance of brain glucose metabolism to thiopental-induced CNS depression in newborn piglets

The transition from mild sedation to deep anaesthesia is marked by the phenomenon of burst suppression (BS). FDG-PET studies show that the cerebral metabolic rate for glucose (CMRglc) declines dramatically with onset of BS in the adult brain. Global CMRglc increases substantially in the post-natal period and achieves its maximum in preadolescence. However, the impact of post-natal brain development on the vulnerability of CMRglc to the onset of BS has not been documented. Therefore, cerebral blood flow and metabolism were measured using a variant of the Kety-Schmidt method, in conjunction with quantitative regional estimation of brain glucose uptake by FDG-PET in groups of neonate and juvenile pigs, under a condition of light sedation or after induction of deep anaesthesia with thiopental. Quantification of simultaneous ECoG recordings was used to establish the correlation between anaesthesia-related changes in brain electrical activity and the observed cerebrometabolic changes. In the condition of light sedation the magnitude of CMRglc was approximately 20% higher in the older pigs, with the greatest developmental increase evident in the cerebral cortex and basal ganglia (P<0.05). Onset of BS was associated with 20-40% declines in CMRglc. Subtraction of the mean parametric maps for CMRglc showed the absolute reductions in CMRglc evoked by thiopental anaesthesia to be two-fold greater in the pre-adolescent pigs than in the neonates (P<0.05). Thus, the lesser suppression of brain energy demand of neonate brain during deep anaesthesia represents a reduced part of thiopental suppressing brain metabolism in neonates.

Comparison of PET metabolic indices for the early assessment of tumour response in metastatic colorectal cancer patients treated by polychemotherapy

PURPOSE

To compare the performance of eight metabolic indices for the early assessment of tumour response in patients with metastatic colorectal cancer (mCRC) treated with chemotherapy.

METHODS

Forty patients with advanced mCRC underwent two FDG PET/CT scans, at baseline and on day 14 after chemotherapy initiation. For each lesion, eight metabolic indices were calculated: four standardized uptake values (SUV) without correction for the partial volume effect (PVE), two SUV with correction for PVE, a metabolic volume (MV) and a total lesion glycolysis (TLG). The relative change in each index between the two scans was calculated for each lesion. Lesions were also classified as responding and nonresponding lesions using the Response Evaluation Criteria In Solid Tumours (RECIST) 1.0 measured by contrast-enhanced CT at baseline and 6-8 weeks after starting therapy. Bland-Altman analyses were performed to compare the various indices. Based on the RECIST classification, ROC analyses were used to determine how accurately the indices predicted lesion response to therapy later seen with RECIST.

RESULTS

RECIST showed 27 responding and 74 nonresponding lesions. Bland-Altman analyses showed that the four SUV indices uncorrected for PVE could not be used interchangeably, nor could the two SUV corrected for PVE. The areas under the ROC curves (AUC) were not significantly different between the SUV indices not corrected for PVE. The mean SUV change in a lesion better predicted lesion response without than with PVE correction. The AUC was significantly higher for SUV uncorrected for PVE than for the MV, but change in MV provided some information regarding the lesion response to therapy (AUC >0.5).

CONCLUSION

In these mCRC patients, all SUV uncorrected for PVE accurately predicted the tumour response on day 14 after starting therapy as assessed 4 to 6 weeks later (i.e. 6 to 8 weeks after therapy initiation) using the RECIST criteria. Neither correcting SUV for PVE nor measuring TLG improved the assessment of tumour response compared to SUV uncorrected for PVE. The change in MV was the least accurate index for predicting tumour response.

The motivations and methodology for high-throughput PET imaging of small animals in cancer research

Over the last decade, small-animal PET imaging has become a vital platform technology in cancer research. With the development of molecularly targeted therapies and drug combinations requiring evaluation of different schedules, the number of animals to be imaged within a PET experiment has increased. This paper describes experimental design requirements to reach statistical significance, based on the expected change in tracer uptake in treated animals as compared to the control group, the number of groups that will be imaged, and the expected intra-animal variability for a given tracer. We also review how high-throughput studies can be performed in dedicated small-animal PET, high-resolution clinical PET systems and planar positron imaging systems by imaging more than one animal simultaneously. Customized beds designed to image more than one animal in large-bore small-animal PET scanners are described. Physics issues related to the presence of several rodents within the field of view (i.e. deterioration of spatial resolution and sensitivity as the radial and the axial offsets increase, respectively, as well as a larger effect of attenuation and the number of scatter events), which can be assessed by using the NEMA NU 4 image quality phantom, are detailed.

A method for model-free partial volume correction in oncological PET

Background

As is well known, limited spatial resolution leads to partial volume effects (PVE) and consequently to limited signal recovery. Determination of the mean activity concentration of a target structure is thus compromised even at target sizes much larger than the reconstructed spatial resolution. This leads to serious size-dependent underestimates of true signal intensity in hot spot imaging. For quantitative PET in general and in the context of therapy assessment in particular it is, therefore, mandatory to perform an adequate partial volume correction (PVC). The goal of our work was to develop and to validate a model-free PVC algorithm for hot spot imaging.

Methods

The algorithm proceeds in two automated steps. Step 1: estimation of the actual object boundary with a threshold based method and determination of the total activity A measured within the enclosed volume V. Step 2: determination of the activity fraction B, which is measured outside the object due to the partial volume effect (spill-out). The PVE corrected mean value is then given by C_mean = (A+B)/V. For validation simulated tumours were used which were derived from real patient data (liver metastases of a colorectal carcinoma and head and neck cancer, respectively). The simulated tumours have characteristics (regarding tumour shape, contrast, noise, etc.) which are very similar to those of the underlying patient data, but the boundaries and tracer accumulation are exactly known. The PVE corrected mean values of 37 simulated tumours were determined and compared with the true mean values.

Results

For the investigated simulated data the proposed approach yields PVE corrected mean values which agree very well with the true values (mean deviation (± s.d.): (−0.8±2.5)%).

Conclusions

The described method enables accurate quantitative partial volume correction in oncological hot spot imaging.

Combined pedunculopontine-subthalamic stimulation in Parkinson disease

OBJECTIVE

To assess the effect of deep brain stimulation (DBS) in the pedunculopontine nucleus (PPN) and caudal zona incerta (cZi)-both separately and in combination-on motor symptoms and regional cerebral blood flow (rCBF) in patients with Parkinson disease (PD).

METHODS

Four patients with bilateral cZi and PPN DBS electrodes were rated with the Unified Parkinson's Disease Rating Scale motor subscale (UPDRS-III) when taking and withdrawn from medication. A block of 16 [(15)O]-H(2)O PET resting measurements of rCBF were performed in 4 different states with patients withdrawn from medication: 1) no stimulation, 2) cZi stimulation alone, 3) PPN stimulation alone, 4) combined PPN/cZi stimulation.

RESULTS

When patients were medicated, combined PPN/cZi stimulation produced a statistically significant improvement in UPDRS-III score compared to cZi stimulation alone. In the "off" medication state, the clinical effect of combined stimulation was not significantly different from that induced by cZi stimulation alone. Concomitant PPN/cZi stimulation had a cumulative effect on levels of rCBF, effectively combining subcortical and cortical changes induced by stimulation of either target in isolation.

CONCLUSIONS

These findings suggest that concomitant low frequency stimulation of PPN and cZi regions induces additive brain activation changes and provides improved control of PD symptoms when medicated.

CLASSIFICATION OF EVIDENCE

This study provides Class IV evidence that concomitant low frequency stimulation of PPN and cZI improves motor symptoms in patients with PD on dopamine replacement. It provides Class III evidence that concomitant low frequency stimulation of PPN and cZi induces additive rCBF changes in motor areas of brain.

Reproducibility of quantitative (R)-[11C]verapamil studies

Background

P-glycoprotein [Pgp] dysfunction may be involved in neurodegenerative diseases, such as Alzheimer's disease, and in drug resistant epilepsy. Positron emission tomography using the Pgp substrate tracer (R)-[¹¹C]verapamil enables in vivo quantification of Pgp function at the human blood-brain barrier. Knowledge of test-retest variability is important for assessing changes over time or after treatment with disease-modifying drugs. The purpose of this study was to assess reproducibility of several tracer kinetic models used for analysis of (R)-[¹¹C]verapamil data.

Methods

Dynamic (R)-[¹¹C]verapamil scans with arterial sampling were performed twice on the same day in 13 healthy controls. Data were reconstructed using both filtered back projection [FBP] and partial volume corrected ordered subset expectation maximization [PVC OSEM]. All data were analysed using single-tissue and two-tissue compartment models. Global and regional test-retest variability was determined for various outcome measures.

Results

Analysis using the Akaike information criterion showed that a constrained two-tissue compartment model provided the best fits to the data. Global test-retest variability of the volume of distribution was comparable for single-tissue (6%) and constrained two-tissue (9%) compartment models. Using a single-tissue compartment model covering the first 10 min of data yielded acceptable global test-retest variability (9%) for the outcome measure K₁. Test-retest variability of binding potential derived from the constrained two-tissue compartment model was less robust, but still acceptable (22%). Test-retest variability was comparable for PVC OSEM and FBP reconstructed data.

Conclusion

The model of choice for analysing (R)-[¹¹C]verapamil data is a constrained two-tissue compartment model.

Partial volume correction strategies for quantitative FDG PET in oncology

Purpose

Quantitative accuracy of positron emission tomography (PET) is affected by partial volume effects resulting in increased underestimation of the standardized uptake value (SUV) with decreasing tumour volume. The purpose of the present study was to assess accuracy and precision of different partial volume correction (PVC) methods.

Methods

Three methods for PVC were evaluated: (1) inclusion of the point spread function (PSF) within the reconstruction, (2) iterative deconvolution of PET images and (3) calculation of spill-in and spill-out factors based on tumour masks. Simulations were based on a mathematical phantom with tumours of different sizes and shapes. Phantom experiments were performed in 2-D mode using the National Electrical Manufacturers Association (NEMA) NU2 image quality phantom containing six differently sized spheres. Clinical studies (2-D mode) included a test-retest study consisting of 10 patients with stage IIIB and IV non-small cell lung cancer and a response monitoring study consisting of 15 female breast cancer patients. In all studies tumour or sphere volumes of interest (VOI) were generated using VOI based on adaptive relative thresholds.

Results

Simulations and experiments provided similar results. All methods were able to accurately recover true SUV within 10% for spheres equal to and larger than 1 ml. Reconstruction-based recovery, however, provided up to twofold better precision than image-based methods. Clinical studies showed that PVC increased SUV by 5–80% depending on tumour size. Test-retest variability slightly worsened from 9.8 ± 6.5 without to 10.8 ± 7.9% with PVC. Finally, PVC resulted in slightly smaller SUV responses, i.e. from −30.5% without to −26.3% with PVC after the first cycle of treatment (p < 0.01).

Conclusion

PVC improves accuracy of SUV without decreasing (clinical) test-retest variability significantly and it has a small, but significant effect on observed tumour responses. Reconstruction-based PVC outperforms image-based methods, but requires dedicated reconstruction software. Image-based methods are good alternatives because of their ease of implementation and their similar performance in clinical studies.

In vivo validation of reconstruction-based resolution recovery for human brain studies

The aim of this study was to validate in vivo the accuracy of a reconstruction-based partial volume correction (PVC), which takes into account the point spread function of the imaging system. The NEMA NU2 Image Quality phantom and five healthy volunteers (using [(11)C]flumazenil) were scanned on both HR+ and high-resolution research tomograph (HRRT) scanners. HR+ data were reconstructed using normalization and attenuation-weighted ordered subsets expectation maximization (NAW-OSEM) and a PVC algorithm (PVC-NAW-OSEM). HRRT data were reconstructed using 3D ordinary Poisson OSEM (OP-OSEM) and a PVC algorithm (PVC-OP-OSEM). For clinical studies, parametric volume of distribution (V(T)) images were generated. For phantom data, good recovery was found for both OP-OSEM (0.84 to 0.97) and PVC-OP-OSEM (0.91 to 0.98) HRRT reconstructions. In addition, for the HR+, good recovery was found for PVC-NAW-OSEM (0.84 to 0.94), corresponding well with OP-OSEM. Finally, for clinical data, good correspondence was found between PVC-NAW-OSEM and OP-OSEM-derived V(T) values (slope: 1.02+/-0.08). This study showed that HR+ image resolution using PVC-NAW-OSEM was comparable to that of the HRRT scanner. As the HRRT has a higher intrinsic resolution, this agreement validates reconstruction-based PVC as a means of improving the spatial resolution of the HR+ scanner and thereby improving the quantitative accuracy of positron emission tomography.

High throughput static and dynamic small animal imaging using clinical PET/CT: potential preclinical applications

PURPOSE

The objective of the study was to evaluate state-of-the-art clinical PET/CT technology in performing static and dynamic imaging of several mice simultaneously.

METHODS

A mouse-sized phantom was imaged mimicking simultaneous imaging of three mice with computation of recovery coefficients (RCs) and spillover ratios (SORs). Fifteen mice harbouring abdominal or subcutaneous tumours were imaged on clinical PET/CT with point spread function (PSF) reconstruction after injection of [18F]fluorodeoxyglucose or [18F]fluorothymidine. Three of these mice were imaged alone and simultaneously at radial positions -5, 0 and 5 cm. The remaining 12 tumour-bearing mice were imaged in groups of 3 to establish the quantitative accuracy of PET data using ex vivo gamma counting as the reference. Finally, a dynamic scan was performed in three mice simultaneously after the injection of (68)Ga-ethylenediaminetetraacetic acid (EDTA).

RESULTS

For typical lesion sizes of 7-8 mm phantom experiments indicated RCs of 0.42 and 0.76 for ordered subsets expectation maximization (OSEM) and PSF reconstruction, respectively. For PSF reconstruction, SOR(air) and SOR(water) were 5.3 and 7.5%, respectively. A strong correlation (r (2) = 0.97, p < 0.0001) between quantitative data obtained in mice imaged alone and simultaneously in a group of three was found following PSF reconstruction. The correlation between ex vivo counting and PET/CT data was better with PSF reconstruction (r (2) = 0.98; slope = 0.89, p < 0.0001) than without (r (2) = 0.96; slope = 0.62, p < 0.001). Valid time-activity curves of the blood pool, kidneys and bladder could be derived from (68)Ga-EDTA dynamic acquisition.

CONCLUSION

New generation clinical PET/CT can be used for simultaneous imaging of multiple small animals in experiments requiring high throughput and where a dedicated small animal PET system is not available.