British Journal of Radiology (2005) Supplement_28, 33-40
© 2005 British Institute of Radiology
doi: 10.1259/bjr/14526714
British Journal of Radiology Supplement_28 (2005),33-40 ©2005 The British Institute of Radiology
Use of combined PET/CT images for radiotherapy planning: initial experiences in lung cancer
P H Jarritt, PhD, FIPEM1,
A R Hounsell, PhD2,
K J Carson, PhD1,
D Visvikis, PhD4,
V P Cosgrove, PhD2,
J C Clarke, MBBS5,
D P Stewart, FRCR3,
V A L Fleming3,
R L Eakin, FRCR3 and
A Zatari, PhD1
1 Northern Ireland Regional Medical Physics Agency, Musgrave and Clark House, Grosvenor Road, Belfast BT12 6BA, UK, 2 Northern Ireland Regional Medical Physics Agency and 3 Clinical Oncology, Belvoir Park Hospital, Belfast BT8 8JR, UK, 4 U650 INSERM, LaTIM, Brest, France and 5 Department of Radiology, Royal Group of Hospitals, Grosvenor Road, Belfast BT12 6BA, UK
 |
Abstract
|
|---|
The potential role of positron emission tomography (PET) in radiotherapy still requires careful evaluation as it becomes increasingly integrated into the radiotherapy planning process. Diagnosis and subsequent radiotherapy planning based solely upon X-ray CT are known to be less sensitive and specific for disease than PET imaging in non-small cell lung cancer. The CT images may not demonstrate the true extent of intrathoracic disease. To overcome this limitation, the direct use of combined PET/CT image data in the treatment planning process has been investigated. A small pilot study of five patients was carried out at the Royal Victoria Hospital, Belfast, following the installation of a GE Discovery LS PET/CT scanner. The initial aims were to investigate the system and to make preliminary clinical evaluations. The key issues that were addressed included: verification of PET/CT alignment, patient position and reproducibility for imaging and treatment; verification of CT numbers on the PET/CT systems for dose calculation; integrity of data transfer; radiation protection of staff; protocols for target volume delineation; and the implications for physiologically-gated PET and CT acquisitions. This paper reviews our practical experience, and technical problems are described.
|
Introduction
|
|---|
Non-small cell lung cancer (NSCLC) is one of the most common causes of cancer-related death in the developed world [1]. Radiotherapy is used only in advanced disease that is apparently localised to the chest. Surgery is the treatment of choice in early disease although radiotherapy can be as effective as surgery [2]. The outcome is poor with conventional radiotherapy; only between 10–20% of patients survive for 2 years. Uncontrolled local disease is the principal cause of death in over 50% of patients and is present in approximately 90% of cases [3]. Planning based on CT alone for radiotherapy treatment has limitations: the true extent of interthoracic disease is not well delineated on CT scans and CT cannot detect tumour in normal-sized lymph nodes or distinguish reactive hyperplasia from tumours. As a result, patients continue to undergo radiotherapy treatment without accurate staging [4, 5]. Inappropriate staging of the patient or inaccurate localisation of the tumour will inevitably mean that dose escalation aimed at improving local control in NSCLC is unlikely to be successful. Functional imaging using 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET) has clearly proven to be useful in the diagnosis, staging and follow-up of NSCLC [2]. The improved sensitivity of FDG-PET has been shown to enable more accurate staging and tumour localisation than previous techniques, leading to more appropriate treatment management. The combination of PET and CT scanning into a single system to provide directly fused PET and CT data brings with it the possibility of the direct use of this combined data in the treatment planning process for NSCLC to improve radiotherapy treatment planning. This development is not without its technical difficulties. In the thorax and abdomen there may be considerable movement of the internal organs and structure in response to respiratory and cardiac motion. The implications of this are basically two-fold. First, the different time frames for PET and CT acquisitions can lead to artefacts in the image appearance and quantification if the wrong attenuation correction is applied at image reconstruction. Second, movement of the radioactive distribution during acquisition will result in a reconstructed image with, at best, a smeared out activity distribution showing poorer contrast within a larger volume than truly exists. This will compromise attempts to delineate treatment volumes or to quantify lesion uptake. These issues are now being addressed by the introduction of respiratory gating systems that will permit segmentation of the acquisition across the breathing cycle to provide accurate attenuation correction coefficients and accurate reconstruction of activity volumes and concentrations.
PET remains a lower resolution imaging technique than CT and this compromises the detection and quantification of activity within small volumes, typically less than 4 ml. With this plethora of technological development and limitations it is essential that a careful and systematic introduction of such technology is undertaken to establish methods that can be globally applied and that have demonstrable patient benefits.
An ethically approved pilot study of five patients has been undertaken, the aim of which was to study the feasibility of using an integrated PET/CT scanner directly for radiotherapy treatment planning (RTP). In this study, acquisition, data transfer and utilisation of combined PET/CT images into the RTP process were investigated. Approval and funding have been granted to extend this work into a larger study investigating the implementation and validation of PET/CT images into RTP. In this article we present our initial experiences, key practical findings and future plans.
|
Pilot study
|
|---|
The pilot project was undertaken using a GE Discovery LS PET/CT scanner. Five patients, who had been referred for a diagnostic PET-CT scan and who were intended for radical radiotherapy, were selected for inclusion into the study. Patients were scanned after injection of 18F-FDG 375 MBq followed by a 45-min uptake period. A standard diagnostic imaging protocol was used and no special breathing instructions were given during the CT acquisition. The PET/CT images were used for diagnostic and staging purposes and reported in the usual way by the radiologists. The images were subsequently transferred to a Theraplan Plus 3D treatment planning computer system. The CT scans were used for planning radiotherapy treatment if the patient was deemed suitable for radiotherapy. In the absence of validation data, the PET images were not used in target volume delineation within the RTP process.
|
The PET/CT planning scan
|
|---|
Patient positioning
Fundamental to the use of images for radiotherapy planning is the requirement to scan the patient in the treatment position. The GE Discovery LS PET/CT scanner has a minimum patient bore of 55 cm. This is significantly less than radiotherapy simulators and large-bore CT scanners specifically designed for radiotherapy planning. This restricted bore creates challenges to patient set-up in the treatment position for image data acquisition. However, it should be recognised that use of CT scanners with comparable bore sizes has been common practice in centres where diagnostic CT scanners are used for treatment planning scans. The current generation of PET/CT scanners have increased bore size; however, there will still be some restrictions on treatment position set-up during imaging.
Radiotherapy treatments are performed with patients on a flat couch, whereas diagnostic scans are usually performed using a concave couch, with the patient lying on a thin mattress. For PET/CT data to be used to guide radiotherapy treatment, the scanner must be equipped with a flat-bed insert, which are now routinely available (Figure 1
). It should be noted that addition of this bed attachment reduces the patient port further and again restricts positioning options.
Immobilisation devices
The use of immobilisation devices in radiotherapy treatment is well established and provides an effective mechanism for the reproducible positioning of patients at each treatment episode. A locally modified Med-TEC thorax immobilisation board and a knee rest were used (Med-TEC, Orange City, IA) in the pilot study. Patients were positioned with both arms above their head to ensure that the arms were outside the treatment fields. The immobilisation board was modified with an additional T-bar grip to enable the patient's arms to be supported above their heads and to reduce the span across the patient's arms to facilitate positioning within the scanner (Figure 2).
|
Planning scan acquisition
|
|---|
The principles and practices already established for RTP were transferred to the acquisition of planning data using the PET/CT system. Two therapy radiographers positioned the patient and provided verification that the position during the data acquisition process could be reproduced during treatment. A previous evaluation of staff radiation doses from routine performing of FDG-PET scanning showed that the major component of the accumulated dose is obtained through close proximity to the patient post injection. To minimise dose to the therapy radiographers, the data acquisition process was implemented as a two-stage process [6].
"Cold" set-up session
Prior to injection of 18F-FDG, the patient was positioned on the couch using the flat-top insert and immobilisation devices as previously illustrated. The patient was carefully positioned to ensure that they would be able to maintain the position for the duration of the imaging process. This was of the order of 40 min while a whole-body PET/CT scan was acquired, as the primary purpose of the scan was for diagnosis and staging. In the subsequent study, a diagnostic scan will be performed independently of the treatment planning session. This will reduce the time for the radiotherapy planning PET/CT to 10–15 min.
During this session the therapy radiographers used the CT scanner positioning lasers to establish anterior and lateral markers on the patient's skin, approximately at the position of the xiphisternum. Full details of the patient's position on the scanner and immobilisation board were manually recorded to permit rapid and accurate repositioning of the patient post injection. The patient was then removed from the PET/CT scanner for the injection and uptake phase. Although use of the internal scanner lasers proved acceptable during the pilot study, further experience has shown that their use can be problematic for the therapy radiographers depending on the shape and size of the patient. This potentially increases the time required to set-up the patient and may lead to increased radiation doses to the radiographers. The use of external room lasers would reduce these problems and is recommended. However, additional quality control testing of these lasers would be required to ensure the external lasers are aligned with the scanner. In addition, the potential effects of sag of the scanner couch during the investigation with the patient being set-up outside the scanner bore and then being moved on the couch into the imaging position must be investigated.
"Hot" set-up session
Following the injection and uptake period, the patient was repositioned by the therapy radiographers using the pre-recorded set-up information. Radio-opaque markers were attached over the skin marks previously identified and the data acquisition process was completed. After the scan, the radiotherapy radiographers permanently marked the patient's skin at the position of the radio-opaque markers to provide reference points between the CT images and the treatment planning system.
|
Staff radiation doses
|
|---|
The therapy radiographers were monitored throughout the pilot project using personal dosemeters; Table 1 provides details of staff doses recorded. The average dose per staff member per patient was approximately 3.0–3.5 µSv [6]. These doses are comparable with those obtained by diagnostic staff throughout the entire imaging process. These doses are not insignificant and will require further consideration if large numbers of patients are to undergo RTP using PET/CT studies. It should be noted that this dose is accumulated by each of the two therapy radiographers owing to the need for process checking.
View this table:
[in this window]
[in a new window]
|
Table 1. Radiation doses received by the therapy radiographers during the hot set-up session and the marking-up session
|
|
|
Image transfer content validation and registration for treatment planning
|
|---|
At the time of the pilot project, the transfer of data between the GE Discovery LS PET/CT scanner and a Theraplan Plus treatment planning system was not straightforward. The CT data were readily transferable using standard CT DICOM [7] objects, and phantom data demonstrated the validity of this transfer process. The PET images could not be directly transferred owing to the limitations of the DICOM interfaces. Instead, the PET images were reformatted using a GE Advantage Windows workstation, which permitted the data set to be transferred as a CT DICOM object rather than a true PET DICOM object. Both data sets were transferred using CD-R media and successfully incorporated into the RTP process. Subsequent to the pilot study, the DICOM interfaces have been improved such that the PET/CT data sets can be directly imported into the treatment planning system. Validation of this data transfer is essential to the integrity of this process. As the PET and CT data sets are inherently fused, no further manipulation of the data sets was required following transfer to the planning computer. However, the way in which the PET data set should be used to define target volumes is undefined and remains the subject of ongoing research.
|
System quality control
|
|---|
The radiotherapy process usually operates within a formal quality management system. The integration of data from systems located and operated remote from the radiotherapy department poses its own unique problems. It is clear that the PET/CT system will need to come within the scope of such a quality management system to ensure, as a minimum, that revalidation and acceptance procedures are in place for the key image parameters of electron density calibration on CT, pixel size for both PET and CT images, and quantitative calibrations for PET uptake values. Each of these may be subject to change following software and hardware maintenance procedures and must be subject to return to service checks if patient doses are not to be miscalculated or volumes of interest incorrectly delineated. These requirements may prove time consuming and costly and their impact should not be underestimated. Of equal importance is the ability to ensure correct identification and labelling of the PET/CT data sets and any associated radiotherapy objects derived from the data such as the tumour volume. This has implications for the unification of patient identifiers between the diagnostic and treatment facilities and indeed the ability to integrate booking systems to provide rapid access to diagnostic services. DICOM worklist management must become a standard facility in these environments.
|
Future considerations
|
|---|
Target volume delineation
Target volume delineation for radiotherapy takes place in a highly regulated and highly organised environment and is usually undertaken by radiation oncologists in conjunction with clinical scientists and treatment planning staff. International protocols define the methodologies to be used for treatment volume determinations [8, 9]. Treatment plans are usually based upon radiographic or CT data acquired as a snapshot data set. The time required for a volume image of the lungs on a modern CT scanner is of the order of seconds and is significantly less than the breathing cycle. The volume image thus provides an indication of the position of the internal organs at a point in time and does not provide information regarding the potential movement of abnormal tissue volumes during the respiratory cycle. For standard radical radiotherapy, this limitation is overcome by the addition of a margin or volume surrounding the CT-defined tumour volume. The extent of this additional margin is governed by experience and data indicating potential excursions for various lesion sites. The anatomical resolution of CT data is such that it has effectively zero influence upon the treatment volume owing to the additional margins that are applied at the discretion of the radiation oncologists. By comparison, the PET data are acquired in a very different time scale and interpretation of the acquired image is more problematic. A PET image is usually acquired over a period of 3–5 min and is therefore imaged over many breathing cycles. The image data therefore portray a distribution that is time-averaged and based upon the residence time of the FDG distribution of each location within the image. It could therefore be argued that no additional margins to account for respiratory motion should be applied to treatment volumes defined by PET data. It should be noted that margins due to patient set-up and subclinical spread of the disease will still need to be included. In addition to this time-averaging effect, the resolution within the PET image is such that there is significant impact upon the quantification of volumes and activity concentrations for small objects owing to the partial volume effect. This effect, combined with motion averaging, means that careful consideration must be given to the characteristics of the image that are used to define treatment volumes with or without the addition of treatment margins. Most published papers have chosen to define regions based upon a percentage of the maximum standardised uptake value (SUV) [10] for the object under consideration or based upon a percentage contour within the image or any absolute value of the SUV [11, 12]. These regions are normally drawn manually and do not immediately address the issues discussed above or indeed signal-to-noise ratio within the images themselves. In addition, the non-target activity levels will impact upon the derived volumes [13]. Despite these uncertainties, PET data are being used within the RTP process and volumes modified are according to the PET distribution [14]. Given these uncertainties, PET data were not used within the pilot study to modify the current treatment planning regimen; however, it was clear that such an approach would have modified treatment volumes in some of the patients studied.
Figure 3 illustrates the problems encountered in the delineation of target volumes based on the thresholding of PET images where volumes can change significantly [15]. Erdi et al [16] illustrated the differences that could be obtained from PET and CT imaging (Figure 4). PET showed involvement of a local node (shown in red), whereas the CT image did not (shown in yellow). The choice of a target irradiation volume remains a complex issue requiring input from a number of clinical specialists to interpret disease likelihoods and treatment impacts.
View larger version (39K):
[in this window]
[in a new window]
|
Figure 3. PET images in a patient with right hilar cancer illustrating problems with accurate delineation of target volumes based on thresholding of PET images. (A) PET threshold is set at 40% of maximum standard uptake value (SUV) of tumour. (B) PET threshold is set at 30% of maximum SUV. If delineating this tumour using PET alone, different threshold settings would result in different target volumes. Reprinted with permission of the Society of Nuclear Medicine from: Bradley JD, Perez CA, et al. Implementing biologic target volumes in radiation treatment planning for non-small cell lung cancer. J Nucl Med 2004;45:96S–101S.
|
|
View larger version (57K):
[in this window]
[in a new window]
|
Figure 4. Wire diagrams showing planning target volume (PTV) delineated from CT (yellow) and from CT + PET (red) for one non-small cell lung cancer patient. Involved paratracheal lymph nodes were detected on PET scan, which were subsequently incorporated into the patient PTV. Reprinted with permission of Elsevier from: Erdi YE, Rosenzweig K, Erdi AK, et al. Radiotherapy treatment planning for patients with non-small cell lung cancer using positron emission tomography (PET). Radiother Oncol 2002;62:51–60.
|
|
The primary effect of respiratory motion is a spreading of the extension of the detected tumour predominantly in the thoracic and to a lesser extent in areas below the diaphragm. However, since the first clinical application of PET imaging in RTP is in lung cancer, the issue of respiratory motion has a predominant position in the accuracy of determination of the functional volume and activity concentration. In particular, the case of PET/CT scanners poses additional challenges since the presence of respiratory motion introduces inaccuracies in the reconstructed images not only as a result of lesion smearing but also as a result of misregistration between PET and CT acquisitions [17, 18]. Since with these hybrid scanners, the CT maps are also used for correction of attenuation effects in the emission data, an extra inaccuracy may be introduced by using non-perfectly aligned CT and PET data sets as a result of respiratory motion [19]. This is illustrated in Figure 5. The left-hand image is a PET scan done with a slow transmission source, using radioactive rod sources where both of the images have been time-averaged to the same extent. This image shows no obvious artefacts in the region of the diaphragm and the observer is certainly not aware of identifiable errors within the image. The central image has been attenuation-corrected using a CT map acquired in a very short time frame. The image demonstrates the classical "banana" artefact sitting just above the diaphragm, just above the liver where the incorrect attenuation map has been used to correct the PET data. The CT map was acquired using a breath-hold technique at end inspiration and thus reflected the lungs at maximum expansion with the PET acquired as a time-averaged image. The image on the right is created using a CT map acquired under a free breathing protocol. Here the artefact is smaller, most probably related to the overall smaller volume of the lungs. The same type of image is obtained using a CT protocol with acquisition at end expiration. It should be noted that these are artefacts due to a mismatch in the geometry of the body during the acquisition of CT and PET data. This effect will be present where there is significant motion of organs or lesions within the field of view, and subsequent errors of attenuation correction will lead to increases and decreases in tracer concentrations at the boundaries of these mismatches. This is illustrated in Figure 6, where the extent of activity distribution with and without physiological gating differs significantly.
View larger version (63K):
[in this window]
[in a new window]
|
Figure 5. Problems with accurate delineation of target volumes as a result of differences in the state of respiration between emission and transmission maps (note changes in the tumour in the mediastinum as well as the artefact areas). Left: attenuation correction using a respiration average transmission map with radioactive rod sources. Centre: attenuation correction using a CT map acquired at end inspiration. Right: attenuation correction using a CT map acquired at end expiration or under free breathing. Adapted from: Visvikis D, Costa DC, Croasdale I, et al. CT-based attenuation correction in the calculation of semi-quantitative indices of [18F]FDG uptake in PET. Eur J Nucl Med Mol Imaging. 2003;30:344–53. Reprinted with kind permission of Springer Science and Business Media.
|
|
View larger version (35K):
[in this window]
[in a new window]
|
Figure 6. Problems with accurate delineation of target volumes as a result of respiratory motion during emission acquisition. (a) Tumour extent during a respiratory average frame. (b) Tumour extent in a respiratory-gated frame. (c) Planning target volume (PTV) from the gated and non-gated PET frames. Reprinted with permission of the Society of Nuclear Medicine from: Nehmeh S, Erdi Y, Ling CC, et al. Effect of respiratory gating on quantifying PET images of lung cancer. J Nucl Med 2002;43:876–81.
|
|
The solutions that have been proposed to date for taking into account the effects of respiratory motion concentrate on the acquisition of respiration-synchronised PET and CT data sets. In the vast majority of the cases the respiratory signal used for synchronisation of the acquired data sets is provided by an external detector. Different detector systems have been proposed, including a transducer or an impedance electrocardiogram monitor measuring changes in abdominal or thoracic circumference [20], a thermistor measuring the temperature of circulating air during patient respiration [21], a spirometer measuring respiratory flow [22], the Varian Real-Time position management (RPM; Varian Medical Systems, Palo Alto, CA), or the Polaris system tracking the displacement of infrared reflective markers in the patient chest [19]. The different measurements carried out by these detector systems provide a respiratory signal that in the majority of cases is used to trigger data acquisition during PET. Alternatively, the provided respiratory signal may be recorded in synchrony with the emission data acquisition, followed by a subsequent re-binning of the acquired data in temporal bins. For the range of physiological triggers evaluated, the majority lead to an accurate and reproducible respiratory signal [22]. A recently proposed alternative method based on the use of image-derived respiratory signal through the acquisition of dynamic data sets or list mode data [23] is currently undergoing clinical validation studies.
Further to the creation of respiration-synchronised PET emission acquisitions, it is equally important to ensure that the CT maps used for attenuation correction are also at the same phase of the respiratory cycle. Few studies have demonstrated the potential for significant errors in quantitative accuracy and artefacts at the level of the diaphragm from the use of CT maps acquired at different phase in the respiratory cycle in comparison with the emission data sets. Such mismatch between PET and CT data sets can also lead to significant changes in lesion volume determination. Proposed methodology for matching PET and CT data sets in terms of respiratory motion involves the acquisition of four-dimensional (4D) CT data sets synchronised with an externally-acquired respiratory signal. The data sets are subsequently sorted to match the breathing phase of the respiratory-synchronised PET data sets [24].
Irrespective of the gating methodology implemented, the emission data acquired in each of the temporal-gated frames are reasonably free of respiration-produced inaccuracies. However, the resulting individual frame images are of reduced resolution and overall quality as they contain only a fraction of the counts available throughout a PET acquisition [25] (Figure 7). Therefore, the need exists for the development of correction methodologies, making use of the gated data sets, to obtain respiration-free PET images using all available data throughout a standard respiration-average PET acquisition. This approach will also remove the need currently existing in terms of significantly increasing the time (over a factor of 3) of gated PET acquisitions to compensate for the presence of reduced statistics in the final reconstructed images. Very limited work is currently available in this domain. First, an emission-driven solution through the combination of respiratory-synchronised emission data sets and an iterative reconstruction algorithm can be envisaged, in a similar fashion to the methodology that has been previously suggested for single photon emission computed tomography (SPECT) cardiac imaging applications [26, 27]. The second option is based on a realignment methodology to "bring" all of the respiratory-synchronised PET data sets to a particular phase in the respiratory cycle. This methodology is potentially applicable to both image and raw data domains, deriving the transformation parameters from the corresponding respiratory motion-synchronised CT frames [23, 28].
View larger version (19K):
[in this window]
[in a new window]
|
Figure 7. Demonstration of image quality degradation as a function of image statistics between respiration average and respiration-gated frames at the level of the lung. Simulated 22 mm (left lung) and 16 mm (right lung) lesions at different levels of the pulmonary field are included. Respiration average frame: (a) 30M total coincidences and (b) 20M total coincidences. Corresponding respiration-gated frame: (a) 10M, (b) 6M, (c) 4M and (d) 2M total coincidences.
|
|
The current generation of PET/CT scanners are capable of performing full dynamic CT acquisitions (4D-CT) and using a respiratory signal to segment the PET data into a number of equal time integrals. Typically, eight to ten time intervals are used covering the respiratory cycle and for a fully gated study, would permit the use of aligned CT and PET data for attenuation correction and dynamic tracer analysis. We have investigated two low-cost devices to permit the acquisition of dynamic data. The first utilising a thermistor placed under the nose of the patient with a signal being derived from the changes in temperature of the inspired and expired air, and the second using a strain gauge placed around the thorax of the patient (Figure 8). Most gating technologies work on the principle of segmenting the data based on a time slice, with data from an end expiration signal being averaged in equal time slots. It is clear that respiratory cycles are very variable and that this approach may not be ideal. The requirement of the gating process is to provide a mechanism of averaging data from the lungs at points of equal lung volume. In this regard, the strain gauge provides a signal that is highly correlated with lung volume and may prove to be an effective mechanism to divide the data consistently based upon lung volume.
The decrease in signal-to-noise ratio in the dynamic image data set will result in increased errors in the delineation of volumes within the PET image that must be balanced against potentially better localisation and volume definition owing to the removal of time averaging for moving objects. The gains from the introduction of this segmentation of the data have yet to be quantified and will clearly vary depending on the size and location of the lesions within the chest.
|
Conclusions
|
|---|
The pilot study investigating PET/CT scanning for radiotherapy planning in NSCLC has highlighted a number of practical issues that must be addressed to facilitate accurate data acquisition. These include patient positioning and compliance, data and system integrity and quality assurance, the impact of respiratory motion and the need to develop detection and correction systems to match advances in radiotherapy delivery systems. Operator dose and the radiation protection of staff must be addressed if it is not to become a limitation to the use of PET data. Beyond these acquisition issues there remains the need for high-quality segmentation software to permit reliable and reproducible target volumes in space and time for both PET and CT data sets.
Another underlying issue that must be addressed is the sensitivity and specificity of the radiotracer used in the PET study. Whilst FDG-PET is known to be more sensitive and specific than CT in the detection of disease, there is still a measurable false-positive and false-negative fraction. The definition of treatment volumes based solely upon the PET image could lead to CT-defined disease being missed or to PET-positive volumes being treated unnecessarily owing to the delineation of non-cancerous physiological processes. These characteristics will differ for each type of cancer at different locations within the body and also with the expertise of the image interpreter. The combination of information from multiple imaging and investigational sources must be understood and optimised as part of this change process for radiotherapy planning. Multidisciplinary teams will be required to implement and interpret combined PET/CT studies for RTP purposes. Clearly defined protocols must be developed and tested through clinical trials aimed at establishing better patient outcomes using these new modalities.
|
References
|
|---|
- Spiro SG, Silvestri GA. One hundred years of lung cancer. Am J Respir Crit Care Med 2005;172:523–9.[Abstract/Free Full Text]
- MacManus MP, Hicks RJ, Ball DL, Kalff V, Matthews JP, Salminen E, et al. F-18 fluorodeoxyglucose positron emission tomography staging in radical radiotherapy candidates with non small cell lung carcinoma: powerful correlation with survival and high impact on treatment. Cancer 2001;92:886–95.[Medline]
- Dillman RO, Herndon J, Seagren SL, Eaton WL, Green MR. Improved survival in stage III non-small-cell lung cancer: seven year follow-up of cancer and leukaemia group B (CALGB) 8433 trial. J Natl Cancer Inst 1996;88:1210–5.[Abstract/Free Full Text]
- Hellwig D, Ukena D, Paulsen F, et al. Meta-analysis of the efficacy of positron emission tomography with F-18-fluorodeoxyglucose in lung tumors. Basis for discussion of the German Consensus Conference on PET in Oncology 2000. Pneumologie 2001;55:367–77 [in German].[Medline]
- Toloza EM, Harpole L, McCrory C. Non invasive staging of non-small cell lung cancer: a review of the current evidence. Chest 2003;123:137S–46S.
- Carson KJ, Hounsell AR, Cosgrove VP, Stewart DP, Grattan MWD, Fleming L, et al. Methodology and issues in the direct use of combined 18FDG-PET/CT images in radiotherapy planning for patients with non-small-cell lung cancer. Eur J Nucl Med 2003;30:S210.
- National Electrical Manufacturers Association. ACR-NEMA Standards Publication PS 3.1-2004. Rosslyn, VA: NEMA.
- International Commission on Radiation Units and Measurements. Prescribing, recording and reporting photon beam therapy. ICRU Report 50. Bethesda, MD: ICRU, 1993.
- International Commission on Radiation Units and Measurements. Prescribing, recording and reporting photon beam therapy, Supplement to ICRU Report 50. ICRU Report 62. Bethesda, MD: ICRU, 1999.
- Black QC, Grills IS, Kestin LL, Wong C-Y, Wong JW, Martinez AA, et al. Defining a radiotherapy target with positron emission tomography. Int J Radiat Oncol Biol Phys 2004;60:1272–82.[Medline]
- Erdi YE, Mawlawi O, Larson SM, Imbriaco M, Yeung H, Finn R, et al. Segmentation of lung lesion volume by adaptive positron emission tomography image thresholding. Cancer 1997;80(12 Suppl.):2505–9.
- Yaremko B, Riauka T, Robinson D, Murray B, McEwan A, Roa W. Threshold modification for tumour imaging in non-small cell lung cancer using positron emission tomography. Nucl Med Comm 2005;26:433–40.[Medline]
- Daisne J-F, Sibomana M, Bol A, Doumont T, Lonneux M, Gregoire V. Tri-dimensional automatic segmentation of PET volumes based on measured source-to-background ratios: influence of reconstruction algorithms. Radiother Oncol 2003;69:247–50.[Medline]
- Bradley J, Thorstad WL, Mutic S, Miller TR, Dehdashti F, Siegel BA, et al. Impact of FDG PET on radiation therapy volume delineation in non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2004;59:78–86.[Medline]
- Bradley JD, Perez CA, Dehdashti F, Siegel BA. Implementing biologic target volumes in radiation treatment planning for non-small cell lung cancer. J Nucl Med 2004;45(Suppl. 1):96S–101S.
- Erdi YE, Rosenzweig K, Erdi AK, Macapinlac HA, Hu YC, Braban LE, et al. Radiotherapy treatment planning for patients with non-small cell lung cancer using positron emission tomography (PET). Radiother Oncol 2002;62:51–60.[Medline]
- Visvikis D, Costa DC, Croasdale I, et al. CT based attenuation correction in the calculation of semi-quantitative indices of 18-FDG uptake in PET. Eur J Nucl Med Mol Imaging 2003;30:344–53.[Medline]
- Beyer T, Antoch G, Muller S, et al. Acquisition protocol considerations for combined PET/CT imaging. J Nucl Med 2004;45:25S–35S.[Abstract/Free Full Text]
- Erdi Y, Nehmeh SA, Pan T, et al. The CT motion quantitation of lung lesions and its impact on PET-measured SUVs. J Nucl Med 2004;45:1287–92.[Abstract/Free Full Text]
- Livieratos L, Stegger L, Bloomfield PM, et al. Rigid-body transformation of list-mode projection data for respiratory motion correction in cardiac PET. Phys Med Biol 2005;50:3313–22.[Medline]
- Wang Y, Baghaei H, Li H, et al. A simple respiration gating technique and its application in high resolution PET. IEEE Trans Nucl Sci 2005;52:125–9.
- Guivarc'h O, Turzo A, Visvikis D, et al. Synchronisation of pulmonary scintigraphy by respiratory flow and by impedance plethysmography. Proc SPIE Medical Imaging 2004;5370:1166–75.
- Visvikis D, Barret O, Fryer T, et al. A posterior respiratory motion gating of dynamic PET images. IEEE Medical Imaging Conference Record 2003;5:3276–80.
- Nehmeh SA, Erdi Y, Pan T, et al. Quantitation of respiratory motion during 4D PET/CT acquisition. Med Phys 2004;31:1333–8.[Medline]
- Visvikis D, Lamare F, Turzo A, et al. Efficiency of respiratory gating for motion correction in PET. J Nucl Med 2005;46:163P.
- Kyme AZ, Hutton BF, Hatton RL, et al. Practical aspects of a data driven motion correction approach for brain SPECT. IEEE Trans Med Imaging 2003;22:722–9.[Medline]
- Lee TS, Segars PW, Tsui BW. Application of 4D MAP-RBI-EM with space-time Gibbs priors to gated myocardial SPECT. J Nucl Med 2005;46:162P.
- Lamare F, Cresson T, Savean J, et al. Affine transformation of list mode data for respiratory motion correction in PET. IEEE Medical Imaging Conference Record 2004.
Top
Abstract
Introduction
