Ung, Yee C. MD, FRCPC*; Bezjak, Andrea MDCM, FRCPC†; Coakley, Nadia MLIS‡; Evans, William K. MD, FRCPC§; the Lung Cancer Disease Site Group of Cancer Care Ontario
Lung cancer is the leading cause of cancer-related deaths in both men and women in Canada.1 Radiation treatment (RT) is indicated for use in approximately 60% of all patients with lung cancer and is used for a variety of intents, including curative, adjuvant, neoadjuvant, and palliative.2 RT is most commonly applied in stage III non-small cell lung cancer (NSCLC), where it is estimated that it might be indicated for as many as 84% of patients.2
External beam radiotherapy (i.e., teletherapy) is the most common form of RT and involves the targeting of high-energy photons (i.e., x-rays) at cancerous tissues to promote malignant cell death. Although healthy cells are better able to repair damage from radiation, RT can kill healthy tissues at sufficient dosage, and epithelial tissues are particularly vulnerable. Tissue scarring can result from radiation exposure and lead to reduced elasticity. This is especially relevant in lung cancer, where critical organs such as the heart, spinal cord, esophagus, and the remainder of the normal lung are often in the vicinity of tumor tissues (i.e., organs at risk [OARs]) and damage to these can be detrimental to the patient. Because of the possibility for significant adverse effects from radiation, radiation oncologists are continually seeking methods to target RT more precisely. The use of positron emission tomography (PET) with radiolabeled [18F]-2-fluorodeoxy-d-glucose (18FDG) PET imaging information is being evaluated as a possible means to improve current RT practices.
RT dosage in lung cancer is generally provided to patients in daily fractions, and a typical dose for a solid epithelial tumor ranges from 60 to 70 Gy, with a fractionation schedule for adults of 1.8 to 2.0 Gy per day. Radiation dosage exposures are commonly described in terms of the percentage of the organ receiving a particular total dose of radiation. For example, V20 lung indicates the percentage of the lungs, excluding the planning target volume (PTV), that received a dose of 20 Gy or more during the course of treatment. The extent to which RT has achieved its objective in killing tumor cells is conveyed by the concept of tumor control probability (TCP). Imaging technologies, specifically planning computed tomography (CT), are used in the RT planning process to delineate tumors and adjacent healthy structures. Traditionally, specialized CT scanners are combined with planning software to virtually simulate the tumor and accurately place x-ray beams. Newer approaches, such as three-dimensional conformal radiotherapy or intensity-modulated RT are expected to further enhance these efforts.
RT planning requires precise definition of the region of the diseased part of the body that is the target of the radiation dose. In current practice, this region or “volume” is defined three dimensionally in accordance with principles articulated by the International Commission on Radiation Units and Measurements. The gross tumor volume (GTV) and clinical target volume (CTV) are clinical-anatomic concepts and refer to the physical space occupied by disease. The GTV is “the gross, palpable, visible or clinically demonstrable location and extent of the malignant growth” and is generally defined by all gross disease identified in scans (e.g., CT, PET, and fused) and through other clinical information.3 The GTV includes the primary tumor and metastatic lymphadenopathy. The CTV contains the GTV and areas where there is a high probability of subclinical malignant disease and typically includes a volumetric extension of the GTV (e.g., a 0.6-0.8 cm margin and/or inclusion of draining lymph node regions).4–6 Unlike GTV and CTV, the PTV is geometric definition that is used directly in targeting a radiation beam. The PTV contains the CTV and margins to account for variability due to internal motion such as respiration in patient setup (“setup margin”) or position of the target for lung tumors (“internal margin”).7 Several algorithms have been proposed to aid in the determination of the PTV, but ultimately it is a clinical judgment that takes into account adjacent topology, specifically the OARs for radiation toxicity.
CT has traditionally been the primary source of anatomic imaging information for target volume selection and delineation in oncology. Nevertheless, CT is limited by the fact that it has diminished resolution for normal soft tissue structures and tumor extent. A number of studies have reported significant variations in the delineations of GTV based on CT data.8,9 There is reason to believe that the tumor metabolic information provided by PET would be valuable in RT planning. Tumor tissues generally exhibit more rapid glycolysis than normal tissues, and the 18FDG tracer allows for the metabolic imaging of this tissue. A number of studies have compared the accuracy of PET in comparison with CT for the purposes of diagnosis and staging in lung cancer.
PET has greater sensitivity and marginally greater specificity relative to CT in specific instances.10–14 This has implications for RT planning in lung cancer. For instance, the systematic review found PET to be superior to CT for mediastinal staging in NSCLC.10 The greater sensitivity of PET is believed to improve the detection of metastatic lymph nodes that CT would have missed. PET may be better able to detect distant metastases and allow for the exclusion of patients from unnecessary radical RT. Conversely, PET may result in the downstaging of CT-false-positive nodes and the exclusion of nonmalignant tissues from the PTV. The benefit of this for patients could be substantial: Graham et al.15 have argued that a reduction of V20 lung by 5 to 17% would reduce the incidence of grade 2 or greater pneumonitis occurring within 24 months of treatment by up to 23%.
Despite this strong theoretical rationale for using PET in RT planning, it is not yet clear that the addition of PET imaging data has a clinically significant impact on planning. Furthermore, assuming there is a benefit to including PET data in planning, the optimal approach to using PET data is not yet established. At present, PET tumor contouring remains unsatisfactory, and there is little standardization in its use. For instance, the delineation of tumor volumes based on a metabolic activity threshold in PET has been shown to vary both by tumor size and the background-to-tumor 18FDG uptake ratio.16 Some clinicians include an area of lower uptake, which some term as the “anatomic-biologic halo,” in the GTV, and one study has shown that including this halo improves coverage of the PTV,17 although, again, the practice is not yet standard.
This systematic review was initiated because of the increasing use and potential importance of PET in this area. This systematic review will provide an evidence-based perspective as to whether planning based on PET and PET-CT imaging data represents an improvement over planning based on CT data alone and inform guidance on its role in RT planning in the lung cancer setting.
PET is an imaging technique that gives high-resolution images based on the use of biologically active compounds, substrates, ligands, or drugs labeled with positron emitters. These radiolabeled agents are processed in vivo in a manner virtually identical to their nonradioactive counterpart, thereby producing images and quantitative indexes of blood flow, glucose metabolism, amino acid transport, protein metabolism, oxygen consumption, and even cell division.
Traditional radiologic imaging (e.g., CT scan and magnetic resonance imaging) is based on structural information and defines disease states based on gross anatomic changes, whereas PET imaging is based on biochemical processes that often precede any gross anatomic distortion. PET imaging is now used primarily in oncological imaging due to the successful application of 18F-FDG. This systematic review will only evaluate the role of 18FDG-PET.
Imaging by PET is based on the detection of 511 keV annihilation photons that are the result of positron, in this case emitted from 18F, colliding with an electron. Photons that are in coincidence are detected by two detectors at 180-degree angle from each other. These photons are considered to have originated from a point source along that axis. All the collected information is then processed into the final image in a two-dimensional or three-dimensional representation that reflects the concentration and distribution of the radioisotope. This creates the image of FDG localization.
The PET image does not provide accurate anatomic information aside from areas of normal physiological uptake (such as the heart, kidneys, and bladder) and soft tissue uptake (e.g., muscle) that can provide an outline of the imaged body. Therefore, advances in imaging technology have combined PET and CT to provide both functional and anatomic information simultaneously, thus improving its accuracy.
Various imaging instrumentation have been used in the studies covered by this systematic review. The instrumentation varies from gamma camera coincidence imaging, to dedicated PET scanner to PET-CT. It is not the intent of this systematic review to evaluate the instrumentation, but different instrumentation will have differing diagnostic performance capabilities.18
The evidence-based series guidelines developed by Cancer Care Ontario's Program in Evidence-Based Care (PEBC) use the methods of the Practice Guidelines Development Cycle.19 For this project, the core methodology used to develop the evidentiary base was the systematic review. Evidence was selected and reviewed by two members of the PEBC Lung Disease Site Group and two methodologists.
This systematic review is a convenient and up-to-date source of the best available evidence on PET in RT planning for lung cancer. The PEBC is supported by the Ontario Ministry of Health and Long-Term Care through Cancer Care Ontario. All work produced by the PEBC is editorially independent from its funding source.
Literature Search Strategy
The Medline (1996 to May 2010), EMBASE (1996 to May 2010), and Cochrane Library (2007, Issue 1) databases were searched for published practice guidelines, technology assessments, systematic reviews, clinical trials, and studies. Reference lists of articles and review articles were scanned for additional citations. The Canadian Medical Association Infobase (http://www.cma.ca/index.cfm/ci_id/54316/la_id/1.htm), the National Guidelines Clearinghouse (http://www.guideline.gov/), and other websites were searched for existing evidence-based practice guidelines. The conference abstracts of The American Society of Clinical Oncology, the American Society for Therapeutic Radiology and Oncology, and the European Society for Therapeutic Radiology and Oncology (2002-2009) were searched for randomized studies. Search terms indicative of lung cancer, RT planning, and PET technology were used. The following MeSH terms were used: lung, (cancer$ or carcinoma$ or neoplasm$) NSCLC, SCLC, positron emission tomography, (PET or positron emission) gamma camera, (tumo$ or target or treatment) (volume$ or portal or plan$ or manage$) (GTV or PTV) (radiation$ or radiotherapy$ or RT) plan$ (Appendix).
APPENDIX. Literature...Image Tools
Study Selection Criteria
The most useful evidence regarding patient outcome. Therefore, studies that reported on any relevant patient outcome, such as survival, recurrence rates, treatment related morbidity, and quality of life, were included.
Nevertheless, there was an a priori expectation that no such data would be available. Therefore, studies that reported data on more technical measures of RT planning were also included, as these data would be useful in evaluating improvements, if any, arising from the incorporation of PET-CT imaging. These measures include changes in the target volume definitions (e.g., GTV), radiation exposures to OARs, tumor dose and TCP, intent of RT or in the management of patients, and rate of geographical misses, and detection of distant metastases.
Articles were selected for consideration in this systematic review of the evidence if they were published reports of studies of any design that reported on aspects of RT planning for patients with lung cancer that incorporated PET imaging data and that compared the impact of consolidated PET-CT imaging data with RT planning done in the absence of PET data. Retrospective studies for which the RT planning was theoretical (i.e., records were reviewed and investigators determined the RT planning that would have occurred had a PET evaluation been done) were included. Studies including patients with multiple disease types (e.g., lung cancer and head and neck cancer) must have reported data for patients with lung cancer specifically or comprise a majority of patients with lung cancer to be eligible for inclusion. Studies reporting data on the impact of PET on clinical management of patients, including not only RT management but also surgical or chemotherapy management, were included if data specific to RT management were reported.
Surveys of clinicians to measure the influence of PET on RT planning were excluded as there is some subjectivity involved in surveys, as were phantom studies. Studies reported in a language other than English were excluded because of a lack of translation resources.
Synthesizing the Evidence
There was considerable inconsistency in the presentation of data across included studies. The nature of the studies and the data they provided did not lend themselves to meta-analysis.
Literature Search Results
Abstracts for 219 studies were retrieved, and of these, 28 journal publications were deemed appropriate for inclusion in this report based on the eligibility criteria outlined earlier. Five studies14,20–23 and two studies of patients with small cell lung cancer (SCLC)24,25 that were relevant to the research questions but did not report data appropriate for summary in this report were excluded. Table 1 outlines the quantity and type of studies identified to answer the questions posed in this review. No practice guidelines, systematic reviews, or meta-analyses were identified, and most studies consisted of small sample size observational prospective or retrospective cohort studies. One randomized trial, the The Impact of Positron Emission Tomography Imaging in Stage Three Non-Small Cell Lung Cancer: A Prospective Randomized Clinical Trial (PET-START) trial, has been reported in abstract form.48
Table 2 presents several descriptive characteristics of the included studies. The type of imaging technologies used in the studies varied: 10 studies evaluated hybrid PET-CT scanners,17,26–31,49,52,53 four studies evaluated gamma camera scanners,32–35 and the remainder evaluated dedicated PET scanners. The combining (or coregistration) of PET and CT images adds additional measurement error to the development of a target volume definition. Coregistration has traditionally been a manual process involving the visual overlay of images with the help of fixed markers, although more recently it has been automated and improved through the use of computer software and hybrid PET-CT devices. Studies using hybrid scanners will have substantially less measurement error due to coregistration.
Observational studies are inherently more susceptible to bias than are randomized controlled trials. In most of the studies included in this review, insufficient detail on efforts to control bias in the treatment planning process was provided, so it was assumed that planning was done by single or in some cases multiple clinicians. The latter approach would lead to interobserver variability in results.
A second bias relates to the sequential integration of imaging data, a practice that may not reflect real-world clinical practice. In all studies, RT plans were developed using CT data alone and then subsequently with PET imaging data included. Clinicians typically evaluate PET-CT data simultaneously, and this may become standard practice as PET becomes more integrated in the RT planning process. These potential biases could be controlled for in studies by having independent evaluators conduct planning on CT and PET-CT data separately. Only three studies in this series used independent evaluators of CT and PET-CT data.17,36,37
A third bias may be present in studies for which the study design was a retrospective case review. The six retrospective studies in this series established a hypothetical PET-based RT plan using PET data obtained for other purposes (e.g., staging).28,30,38,39,49,51 These studies may be biased relative to prospective studies in that investigators may have been less conservative in their planning definitions knowing the definitions would not be applied in real patients. One prospective study was also vulnerable to the same bias in that a nontreating physician conducted the PET-based RT planning.36 Details on the clinicians involved in planning (e.g., whether they were treating physicians or not) were not provided in most studies, and others might be subject to the same threats to validity.
All studies evaluated patients with NSCLC (and in three studies, patients with SCLC, and with “lung cancer” without further specification were also included) for whom radical RT was deemed appropriate on the basis of conventional imaging data. Two studies reported including patients with recurrent NSCLC,32,40 and one retrospectively reviewed patients who were at high risk for tumor recurrence after pneumonectomy.41 One study only included patients with positive lymph nodes.39 Only limited details on specific pathologic features were included in the remainder of the studies in this series.
The studies included in this systematic review reported data on a total of 1054 patients. Samples ranged in size from 5 to 153 patients, with a mean of 38 patients and a median of 59 patients. Most studies provided basic summary statistics only (e.g., means, proportions, and ranges) and did not report results from statistical testing (e.g., p values and confidence intervals).
Measures and Outcomes
The PET START Trial
The PET START trial (NCT00136864) randomized patients to either standard combined modality therapy for stage III NSCLC or PET imaging before combined modality therapy with curative intent. Data from this trial were reported in abstract form at the 2009 American Society of Clinical Oncology Annual Meeting.48 The primary outcome was the proportion of patients who did not receive combined modality therapy because their tumor was upstaged to stage IV or their intrathoracic tumor was too extensive for radical RT. The abstract provides insufficient detail to assess the methodological quality and potential for bias in the trial with respect to randomization method, blinding, and the balance of prognostic factors between the arms. The trial was originally designed to enroll 400 patients who had undergone conventional staging for lung cancer and were found to have stage III NSCLC. Nevertheless, it was reported that the trial was stopped after a planned interim analysis in November 2008 because of superior efficacy on the PET arm. The trial was stopped on the recommendation of the data and safety monitoring board. In the abstract, data were available for 289 patients; 15% of patients in the PET-CT arm achieved the primary outcome compared with 2.7% in the CT arm (p = 0.0002). No data were reported on other outcomes in the abstract, but final analysis is pending.
The Impact of PET on Target Volume Definitions
As expected, no studies were identified that reported on patient outcomes such as survival, recurrence, treatment-related morbidity, or quality of life. Therefore, this review will concentrate on the technical data reported by the included studies.
The key technical measures (e.g., changes in GTV or PTV) considered in this review were not reported consistently across the studies; only one study reported on all the measures considered relevant in this review as outlined in the study section criteria.36 For most measures, the data reported were also inconsistent across studies: for example, some reported the number of patients experiencing a change in GTV and not the mean change in GTV. The following results are comprehensive and have included any data reported in the studies and, if possible, summary values calculated from raw data provided in tables.
Changes in target volume definitions presented in this section represent net changes: in some cases, it is likely that PET data resulted in volume decreases due to a better differentiation of benign masses and increases due to the identification of involved mediastinal nodes missed by CT. One study reported simultaneous GTV increases and decreases in 6 of 20 patients, although this study was an exception as most reported only net changes.28
Gross Tumor Volume.
Eighteen studies including a total of 587 patients reported changes in GTV as a result of the inclusion of PET data in RT planning (Table 3).17,26–30,32,36,38,40,42–44,49,50,52,53 The magnitude of GTV changes was reported in 11 of 18 of these studies26,28,29,32,36,38,43,44,49,50,53; one other study reported the proportion of patients experiencing a change of more than 25%.17 In the nine studies26,28,29,32,36,43,49,50,53 that reported the magnitude of the increase in GTV, the mean increase per patient in patients with an increase ranged from 10.6 to 153% (median study increase 49%); the greatest increase reported for any single patient was 735%. In the 11 studies26,28,29,32,36,38,43,49,50,52,53 that reported the magnitude of the GTV decrease in affected patients, the mean decrease per patient across those studies ranged from 13.9 to 71% (median study decrease 40.5%); the greatest decrease reported for a single patient was 143%.
Planning Target Volume.
Ten studies including a total of 262 patients reported changes in PTV as a result of the inclusion of PET data in RT planning (Table 4).17,29,32,33,36,37,39,41,44,51 The magnitude of PTV changes was reported in nine of these studies.17,29,32,33,36,37,39,44,45 In the four studies29,32,35,36 that reported the magnitude of the PTV increase, the mean increase per patient across those studies ranged from 7 to 159% (median increase 27%); the greatest increase for any single patient was reported as 381%. Three other studies17,37,44 reported increases of PTV per patient of 10% or more in patients with an increase. In the six studies17,29,32,36,39,45 that reported the magnitude of PTV decrease, the mean decrease per patient was less than 29% across all studies; the greatest decrease for any single patient was 70%. Three other studies33,37,44 reported decreases of PTV per patient of 3% or more in affected patients.
Few studies provided specific details on the inclusion of PET-positive tissue that had been missed in CT-based planning; most reported aggregate data on GTV expansions (reported earlier). Nevertheless, specific instances of such geographic misses were reported in three studies. Mah et al.33 reported that, in 5 of 23 patients suitable for radical RT, FDG-avid nodes were detected within 5 cm of the primary tumor; these nodes had not been included in the CT-based GTV. Similarly, in a study by Lewandowska et al.28 PET identified CT-occult mediastinal nodal metastases in 9 of 20 cases, which were not included in the CT-based GTV. In both of these studies, the GTVs were expanded to incorporate these nodes. MacManus et al. reported that in 3 of 10 cases, regions were located entirely outside the CT PTV. These areas would not have been contained within the target volume if the treatment was delivered using the CT plan alone.44
The Impact of PET on Organs at Risk Radiation Exposure and Dose
Esophageal Exposure and Dose.
Five studies reported on changes in esophageal radiation exposure due to PET in a total of 166 patients (Table 5).27,31,32,34,38 The mean percentage of the esophageal volume exposed to a radiation dose of 50 to 55 Gy (V50eso and V55eso) per patient decreased in three studies27,31,38 and increased in two studies32,34; the range of changes in the studies was from −10.4 to 4.5%. Two of the three studies reporting a decrease in esophageal exposure indicated that the mean decrease was statistically significant with p values less than 0.005.31,38
Six studies including a total of 179 patients reported on the impact of PET on the maximal esophageal radiation dose (Table 6). The mean dose per patient received by the esophagus decreased in four studies27,31,38,51 and increased in two studies32,36; the range of changes in the studies was from −8.8 to +6.1 Gy. In one study, a decrease of 6.1 Gy was reported to be statistically significant.38
Lung Exposure and Dose.
Ten studies including a total of 267 patients reported data on changes in lung radiation exposure due to PET,27,31,32,33,34,36-39,44 and seven of these provided data on the number of patients experiencing a change or the mean value of the change (Table 5).27,29,31,32,34,36,38,39 The greatest reported increase in V20 lung for any single patient was 2000%; the greatest decrease was 100%. Four of these studies32,34,36,39 reported a change in V20 lung in between 46% and 100% of the studied patients. Three studies31,38,39 reported a statistically significant decrease in V20 lung across the studied patients. Two studies33,44 reported that changes in V20 lung were not statistically significant but did not provide data.
Four studies including a total of 154 patients reported on the impact of PET on the maximal lung radiation dose (Table 6). The mean dose received by the lungs decreased in three studies27,31,38 and increased in one study26; the range of changes in the studies was from −5.1 to +1.6 Gy. One study reported a statistically significant decrease in dose.36
Impact of PET on Total Radiation Dosage and Tumor Control Probability
Modifications of the total radiation dose to the tumor taking account of the need to limit the radiation dose to the lung, esophagus, and spinal cord were reported to show an increase in two studies by a mean of 13.7 Gy (from 55.2 ± 2.0 Gy to 68.9 ± 3.3 Gy with PET, p = 0.002) across 21 patients (including 15 stage III) and a mean of 15 Gy (from 56.0 ± 5.4 Gy to 71.0 ± 13.7 Gy, p = 0.038) across 21 stage III patients (Table 7). In both studies, patient-specific details (e.g., the number of patients for whom the dose increased) were not reported.
TCP increased in the two studies reporting on this measure (Table 7). In one study, the TCP increased 17.7% (from 6.3 ± 1.5% to 24.0 ± 5.6%, p = 0.01) among 21 patients,31 and in the other, it increased 8.6% (from 14.2 ± 5.6% to 22.8 ± 7.1%, p = 0.026) among 21 patients.38 In this latter study, the change was reported as being statistically significant (p = 0.026).
Impact of PET on Clinical Management and Patient Outcomes
Changes in of RT Intent.
Six studies reported that the inclusion of PET imaging information in RT planning resulted in the detection of distant metastases (Table 8).32,33,35,36,42,44 The proportion of patients for which distant metastases were identified ranged from 8 to 25% (median study identification rate 17.5%) across these six studies.
Eleven studies reported on whether PET information resulted in a change from curative to palliative RT intent.29,32,33,35,36,40–43,46,47 The percent of patients for whom the intent of RT was changed ranged from 8 to 41% of patients across the 11 studies. Specific reasons for the change in patient management were not consistently provided in studies, although several cited more “extensive disease” or distant metastases.
Proponents of PET believe that PET has value in the clinical management of lung cancer by producing more accurate diagnosis and staging, lower rates of futile thoracotomies, and better clinical management decisions leading to improved patient outcomes. This optimism for PET extends to its role in RT planning for lung cancer. Many clinicians believe that PET contributes to the identification of CT-occult disease, particularly mediastinal lymph nodes, and leads to the beneficial expansion of target volumes. Nevertheless, the resolution of PET is not sensitive enough to detect microscopic disease. The high sensitivity of PET has been demonstrated to appropriately exclude patients from radical therapy when distant metastases are present. There is growing consensus that PET has a greater specificity to exclude nonmalignant areas, for example, in differentiating atelectasis, and that this can appropriately reduce target volumes and radiation exposure to patients. The intention of this review was to systematically evaluate the available evidence related to these and related issues and to determine what role, if any, PET should play in RT planning for patients with lung cancer.
The PET-START trial48 is the only randomized trial reported to date that addresses PET-CT for treatment planning in NSCLC. Unfortunately, this trial has only been reported in abstract, with insufficient detail to fully assess its quality and potential for bias. Nevertheless, once this trial has been published in a peer-reviewed publication, it will likely report significant data that may address at least some of the issues described in detail later.
The review of the available literature showed that a large proportion of patients experienced changes in target and planning volumes through the use of PET imaging data (see Tables 3 and 4 and accompanying text). Two studies reported on the PET-based detection of geographic misses that resulted in increases in target volumes. Although in some cases the changes in volume are minor and would not be considered clinically relevant, on the whole, they are substantial. Increases and decreases of greater than 10% in both GTV and PTV were commonly reported across all studies. What is not clear from these studies is whether the PET-based changes in volumes were truly appropriate and led to better outcomes. Very few studies have confirmed through surgical pathology whether the changes in the RT field were appropriate, because biopsy correlation is often not possible in locally advanced unresectable NSCLC. There is some reported data on clinicopathological correlation.54–59 Nevertheless, assuming that the majority of measured change was beneficial, these values suggest that PET is contributing to both the exclusion of nonmalignant tissue and the inclusion of CT-occult tissue in RT planning.
Changes in volume size have the potential to produce corresponding changes in organ radiation exposure if tissue is included in or excluded from the PTV. The limited data available suggest that the addition of PET to RT planning is more likely to decrease the dose to the esophagus than to increase it. Two of six studies27,31,32,34,38,51 reporting esophageal exposures (V50-55eso) reported statistically significant decreases (−10.4%, p < 0.005, and −8.7%, p = 0.004, respectively).31,38 Changes in total radiation dosages to the esophagus were variable across the studies, although one study did report a statically significant (p = 0.004) decrease of 6.1 Gy.31,38
The available data regarding the effect of PET in RT planning on dose to lung tissue are mixed. Although substantial numbers of patients experience a change in V20 lung (between 42% and 100% of patients across four studies),31,36,38,39 these changes involve both increases and decreases. Nevertheless, three studies31,38,39 did report statistically significant reductions in V20 lung. The data do suggest that PET does reduce lung dose, with three of four studies31,36,38,39 reporting decreases (range of changes −5.1 to +1.5 Gy) and one of these reporting a statistically significant decrease.38
Changes in the total administered radiation dosage and TCP were reported in two studies only.31,38 In both studies, the effect of PET was to increase the total radiation dosage administered to patients (+14, 15 Gy) and to increase the TCP of RT (+18%, 9%); in both studies, the changes were statistically significant (p < 0.04, both studies). On the assumption that PET allows for more accurate administration of radiation to malignant structures, the reported net increases in radiation dosage suggest that PET may contribute to more effective RT.
There are data that suggest that the incorporation of PET into RT planning has an impact on the management of patients. In 12 studies,29,32,33,35,36,40–44,46,47 PET was reported to detect distant metastases in 8 to 25% of patients and change the intent of RT from curative to palliative in 8 to 41% of patients. The exclusion of patients with distant metastases from undergoing futile radical radiation therapy provides the clearest benefit in patients with stage III NSCLC for whom RT is indicated.60,61
There are no data available to date that show an impact of PET-based RT planning on patient outcomes such as survival or local recurrence rates. If PET is used in the determination of disease extent, it is important to confirm that areas of 18FDG uptake in mediastinal nodes or in distant, particularly isolated, sites are confirmed histologically or cytologically, so that patients are not inappropriately denied potentially curative therapy.
This systematic review highlights the limitations of the available evidence. There is rather poor consistency in reporting among the studies evaluating PET in RT planning. The measures and outcomes described in studies vary considerably, and the corresponding results are reported in inconsistent manners. This heterogeneity in analysis and presentation is confusing to clinicians seeking guidance to use this literature to inform treatment planning. In the individual studies, there is rarely independent evaluation of the CT and PET-CT imaging data to preclude bias. Investigators conducting research in this area should evaluate the measures and outcomes considered in this systematic review to facilitate comparison of future studies. Such practices will allow for the optimal use of research findings.
There are a number of issues regarding the use of PET in RT planning for which there is little or no evidence to inform clinical practice. Atelectasis is known to contribute to interobserver variability in treatment planning in NSCLC.54 Nevertheless, it is not known whether PET contributes to or diminishes the interobserver variability seen in the delineation of target volumes in this situation. The exclusion of atelectatic tissue by PET is a reason to believe that PET may reduce variability, but this supposition has not been widely evaluated in empirical studies. Two studies incorporated interobserver comparisons in their study designs: one found that PET produced greater concordance between observers in volumes17 and the other found that the effect of PET varied by observer.33
The optimal PET intensity measure for defining the tumor's edge remains unclear. Some studies report using regressive threshold functions,27 but the majority report using a fixed threshold (percentage of the maximal standardized uptake value intensity [e.g., 40-50%]) or do not report the intensity measure used at all. Some argue that regressive or lower fixed-threshold values (10-20%) are preferred.62,63 In addition, some authors advocate contouring the distinct “halo” seen around the area of maximal intensity of PET because of the clinical ease of this approach and the lesser interplanner variability it generates.17 There has been no rigorous evaluation to determine an optimal threshold to date, and there are technical aspects of PET that impede the use of the technology. The evaluation of the lungs by PET is affected by respiratory motion, which generates a degree of measurement error thereby complicating coregistration. Hybrid PET/CT technologies help to reduce these errors but do not remove them altogether. No studies in this series compared hybrid PET/CT with dedicated PET devices.
Clearly, higher quality evidence is needed to guide clinical and policy decision making regarding the use of PET in RT planning. This evidence should be generated by well-designed studies, which typically require large numbers of patients and appropriate technological resources, including high-quality PET scanners. A major study in patients with lung cancer has recently been completed in Ontario48 and will address at least some of the questions covered by this review.
Data from a number of small, nonrandomized studies suggest that the inclusion of PET imaging in the planning process produces modifications in RT planning that may be beneficial. These changes include changing the intent of treatment from curative to palliative in a substantial proportion of patients and changes in target and planning volumes. In many cases, these changes are substantial and clinically significant, although it is not certain that these changes result in better clinical outcomes. Data from these studies also suggest that PET has a small but consistent protective effect on the lungs and esophagus, and a few studies confirm a benefit for PET in terms of increasing the total dose and TCP. PET may be most useful in those cases where there is a large area of lung opacification that may be due to tumor and/or atelectasis/pneumonitis secondary to airway obstruction.
These data, taken as a whole, are highly suggestive of a benefit of PET in RT planning in lung cancer, and further evaluation of PET for this purpose is warranted. PET should continue to be used cautiously as part of research protocols, bearing in mind current uncertainties and evolving knowledge. Clinicians should be particularly mindful of situations in which PET is known to produce false-positive results (e.g., presence of inflamed lymph nodes due to pneumonitis). When performing RT planning, clinicians should take into consideration the technical specifications of the PET scanner being used as these may modify the utility of the device for RT planning purposes.
The Program in Evidence-Based Care (PEBC) is a provincial initiative of Cancer Care Ontario supported by the Ontario Ministry of Health and Long-Term Care through Cancer Care Ontario. All work produced by the PEBC is editorially independent from its funding source.
2.Tyldesley S, Boyd C, Schulze K, et al. Estimating the need for radiotherapy for lung cancer: an evidence-based, epidemiologic approach. Int J Radiat Oncol Biol Phys 2001;49:973–985.
3.International Committee on Radiation Units and Measurements (ICRU). Prescribing, Recording, and Reporting Photon Beam Therapy. Bethesda, MD: International Committee on Radiation Units, 1993.
4.Chan R, He Y, Haque A, et al. Computed tomographic-pathologic correlation of gross tumor volume and clinical target volume in non-small cell lung cancer: a pilot experience. Arch Pathol Lab Med 2001;125:1469–1472.
5.Giraud P, Antoine M, Larrouy A, et al. Evaluation of microscopic tumor extension in non-small-cell lung cancer for three-dimensional conformal radiotherapy planning. Int J Radiat Oncol Biol Phys 2000;48:1015–1024.
6.Li WL, Yu JM, Liu GH, et al. [A comparative study on radiology and pathology target volume in non-small-cell lung cancer.] Zhonghua Zhong Liu Za Zhi 2003;25:566–568;[Abstract in English].
7.Wambersie A, DeLuca P, Whitmore G. Prescribing, recording and reporting electron beam therapy; 2 volumes. J Int Comm Radiat Units Meas (ICRU) 2004;4:25–37.
8.Caldwell CB, Mah K, Ung YC, et al. Observer variation in contouring gross tumor volume in patients with poorly defined non-small-cell lung tumors on CT: the impact of 18FDG-hybrid PET fusion. Int J Radiat Oncol Biol Phys 2001;51:923–931.
9.Van de Steen J, Linthout N, de Mey J, et al. Definition of gross tumor volume in lung cancer: inter-observer variability. Radiother Oncol 2002;62:37–49.
10.Ung YC, Maziak DE, Vanderveen JA, et al. 18Fluorodeoxyglucose positron emission tomography in the diagnosis and staging of lung cancer: a systematic review. J Natl Cancer Inst 2007;99:1753–1767.
11.Birim O, Kappetein AP, Stijnen T, et al. Meta-analysis of positron emission tomographic and computed tomographic imaging in detecting mediastinal lymph node metastases in nonsmall cell lung cancer. Ann Thorac Surg 2005;79:375–382.
12.Toloza EM, Harpole L, McCrory DC. Noninvasive staging of non-small cell lung cancer: a review of the current evidence. Chest 2003;123(Suppl 1):137S–146S.
13.Gould MK, Kuschner WG, Rydzak CE, et al. Test performance of positron emission tomography and computed tomography for mediastinal staging in patients with non-small cell lung cancer: a meta-analysis. Ann Intern Med 2003;139:879–892.
14.Fischer BM, Mortensen J, Højgaard L. Positron emission tomography in the diagnosis and staging of lung cancer: a systematic, quantitative review. Lancet Oncol 2001;2:659–666.
15.Graham MV, Purdy JA, Emami B, et al. Clinical dose-volume histogram analysis for pneumonitis after 3D treatment for non-small cell lung cancer (NSCLC). Int J Radiat Oncol Biol Phys 1999;45:323–329.
16.Yaremko B, Riauka T, Robinson D, et al. Threshold modification for tumour imaging in non-small-cell lung cancer using positron emission tomography. Nucl Med Commun 2005;26:433–440.
17.Ashamalla H, Rafla S, Parikh K, et al. The contribution of integrated PET/CT to the evolving definition of treatment volumes in radiation treatment planning in lung cancer. Int J Radiat Oncol Biol Phys 2005;63:1016–1023.
18.Humm J, Rosenfeld A, Del Guerra A. From PET detectors to PET scanners. Eur J Nucl Med Mol Imaging 2003;30:1574–1597.
19.Browman GP, Levine MN, Mohide EA, et al. The practice guidelines development cycle: a conceptual tool for practice guidelines development and implementation. J Clin Oncol 1995;13:502–512.
20.Hebert ME, Lowe VJ, Hoffman JM, et al. Positron emission tomography in the pretreatment evaluation and follow-up of non-small cell lung cancer patients treated with radiotherapy: preliminary findings. Am J Clin Oncol 1996;19:416–421.
21.Kiffer JD, Berlangieri SU, Scott AM, et al. The contribution of 18F-fluoro-2-deoxy-glucose positron emission tomographic imaging to radiotherapy planning in lung cancer. Lung Cancer 1998;19:167–177.
22.Munley MT, Marks LB, Scarfone C, et al. Multimodality nuclear medicine imaging in three-dimensional radiation treatment planning for lung cancer: challenges and prospects. Lung Cancer 1999;23:105–114.
23.Nestle U, Walter K, Schmidt S, et al. 18F-deoxyglucose positron emission tomography (FDG-PET) for the planning of radiotherapy in lung cancer: high impact in patients with atelectasis. Int J Radiat Oncol Biol Phys 1999;44:593–597.
24.Bradley JD, Dehdashti F, Mintun MA, et al. Positron emission tomography in limited-stage small-cell lung cancer: a prospective study. J Clin Oncol 2004;22:3248–3254.
25.Kamel EM, Zwahlen D, Wyss MT, et al. Whole-body (18)F-FDG PET improves the management of patients with small cell lung cancer. J Nucl Med 2003;44:1911–1917.
26.Gondi V, Bradley K, Mehta M, et al. Impact of hybrid fluorodeoxyglucose positron-emission tomography/computed tomography on radiotherapy planning in esophageal and non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2007;67:187–195.
27.Grills IS, Yan D, Black QC, et al. Clinical implications of defining the gross tumor volume with combination of CT and 18FDG-positron emission tomography in non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2007;67:709–719.
28.Lewandowska A, Windorbska W, Morgas T. Radiation treatment planning using positron emission tomography for patients with non-small cell lung cancer. Nowotwory 2006;56:259–314.
29.Brianzoni E, Rossi G, Ancidei S, et al. Radiotherapy planning: PET/CT scanner performances in the definition of gross tumour volume and clinical target volume. Eur J Nucl Med Mol Imaging 2005;32:1392–1399.
30.Hong R, Halama J, Bova D, et al. Correlation of PET standard uptake value and CT window-level thresholds for target delineation in CT-based radiation treatment planning. Int J Radiat Oncol Biol Phys 2007;67:720–726.
31.De Ruysscher D, Wanders S, Minken A, et al. Effects of radiotherapy planning with a dedicated combined PET-CT-simulator of patients with non-small cell lung cancer on dose limiting normal tissues and radiation dose-escalation: a planning study. Radiother Oncol 2005;77:5–10.
32.Deniaud-Alexandre E, Touboul E, Lerouge D, et al. Impact of computed tomography and 18F-deoxyglucose coincidence detection emission tomography image fusion for optimization of conformal radiotherapy in non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2005;63:1432–1441.
33.Mah K, Caldwell CB, Ung YC, et al. The impact of (18)FDG-PET on target and critical organs in CT-based treatment planning of patients with poorly defined non-small-cell lung carcinoma: a prospective study. Int J Radiat Oncol Biol Phys 2002;52:339–350.
34.Giraud P, Grahek D, Montravers F, et al. CT and (18)F-deoxyglucose (FDG) image fusion for optimization of conformal radiotherapy of lung cancers. Int J Radiat Oncol Biol Phys 2001;49:1249–1257.
35.Roman MR, Rossleigh MA, Angelides S, et al. Staging and managing lung tumors using F-18 FDG coincidence detection. Clin Nucl Med 2001;26:383–388.
36.Bradley J, Thorstad WL, Mutic S, 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.
37.Schmücking M, Baum RP, Griesinger F, et al. Molecular whole-body cancer staging using positron emission tomography: consequences for therapeutic management and metabolic radiation treatment planning. Recent Res Cancer Res 2003;162:195–202.
38.van Der Wel A, Nijsten S, Hochstenbag M, et al. Increased therapeutic ratio by 18FDG-PET CT planning in patients with clinical CT stage N2-N3M0 non-small-cell lung cancer: a modeling study. Int J Radiat Oncol Biol Phys 2005;61:649–655.
39.Vanuytsel LJ, Vansteenkiste JF, Stroobants SG, et al. The impact of (18)F-fluoro-2-deoxy-D-glucose positron emission tomography (FDG-PET) lymph node staging on the radiation treatment volumes in patients with non-small cell lung cancer. Radiother Oncol 2000;55:317–324.
40.Kalff V, Hicks RJ, MacManus MP, et al. Clinical impact of (18)F fluorodeoxyglucose positron emission tomography in patients with non-small-cell lung cancer: a prospective study. J Clin Oncol 2001;19:111–118.
41.Roberts KB, Manus MP, Hicks RJ, et al. PET imaging for suspected residual tumour or thoracic recurrence of non-small cell lung cancer after pneumonectomy. Lung Cancer 2005;47:49–57.
42.MacManus MP, Hicks RJ, Ball DL, et al. F-18 fluorodeoxyglucose positron emission tomography staging in radical radiotherapy candidates with nonsmall cell lung carcinoma: powerful correlation with survival and high impact on treatment. Cancer 2001;92:886–895.
43.Ceresoli GL, Cattaneo GM, Castellone P, et al. Role of computed tomography and [18F] fluorodeoxyglucose positron emission tomography image fusion in conformal radiotherapy of non-small cell lung cancer: a comparison with standard techniques with and without elective nodal irradiation. Tumori 2007;93:88–96.
44.MacManus M, D'Costa I, Everitt S, et al. Comparison of CT and positron emission tomography/CT coregistered images in planning radical radiotherapy in patients with non-small-cell lung cancer. Australas Radiol 2007;51:386–393.
45.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.
46.Messa C, Ceresoli GL, Rizzo G, et al. Feasibility of [18F]FDG-PET and coregistered CT on clinical target volume definition of advanced non-small cell lung cancer. Q J Nucl Med Mol Imaging 2005;49:259–266.
47.Hicks RJ, Kalff V, MacManus MP, et al. (18)F-FDG PET provides high-impact and powerful prognostic stratification in staging newly diagnosed non-small cell lung cancer. J Nucl Med 2001;42:1596–1604.
48.Ung YC, Sun A, MacRae R, et al. Impact of positron emission tomography (PET) in stage III non-small cell lung cancer (NSCLC): a prospective randomized trial (PET START). J Clin Oncol 2009;27:[Abstract] 7548.
49.Spratt DE, Diaz R, McElmurray J, et al. Impact of FDG PET/CT on delineation of the gross tumor volume for radiation planning in non-small-cell lung cancer. Clin Nucl Med 2010;35:237–243.
50.Feng M, Kong FM, Gross M, et al. Using fluorodeoxyglucose positron emission tomography to assess tumor volume during radiotherapy for non-small-cell lung cancer and its potential impact on adaptive dose escalation and normal tissue sparing. Int J Radiat Oncol Biol Phys 2009;73:1228–1234.
51.Vinod SK, Kumar S, Holloway LC, et al. Dosimetric implications of the addition of 18 flurordeoxyglucose-positron emission tomography in CT-based radiotherapy planning for non-small cell lung cancer. J Med Imaging Radiat Oncol 2010;54:152–160.
52.Hanna GG, McAleese J, Carson KJ, et al. (18)F-FDG PET-CT simulation for non-small-cell lung cancer: effect in patients already staged by PET-CT. Int J Radiat Oncol Biol Phys 2010;77:24–30.
53.Kruser TJ, Bradeley KA, Bentzen SM, et al. The impact of hybrid PET-CT scan on overall oncologic management, with a focus on radiotherapy planning: a prospective, blinded study. Technol Cancer Res Treat 2009;8:149–158.
54.Nestle U, Walter K, Schmidt S, et al. 18F-deoyxglucose positron emission tomography (FDG-PET) for the planning of radiotherapy in lung cancer: high impact in patients with atelectasis. Int J Radiat Oncol Biol Phys 1999;44:593–597.
55.Faria SL, Menard S, Slobodan D, et al. Impact of FDG-PET on radiotherapy volume delineations in non-small cell lung cancer and correlation of imaging stage with pathological findings. Int J Radiat Oncol Biol Phys 2008;70:1035–1385.
56.Wu K, Ung YC, Hornby J, et al. PET CT thresholds for radiotherapy target definition in non-small cell lung cancer: how close are we to the pathological findings. Int J Radiat Oncol Biol Phys 2010;77:699–706.
57.Dahele M, Hwang D, Peressotti C, et al. Developing a methodology for three-dimensional correlation of PET-CT images and whole mount histopathology in non-small cell lung cancer. Curr Oncol 2008;15:62–69.
58.Stroom J, Blaauwgeers H, van Baardwijk A, et al. Feasibility of pathology correlated lung imaging for accurate target definition of lung tumors. Int J Radiat Oncol Biol Phys 2007;69:267–275.
59.van Baardwijk A, Bosmans G, Boersma L, et al. PET CT based auto-contouring in non-small cell lung cancer correlates with pathology and reduces interobserver variability in the delineation of the primary tumor and involved nodal volumes. Int J Radiat Oncol Biol Phys 2007;68:771–778.
60.MacManus MP, Wong K, Hicks RJ, et al. Early mortality after radical radiotherapy for non-small cell lung cancer: comparison of PET staged and conventionally staged cohorts treated at a large tertiary referral center. Int J Radiat Oncol Biol Phys 2002;52:351–361.
61.MacManus MP, Hicks RJ, Matthews JP, et al. High rate of detection of unsuspected distant metastases by PET in apparent stage III non-small cell lung cancer: implications for radical radiation therapy. Int J Radiat Oncol Biol Phys 2001;50:287–293.
62.Black QC, Grills IS, Kestin LL, et al. Defining a radiotherapy target with positron emission tomography. Int J Radiat Oncol Biol Phys 2004;60:1272–1282.
63.Biehl KJ, Kong FM, Dehdashti F, et al. 18F-FDG PET definition of gross tumor volume for radiotherapy of non-small cell lung cancer: is a single standardized uptake value threshold approach appropriate? J Nucl Med 2006;47:1808–1812.