| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Radiology, Mayo Clinic, 200 First St. SW, Rochester, MN
55905.
2 Department of Pulmonary and Critical Care Medicine, Mayo Clinic, Rochester, MN
55905.
Received May 11, 2004;
accepted after revision September 25, 2004.
Address correspondence to R. M. Lindell.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. PET findings were negative in 32% of the cases of non-small cell carcinomas that were detected on screening CT in a high-risk patient population. These tumors were small, low-grade, or both. The most common histology was bronchioloalveolar cell carcinoma. The role of PET in evaluating screening-detected indeterminate noldules in a high-risk population may be more limited than in a general population.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chest radiography and sputum cytology were examined as screening tools in the 1970s through large trials at Johns Hopkins, Memorial Sloan-Kettering, and Mayo Clinic [1-3]. Unfortunately, these studies did not show that screening chest radiography or sputum cytology improve disease-specific mortality of patients with this entity. More recent studies have shown that chest CT is more sensitive in detecting lung cancers than chest radiography [4, 5]. This finding has led to multiple international studies that are currently under way examining CT as a screening tool for lung cancer [6-10].
Our institution has been participating in a prospective trial sponsored by the National Cancer Institute (NCI) in which CT is being assessed as a screening tool for non-small cell lung carcinomas. Preliminary results have shown that screening chest CT may be limited by a large number of false-positives because CT detects a large number of benign noncalcified nodules [10]. For example, over 2 years, 2,832 noncalcified nodules were detected in 1,049 participants, of which only 40 were identified as lung cancer [11]. To reduce the number of patients with false-positive findings who undergo intervention, radiologists must consider patient history and nodule morphology and growth. In addition, diagnostic tests such as pulmonary nodule CT enhancement or PET may be helpful.
PET detects elevated glucose metabolism that is often present in malignancy. It can be used to evaluate an indeterminate nodule and, if cancer is present, to help stage a tumor. Studies have shown two important limitations of PET. The first limitation is a correlation between false-negative PET scans and small tumor size, even with less differentiated malignancies. Specifically, Imdahl et al. [12] found a PET failure rate of 27% in lesions that were 1 cm or smaller, 10% in lesions between 1 and 2 cm, and 12% in lesions larger than 2 cm. Likewise, Lowe et al. [13] found an 80% sensitivity in lung nodules that were 7-15 mm. Current PET technology is likely inaccurate in discriminating nodules smaller than 7 mm. The second limitation is that well-differentiated lung carcinomas, such as bronchioloalveolar cell carcinomas and carcinoid tumors, have been shown to be PET-negative in prior studies [12, 14, 15]. Marom et al. [15] looked at 183 non-small cell T1 carcinomas (3 cm) in a general adult patient population on PET and reported a 5% false-negative rate.
To our knowledge, no prior PET studies of solitary nodules have concentrated specifically on a high-risk population undergoing lung cancer screening. The purpose of this study was to retrospectively review PET results in lung cancers detected during the NCI-sponsored lung cancer screening trial in an effort to evaluate PET in the detection of lung cancer in a high-risk screening population.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To date, 62 non-small cell lung carcinomas have been found in 1,520 subjects. Of those 62 carcinomas, 20 patients had a preoperative PET scan. All PET scans were ordered at the discretion of board-certified pulmonologists with a special interest in thoracic oncology. Our study group was composed of these 20 patients who had a biopsy-proven non-small cell lung carcinoma and a preoperative PET scan. Histologic and staging data were obtained from the surgical and pathology reports. Clinical notes were reviewed.
CT was performed on an MDCT unit (Light-Speed QX/i, GE Healthcare) using 5-mm slice width, 3.75-mm reconstruction, high-speed mode, 1.5-mm pitch, 0.8 sec/rotation exposure, 30-mm table feed/rotation, 120 kVp, and 40 mA. Screening prevalence CT scans were obtained followed by incidence CT scans every 12 months for 5 years. Scans were originally interpreted at a computer workstation using the cine mode. Original reports were reviewed, and then the nodule morphology and size were examined on film.
A nodule was defined as a focal round or oval opacity that measured less than 3 cm in its greatest diameter. Nodule size was determined by the average of two diameters measured on an axial image: the diameter of the longest horizontal axis and the maximum diameter perpendicular to it. Nodules were described as solid, ground-glass, or semisolid. If a nodule obscured the underlying lung architecture, it was defined as solid. If a nodule was entirely composed of increased attenuation that did not obscure the underling lung architecture, it was described as ground-glass. If a nodule had both solid and ground-glass components, it was referred to as semisolid.
Nodules were evaluated for air bronchograms or cavitation. Margins of soft-tissue nodules were characterized as smooth, irregular, or spiculated. A nodule was defined as an incidence nodule if the nodule was not reported on the participant's first screening CT examination and as a prevalence nodule if it was reported on the first screening CT examination. If an incidence nodule was detected retrospectively on the first screening examination, this was noted separately.
Reports from PET scans were retrospectively reviewed. PET was performed using either a PET scanner (Advance, GE Healthcare) or a PET/CT scanner (DLS, GE Healthcare). Fluorine-18 FDG (18F-FDG) produced at an in-house cyclotron was administered IV with a minimum dose of 15 mCi (555 GBq) and a maximum dose of 20 mCi (740 GBq) or 0.214 mCi/kg (7.9 GBq/kg). Patients rested quietly for 60 min before scanning. Emission scans were acquired using 4-6 min per bed position usually with 6-9 bed positions per patient. The transmission scans from the PET scanner were acquired using a rotating gallium-68 source and 5 min per bed position. The transmission scans from the PET/CT scanner were acquired at 140 kVp and 80 mA with a 4-MDCT scanner and 0.5-sec rotation time. CT scans were reconstructed at 4.25-mm thickness in the axial plane to match the PET scans. Scans were originally interpreted using an Advance workstation (GE Healthcare) and a Windows NT workstation (Microsoft) running Entegra software (GE Healthcare) that were capable of multiplanar reconstructions with multiple color schemes.
Scans were considered positive when the nodule uptake was greater than mediastinal uptake. Standardized uptake values (SUVs) were noted if present in the original report. An SUV of greater than 2.5 indicated a high probability of malignancy. If a tumor had been scanned more than once with PET, the date closest to the resection date was used.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Histology
Of the 22 non-small cell lung cancers, the cell types were adenocarcinoma
(n =9); bronchioloalveolar cell carcinoma (n =4); squamous
cell carcinoma (n = 6); non-small cell carcinoma with squamous
features (n =2); and non-small cell carcinoma, not otherwise
specified (n = 1) (Table
1). Two patients had two lung cancers. One patient had two
synchronous non-small cell carcinomas with squamous features in the same lobe
(Figs. 1A,
1B, and
1C). The other patient had a
grade 2 adenocarcinoma resected and then, 1.5 years later, a grade 4 squamous
cell carcinoma resected.
|
|
|
|
Positive PET Scans
In 14 of the 22 tumors, PET findings were positive. Histology, grade, size,
and CT appearance are listed in Table
1. The average nodule size was 10 mm. Most nodules were solid,
whereas the three grade 4 squamous cell carcinomas were semisolid. The one
semisolid nodule had a tiny internal radiolucency, but it was difficult to
tell if it was an air bronchogram or cavitation. No nodules were of
ground-glass attenuation. Margins were either irregular or spiculated, none
smooth.
The SUV of one of the PET-positive tumors was reported. A grade 4 squamous cell carcinoma that was 10 mm in diameter had an SUV of 3.1, indicating a high probability of malignancy.
Negative PET Scans
Of the 22 non-small cell carcinomas, PET findings for seven (32%) were
negative (Figs. 2A, and
2B). Tumor histology, grade,
size, and CT appearance are outlined in
Table 1. The average nodule
size was 10.5 mm. These tumors were ground-glass, semisolid, or solid. Most
tumor margins were irregular or spiculated; only one tumor was found to have
smooth margins. The two adenocarcinomas and the one squamous cell carcinoma
showed mild uptake on PET, but not enough to meet the criteria for positive
PET findings (Figs. 3A and
3B).
|
|
|
|
Of interest is that despite the fact that PET findings were negative, five of the nodules (7-mm grade 1 bronchioloalveolar cell carcinoma, 14-mm grade unknown bronchioloalveolar cell carcinoma, 11.5-mm grade 2 bronchioloalveolar cell carcinoma, 8.5-mm grade 2 adenocarcinoma, and 6.5-mm squamous cell carcinoma) were surgically removed without further CT follow-up because they were enlarging and morphologically suspicious on CT. The 11.5- and 14-mm bronchioloalveolar cell carcinomas had negative findings on one prior PET scan. The histology of the 11.5- and 7-mm bronchioloalveolar cell carcinomas was accurately suspected by the pulmonologist and thoracic surgeon based on the negative PET scan and ground-glass attenuation. The other two nodules with negative PET findings (8-mm bronchioloalveolar cell carcinoma and 11-mm adenocarcinoma) were followed up with CT and then removed because of continuing enlargement.
Indeterminate PET Scan
In one tumor, PET findings were indeterminate. As shown in
Table 1, this tumor was a 10-mm
grade 4 adenocarcinoma that was solid and spiculated. The PET report said that
most of the SUV measurements were less than 2.0 but that a few were slightly
greater than 2.5. This nodule was then followed up with CT and removed for
continuing enlargement and suspect morphology.
Prevalence Versus Incidence Tumors
As shown in Table 1, most of
the 14 tumors with positive PET findings were incidence tumors (n =
9). One of the incidence tumors was present on the first screening CT in
retrospect. Contrarily, most of the seven PET-negative scans were present on
the first screening examination (n = 6). Two of these were seen only
in retrospect. Interestingly, both of these were ground-glass nodules. The one
PET-negative tumor that developed since the first screening examination was
the squamous cell carcinoma. The indeterminate PET nodule was an incidence
tumor.
Staging
Pathology staging was available for 21 of the tumors, and the stages were
as follows: IA (n = 15), IB (n = 1), IIA (n = 1),
IIIA (n = 1), and IIIB (n = 3). The two synchronous
non-small cell carcinomas with squamous features in the same lobe were stage
IIIB. CT staging was inaccurate in three of the 21 tumors. PET inaccurately
staged two of the 14 PET-positive cases. In the stage IIIA case, a right
paratracheal lymph node was missed on CT but was accurately detected on PET.
Both CT and PET missed a peribronchial intrapulmonary lymph node metastasis in
the stage IIA case. As would be expected, CT was unable to detect microscopic
visceral pleural involvement in the stage IB case, which was the case with
indeterminate findings on PET.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3 cm) in a nonscreening patient population.
The reason for this higher rate of false-negative PET scans in our study
appears to be twofold. First, the cancers were smaller in our study. The mean
tumor size in our study was 10 mm, whereas it was 20 mm in the study by Marom
et al. Second, there was a higher percentage of low-grade tumors in our
screening population. Specifically, the percentage of bronchioloalveolar cell
carcinomas compared with all non-small cell carcinomas in our study was 18%
compared with 6% in their study. Although larger scale studies are needed, it is not surprising that a higher percentage of small tumors and low-grade tumors was detected on CT in a lung cancer screening population than in a nonscreening patient population because these tumors are more often difficult to detect radiographically and are subclinical. It follows that the limitations of PET in detecting small tumors and low-grade tumors are amplified in a screening population. Specifically, PET appears to be less helpful in a lung cancer screening population because false-negative results appear with greater frequency.
On the other hand, the fact that the PET findings were negative may provide useful information about the aggressiveness of a tumor. The level of 18F-FDG uptake has been shown to correlate with the tumor doubling time [16]. Furthermore, a higher 18F-FDG uptake has been shown to correlate with shorter patient survival [15, 17-19]. Marom et al. [15] observed that all of the PET-negative lung cancers in their study were stage IA at the time of resection. Likewise, our seven PET-negative cases were all stage IA. Therefore, PET-negative lung cancers may be less aggressive and correlate with longer survival. A negative PET scan may in the future be of use to the clinician when deciding risk versus benefit in the resection of an indeterminate or suspicious nodule. However, the numbers in our study are too small to make any definitive statement about these possibilities.
Steinert et al. [20] showed that PET is superior to CT in nodal staging of non-small cell cancer. This possible benefit appears to be less important in a CT screening population because tumors seem to have a higher PET-negative rate and tumors are detected at earlier stages. In our study, PET accurately staged 12 (86%) of the 14 PET-positive tumors and CT likewise accurately staged 18 (86%) of the 21 tumors.
In keeping with previously published results [12, 14, 15], all bronchioloalveolar cell carcinomas in our study were PET-negative. It is interesting to examine the other PET-negative tumors and the PET-indeterminate tumor in comparison with the PET-positive tumors. The tumors were of similar average size: PET-positive tumors, 10 mm; PET-indeterminate, 10 mm; and PET-negative, 10.5 mm. More specifically, the PET-negative 6.5-mm grade 3 squamous cell carcinoma was 0.5 mm smaller than a similar appearing PET-positive nodule. Because PET is inaccurate for detecting nodules smaller than 7 mm, this result is not surprising. However, the 10-mm grade 4 adenocarcinoma with indeterminate findings on PET was similar in size and was higher in grade than the adenocarcinomas with positive PET findings, so the indeterminate PET result for that nodule is somewhat surprising. The PET-negative adenocarcinomas were similar in grade and size to the PET-positive adenocarcinomas but, unlike them, were of a semisolid consistency. In fact, 11 (79%) of the 14 PET-positive tumors were solid, whereas only two (29%) of the seven PET-negative tumors were solid.
The presence and extent of ground-glass attenuation within a nodule may be useful when evaluating an enlarging nodule detected on screening CT. Because bronchioloalveolar cell carcinoma is often PET-negative [12, 14, 15] and has a high rate of ground-glass attenuation morphology [21, 22], it may be reasonable to discourage obtaining PET scans of ground-glass-attenuation nodules to decrease the rate of false-negative PET scans. In our study, two of the three bronchioloalveolar cell carcinomas appeared as ground-glass-attenuation nodules on CT. No other tumors appeared as ground-glass-attenuation nodules on CT. Observing such ground-glass-attenuation nodules for interval growth rather than evaluating them on PET scans may be a better strategy.
A larger percentage of PET-negative tumors were present on the first screening examination. Six (86%) of seven PET-negative tumors were present on the first examination versus six (43%) of 14 of the PET-positive tumors. This could be due to the fact that the PET-negative tumors tended to be a lower grade and, therefore, have a longer subclinical course.
Our study is limited by the fact that PET scans were ordered at the discretion of pulmonologists rather than as part of a randomized prospective trial. There may be some bias from the information used by the physician in selecting whether a nodule was evaluated on PET. In addition, the numbers in our study are too low to make statistically valid conclusions. Our study was also limited by the technology that was available at the time the examinations were performed. Improved PET resolution and PET/CT fusion may improve PET of screening CT-detected lung nodules.
PET may be useful in the evaluation of screening CT-detected pulmonary nodules, especially when positive. However, in our high-risk population, PET was negative in 32% of the non-small cell lung carcinomas detected on screening CT. The limitations of PET in detecting small tumors and low-grade tumors were amplified in this population because screening seems to inherently detect a higher percentage of these tumors. The impact of detecting these small, low-grade tumors on morbidity and mortality is not yet known, but the poor sensitivity of PET in detecting these tumors raises questions about its usefulness in the preoperative evaluation of nodules detected on CT screening. At this time, if a CT screening-detected nodule is PET-negative, continued follow-up of the nodule is warranted to assess for interval growth.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. McWilliams and J. Mayo Computed Tomography-detected Noncalcified Pulmonary Nodules: A Review of Evidence for Significance and Management Proceedings of the ATS, December 15, 2008; 5(9): 900 - 904. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. K. Gould, J. Fletcher, M. D. Iannettoni, W. R. Lynch, D. E. Midthun, D. P. Naidich, and D. E. Ost Evaluation of Patients With Pulmonary Nodules: When Is It Lung Cancer?: ACCP Evidence-Based Clinical Practice Guidelines (2nd Edition) Chest, September 1, 2007; 132(3_suppl): 108S - 130S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. B. Markowitz, A. Miller, J. Miller, A. Manowitz, S. Kieding, L. Sider, and A. Morabia Ability of Low-Dose Helical CT To Distinguish Between Benign and Malignant Noncalcified Lung Nodules Chest, April 1, 2007; 131(4): 1028 - 1034. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Schoder and M. Gonen Screening for Cancer with PET and PET/CT: Potential and Limitations J. Nucl. Med., January 1, 2007; 48(1_suppl): 4S - 18S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Lobrano Partnerships in Oncology and Radiology: The Role of Radiology in the Detection, Staging, and Follow-up of Lung Cancer Oncologist, July 1, 2006; 11(7): 774 - 779. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
