| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Medicine, Faculty of Health Sciences, McMaster University, Hamilton, Ontario, Canada; and 2 Department of Internal Medicine and the Mallinckrodt Institute of Radiology, Washington University School of Medicine, St. Louis, Missouri
Correspondence and requests for reprints should be addressed to Myrna Dolovich, P.Eng., McMaster University Health Sciences Centre, Room 1V16, 1200 Main Street, West Hamilton, ON, L8N 3Z5 Canada. E-mail: mdolovic{at}mcmaster.ca
ABSTRACT
This article reviews the potential use of positron emission tomography (PET), alone and in combination with computed tomography, for evaluating the severity of disease in cystic fibrosis. PET scanning using injected 18F-fluorodeoxyglucose provides visual and quantitative information for the rate at which glucose is taken up by the lung, a process that should relate to the presence of inflammation and reflect the extent of the disease. The computed tomography scan gives highly accurate density and anatomic information to locate areas of inflammation seen on the PET scan, increasing the accuracy of the interpretation. Until recently, the scanners have been single systems, often located in separate hospital departments. Combined systems are now commercially available, with major advantages for patients and in the quality of analytical information obtained for interpretation by the physician. The use of 18F-fluorodeoxyglucose uptake and PET scanning has been suggested as a biomarker of progressive pulmonary inflammation in cystic fibrosis. Although promising, the data so far are limited. Further studies will be needed to validate this measurement for this purpose.
Key Words: 18F-fluorodeoxyglucose • computed tomography • cystic fibrosis • inflammation • positron emission tomography
The use of positron emission tomography (PET) to investigate lung physiology as well as biochemical events in the body has provided valuable information about the extent and progression of disease (1–3). PET is used to measure the distribution and pharmacokinetics of drugs radiolabeled with molecular markers (11C, 15O, 13N, 18F) (4), and is increasingly being used in drug development as a means of establishing "proof of concept" (5, 6). The substantial number of publications in the field of molecular imaging and biology in the last several years serve as an indicator that imaging provides an important diagnostic tool to clinical medicine. In nuclear medicine, PET scanning after injected 18F-fluorodeoxyglucose (18FDG) is used routinely to detect and monitor active tumors; research applications specific to investigating inflammation in the lung have successfully demonstrated increased 18FDG uptake in several chronic respiratory diseases in which inflammation plays a major role in the disease process (sarcoidosis, idiopathic pulmonary fibrosis [IPF], chronic obstructive pulmonary disease) (7–12) and cystic fibrosis (CF) (11, 12). Table 1 lists those studies in which 18FDG uptake was shown to be increased in the presence of lung injury, infection (13–17), and three studies in patient populations in which uptake was not increased despite the presence of inflammatory cells in the lung (7, 12, 13). Other clinical research applications using PET imaging of the lung include detection of infection, localization of receptors, measurement of drug dose and distribution, pharmacokinetics, and ventilation–perfusion (3, 18–20). Whether measurements of 18FDG uptake in the CF lung can provide quantification of disease status, reflect the change in status over time, and be used as a measure of efficacy in clinical trials investigating responses to novel therapies is a current topic of discussion. Follow-up PET scans may be done clinically to monitor patients with cancer, but there are currently few research studies in the literature addressing the benefit of longitudinal PET imaging in CF. More serial data need to be obtained before the information from the PET scan can be accepted as a meaningful clinical outcome measurement in this population.
|
Obtaining a computed tomography (CT) image of the lung has provided physicians with accurate anatomic markers to help identify regions of FDG uptake seen on the PET scan (21). This has, in turn, contributed to increased accuracy in the reporting of PET scan findings (22). The preferred scanner for acquiring this information appears to be the integrated PET–CT system rather than information from separate scanners.
Clinical introduction of integrated PET–CT scanners occurred in 2001 (23), and, over the last several years, these hybrid scanners have gained widespread acceptance such that they are now the standard PET imaging tool (24). Approximately 80% of new installations are PET–CT systems. The technical details for these scanners can be found in numerous papers in the literature (25).
The dual-purpose machines allow the CT and PET scans to be done sequentially and with the patient positioned only once on the same scanner table. This feature, plus the ability to obtain both these scans in a shorter period of time, along with improved software algorithms for coregistration of the two image data sets, are major benefits for both patients and physicians interpreting the scans. The increased accuracy in anatomic localization of the functional (18FDG) signal provided via information from the CT scan (21, 26) has reduced the number of reported equivocal findings on PET scans, both in adult and pediatric patients with cancer (27–29), and particularly when 18FDG uptake is low (30), as is common when imaging the thorax. Although there is more precise coregistration of the PET and CT information, a recent study in two groups of patients, comparing registered data sets from thoracic images acquired with a PET–CT versus separate CT and PET scanners, showed no significant difference in alignment of the images. However, this was only the case when the individual PET and CT scans were obtained on the same day and were matched for body positioning and breathing pattern during the two acquisitions (31).
Centers without a PET–CT must either interpret PET scans without CT information, or use CT scan information acquired on a separate device. However, there are inherent problems with visually integrating images from two separate scanners and coregistering two datasets (32). In the absence of CT, the alternative procedure for locating anatomic landmarks is to rely on the lower-resolution transmission scan acquired with each PET scan. With an integrated PET–CT scanner, there is no need for the transmission (radioactive) source. The CT provides the attenuation map for the PET scan in less time and with greater accuracy (33).
Combination PET–CT systems can provide two types of CT scans: a low-resolution scan to obtain the attenuation factors to be applied to the PET image, and a higher-resolution scan for individual airway assessment. The latter would likely be the protocol selected to monitor CF lung status over time, and particularly when done in conjunction with a PET scan. In addition, as the uptake of 18FDG may not be uniform throughout the lung, focal points of greater uptake signaling regional inflammation may be more readily located when linked to regional lung anatomy with the information from the high-resolution CT scan; this despite the difference in spatial resolution between PET (5 mm) and CT (
1 mm). In CF, it is possible that the changes observed on the PET scan would be evident earlier in the course of the disease, and perhaps before spirometric changes or changes in other indices of pulmonary status (34–36).
IMAGING OPTIONS FOR INVESTIGATIONS OF LUNG PHYSIOLOGY
There are many published research studies investigating changes in airway caliber with disease and with the effect of various types of therapeutic interventions. Radiotracers with conventional 2D planar imaging, 3D single photon emission computed tomography (SPECT), PET, and CT, and, more recently, magnetic resonance imaging have been used to obtain qualitative and quantitative information from the lung, addressing a variety of questions. High-resolution CT has proven to be extremely useful in evaluating changes in airway geometry and identifying areas in the CF lung with abnormal pathology. As mentioned previously here, some of these abnormalities seen on CT appear to precede changes in pulmonary function indices measured by conventional clinical tests.
PET has been used to define biological events in a number of diseases and to explore physiologic processes in the lung using both animal models and human subjects (18, 37, 38). PET scanning contributes to diagnostic decisions in a number of disease categories in which the imaging outcomes have been shown to correlate with clinical measurements, including oncology, infection, neurology, and neuropsychiatry. Although there are many research studies using PET to investigate lung physiology (39), only a few of these studies have been undertaken in either adult or pediatric patients with CF. The lack of functional PET studies in CF is in contrast to the number of studies investigating the use of CT as an outcome measure in CF (34, 36, 40, 41). Currently, there are no published studies in CF evaluating lung function using PET–CT as the imaging modality.
METHODS FOR QUANTIFIFYING 18FDG UPTAKE
There does not appear to be universal agreement among investigators as to the best method for deriving the rate of uptake of 18FDG in tissues using data acquired from PET scans. Chen reanalyzed data from a previous study in acute lung injury, comparing four methods for calculating 18FDG uptake (42). The influx constant (Ki), representative of 18FDG uptake, was calculated using a three-compartment model (43), the recommended gold standard approach, and by a graphical analysis first introduced by Patlak (44, 45). Correlation between these two methods was high (r2 = 0.97). Although a correction for the initial volume of distribution of the tracer in blood and tissue is often applied to the Ki calculation (7, 13), this did not prove useful in the acute lung injury data nor in differentiating 18FDG uptake between the four groups of subjects with CF defined in her recent CF study (11). The fourth method is the calculation of the tissue-to-plasma ratio of radioactivity, shown through compartmental analysis by Chen to be highly correlated with uptake (r2 = 0.96). Values are read from the dynamic lung and blood curves at the time that tissue activity is maximal and circulating activity in blood minimal.
The simplest and currently most widely used method for quantifying 18FDG uptake in nonlung applications is the calculation of standardized uptake values (46). This metric is the radioactivity measured per unit volume in a specified region, normalized for injected radioactivity per body mass. Established values are available in the literature for comparison. Although Chen reported a moderate correlation with Ki (r2 = 0.39), the standardized uptake values approach is not without issue (47, 48). A two-compartment model proposed by Laffon and colleagues also involves calculating the ratio of radioactivity in tissue to that in plasma at similar time points to those used in the tissue-to-plasma ratio calculation (49).
PET IMAGING AND 18FDG UPTAKE: A MARKER OF LUNG INFLAMMATION?
Numerous case series document that PET imaging with the tracer, 18FDG, can detect lung inflammation in vivo. In CF, increased 18FDG activity detected in the lung should be a reflection of chronic neutrophilic inflammation. Two studies in CF and one in non-CF disease (bronchiectasis) have been published (11–13). In the CF studies, different approaches to defining the populations studied resulted in different results. Chen and colleagues stratified 20 patients into 3 groups on the basis of their rate of lung function decline over the previous 4 years (11). Only the subgroup of "rapid decliners" had significantly elevated values for 18FDG uptake compared with healthy volunteers. In the study by Labiris and colleagues (12), despite high sputum neutrophil counts in the patients with CF, their 18FDG uptake was on average not significantly different from that of published controls for glucose use in normal subjects (50), but was significantly less than that measured in their positive control subjects—namely, patients with inactive sarcoidosis. However, in a post hoc analysis, the few subjects that fit Chen's definition of a rapid decliner did show an increased uptake of 18FDG (unpublished data).
As most subjects in these two studies had glucose uptake values similar to normal control subjects, measurement of 18FDG uptake may not be sufficiently sensitive to detect airway inflammation in the majority of patients with CF. In the Chen study, 8/20 subjects with CF had uptake values less than 2 SD above their normal control value. In addition, there was considerable overlap in the individual values among the three patient groups (11). When the net rate of 18FDG uptake in the lung, expressed as a rate constant indicating the transfer of radioactivity between blood and lung compartments (Ki, ml blood/ml lung/min) for each of the subjects, was plotted versus FEV1 in liters or as FEV1% predicted, the correlation was moderate (r2 = 0.36) and the ability of the uptake measurement, Ki, to discriminate between the stable and intermediate groups was poor (Figure 1). As the mean (SD) FEV1% predicted for these two groups of subjects with CF was similar, it is possible that using FEV1 to provide the clinical correlate with Ki contributed to the inability to detect significant differences between the stable and intermediate groups. In addition, 8/20 of the subjects with CF were receiving inhaled steroids and 2/20 were receiving oral steroids at the time of the measurement which may have reduced Ki, as has been demonstrated in other patient populations (9). Perhaps by obtaining data from more patients with CF and including measurements from younger patients and/or patients not already receiving antiinflammatory therapy, a stronger relationship between FDG uptake and disease severity may become obvious.
|
These observations raise several interesting questions for future research. Would the projected course of the disease of those patients with CF with normal uptake values be different from those with high values? Would the response to antiinflammatory therapy be similar for all patients with CF with high uptake values? What is the effect of steroid therapy on Ki in this population (51–53)? It is also important to note that, because of restrictions on imaging children for research purposes, subjects in both studies were older than 18 years of age. Therefore, it is currently not known whether uptake would be different in younger patients, or whether a change in uptake to higher values in children might indicate the onset of a rapid decline in pulmonary function.
Clearly, more information needs to be obtained before the 18FDG uptake measurement can be used to guide clinical management in CF or to support its use as an outcome measure in clinical trials evaluating current and new therapies for CF. In addition, standardized protocols need to be developed and validated, both for acquisition and data analysis. A recommendation for all studies being undertaken would be to include healthy, age-matched control subjects for comparison. Clearly, this introduces some ethical concerns in regard to radiation exposure to naive subjects (54).
As mentioned previously here, reproducibility of the measurement needs to be established, as does the threshold for change indicating a decline or improvement in the patient's status, with or without therapeutic intervention. The effect of treatment on FDG uptake has not been widely studied. In the Labiris study, no change in uptake was measured after 28 days of inhaled tobramycin—but neither was there a reduction in sputum neutrophils after treatment. Other therapies may be more successful in decreasing airway inflammation, and, thereby, effecting a change in FDG uptake. To be useful as a clinical tool, changes in 18FDG uptake need to show some correlation with clinical outcomes (e.g., pulmonary function, sputum neutrophils [55], number of exacerbations requiring hospitalization, etc.). Using CT in conjunction with PET may increase the sensitivity of the PET measurement by identifying regional differences and providing a parallel score to assess the disease at the same point in time. However, to date, there are no published studies of the lung in patients with CF using the integrated scanners.
MONITORING LUNG VENTILATION AND AIRWAY CALIBER USING INHALED 18FDG
Deposition of aerosol in the lung is dependent on aerosol particle size, ventilatory parameters, and airway geometry. Extra-fine aerosols ( 1 µm volume median diameter) are routinely used to assess lung ventilation clinically, whereas coarse aerosols (< 6 µm) can reflect the distribution of a therapeutic aerosol in the lung. In patients with airway disease, deposition is nonuniform compared with that in the healthy lung, with aerosol concentrated on central airways rather than peripheral airways (56). We have used nebulized 18FDG and PET imaging to map differences in regional lung deposition and distribution as a function of aerosol droplet size, changes in airway caliber, and ventilation due to treatment or after methacholine challenge testing (57). These investigations address a specific research question, and usually involve small numbers of subjects. With the exception of diagnosing pulmonary emboli using a submicronic aerosol, the measurement of lung deposition and distribution of inhaled tracer aerosols is not currently established as a clinical diagnostic test.
In our studies, PET imaging of the lung after inhalation of extra-fine and coarse aerosols of 18FDG in patients with CF showed a marked nonuniform pattern compared with that in healthy volunteers (Figure 2). To obtain regional data, each lung was partitioned into 10 3D shells concentric about the hilus (58). Less activity was distributed to peripheral shells in CF, and greater heterogeneity was seen in the distribution of 18FDG, more so for the coarse than for the extra-fine aerosol (58). In the subjects with CF, unlike the healthy volunteers, no difference was seen in the regional distribution between the two aerosols, likely reflecting the major influence of airway caliber changes on deposition. There are many studies using planar imaging that demonstrate changes in aerosol deposition and clearance with airways disease (59, 60), but few studies correlating aerosol deposition and disease progression. Corrections for age and gender from predictive models (61, 62) or lung imaging studies (63) would need to be applied to these data to isolate the changes due to variations in morphology.
|
The standard PET protocol for acquisition of FDG uptake data usually requires 45–60 minutes of scanning time, with the patient lying still on the narrow scan table. Inevitably, there will be patient movement, movement of internal organs, and respiratory motion during scanning, all resulting in alignment errors and giving rise to fusion errors between CT and PET. As both cardiac and respiratory gating are options available with PET but not with CT, artifacts due to motion can be minimized by asking patients to breath-hold at FRC during the 10–15 seconds required to obtain the CT scan of the lung. However, maintaining a breath-hold for even 10 seconds could be difficult for those patients with severe lung disease. Misalignment between PET and CT may be adjusted by comparing activity in other organs, such as the kidneys and liver.
There are interobserver differences in scoring and measuring airway dimensions from CT scans, and these have been addressed in a number of recent studies (64, 65). For some component scores, interobserver variability was found to be high. As the methods for calculating 18FDG uptake are operator dependent, the issue of interoperator variation in processing the PET data also needs to be addressed.
RADIATION EXPOSURE AND DOSE
The increased radiation exposure from the CT scan compared with that from a transmission scan is a recognized concern, particularly for pediatric patients. The PET scan results in an exposure of approximately 7 mSv, and this can increase by 1.5–4.7 mSv with the CT scan, depending on the setup parameters (66). Although using CT or PET–CT to track disease progression in CF may only require a scan every other year, or a combined scan every 4 years, nevertheless, the cumulative radiation dose from multiple procedures needs to be considered (67).
CONCLUSIONS
Results of preliminary research using PET and CT in the assessment of CF are encouraging, and do provide the basis for further investigations. But it is obvious that more data in a greater number of patients with CF need to be collected to be able to judge the value of the measurement in tracking disease status over time or predicting responses to existing or innovative therapies. With PET–CT, there is the opportunity to score the CT at the same time as the measurement of 18FDG uptake. This is clearly an advantage, and provides a comprehensive dataset along with other clinical information obtained on the patient.
Where do we go from here to develop further the possible role of PET or PET–CT as imaging procedures for monitoring disease in CF?
FOOTNOTES
Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form November 29, 2006; accepted in final form March 27, 2007)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
